Studies in Surface Science and Catalysis 65 CATALYSIS AND ADSORPTION BY ZEOLITES
This Page Intentionally Left Blank
Studies in Surface Science and Catalysis Advisory Editors: 6. Delmon and J.T. Yates Vol. .65
CATALYSIS AND ADSORPTION BY ZEOLITES Proceedings of ZEOCAT 90, Leipzig, August 20-23, 1990
Editors
G. Ohlmann Zenrralinsrirur fur Physikalische Chemie, Rudower Chaussee 5, Berlin Adlershof, 0- 1 7 99 FRG
H. Pfeifer Universirar Leipzig, Sekrion Physik, Linnhstrasse 5, Leipzig, 0-70 10 FRG and
I?.Fricke Zenrralinstirur fur Physikalische Chemie, Rudo wer Chaussee 5, Berlin Adlershof, 0- 1 199 FRG
ELSEVIER
Amsterdam - Oxford - New York - Tokyo
1991
ELSEVIER SCIENCE PUBLISHERS B.V. Sara Burgerhartstraat 25 P.O. Box 2 1 1, 1000 AE Amsterdam, The Netherlands Distributors for the United States and Canada:
ELSEVIER SCIENCE PUBLISHING COMPANY INC 655, Avenue of the Americas New York, NY 10010, U S A .
ISBN 0-444-89088-2
0Elsevier Science Publishers B.V., 199 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the publisher, Elsevier Science Publishers B.V./ Academic Publishing Division, P.O. Box 330, 1000 AH Amsterdam, The Netherlands. Special regulationsfor readers in the USA - This publication has been registered with the Copyright Clearance Center Inc. (CCC), Salem, Massachusetts. Information can be obtained from the CCC about conditions under which photocopies of parts of this publication may be made in the USA. All other copyright questions,-including photocopying outside of the USA, should be referred to the publisher. No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any rnethods, products, instructions or ideas contained in the material herein. This book is printed on acid-free paper. Printed in The Netherlands
V CONTENTS Preface.........................................................XI International Scientific Committee, National Organizing Committee, Sponsors
...........................................
XI11
INVITED LECTURES Catalysis on ZSM-5 zeolites modified by phosphorus G. Ohlmann, H.-G. Jerschkewitz, G. Lischke, R. Eckelt, B. Parlitz, E. Schreier, B. Zibrowius, E.Lijffler...........l New directions in zeolite catalysis J. Weitkamp
...............................................
21
Zeolites as catalysts for alkane oxidations R.F. Parton, D.R.C. Huybrechts, Ph. Buskens, P . A . Jacobs..47 Sorption and separation of binary mixtures of CH4, N, and COz in zeolites L.V.C. Rees...............................................61 Use of ZSM zeolites in the liquid phase separation of alcohols R. Schollner, W.-D. Einicke.......................... .....75 Basic principles and recent results of 'H magic-angle-spinning and pulsed field gradient nuclear magnetic resonance studies on zeolites H. Pfeifer, D. Freude, J. Karger
..........................
89
Spectral study of lewis acidity of zeolites and of its role in catalysis V.B. Kazansky........ ~ 1 7
....................................
Comparative measurements on acidity of zeolites H.G. Karge.
..............................................
133
Acidity and basicity in zeolites D. Barthomeuf.............................................l57 Matrix vs zeolite contributions to the acidity of fluid cracking catalysts R. v. Ballmoos, C.-M.T. Hayward
..........................
171
VI ZEOSORB HS-30 - a template-free synthesized pentasil-type zeolite K.-H. Bergk, W. Schwieger, H. Furtig, U. Hadicke..
....... 185
Selective conversion of syngas to hydrocarbons by zeolites H. Tominaga, K. Fujimoto, T. Tatsumi........... ..........203 '"Xe NMR of adsorbed xenon for the determination of void spaces Q. Chen, M.A. Springuel-Huet, J. Fraissard... 219
............
Diffusion of hydrocarbons in A and X zeolites and silicalite D.M. Ruthven, M. Eic, Z. Xu..............................233
- Present - Future ...........................................
Atlas of zeolite structure types: Past W.M. Meier....
247
Zeolites as membranes: The role of the gas-crystal interface R.M. Barrer..............................................257
The roles of metal and organic cations in zeolite synthesis D.E.W. Vaughan..
.........................................
275
Temperature dependence of nucleation of zeolites in alkaline aluminosilicate gels in hydrothermal crystallization conditions S.P. Zhdanov, N.N. Feoktistova, L.M. Vtjurina 287
............
SUBMITTED PAPERS Catalvsis Synthesis of piperazine and triethylenediamine using ZSM-5-type zeolite catalysts J. Weitkamp, S . Ernst, H.-J. Buysch, D. Lindner..........297 Diffusion effects on the kinetics of toluene methylation and xylene isomerization on HZSM-5 zeolites F. Bauer, J. Dermietzel, W. Jockisch......
............... 305
Iron-containing ZSM-5 type zeolites used in the coupled methanol-hydrocarbon cracking (CMHC) A. Martin, S. Nowak, B. Lucke, W. Wieker, B. Fahlke......315 Dispersion dependent selectivities of syngas conversion on faujasite encapsulated Pt, Pd or Ir N.I. Jaeger, G. Schulz-Ekloff, A . Svensson...............327
VII
Contribution of 13C N M R spectroscopy to the analysis of surface compounds formed in the transformation of acetone on zeolites V. BosaEek, L. Kubelkova, J. Novakova....................337 Isopropylation of benzene over large pore zeolites A.R. Pradhan, B.S. Rao, V.P. Shiralkar
...................347
EXAFS study of local structure of Pt-Cr clusters in pentasils in relation with their reactivity in lower alkanes aromatization E.S. Shpiro, R.W. Joyner, G.J. Tuleuova, A.V. Preobrazhensky, O.P. Tkachenko, T.V. Vasina, O.V. Bragin, Kh.M. Minachev 357
...........................................
Splitting of methane into the elements over nickel containing ZSM-5 catalysts J. Hoffmann, R. Bauermeister, B. Hunger, K. Hantel, G. Ullmann, K.-H. Steinberg, H. Siege1
...................367
Sulfur tolerant Ni-Mo-Y-zeolite catalysts for water-gas shift reaction M. Zianiecki, W. Zmierczak...........................
.....377
The influence of cations on the alkylation of toluene with ethylene over modified ZSM-5 zeolites J. Eejka, B. Wichterlova, G.L. Raurell...................
387
Multinuclear MAS NMR studies on coked zeolites H-ZSM-5 H. Ernst, D. Freude, M. Hunger, H. Pfeifer...............397 Coke oxidation in HZSM-5 zeolites. Intermediates, final products and reformation of OH groups and void volumes J. Novakova, L. Kubelkova................................405 Influence of the conditions of dealumination and the partial extraction of non-framework aluminium on the properties of ZSM-5 catalysts G . Ohlmann, H.-G. Jerschkewitz, G . Lischke, R. T. Gross, B. Parlitz, I. Schulz, K. Wehner, D.
effect of catalytic Eckelt, Timm... ...415
Spectroscopic and catalytic investigations of hydrothermally dealuminated ZSM-5 E. Loffler, L.M. Kustov, V.L. Zholobenko, Ch. Peuker, U. Lohse, V.B. Kazansky, G. Ohlmann......................425
VIII
Sorption and diffusion Computer modelling of p-xylene sorption in ZSM-S/silicalite-l K.-P. Schroder
...........................................
435
Microdynamics of guest molecules in zeolites studied by quasi-elastic neutron scattering and NMR pulsed field gradient technique H. Jobic, M. Bee, J. Caro, M. Bulow, J. Karger, 445 H. Pfeifer
...............................................
Permeability studies on a silicalite single crystal membrane model E.R. Geus, A.E. Jansen, J.C. Jansen, J. Schoonman, H. van Bekkum.......
.....................................
457
The adsorptive and the catalytic diffusion of 2,3-dimethylbutane in large crystals of (aluminated) silicalite ......... 467 P. Voogd, H. van B e k k u m . . . . . . . . . . . . . . . . . . . . . . . . . Transport phenomena and reactions in 13X type zeolites M. Biilow, W. Hilgert, G. Emig
............................
479
Svnthesis and structure The effect of various physical and chemical parameters on the synthesis of ZSM-5 for propene oligomerization C.T. O'Connor, S. Schwartz, M. Kojima
....................
491
On controlled growth of SAPO-5 molecular sieve crystals of different sizes and shapes G. Finger, J. Kornatowski, J. Richter-Mendau, K. Jancke, M. Bulow, M. Rozwadowski......................501 Approximate assignment of vibrational frequencies of the NaX framework E. Geidel, H. Bohlig, Ch. Peuker, W. Pilz................511 The topological structure representation of zeolites B. Muller................................................521 Studies of secondary synthesis on modified pentasil zeolites W. Reschetilowski, W.-D. Einicke, B. Meier, E. Brunner, H. Ernst...........................
......................
529
IX Incorporation of silicon into the framework of SAPO-5 studied by N M R and IR spectroscopy B. Zibrowius, E. Loffler, G. Finger, E. Sonntag, M. Hunger, J. Kornatowski
................................
537
On the synthesis and structure of A1P04-14 B. Zibrowius, U. Lohse, K. Szulzewsky, H. FichtnerSchmittler, W. Pritzkow, J. Richter-Mendau.......
........549
Y-zeolite treated with Sicl, vapour. Structure and properties G.M. Telbiz, A.I. Prilipko, I.V. Mishin..................563 Characterization and nucleation of Na,TPA-ZSM-5 zeolite with different aluminium content G. Golemme, A. Nastro, J.B. Nagy, B. Subotic, F. Crea, R. Aiello................................................573 New molecular sieve - vanadium silicalite ms-5 J. Kornatowski, M. Sychev, V. Goncharuk, W.H. Baur.......581 Multinuclear NMR study of the crystallization of SAPO-37 N. Dumont, T. Ito, J.B. Nagy, Z. Gabelica, E.G. Derouane......................
......................
591
Metastability of zeolites in tetraethylammonium media F. di Renzo, A. Albizane, M.-A. Nicolle, F. Fajula, F. Figueras, Th. des Courieres...........................603 Synthesis of ferrierites with high gallium content M.A. Camblor, J.A. Martens, P.J. Grobet, P . A . Jacobs.....613 Synthesis of artificial zeolite-like mountainite M.K. Babayev, D.M. Ganbarov, D.B. Tagiyev
................ 623
The distribution of iron in ZSM-5 type iron containing zeolites and ferrisilicates: acidic and catalytic properties G. Vorbeck, M. Richter, R. Fricke, B. Parlitz, E. Schreier, K. Szulzewsky, B. Zibrowius.................631 Zeolite ZSM-57: Synthesis, characterization and shape selective properties S . Ernst, J. Weitkamp 645
....................................
X
Aciditv New data on the structure and properties of acidic sites in HZSM-5 zeolites: IR-spectroscopic studies and non-empirical quantum chemical calculations I.N. Senchenya, V.Yu. Borovkov...
........................
653
FTIR in-situ investigation of zeolite activation R. Salzer, B. Ehrhardt, J. Dressler, K.-H. Steinberg, P. Klaeboe.......
663
........................................
IR spectra of CO adsorbed at low temperature (77 K) on titaniumsilicalite, H.-ZSM5 and silicalite A. Zecchina, G. Spoto, S . Bordiga, M. Padovan, G. Leofanti, G. P e t r i n i . . . . . . . . . . . . . . . . . . . . . . . . . . ........671 The properties of boralites of various boron contents J. Datka, A. Cichocki, Z. Piwowarska.................
....681
An electrostatic model to predict the infra red characteristics of zeolite hydroxyl groups after adsorption of aromatics P.J. O'Malley 6ag
............................................
The influence of the acidity of zeolites on the formation of unsaturated carbenium ions I. Kiricsi, H. Forster, G . Tasi, P. Fejes
................ 697
Author Index................. Subject Index.............
.................................. .....................................
707 711
Studies in Surface Science and Catalysis (Other volumes in the series) 715
........................................................
XI
PREFACE There is a permanently growing interest, worldwide, in zeolite science and technology. Especially in Europe it has become an established tradition to hold specialized zeolite meetings which occur sandwiched in between the large International Zeolites Conferences. The ZEOCAT 90 Conference held in Leipzig, is the latest in a series of successful European zeolite conferences which have been held previously in Wurzburg, Nieuwpoort, Siofok, Prague and Bremen. The range of synthesized zeolites and zeolite-related molecular sieve materials is constantly expanding. The variety of structures comprises high- and low-silica zeolites, micro-porous metallosilicates, alumino-phosphates and crystalline silica polymorphs. These newly developed, modified or improved zeolites and other molecular sieves have opened up new horizons by their application as shape-selective catalysts and adsorbents, as cation exchangers or as high-tech materials. Current topics of interest are the newly developed catalytic applications of zeolites in the synthesis of fine chemicals, the use of zeolites as a new material for sensors or solid electrolytes and the setting up of sophisticated zeolite-based separation processes including molecular sieve membranes. In order to master the promising potential offered by the new zeolites, the links between the various fields of molecular sieve research are growing closer. This is especially true for merging synthesis and characterization by such powerful methods as solid state NMR and Xray or neutron diffraction, as well as for adsorption, diffusion and catalytic studies. A new trend is the growing interest in the use of computational and theoretical approaches to explain synthesis, sorption, diffusion and catalytic data. ZEOCAT 90 was comprised of the many different and diverse aspects of the rapidly expanding field of zeolite science and technology. The invited plenary lectures of this conference summarize our current knowledge and address topical areas of zeolite research. New interesting findings produced by fundamental research have led to new industrial applications and these developments have, in their turn, inspired new directions for fundamental research. The deliberate control of crystallization processes has led to the production of many zeolites which do not occur naturally. Chemically
XI1
and physically controlled treatments have been used to tailor shapeselective zeolite catalysts and adsorbents for industrial use. Newly developed physical and chemical spectroscopic and diffraction methods have given us new insights into the exciting world of molecular interactions of solid surfaces. The editors are convinced that ZEOCAT 9 0 contributed, like its European predecessors, to the successful interdisciplinary exploration of zeolites and to finding new pathways for their application in science and technology.
G. dhlmann, H. Pfeifer, R. Fricke (Editors)
XI11 ZEOCAT 90 - Leipzig International Scientific Committee R.M. Barrer, London D. Barthomeuf, Paris H.K. Beyer , Budapest H. Bremer, Berlin P. Fejes, Szeged J.P. Fraissard, Paris J. Haber, Krakow P.A. Jacobs, Heverlee H.G. Karge, Berlin V.B. Kazansky, Moscow
W.M. Meier, Zurich W.J. Mortier, Heverlee C. Naccache, Villeurbanne L.V.C. Rees, London W. Schirmer, Berlin G. Schulz-Ekloff, Bremen J. Weitkamp, Stuttgart B. Wichterlova, Prague S.P. Zhdanov, Leningrad
National Orsanizina Committee Chairmen: G. Ohlmann, H. Pfeifer Secretary: H. Miessner, B. Staudte H. Bremer, Berlin M. Bulow, Berlin D. Freude, Leipzig R. Fricke, Berlin J. Karger, Leipzig H. Lieske, Berlin SI)onsors AKZO Chemicals B.V. BASF AG BAYER AG Chemie AG Wolfen-Bitterfeld Degussa AG Deutsche Shell AG EXXON Chemical Holland-BCT Hoechst AG Intern. Zeolite Ass. (IZA) Leuna-Werke AG PCK AG Schwedt PQ Corp.
W. Schirmer, Berlin R. Schollner, Leipzig H. Stach, Berlin K.-H. Steinberg, Leipzig J. Volter, Berlin
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V.. Amsterdam
1
G OHLMANN, H.-b.JtHSCHKEWlIL, b.LlSCHKt, H.ECKtLT, b.PARLIIL, t S C H R k l E R . B / I K R O W i I I S . 6 InFFLER
CentraJ I n s t i t u t e o t P h v s i c a l Chemistry o f t h e Academy o f Sciences R:idower u m i s s e p 5 . 11'JY M R L IN-ADLtRSHOF AbS1 R A L ! M o d i r v i n s (J teinplcite-tree s y n t h e s i z e d and h y d r o t h e r m a l l y dealuminuted LSIY-5 t v o e z e o l i t e w i t h o r t h o p h o s D h o r i c a c i d v i e l d s c a t a l y s t s t o r t h e h i g h - t e m p e r a t u r e MIO-reaction which a r e s u p e r i o r t o t h e n i e i ' e l y aealuniiiiot t'cl z e o l i t e s i n r e s p e c t o f Cz-Cd-olef i n s e l e c t i v i . t i e c nncl d e a c t i v a t i o n t i m e s . F o r m a t i o n o f methane, a r o m a t i c s ond coke i s : ; t i - o n y I v suppressed. I h e mechanism o f t h e P-modificnr t o n 3 5 char;ic,teriz;ld by on n t l e n s t partially r e v e r s i b l e c h e m i c a l i n t e i - i l i . t i u i 1 o i p h l i s E t j o i ' l c a c i d w i t h t h e sti-onq a c i d s i t e s and by n c t ~ i i : y e i.11 ; . o n c e n t r u t i n n and n a t u r e o t weak a c i d s i t e s which a l s o contrtbute f o c a t a l y s i s . A f t e r P-modification, dealumination of t h e z e o l i t e tramework by steoiaing i s i n h i b i t e d .
';inre t h e work n t Kciedinq and R u t t e r ( 1 ) who f i r s t hnvp shown t h e p o s i t i v e e t f r c t o f t h e PhosDhOrus m o d i t i c a t i o n o f ZSM-5 c a t a l v r t s on t h e i r r r r : a l v t i c p r o p e r t i e s i n t h e MIO-process and soniewhat l a t e r on t i l e P o c o - s e l e c t i v e a l k y i a t i o n o t t o l u e n e w i t h methono1 ( L ) , numeruus s t u d i e s have been aevoted t o t h e problem o t t h e niec;huiiisin o t ttiis i a o d i r i c a t i o n ( 3 - 11). KUt?dillg uiicl B u t t e 1 pruposed i n t h e i r Paper bonding o t PhosPhorLls t o the z e o l i t e framework t h r o u g h oxygen what f o r m a l l y i s equival e n t t o t h e reoiacpinent o f t h e h r i d g e d h y d r o x y l groups hv a HZP04QI'OUD bv means o t I P D mensuremmts o f NHs t h e y tound a s i g n i f i r a n t l v i n r r e a s f t l number o f a c i d s i t e s . which a r e weaker i n s t r e n u t h compared t i l t h those i n t h e unmodi t i e d z e o l i t e . Vedrine e t a 1 ( 4 ) s t z t e d t h a t Phosphorus n e u t r a l i z e s a c i d s l t e s
p r i m a r i l v u t t n e e n t r a n t - e a t t h e LSM-5 channels whereas t h e s t r o n aest n c i a s i t e s o r e n o t m o d i i i e d . Consequentlv t o r e x p l a n a t i o n o t t h e observed i n c r e n s e n f C2-C4-.nlefin selectivities. o slight aocl; : i r . a t i n n i n chnnnel r;ize und i n c r e a s e d t u r t u o s i t v due t o phos~ h o i ~ i i : ;comooirnd i s p r e f e r e d , compared t o changes i n a c i d s t r e n g t h . U e r e w i n s k i und Haber (6) found by NHZ-lPD t h a t m o d i f i c a t i o n w i t h r I. inie t n y 1piio 5 uii i t e ond d iilmmon i umhyd r agenp hos p ha t e e 1i m i n a t es t h e Hrrjn:;ted acirl s i . t e s and a l s o s i ~ n I f I c a n t decreases l~ the content u r W C ~ C I K C ~ Iricic'l . S I tes B~isecl on P~IISC~I~~I~Y experiments w i t h q u i n o l i n e and t r i m e t h y l p h o s p l i i t e Nunoil et o i ( L , j r;ont:liirierl, t h a t t h e most i m p o r t a n t e f f e c t o f u n o s ~ h ~ ~I:;r ~ ~r h; t : ooir,oning n t t h e s t r n n g ocirf site:, o n t h e e x t e r r i o l :;urtnr.f r i t t h e i ' . r , v s t a l s . K:IIl. Maiiit~c~arid Yii;hinio ( i i J ) s t u d i e d t h e shupe s e l e c t i v e a l k y l u t i o n n t ethvlhenzene W I tll e t h a n o l or1 P - m o d i f i e d ZSM-5 c a t a l v s t s and claim t h n t t h e S L I P P ~ ~ S S I O o~ i Para-di.ethvlbenzene L s o m e r i z a t i o n i s achieved by I'eduLing t h e amount o t s t r o n s a c i d s i t e s r a t h e r t n a n tiy r e ~ ~ u c i nthe a e t t e c t i v e pnre diameter o f LSM-5. S h i u e r u I k ( i 1 et r i l . ( 1 2 ) orepnred a c a t a l y s t f o r methanol conver(;oIir;i?,t i n u ot H'-LSM-5 oel l e t i z e d and m o d i t i e d w i t h sion onrl HPU4&- A c c o r d i n g t o t h e i r c o n c l u s i o n t h e b i n d e r reduces t h e a c i d s i t e s an t h e e x t e r n a l s i i r t u c e CJt t h e serves 0s a source o f g r a d u a l s u p p l y o f HPUqA-, which ocirl s i t e s arid r e t a r d s d e a l u m i n a t i o n . Most comprehent h e mechanism o f P - m o d i f i c a t i o n has been problein o f the wnrks n t l e r c h e r e t n l ( 7 . 8 . 1 4 , 1 5 ) . I n t h e i r w r I :FIw,i; 1, ( 8 ) tl;ev hnd develooed a concept o f c o n t r o l l e d decrease . * ;;i- -i,.; .?,i:-~r,ath hv c n n v e r s i n n o f s t r o n g a c i d s i t e s i n t o weakely i i ( i d oiler,. u i t l l o i i i alteration n t t h e OVel-all acid-base P r o p e r t i e s . 111 t t i , L: i;r:est p u b l i c a t i o n s they come t o t h e c o n c l u s i o n , t h u t t i e u ( i n e : i ~ o i t i l e zeolite w i t h H3PU4 and t r i i n e t h u l p h a s p h i t e decrea~ e ! tire , CcjiiLenti'ution O T s t r o t i s a c i u s i t e s e s s e i i t i u i l y by dealuniir x i T i n i i , I n t h e case n t t r i m e t h y l p h o s p h i n e as t h e m o d i f y i n g agent thev rniiiid ( 1 5 ) ('I hl.ocking o f s t r o n q a c i d s i t e s up t o u r e a c t i o n teinc;t?rotiir~C i t 0 O O n C . Ahc,ve t h i s t e m p e r a t u r e t h e phosphine i s de?,rjrtji?<j. i r ? q c i i e r o I 1 . i ~t h e o r i g i n a l a c i d i c p r o p e r t j . e s o f H I - L S M - 5 . Lumniar-iziriy t h i s o v e r v i e w i t may be s t a t e d , t h a t decreuse o t ;troriy i l i i i l i t y atiJ i ' t ' a u c t i o n o f p o r e d i a m e t e r a r e i n o s t l y d i s c u s s e d i n ui'tJei tc, e x p l a i n t h e e t i e c t o t phosphorus m o d i f i c a t i o n . the I e i c l i ~ v t ' 1 l l l I J l J I ~ t ~ i l oL r~ these r o c t o r s is a i t r e r e n t l y e v a l u a t e d and rfww-ids G I I t h e s t t i d i e d r e a c t i o n . C o n t r a d i c t o r y a r e t h e views o f v n r i o i i s nilthrbrs on t h e wov. i.n which s t r o n g a c i d s i t e s a r e decrenL i .
3
sed, by i n t e r a c t i o n w i t h t h e b r i d g e d h y d r o x y l groups, o r by u a r t i a l deaiumination. i n t h i s paper we d e s c r i b e t h e combined e f f e c t o f h y d r o t h e r m a l d e a l u m i n a t i o n and phosphorus m o d i f i c a t i o n on t h e s e l e c t i v i t i e s and d e a c t i v a t i o n rimes i n t h e high-temperature MTO-reaction. F o r a l l experiments t h e ZSM-5 type. template-free synthesized z e o l i t e HS-30 o f Chemie AG B i t t e r f e l d , has been used. TPD o f ammonia, I R and :*1P--.27AI-MAS-NMR soectroscopy were employed t o g e t deeper i n s i g h t i n t o t h e mechanism o f t h e z e o l i t e m o d i f i c a t i o n by o r t h o ptiosplioric a c i a . l h e work was p a r t o f a c o o p e r a t i o n w i t h t h e Researcn and Uevelopment D i v i s i o n o f Leuna-Werke-AG.
f r e P a r o r i o n v t soimles P e n t a s i l t y p e LSM-5 z e o l i t e (HS-30, Si/Al=19, Chemie AG B i t t e r t e l d ) i s c o n v e r t e d I n t o t h e H*-form by a f o u r t i m e s r e p e a t e d t r e a t m e n t w i t h a 0 . 2 N so1.ution o f HN03 i n a s i m u l e b a t c h Proced i i r e . For d e u l u s i n a t i o n . samples o f t h e H+-form a r e exposed t o a stream o t a i r and steam (pwater.=90 kPa) a t 400, 410 and 700°C. Framework aluminiirm ( A h ) i s determined u s i n g t h e ammonium exchonge method. For p a r t i a l e x t r a c t i o n o f non-framework aluminium t A 1 1 . i ~ ) torined d u r i n g deUlUminatiOn, samples a r e t r e a t e d w i t h 2 N HN03 a t 105°C f o r t w o h o u r s . E x t r a c t e d aluminium was d e t e r m i n e d o n a l y t i c a l l v i n the s o l u t i o n o f t h e e x t r a c t i n g agent. C a t a l y s t p a r t i c l e s were oreoared by m i x i n q t h e HS-30 samples w i t h Aerosil-ZGO and water and subsequent e x t r u d i n g , t h e ready made e x t r u d a t e s (d = 1 t o 2 mm) c o n t a i n i n g 6 5 % by w e i g h t o t z e o l i t e . P-modir i c a t i o n was c a r r i e d o u t by i m p r e g n a t i o n w i t h aqueous p h o s p h o r i c a c i d . i h e c o n t e n t o f phosphorus f o r c a t a l y s t s based upon e x t r a c t e d HS-30 was h e l a somewhat lower t a k i n g i n t o c o n s i d e r a t i o n t h e i r di.minished c o n t e n t s o f t o t a l . alilminium. F i n a l l y c a t a l y s t samples were c o n d i t i o n e d b y steaming a t 700°C f o r 1 h o u r .
Ammonium exchange z e o l i t e s a r e exchanged by NH4*-ions by l h e H I - i o n s o t LSM-5 treatment w i t h an aqueous s o l u t i o n o f an ammonium s a l t . F o r t h i s Purpose t h e sample i s p l a c e d on a 64 s l a s t r i t (approx. 4s) and exposed t o CI tiow o f 50u m l o t an aqueous 0,2 N NH4(CHaCOO) s o l u t i o n t o r b Iluurs a t /c) t o 80°C. A t t e r c a r e f u l washing, t h e c o n t e n t
4
o f ammonium i s d e t e r m i n e d by t h e K j e l d a h l method. D e t e r m i n a t i o n o f ahasphorus i n o w ous s o l u t i o n s Ihe suiiif d e v i c e a s i n t h e ammonium exchanse i s used t o a n a l y z e phosphorus removed from P-modified samples by e l u t i o n w i t h h o t w a t e r . Q u a n t i t a t i v e d e t e r m i n a t i o n o f t h e e x t r a c t e d Phosphorus i s based on t h e p h o t o m e t r i c molvbdo-vanada-phosphoric a c i d method. Temper a t w r o g r a m m e d des o r p t i o n o f am0n i a (NH 3-TPD) A t t e r - c l e a n i n g z e o l i t e samples by h e a t i n g t o 500°C i n an He 9ns stream ( v = 3 m l / s ) f o r one hoirr, ammonio i s odsarbed a t 120°C from a He gas stream c o n t a i n i n g 3 Vo1.-I o f N H 3 . A f t e r f l t r s h i n g by p u r e He a t 120°C f o r two ho:irs, d e s o r p t i o n o f ammonia up t o 500°C 1 s s t a r t e d crlow f a t e o f h e l i u m : 1 ml/s; heat r a t e l P C / m i n ) . I h e c o n c e n t r a t i o n o t NH3 i n t h e e x i t gas i s d e t e r m i n e d u s i n g a thernioc o n d u c r i v i t y c e l l . I h e c u r v e o f t h e desorptogram is r e c o r d e d DY a c o n v e n t i o n o l d e v i c e . A d d i t i o n a l l y ammonia i s absorbed from t h e e x i t acts stream SY on 0 . 1 N HaS04 s o l u t i o n . I t s t o t a l amount i s determined by back t j t r a t i o n o f excess s u l f u r i c o c i c l .
T R Soectroscooy LSM-5 z r o l i t e samples a r e compacted t o t h i n s e l f s u p p o r t i n g and p l a c e d i n a q u a r t z 1 R c e l l . The waTe1-s tiouohlu 7 mg/cmz) worers o r e c u r c i n e a a t 400°C i n vacuo. A f t e r c o o l i n g t o 200°C, p y r i o i n e vapour i s uailiitted i n t o t h e system f o r 30 m i n u t e s . A f t e r w u r ~ t s , t l i e c e l l 1 s degusbecl and evacuated t o e l i m i n u t e p h v s i s o r b e d P v r i d l nf Transmission s p e c t r a a r e r e c o r d e d i n t h e range from 4000 t o 3300 cm-' b o t h h e t o r e and a f t e r p y r i d i n e n d s o r p t i n n . u s i n g o SDecord M 85 spectrometer o f C a r l Z e i s s Jena w i t h a 4 c m - l r e s o l u t ion. Diffuse r e t l e c t a n c e s p e c t r a a r e r e c o r d e d w i t h t h e t - T - 1 H t e c h n i q u e t h e r e s o l u t i o n o f which i s 4 cm-I. The u s i n g u s p e c i a l aevice, z e o l i t e Powdei i s p r e t r e a t e d f o r t o u r h o u r s a t 450°C and a t 10- ' Po
L 7 A l- and P-MAS-NMK--measurement s I h e NMK s p e c t r a were o b t a i n e d on a BRUCKER MSL4OO m u l t i n u c l e a r spectronieter o p e r a t i n g a t u f i e l d o t 9 , 4 r w i t h a s t a n d a r d BRUCKER d o u b l e - b e a r i n g MAS p r o b e , I h e z i r k o n i u o x i d e r o t o r s were spun near
5
5 kHz w i r h a r y n i t r o g e n as d r i v i n g gas. About 200 my o f sample m a t e r i a l a r e f i l l e d i n t h e i n n e r r o t o r volume o f about 0 . 3 5 cm:’. T y p i c a l l y , 1500 t o 1800 f r e e i n d u c t i o n decays ( F I D ’ s ) were accumulated p e r sample f o r b o t h n u c l e i . The p u l s e w i d t h s were 3 . 9 11s (n/U-pulse) w i t h r e p e t i t i o n times between 5 t o 30 s and 0.6 11s (n/l2-puise) w i t h a r e p e t i t i o n time o f 1 s f o r a t 362 MH7 and f o r 2 7 A l a t 104 MHz, r e s p e c t i v e l y . The c h e m i c a l s h i f t s a r e g i v e n on t h e d s c a l e w i t h a maximum u n c e r t a i n t y o f fl Ppm. I n o r d e r t o o b t a i n r e l i a b l e q u a n t i t a t i v e r e s u l t s by t h e L’7Al-MAS;NMR measurements, t h e samples were h y a r a t e d i n an e x s i c c a t o r i n an atmosphere s a t u r a t e d wi.th water f o r more t h a n 20 h o u r s . Ihct ahs o l u t e i n t e n s j t v mode (AI-mode o f t h e DISMSL-software package) was used t o r e x p e r i m e n t a l d a t a a n a l y s i s . La t a l v s i s C a t a l y t i c experiments were c a r r i e d o u t i n a t u b u l a r f i x e d bed r e a c t o r (6 = 12 mm) u s i n g a m i x t u r e o f methanol/Nz ( m o l e c u l a r r a t i o = 1 : l ) as r e a c t i o n feea ( c o n d i t i o n s o f r e a c t i o n see t a b l e 1 ) . Products were analyzed DY an o n - l i n e o p e r a t e d guscnromotogrophic device C a l c u l a t i o n o f t h e c o n c e n t r n t i n n from a n a l v t i c a l d a t a i s based on i n d i v i d u a l c a l i b r a t i o n o f t h e components
lABLE I : Parameters o f r e a c t i o n ~ _ _ t enipe r a t u r e :
c o n c e n t r a t i o n o t methanol: c a t a l y s t volume: c a t a l y s t mass ( a v e r a g e ) : teed t-are ( i n i x r u r e ) : charge GHSV ( m i x t u r e ) . LHSV(methano1):
~
560°C 50 Val-% 12 m l 5.b 9 4 . 0 m l / s (Sir)
1200 m l / m l c a t 1 m 1/ml a t
I vs t I
I.
*h *h
I h e t i m e on stream when methanol f i r s t appears i n the e x i t gas o f t h e r e a c t o r i s clenGted as t 1 m e o t d e a c t i v a t i o n C a t a l y s t s a r e named a c c o r d i n g t o t h e g e n e r a l formula D X P v . x d e n o t i n g the b i / A l ~ r a t i o and Y t h e c o n t e n t o t pnosphorus us weight p e r c e n t .
6
RESUL LS .MRDILSCUSSIQN R e c e n t l y we have shown ( 2 0 ) , t h a t d e a l u r n i n a t i o n o f z e o l i t e HS-30, used as c a t a l y s t i n t h e M'10-reaction a l t e r s t h e s e l e c t i v i t i e s o f l i g h t o l e f i n s and o f a r o m a t i c s i n a s i m i l a r way as was found by Chans e t a 1 . ( 1 9 ) f o r ZSM-5 c a t a l y s t s s y n t h e s i z e d w i t h v a r i o u s S ~ / A I F r a t i o s . f o r t h e r e a c t i o n c o n d i t i o n s chosen i n t h i s worK t n e e t t e c t o t w r y i n g c a t a l y s t S i / A l F r o t i o on t h e s e l e c t i v i t i e s o t cZ-c4- o i e T i n s , L 6 - b - U r o l n a t i c s and methone as w e l l a s on t h e d e a c t i v a t i o n t i m e i s demonstrated i n F i g . 1 . S e l e c t i v i t i e s o f l i g h t o l e t i n s i n c r e a s e UD t o almost 70% wh,ile t h e f o r m a t i o n o f a r o m a t i c s ond methane i s d i s t i n c t l v decreased UP t o c a t a l ~ s t S i / A l F o r 2 6 8 . lhese changes a r e accompanied by ii inarked ~ r o l o i i ~ a t i o n o i t h e d e a c t i v a t i o n t i m e . As d e s c r i b e d i n a s e p a r a t e paper ( 7 0 ) a remarkable i n c r e a s e o f t h e d e a c t i v a t i o n t i m e may be n t t a i ilea by e x t r a c t i o n o t t h e non-tramework aluminium u t t e r s u i t i c i e n t IY l o n g d e a l u m i n a t i o n a t mediuin t e m p e r a t u r e s . F u r t h e r dealuminat i n n , however, does n o t r e s u l t i n a b e t t e r c a t a l v s t . O l e f i n s e l e c t i v i t i e s t e n d t o decrease a g a i n , t h e some h o l d s t r i i e o f t h e dent:-t i v a t i . o n t i m e . W h i l e t h e s e l e c t i v i t i e s a t a r o m a t i c s a r e n o t sn
I0
6C
50
m \
40 Y
r(
c
Y r(
10 m 9 #
! 20 L 10
k
,so '61
'9 6
268
102
1
-,immntics
0
1.25
-time o r dcactiv.
F i g . l . E t t e c t o f d e a l u m i n a t i o n on p r o d u c t s e 1 e c : i v r t i e s and O e a c t i v a t i o n t i m e compured w i t h u d d i t i o n a l phosphorus inod1 t i c a t i o n
7
mucn affected, a significant growtn o f methane - in particular on the catalyst with S i / A l F 480 - is observed. lhus maxima o f olefin selectivity and a t deactivation time, attainable by dealumination. seem to exist. Concerning the deactivation time this behaviour is to be expected since the concentration of strong acid sites hecomes very sma 1. lhe behaviour of the oiefins, however, is less plausible. lhe Turtner decrease of strong acid sites should result in a further loss ot catal.ytic activity and consequently favour the accumulat on of the lower olefins as the primory products in the reaction sequence. Under the severe reaction conditions the light olefins - ethene in particular - are partly formed a l s o by crackina of higher olefins. A strong suppression of these cracking reactions by the decrease of strong acid sites as compared with tne diminished rate of the overall reaction may therefore contribute to the changes in olefin selectivity. Indeed. the portion of ethene in the c 2 - c 4 olefin fraction decreases from 30% for D 2 6 8 to around 18% for 0 4 8 0 . B e s i w s , one has to taKe into consideration, thflt Si/AlF ratios above 270 can be obtained with reasonable times o f hydrothermal treatment only at temperatures o f 700 - 750 " C . Nevertheless no loss of cristallinity could be detected by X-ray diffraction tor the sainples treated at this severe conditions. However, they undergo a phose transformation from orthorhombic to monoclinic. As wns reported ( 2 1 ) this phase transition is indeed aependent on sl/ulF ond temperature. It i s not clear, however, wtietner this i s ot importance for the observed reversed olefin selectivities or not. For further improving the selectivity of light olefins. suppressing the tormation o t aromatics and increasing the deactivation time, the modirication with phosphorus, f i r s t used by Sutter and Kaeding ( 1 ) with non-dealuminated LSM-5, was chosen. Ihe results achieved arter systematical study are examplitied by the catalyst b 2 6 8 f 1 . 2 5 in k i g . 1 . i n order to rina tile conditions Tor preparing catalysts with optimum activity, maximum light oletin selectivity and deactivation time. the influence o f qrowing amounts o f phosphorus at different degrees of dealumination on the product distributlon of methanol coilversion had to be studied. As shown in Fig.2 UP t o 2% phosphorus the selectivities O T L 2 - i 4 oieTins grow on a level which i s the higher the greater the Si/AiF ratio, except for very hign S i I A 1 ( 1 ) 3 0 0 ) . Aocive 2% i.' the octivities 01 the catalysts drop, dimetnvlether appears right from the beginning o f the reaction and the selectivities o f the CZ-C4 olefins ore drastically decreased.
8
ring
only
hydrocarbons
as reoc-
sz
t i o n p r o d u c t s . Quite t h e o p p o s i t e b e h o v i o u r show t h e s e l e c t i v i t i e s o f t h e a r o m a t i c s . They f a l l w i t h g r o w i n g c o n t e n t o t phosphorus, but agoin the catalysts with d i T r e r e n t S i / A l F r a t i o s have d i f ferent levels o f s e l e c t i v i t i e s , t h e lowest t o r the c o t o l ~ s t s w i t h the highest degree o f dea l u m i n a t i o n . Here t h e samples w i t h S i / A l i - = 3 3 0 e x h i b i t no exc:ept i o n a l behoviour. [he suwr'es-
S %
10
Bo
7 0 .
I
/.
67*57
/,P$y M,
*/.,'
/
\
yp! \ ,./
! I
ea 1'
,/'
!' .--__-.J I,';
m-
50.
-. ..
. I
\.
'i-330
I
I
\
40.
I
'
.
9 8 .
1 -
6 5 4 -
J 2
.
1
-
.'
I 0:5
i.0
1.5
2:O
2.5
IP
k i g . 5 , t t i e c t o t Phosphorus s e l e c t i v i t y O T aroinotics l o r various S i / A l F r a t i o s
0.5
1.0
1,5
2.0
2.5
3.0
S P
k l Q . 4 . tlrei1 O T P1105PtlOfU5 on s e l e c t i v i t y 0 1 inetnune for varioiis S i / A l t . rfltins
9
p u r t i c u l n r i n the r e g i o n between 1 . 5 and 2 . 0 % P . i t i s g e n e r a l l y acceated now. t h a t t h e higher o l e t i n s a r e formed mainl v hv n l k v l n t i o n o f t h e l o w e r ones w i t h m e t h a n o l o r d i n i e t h ~ l e t h e r and t o wme degree a l s o hv t h e i r o l i g o m c r i z a t i on. tthene i s the primary p r o d u c t o f t i r s t C-Cbond toi-tnation, b u t un-
sx 4
,,,-
I
*,
i4,
‘... ..__,
\
1 ’
”-
./’
-__a
x Ethene
rO‘
;.-*-.
I’
20.
Si/A1=107
l
.”
P
c) Propene
,/a
/d
Butene
0 C,-aliphatics
yi\x II \ ,! lo-
---*--‘ *
10
ably also for crackinn. while catalyst ability for the alkvlation o f oiefins by methanol is evidently less affected. lhis conclusion i s supported by an experimerit with cofeeding of Propene to the methanol stream durjng the MTO reaction. A higly dealuminated cotalyst sample with very small concentration of strong acid Bransted s i t e s , additionallv modified with 0.5% phosphorus was chosen for this purpose. The most striking effect of cofeeded propene. as f o l l o w s troin table 2, i s the distinct increase of methanol conver-
WHSVIh- I ) Methanol WHLV(h-') Prnoene Methanol conversion ( X I
1,55
1,55
0.00
0,46 98,7
90.3
Products
+8.4
(% V o l . )
Methane
1,65
0.81
-0,84
E.f ha tie
0,05 0.31 0.35
0,05 0,49
0.14
0,zo
0,oo +0,18 +o. 19 + O . 06
0,81
0,96
+0,15
Prooane I -Butane N- Hu tflne La-Aliphatics senzene I oluene Xylenes Cs-Aromatics
0.54
0,oo
0,04
0,lO
0,16
+0,04 +0,06
0,3Y
0,38 0,06
-0,Ol -0,015
0,075
sion, which most likely may be explained as the result of its consumption in the alkylation of added propene. Indeed the concentrations of all Ca+-products. except for Cs+-aromatics. are
11
increusecl. I n e t a c t , t n a t t h i s i s v a l i d a l s o f o r t h e c o n c e n t r a t i o n o f ethene i s i n d i c a t i v e o f t h e a t l e a s t p a r t l y m a i n t a i n e d c r a c k i n g a c t i v i t y o t the catalyst. To g e t deeper i n s i g h t i n t o t h e mechanism o f m o d i f i c a t i o n by phosphorus t h e i n t l u e n c e of growing amounts o f phosphorus on t h e con'*O c e n t r a t i o n o f weaK and 0-3 strong a c i d s i t e s was s t u d i e d by NHS-TPD. I h e 7mC.6 IPD t r a c e s o f ammonia de- ,+ s o r p t i o n show characteOB: r i s t i c a l l y o f ZSM-5 ( 2 2 ) two maxima, one f o r weak 'D! z acid sites of ditferent n a t u r e a t about 240°C and ' 0 '1#5 ' 0 '1.5 Po '1.0 '0.5 '2.5 '0.5 p ~ , 2 ~ '0.5 '2.5 a h j g h temperature peak -ti2*-&"-3,07the o t 420°C r e f l e c t i n strong Bronsted acid F i g . 6 . E f t e c t o t P on weak and s t r o n g sites (bridged hydro(hatched p a r t ) a c i d i t y f o r x v l groups zSi.-p--Al=). various Si/AlF of the z e o l i t e il i h e i r c o n c e n t r a t i o n c o u l d De determined q u a n t i t a t i v e l y f o r SUtTlc i e n t l v , separated peaks by t h e i r peak a r e a s . Otherwise, t n e amount o t s t r o n g a c i d si.tcts was determined s e p n r a t e l y from s p e c i a l exaeriments a f t e r NH3 a d s o r p t i o n a t 2500C. The r e s u l t s a r e demons t r a t e d i n F i g . 6 . The f i r s t column o f each group ( P O ) r e p r e s e n t s the data f o r the unmoditied c a t a l y s t s . Obviously both type o f s i t e s decrease w i t n d e a l u m i n a t i o n , a t l e a s t . above S j / A 1 ~ = 2 8 . A f t e r m o d i f i c a t i o n w i t h g r o w i n g amounts o f phosphorus, t h e concent r a t i o n o f s t r o n g Bronstea a c i d s i t e s i s d r a s t i c a l l y d i m i n i s h e d by a f a c t o r o f 10 t o 5 depending on S i / A l F r a t i o t h e most pronounced e f f e c t b e i n g ochi.eved w i t h 0 . 5 % phosphorus. a l r e a d y . F u r t h e r i n crease o f P-content r e s c i l t s i n p r a c t i c o l l v no change o f t h e i r conc e n t r a t i o n , which, i f any. l i e s w i t h i n t h e margins o f e x p e r i m e n t a l e r r o r ( * 0 . 0 0 5 mmol/g N H 3 ) . T h i s i s i n p a r t i c u l a r v a l i d f o r t h e c a t a l y s t s based on 068 and 0107. I-he c o n c e n t r a t i o n o f weak a c i d s i t e s i s reduced d i f f e r e n t l y , dePendinQ on t h e P-content and on the s l / U l F r a t i o 0s w e l l . t a c h o r t h e samples oased on D2e and D ~ H seem t o have a maximum c o n c e n t r a t i o n a t 1 . 5 % P w h i l e f o r t h e cat h e l o w e s t c o n c e n t r a t i o n o f weak a c i d s i t e s tfllvsts DlOiP0.s-2.5 i s o b t a i n e d w i t h 0 . 5 % P . I n t h i s case f u r t h e r i n c r e a s e o f P-conA
8
0"
12
t e n t i s accompanied bv a c o n t i n u o u s i n c r e n s e o f t h e c o n c e n t r a t i o n o f weak a c i d s i t e s . r e a c h i n g w i t h 2 . 5 % P an even h j q h e r v a l u e t h a n w i t h o u t phosphorus. l h i s complex dependency is a c l e a r i n d i c a t i o n o t a change i n t h e n a t u r e o f weak a c i d s i t e s w i t h P - m o d i f i c a t i o n . some o f them h e i n s removed, o t h e r s newlv c r e a t e d . I h e f a c t t h a t P-content above 1 - 1 . 5 % t o r a l l S ~ / A I Fr a t i o s p r a c t i c a l l y t h e same c o n c e n t r u c i o n o f s t r o n g a c i d s i t e s is observeo l e a d s t o t h e c o n c l u s i o n , t h a t t h e d i s t i n c t d i f i e r e n c e s i n Product s e l e c t i v i t i e s . o b t a i n e d i n t h j s ranqe o f P - c o n t e n t s . s h o u l d he due, a t l e a s t p a r t l y . t n t h e d i f f e r e n c e s i n n n t l r r e and c a n c e n t r a t i o n o f t h e w e a k a c i d s i t e s . rhus. weak a c i d s ' i t e s n t t h i s high-temperature c o n d i t i o n s a r e l j k e l v t o p a r t i c i p o t e i n c a t a l v sis. O f course, an a d d i t i o n a l shape s e l e c t i v i t y e f f e c t caused bv t n e growing amount o t P-compounds d e p o s i t e d i n s i d e t h e channels cannot iie r u i e u o u t . l h e q u e s t i o n i s now. how t h e decrease o f s t r o n g nnd weok a c i d s i t e s caused by phosphorus m o d i f i c a t i o n can he e x o l a i n e d As n l ready mentioned. t h i s problem i s s t i l l d i s D u t e d . h u t t h e n i i t h n r s OT most r e c e n t p u b l i c a t i o n s i n t e r o r e t t h i s e t f e c t e s s e n t i n l l v hv d e a l u m i n a t i o n o t t h e z e o l i t e . I n case or a c n e m i c a l i n t e r a c t i o n o f p h o s p h o r i c n c i d w i t h t h e b r i d g e d h v d r o x v l 4roiIPs as t h e a l t e r n a t i v e t o d e a l u m i n a t i o n , one would expect a c e r t n i n degree o r t n e columns r e p r e s e n t a g a i n t h e sulil of reversibility. i n Fig./. wenk and strong acid s i t e s o f a non-dealuminnt e d z e o l i t e sample. ThF! m o d i f i c a t i o n o f t h i s onrl sample w i t h 2 . 5 % P treatment for 1 hour i n a He- stream n t 500 O L causes a d i s t i n c t decrease o t s t r o n q nc;.tl sites. A T t e r e x t r a c r i o n o i phosp n o r i c a c i d w i t t i water u t 80 C'C t o r b h o u r s , nowever, t n e former concent r a t i o n o f strong uciu Fig.7.Changes o f a c i a s i t e s c o n c e n t r a sites i s fully restorea. t i o n of a P-modified s a m p l e ( D ~ ~ P ~ . o ConseqitPnt1.v ) the rlisa f t e r various treatments (hatched onpearence o f s t r o n a a c i d p a r t o t columns: c o n c e n t r a t i o n o f s i t e s has t o be unders t o o d as t h e r e s u l t o f strong acid s i t e s ) .
13
reversible
chemicnl i n t e r a c t i o n o f p h o s p h o r i c a c i d w i t h t h e b r i g -
df*d h y d r o x v l q r o i t p s and cannot be due t o d e a l u m i n a t i o n o f t h e zeo-
. I i , t e . I f t h e P - m o d i f i e d sample p r i o r t o e x t r a c t i o n i s steamed f o r 4 hours o t 40u "i r e s p . t o r 0 , 5 hours a t 700 "C a f u r t h e r decrease o r s t r o n g and weak a c i d s i t e s occurs (see F i g . 7 ) . The same e x t r a c t i o n pi-i.rr.edure, employed t o t h i s samples, a g a i n r e s t o r e s t h e aci.d s i t e s b u t the o r i g i n o l c o n c e n t r a t i o n o f t h e s t r o n g a c i d s i t e s i s nnt n t t n i n e r l l r r e v e r s i b l e changes. o b v i o u s l . . ~d e a l u m i n a t i o n , have token olnce. I t tins t o b e noted, however, t h a t a f t e r steaminq i n c o n t r a s t t o t h e i n e r e l y thermal o r e t r e a t m e n t i n He a t 500 " C . phosp h o r i c u c i a c o u l d be e x t r a c t e d o n l y by 89% and 60% r e s p e c t i v e l y . l h e c ~ u e s t i o n o r i s e s us t o whether the Presence o t t h e p h o s p h o r i c nLi.d 1.n the z e o l i i e channels i a v o u r s t h e process o f d e a l u m i n a t i o n o r n o t . i h e ciiogront i n t-19.8. shows t h e c o n c e n t r a t i o n o t weak and s t r o n g a c i d s i t e s o f t w o HS-30 samples one un1.1 m o d i f i e d and t h e other 1.0 0.9 modified w i t h 2,5% P o f 0.8 t e r steaming a t eqiinl. c 0.7 condi t i o n s and subsealtent 'b 0.6 e x t r n c t i o n o f ohosohorus .+ 0.5 from t h e m o d i f i e d sample. gE 0.4 The concentrations o f v 0.3 $0.2 strong acid s i t e s d i f f e r z 0.1 by a f a c t o r o f t w o i n t a vour o f t h e P-inoditied i i ~ . 8 . A c i u i 1cnuiiges ~ O T unmoditied sample. Consequently t h e U i i i l i'-InOC)iTieCI Hb-50 ( Z , 5 X P ) o t t e r modification o f strong steominu und e x t r a c t j o n o t P . a c i d B r o n s t e d s i t e s by phosphoric a c i d excerts a p r o t e c t i n n e t t e c t on t h e A1F-atoms r a t h e r t h e n f a v o u r i n s t h e j r t r o m t h e Traniework. I h o t e x t r a c t i o n o f phosphorus l e a d s exvu1:ion t o i u t u l y s t s w i t h p r u p e r t i e s approaching a g a i n t h o s e o f u n m o d i f i e d i ; u t a l y s t s i s cleinonstratecl i n F i g . 9 . I h i s h o l d s t r u e a l t h o u g h , due t u 11io1-2sever-e steuiiiiriy o t t h i s sample, o n l y about 16% o i pnosphor u s c o u l d be removed. A complete r e s t o r a t i o n o f t h e o r i g i n a l p r a a e r t i e r , o t IJnmOditied somples t h e r e f o r e c o u l d n o t be expected. Neverthelexr, t h e t o c t o t r e o p p r o a c h i n g t h e former p r o p e r t i e s i n d i c o t e ! : t h a t t h e v e r v molecules o t p h o s p h o r i c a c i d which hove i n t e r n c t e d w i t h t h e s t r o n a o c i d si.tes can e a s i l y be removed by e x t r a c ~ 1 0 1 1 . i h e gi.eiitt:r p o r t o f phosphoric a c i d m o l e c u l e s must e x i s t i n s i d e t h e channels i n a c l i t f e r e n t s t a t e , o b v i o u s l y c h e m i c a i l y
-
14
70
30 20 10
k1g.Y. Product s e l P c r i v i t i e s a f t e r P - e x t r a c t i o n compared w i t h [inm o d i t i e d nnd m o d i t i e d samDles bonded w i t h t h e non-framework a l i i m i n i u m k u r t n e r s u p p o r t and a d d i t i o n a l i n f o r m a t i o n on t h e mechanism o f Pm o d i t i c . a t l o n wete o b t a i n e d by means o f 1R- ond MAS-NMR-spectroscoPV I h e T i r s f two s p e c t r a i n F i g 10 a r r a n g e d one upon a n o t h e r , i l l u n t r n t e t h e e f f e c t o t P - m o d i f i c a t i o n ( 2 5% P, w i t h o u t steamiiiq)
3610 I
k i s . i U . I K - l r a n s l n i s s i o n s p e c t r a o f HS-30 samples o t t e r d i f f e r e n t o r e t r e a t m e n t s . Kegion o t O H - s t r e t c h i n g f r e q u e n c i e s .
15
nn t h e i n t e n s i t v ,if t h e s t r o n a hand o t 3610cm-i which is g e n e r a t e d bv t h e b r i d g e d OH-arnilos i n t h e 7 e o l i . t e . 1-he d r a s t i c decrease o f t h i s hnnd i s riuoarent. The next two s p e c t r a demonstrate how t h e i n t e n s i t i e s and t h e n a t u r e o f OH-groups a r e changed by steaming and SIJbSeClUent e x t r a c t i o n o f phosphorus. besides t h e d i s t i n c t l y aecreased band a t 3610 cm-I, t h e appearance n f n new hand nt 3 6 6 5 c m - I , which hos been assigned t o A1-OH groups o f non-framewnrk aluminium (23). i s r e g i s t e r e d . ConsequentI v . steomina f o r 5 hours a t 400°C causes a o a r t i a l d e a l u m i n a t i o n . Ihe l a s t two s p e c t r a show. what happens w i t h t h e u n m o d i f i e d samole a f t e r t h e sume steaming p r e t r e a t m e n t compared w i t h t h e P - m o d i f i e d one ( t h e t o p s p e c t r u m ) . l h e decrease o f t h e s t r o n g B r o n s t e d s i t e s is s z g n i r i c a n t i y g r e ( l l e r , d e a l u m i n a t i o n proceeded t o a l a r g e r degree. I h u s . t h e p r o t e c t i n g e f f e c t o f phosphorus i s c o n f i r m e d a l s o bV t h e IR-dfltfl. lhe i n t e r a c t i o n a t ohosohoric a c i d w i t h non-framework aluminium i s shown i n F 1 9 . 1 1 . M o d i r y i n g a deaJuminated sample w i t h P p r a c t i c n i l v r e s u l t s i n the disauuearance o f t h e band a t 3 6 6 5 cm-I , which reappears, however. a r t e r severe steaming a t 7OO0C, a p p a r e n t l y as u consequence o t f u r t h e r d e a l u m i n a t i o n and d e f i c i e n c y o f phosphor u s t o r c o i w l e r e oonding t h e newly c r e a t e d non-framework alum].-
3740
P at 120°C 37iC
/ :::5 -b-------L-
3900
I k i 8
at 700°C +steamed lh
Il.trrect
OT
P on 1 R -
hnnu i n t e n s i t i e s a t 5 b 6 3 and 5610 cm-'
3800
3700
3600
--
3500 ?/Em
3600
-'
Fig.lZ.IR-dittuse reflectance spectra o f dealuminated HS-30 b e f o r e ( f u l l l i n e ) and a f t e r P - m o d i f i c a t i o n
16
niiim. l h e i n t e n s e i n t e r a c t i o n w i t h t h e s t r o n g a c i d s i t e s and t h e OH-QroUPS o r non-TrameworK a l u m i n i i l m can be concluded a l s o from the d i f t u s e r e t t e c t a n c e IR-spectra presented i n F i g . 1 2 . A dealuh l z , r u l l l i n e spectrum) shows t h e t y p i c a l bands minuted sample Tor o i l o b s e r v e a b l e OH-groups as a s s i g n e d b e f o r e . A f t e r m o d i f i c o t i o n w i t h 1 . 5 % P t h e c h a r a c t e r i s t i c bands o f t h e s t r o n g a c i d s i t?-, (361.0 cm-l) rind t h e A ~ N F - O H groitos ( 3 3 6 5 cm-') a r e s i q n i f i c n n t i v reduced. l n t e r e s t i n s l v t h i s spectrum r e v e a l s a s h o u l d e r o f rJ twitid a t 3670 cm-' - a frequency r e g i o n , where P-OH-groups show absorption. i h i s muy be c o n s i d e r e d as an i n d i c a t i o n o f t h e i r p r e sence. Wh;le i t i s c l e a r trom t h e s e d a t a , t h a t p h o s p h o r i c a c i d r e a c t s w i t h t h e b r i d a e u h v d r o x v l groups and t h e non-tramework a l u m i n i u n i CIS w e l l . t h e i n t o r m a t i o n about t h e n n t u r e o f t h e p r o d u c t o f t h i s i n t e r a c t i o i i remains r a t h e r D o o r . l h e r e f o r e t h e P - m o d i f i e d sarnolea o d d i t i o n o l l v :;tiidie!d hv :I1P- and "'Al-MAS-NMR spectroscopv. I h e :"P-MAS-NMR spectrum o f a d e a l u m i n a t e d P-modified sample (spectram A i.n k i q . 1 3 ) e x h i b i t s a t l e a s t t h r e e d i f f e r e n t resonance i i n e s . one o t them ( a t =1 ppm) corresponds t o t h e o r i g i n a l o r t h o ohosphorlr: o c i d . t h e second a t - 1 5 ppm r e p r e s e n t s c h e m i c a l s h i f t s u s u a l l y o b t a i n e d t o r s h o r t c h a i n polyphasphates (241, and t h e r e inoinina l i . n e n t nboitt -;."-I opm corresoonds t o t h e range c h a r a c t e r i ~ ~ -, :;tic o f ~;IiiminiUr;i ahos- 3 r p - NMR phiites ( 2 5 ) . Residues u l A o r t h o u n o s p h o r i c o c i u and D95p1,6 Y.:traotion (80.C) dried a t 120% snort c h a i n Doiyuhosphate s ~ e ~ i e fart? , removea by e x t r u c t i o r i w i t t i wuter spzctt'uin ils L ii n s k Shown i g . 1 5 by 111
at l h 100% eteame:
A-A ax'raction
(80.C)
which t t i e col-I-esPunuinY - ," s i g t i a l s a r e absent . u n i y t n e i i n e cnurocteristic ot alumiilluni0 0 -0 -10 0 3 -0 -rG pnosphates, t h e pr'ouuct 71 0: i n t e r a c r i o n o r phosk i q 13.="P-MAS-NMR s p e c t r a o t P - m o d i t l e d . u h o r i c a c i d w i t h nondealurninated samples o t t e r v a r l o t i r f I ninework oluminium pretrmtments r emoins ITt h e sample i s steamed t o r 1 hour a t 100 " C (spectrum b) t h e s i ~ n a i s a t 1 cliiu - 1 5 ppin d i s a p p e u r and new l i n e s a t -3Y and -4b pplli appear U b v l o u s l y t h e p h o s p h o r i c anu POlyPhOSPhOt~lC a c i d
17 species undergo f u r t h e r condensation r e a c t i o n s , s i n c e t h e new lines may be i n t e r p r e t e d as caused by h i g h l y condensed p o l y ~ h o s h a t e species ( 2 6 ) . A chemical s h i f t a t -46 ppm is known f o r t h e branc h i n g groups i n P 4 h O ( 2 4 ) . These s p e c i e s can be removed by h o t water e x t r a c t i o n as is r e f l e c t e d by spectrum D, i n which t h e c o r r e s p o n d i n g l i n e s almost c o m p l e t e l y v a n i s h . Unchanged remains t n e s i g n a l a t -30 ppni c h a r a c t e r i s t i c or aluminium phosphate. Inus, d i r e c t evidence i s g i v e n f o r t h e i n t e r a c t i o n p r o d u c t o f pnosphoric a c i d w i t h non-framework aluminium. U n f o r t u n a t e l y . from these d a t a t h e e x i s t e n c e of an i n t e r a c t i o n o r o d u c t o t p h o s p h o r i c a c i d w i t h b r i d g e d hvdroxv.1 groups cannot be concluded. However. t n e "'Al-MAS-NMR p r o v i d e s o d d i t i o n a l i n d i r e c t evidence t o r t h e e x i s t e n c e O T such a p r o d u c t . The spectrum on t o p of F i g . 1 4 shows t h e c h a r a c t e r i s t i c l i n e o f A ~ Fa t 5 5 ppm and a s m a l i amount o f o c t a h e d r o l l y c o o r d i n a t e d A I N F a t -1 ppm. P - m o d i T i c a t i o n o f t h i s sample and steaming T o r 3 hours s i g n i f i c a n t l y l o w e r s t h e i n t e n s i t y o f t h e A l l . s i g n a l . The s i g n a l a t about -10 ppm i s c h a r a c t e r i s t i c o t o c t a h e d r a l l y c o o r d i n a t e d aluminium ( 2 7 ) I n aluminophosphates, which r e s u l t from t h e r e a c t l o n o f orthoDhosPhoric a c i d w i t h A ~ N F . I h e secona new s i g n a l a t 40 PPm stems from t e t r a h e d r a l l y c o o r d i n a t e d A ~ N F a l s o p r e s e n t as alulninophosphutes. Other t e t r a h e d r a l l y c o o r d i n a t e d A ~ N F species can c o n t r i b u t e o n l y t o a s m a l l e x t e n t . dy means o f n u t a t i o n NMR spectroscopy (28, t h e quadrupoie i n t e r action o t the nuclei g i v i n g r i s e t o t h i s l i n e was shown t n be s m a l l . l h e soectrum a t t h e bottom o f F i g . 1 4 of demonstrates. t h a t e x t r a c t i o n phosphoric a c i d i n c r e a s e s c o n s i d e r ably the i n t e n s i t v of the Ah-signal a t 5 5 w i n . A s r e u i u i n i n a t i o n a t these I . . . . ,. . _I.. . . . I . . c o n d i t i o n s is u n l i k e l y t o occur, D22P1 , 6 (>h/7W°C) t n i s pnenonienoii i s i n u i c a t i v e TOI- o very close i n t e r a c t i o n o r r w i t h , . . . . , . . . . I . . . . , AIP, causing a diStOrtiOn O T i t s Syininetry, Which i s e i i m i n a r e c l DY t h e D2*P,,(, eluted removal of P . so, t n e i n t e n s i t y p t t h e A1F-signal i s a t l e a s t p a r t l y reduced by t h i s i n t e r a c t i o n . $00 50 0 -50 ppo F i n a l l y a few words concerni.ng t h e n a t u r e o f weak a c i d s j . t e s . I t has F i g 14.z7A1-NMH s p e c t r a o f been P o i n t e d o u t a l r e a d y . t h a t P- DzzHS-30 b e f o r e ana n f t e r P m o d i f i c o t i o n changes n o t o n l y t h e i r m o d i t i c a t i o n and - e x t r a c t i o n
A
-
18
c o n c e n t r a t i o n b u t a l s o t h e i r n a t u r e . Some i n c l i c a t i o n as t o t h i s p o i n t was o b t a i n e d trom t h e temperature-programmed d e c o m p o s i t i o n o f t h e NH-~*-exchanaed, P - m o d i f i e d samples.
c h
,. ?
d
I
2M
F i n . l 5 . T e m p e r a t u r e - ~ r o g r a m m e d d e c o m p o s i t i o n o f an NH4+-exchanged u n m o d i f i e d ( l e f t ) and P - m o d i f i e d somple. From F i a . 1 5 i t appears, t n a t i n d i s t i n c t i o n t o t h e u n m o d i f i e d samp l e , t h e P-modified sample b e s i d e s t h e h i g h - t e m p e r a t u r e PeaK exnib i t s a second one w i t h a maximum a t about 280 " C . l h i s peok may be assigned t o a c i d s i t e s w i t h medium s t r e n g t h . [ h e i r p r o p e r t y t o he ion-exchangeable f a v o u r s t n e i r i n t e r p r e t a t i o n us f i r o n s t e d a c i d s i t e s . Whether t h e i r e x i s t e n c e i s due t o t h e i n t e r o c t i o n p r o d u c t of phosphorlc acid w i t h b r i d g e d h v d r o x y l groups o r w i t h non-framework oluminium i s d i f f i c u l t t o d e c i d e w i t h o u t f u r t h e r experimental data.
D e a l u r i n a t i o n oT LSPI-5 t y p e HS-30 z e o l i t e bv h y d r o t h e r m a l t r e a t ment decreases t h e s t r o n g a c i d S i t e s and - c o n s e q u e n t l y i s ossoc i a t e d w i t h a d i m i n i s h i n g o t c o t u l y t i c a c t i v i t y . C o n s e c u t i v e orod u c t s o f t h e o l e f i n s a r e suppressed and t h e Cz-C4-olefin s e l e c t i v i t y i s r a i s e d . I t s v a l u e seems t o he l i m i t e d . however. t o about b 5 X . M o d i t i c a t i o n of P r e v i o u s l y d e a l u m i n a t e d samoles w i t h o r t h o phosphoric a c i d i n c r e a s e s t h e s e l e c t i v i t y o t cZ-c4 o l e t i n s UP to 80%, s u p p r e s s i n g t o r m a t i o n o f o r o m o t i c s and coking. D e a c t i v a t i o n times a r e s i g n i t i c a n t l v Prolonged (UP t o 1 0 0 h ) . A l l c a t a i y t i c
19
parameters depend bn t h e o r i g i n a l degree O T d e a l u m i n a t i o n and UP t o a c e r t a i n l e v e l at S i / A l = 250-300 t h e y a r e improved. M o d i f i c a t i o n w i t h o r t h o p h o s p h o r i c a c i d causes a f u r t h e r decrease o f s t r o n g a c i d s i t e s by i t s r e v e r s i b l e i n t e r a c t i o n w i t h t h e s e s i t e s . This i n t e r a c t i o n seems t o p r o t e c t AlF-atoms from b e i n g expulsed trom t h e framework under h y d r o t h e r m a l c o n d i t i o n s . The observed t u r t h e r d e a l u m i n a t i o n o f P - m o d i f i e d samples a f t e r severe hvdrotherinal t r e a t m e n t a p p a r e n t l y i n v o l v e s P r e f e r e n t i a l l y unprotected Ah-atoms. C o n c o m i t a n t l y w i t h t h e decrease o f s t r o n g a c i d s i t e s t h e c o n c e n t r a t i o n o f weak a c i d s i t e s i s changed and t h e s i t e s a r e a l t e r e d i n t h e i r n a t u r e . A new. ion-exchanseoble a c i d s i t e w i t h medium a c i d i c s t r e n g t h has been d e t e c t e d . These chnnaes a r e p r e d o m i n a n t l y due t o an i n t e r a c t i o n o f t h e p h o s p h o r i c a c i d w i t h t h e non-framework aluminium, which r e s u l t s i n a k i n d o f a l u minium phosphate. Ihe v e r y slnail diTTerences o t t h e c o n c e n c r a t i o n o i s t r o II g a c i d s i t e s above 1%P a l o n e h a r d l y e x p i o i n t n e Uependence 0 1 sel e c t i v i t y on t h e o r i g i n a l degree o t d e a l u m i n a t i o n . The more oronounced d i f f e r e n c e s i n t h e c o n c e n t r a t i o n o f w e a k a c i d sites favour t h e i d e a o f t h e i r p a r t i c i p a t i o n i n c a t a l y s i s a t these cnnd i t i o n s . An a d d i t i o n a l shape s e l e c t i v i t y e t r e c t , can n o t be e x c l u ded and seems t o be even l i k e l y f o r coke t o r m a t i o n .
I'he a u t h o r s express t h e i r g r a t i t u d e t o Leuna-Werke-AG t o r fund i n g P a r t s o f t h i s w o r k . They a r e i n p a r t i c u l a r i n d e p t e d t o Or-. K.Wehner and D r . D 3 i m m f o r t h e i r Permanent i n t e r e s t i n th1.s work and s t i m u l a t i n g d i s c u s s i o n s . The a u t h o r s thank M r s I n a Schulz t o r t h e d e t e r m i n a t i o n o f coke on spent c a t a l v s t s .
REF LRENCtS 1 W . W Kaedinq Und S 8 . B u t t e r . J . C o t a 1 . b l . 155(1Y80) W.W.Kaeding, C.Lhu; L B.Youna. B . W e l n s t e l n , anu s . A . d u t t e r , J
2
5 4 5
Cata1.67, 159(1981) L.6.Young, S . U . B u t t e r . and W.W.Kaeuing, J . ~ u t a l . / b . 41811Y8L) J.C.Vedrine, A Auroux, P . D e j a i f r e , V Oucarme, H.Hoser. s Lnou. J . C a t a l . / 3 . 14/(1982) J Nunan. J C r o n i n . and J.Cunningham, J . C a t a 1 . 8 7 . I7(1984)
20
6 M . D e r e w i n s k i . J . Haber, J . P t a s z y n s k i , V . P . S h i r a l k a r , and S . U z w i g a j . S t u d . S u r t . S c i . C a t a l . 1 8 , 209(1984) 1 J . A . L e r c h e r , ti. Rumplmayr, H . N o l l e r , A c t a Phys .Chem. 3 1 , 7 1 (1985) 8 J . A . L e r c t i e r , 6 . Rumplmavr, UPPI. C a t u l . 2 S , 215(1986) 9 A.Hahman, G.LemaY, A.Adnot, S . K a l i a s u i n e , J . C a t a 1 . 1 1 2 . 4 5 3 ( 1 9 8 8 ) 10 J.H.Kim, S.Namba. l-.Yasnima, Bull.Chem.Soc.Japan 61, l O S l ( 1 9 8 8 ) 11 F . L o n v l , J.Enge1hardt and D . K a l l o ; P . A . J a c o b s nnd R . A . v a n S a n t e n ( t d i t o r s ) , Z e o l i t e s : F-acts, F i g u r e s . F u t u r e 1988 t l s e v i e r Pub. B . V . Amsterdam. 1 3 5 / 1%S . 1 k a i , and M . Okamo t o , H .N i s h i o k a , r . Mivainoto, K . Mat s u z a k i , K . S u z u k i , Y.Kivazumi, T.Sano, S.Shin, A ~ p l . C a t a l . 4 9 , 143(1989) 1 5 H . V i n e k , G . Rumplmavr, ana J . A . L e r c h e r , J . C a t a l , 1 1 5 , 2 Y 1( 198Y) 14 A , J e n t y s , ci. Huiiwimavr, and J . A . L e r c h e r , A p p l . C a t o l . 5 5 , 2 9 9 ( 1 9 8 9 ) 1 5 G.Ruinplmavr. ~ . A . ~ e r c n e rL,e a l i t e s , l O . 283(1990) 36 G . S e o . R.Rvoo, J . C a t a 1 . 1 2 4 , 224(1990) 1/ M . Ko j ima , F . 1.et ebvre, flnd Y . Ben Taa r it , J . Chem . Soc . Fnrad . 1 r a i l s . 86(4), /5/(1YYO) 18 G.ohlmann, H . - G . J e r s c h k e w i t z ,
G . L i s c h k e . B . P a r l i t z , R. t c k e l t , M . R i c h t e r , L . ~ . L n e n i i e28, 5 , l b l ( 1 9 8 8 ) 1 9 c.D.Chans, C . T.--W.Chu, and R . F . Socha. .I . i o t a l . 8 6 . % 8 9 (1984) 20 li.ohlinann, H.-G.Jerschkewitz, G . i i s c h k e , R . k c k e l t , r . G r o s s , b . P a r i i t z , I . S c h u l z , K.Wehner, and D . l i m m , t h i s e d i t i o n O T Stud.Surf.Sci.Cata1. 2 1 . I . R .Anderson. K . Fnqer, T.Mole, and R . A . R n j a d h v a k s h i , . I . C a t n l . S 8
118 (1919) 2 5 t.LotTler, Ch.Peuker. H . - G . J e r s c h k e w i t z . C a t a l . lodclv 3 . 415(1988) 24 A . -R . Grimmer and U . HaUDeIlrel S s e r , i hein. P h v s . Ler t . Y 9 . 48 / ( 19 8 S ) i 3 U . f ” l u i i e r , t .Junn, b . ~ a O w and i ~ U.Hauoenreisser, C n e m . r h y s . L t ? t t . lU5, 55L(1984) Zb 1 .M.Duiican and D.C.Doualass. Chern.Phys.8/. 5 ? 9 ( 1 9 8 4 ) Z/ D . M u l l e r , 1 .Grilnze. t . H o l l a s and G.Ladwig. 7 . a n o r q . f l l l ~ . C . h e m . 500. 80(1985) 28 6 . L i t ) r o w i u s nnd W.Storek, u n p u b l i s h e d r e s u l t s .
21
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
New Directions in Zeolite Catalysis Jens Weitkamp Institute of Chemical Technology I, University D-7000 Stuttgart 80 (Federal Republic of Germany)
of Stuttgart, Pfaffenwaldring 55,
Summary It is likely that the catalytic performance of a much broader variety of zeolites and related microporous materials will be routinely scrutinized in the nineties. In shape selective catalysis, reactions of lar er molecules will play an increasing role. Even in petroleum refining and the manufacture o f commodity petrochemicals, there is room for novel a plications of zeolite catalysts. Two examples, viz. dewaxing by isomerization and isobutane butene alkylation are discussed in some detail. The use of zeolite catalysts for the synthesis o organic intermediates and fine chemicals continues to be a thrust area. It is desirable that synthetic organic chemists routinely employ molecular sieve catalysts. Stereoselective catalysis in zeolites is still an underdeveloped field. The same holds for base catalysis in zeolites, especially shape selective base catalysis, but new approaches can be perceived in this area. Coupling of the exothermic MTO process and the endothermic hydrocarbon cracking on ZSM-5 type catalysts allows an almost thermoneutral production of lower olefins. The role of zeolites as hosts for a variety of guests is recognized by an ever increasing number of researchers. Ship-in-the-bottle catalysts or shape selective photochemistry are just a few examples for host/guest chemistry in zeolites,
P
Introduction and scope About thirty years after its first industrial application, catalysis on zeolites continues to be a most dynamic and promising field of research. The advantages of microporous crystalline materials over more conventional solid catalysts are nowadays recognized by an ever increasing number of scientists. Among these advantages are (i) the strictly regular pore size and pore architecture; (ii) pore widths of molecular dimensions which enable shape or size selective catalysis; (iii) acid surface properties with the possibility to tune the nature (Br6nsted versus Lewis acidity), density, strength (or strength distribution) and location (e. g., inside the pores or at the external surface) of the acid sites within certain ranges; (iv) the high thermal stability; (v) the possibility to regenerate deactivated catalysts, e. g., by burning off carbonaceous deposits; (VI)the capability of zeolites to act as hosts for a variety of guests with catalytically attractive properties, such as transition metal ions, small metal clusters or highly selective transition metal complexes or chelates. Indeed, through host/guest chemistry, zeolite science has many features in common with the modern supramolecular chemistry. Although nowadays, an impressive number of industrial processes are operated with zeolite catalysts, the riunzber of strLictiiru//y different zeolites accepted as commercial catalysts has remained surpri.ying/y.mu//:To our knowledge, the zeolites utilized so far as industrial catalysts are restricted to faujasite, mordenite, ZSM-5 and, probably, erionite. The domain of zeolite Y is petroleum refining where it has been used on a very large scale in fluid catalytic cracking (FCC) [l-41and hydrocracking [5-9].-Mordenite is applied in
22
the isomerization of light alkanes for octane enhancement of gasoline [5,8-101. Zeolite ZSM-S is the prototype of shape selective catalysts. Its realm is the manufacture of commodity petrochemicals, such as ethylbenzene or para-xylene [ 11-13]. Moreover, ZSM-5 based catalysts are applied i n the production of gasoline from methanol [13-151, in the M-Forming process [16,17] for octane improvement of reformer gasoline and in dewaxing of middle distillates and lubricating oils [12,13]. For the latter process, mordenite based catalysts have ;iIso been employed [ lS,l9]. Finally, ZSM-S is used as an additive in some F C C catalysts for octane enhancement of the gasoline produced [20,21]. The titanium containing analogue of ZSM-5, named TS-1 [22,23], is employed for the hydroxylation of phenol to hydroquinone/catechol mixtures. Erionite was the catalyst in the old Selectoforming process [24,2S] and is reported to be used again, in an improved form, i n the German Democratic Republic for octane enhancement of reformate [26]. Until several years ago, most fundamental studies on zeolite catalysis, too, were restricted to ;I very small number of structural types, again mostly faujasites, mordenite and ZSM-S. Lately, this situation has changed significantly: The catalytic behavior of an ever increasing number of alumosilicates and related microporous materials is being scrutinized, which stems, in part, from the remarkable progress in hydrothermal synthesis and structure determination. Likewise, the recent progress in post-synthesis modification techniques has a strong impact on catalysis. It is another most important trend that zeolite catalysis spreads into almost all branches of chemistry with a remarkable boom in the manufacture of organic intermediates and fine chemicals. The present review paper is intended to describe some selected developments in zeolite catalysis which, in the author’s opinion, deserve particular attention in the nineties. Hroader variety of microporous materials More and more groups which have been traditionally active in zeolite catalysis are adopting the techniques of hydrothermal synthesis with the benefit that they now dispose of a broad assortment of zeolitic materials, in addition to the few ones which are commercially available. In our opinion the alumosilicate zeolites listed in ‘Table 1 will receive particular attention in catalysis over the coming decade. Even today, materials like Beta [27-311, Omega [32-35], L [35-371, offretite [3 1,351, EU-1 [38-401, ZSM-12 (35,411, ZSM-22 [38,42-441, E M - 2 3 [38,42,43] and ZSM-48 [35,38,45] are known to possess attractive catalytic properties for many applications. It is likely that, in the nineties, several other alumosilicates with eight-, ten- o r twelve-membered ring or even larger pores have to be added to this list. One example could be ZSM-18 with its unique pore system consisting of pseudo-unidimensional, puckered 12-membered ring channels with side pockets capped by 7-membered rings [46]. The zeolites enumerated in Table 1 cover a n appreciable range of pore widths from ca. 0.5 to 0.8 nm which is the relevant range for most shape selective conversions. As a relative nie:isure for the effective pore width of microporous materials, indices derived from certain cat:ilytic test reactions have been found to be very useful. The oldest and best known example is Mobil’s Constraint Index [47]. Other useful indices for this purpose are the Refined or
23
TABLE 1 Aluniosilicate zeolites with particular potential in catalytic research. (The dotted line separates zeolites with 10- and 12membered*ring pores. CI denotes the Constraint Index [47], the* data for CI were taken from [48]; CI is the Refined Constraint Index [49,50], the CI data were taken from [51]; SI stands for the Spaciousness Index [52,53], the SI data were taken from [53]). ZEOLITE
IZA CODE
RELATED STRUCTURES
Theta-1 ZSM-23 Ferrierite ZSMJ ZSM-48 ZSM-11
TON M7T FER MF1
ZSM-12 EU-1 Offre tite Mordenite Omega L Beta ZSM-20 Y
CI
CI*
MEL
ZSM-22, ISI-I, KZ-2, Nu-10 EU-13, ISI-4, KZ-1 ZSM-35, ISI-6, NU-23,FU-9 Silicalite 1, Nu-4, Nu-5 EU-2, EU-11, ZBM-30 Silicalite 2
_____.
. . . . . . . . . . . . . . . . . . . . . . . . . .
MTW EUO OFF MOR MAZ LTL
CZH-5, Nu-13, Theta-3 ZSM-50, TPZ-3
FAU
Mazzite, ZSM-4 NU-2 Intergrowth of BSS1) and FAU Faujasite, X
7.3 9.1 4.5 6- 8.3 3.5 5 - 8.7 2.3 2.1 3.7 0.4 0.5 0.6- 2.0 0.5 0.4- 0.5
14.4 10.8 10.3 6.8 5.2 2.7 2.2 1.8 1.8 1.2 1.0 1.4 1.3
SI
w l
-1 w 1 w l Fa1 u l
____ 3 5 5.5 7.5
16.5 13- 19 20 21
1)Code for Breck’s Structure Six Modified Constraint Index which was defined by Jacobs et al. [49,50] and the Spaciousness Index which was introduced by Weitkamp et al. [52,53]. Wherever available, these three indices are given for the zeolites listed in Table 1. From top to bottom, i. e., with increasing effective pore width, both the Constraint Index and the Refined Constraint Index decrease whereas the Spaciousness Index increases. A number of elements other than silicon and aluminum can occur on tetrahedral sites in the crystal lattice of microporous solids [54]. This way, a huge number of materials result, part of which possess structures known from the alumosilicates. An example is titanium silicalite, TS-I, with the crystal structure of ZSMd and very interesting catalytic properties in selective oxidation reactions [22,23,55-571. Other members from this familiy of materials possess completely new structures. This is particularly true for certain alumophosphates (AIPOds), silicoalumophosphates (SAPOs) and the microporous solids derived from them by
24
incorporation of additional elements, such as titanium, manganese, cobalt or iron into the lattice [58,59]. The potential of alumophosphate based molecular sieves in catalysis has rcccntly been reviewed by Rabo et al. [60]. I
I
( c a . 0.8nm I
lea. 1.2 nm 1
Fig. 1. Structures of AIP04-11, AIP04-5 and VPI-5. Within the frimily of AIP04s, very large and super-large pore molecular sieves have recently been identified. Davis et al. recognized that V P I J is a material with unidimensional channels formed from 18-membered rings with an approximate diameter of 1.2 nm [61-631. The structure of VPI-5 is shown in Fig. 1 along with the related structures of A1P04-5, a material with unidimensional 12-membered ring pores, and AIP04-l 1, a material with unidimensional 10-membered ring pores. It is seen that the cross section of the pores in AIP04-I 1 is strongly elliptic which renders the material particularly interesting for shape selective catalysis. The discovery of the existence of super-large pore molecular sieves might be a landmark in zeolite catalysis. Such materials are of great interest for the catalytic conversion of feedstocks containing large molecules as, e. g., petroleum residues. AIPO,s, however, possess very little o r n o acidity. As soon as silicoalumophosphates or zeolitic :ilumosilicates with super-large pores can be synthesized this will have a tremendous impact o n catalysis. Another material, named MCM-9 [64,65a], does contain Si-VPI-5 but it is apparently not a pure phase. A third molecular sieve material with very large pores, i. e., channels formed by more than 12 tetrahedra, is AIPO4-8 [65a] the structure of which has been determined very recently [6Sb]. A1P04-8 possesses unidimensional channels formed from 14-membered rings with the approximate dimensions 0.79 x 0.87 nm [65b]. Shape selective catalysis: Towards larger molecules Shape selectivity effects can occur if the pore width of the catalyst is in the same order as the dimensions of reactant molecules, transition states, intermediates and/or product molecules. The principles of shape selective catalysis in zeolites have been discussed in much detail in prior review articles, e. g. [66, 671. In the past, almost all shape selectivity effects were
25
related to the conversion of aliphatic compounds or inononuclear aromatics [68] in tenmembered ring zeolites, typically HZSM-5. We perceive more and more attempts to make use of shape selective catalysis for the production of larger molecules, e.g., valuable derivatives of biphenyl or naphthalene. 4,4’-Diisopropylbiphenylwhich is an intermediate for liquid crystals could be produced from propene and biphenyl in high yields on dealuminated H-mordenite [69]. Other 12-membered ring zeolites like HY, H L or H-offretite which were tested for comparison gave very poor results. So did the original mordenite sample with an Si/AI ratio of 5. Upon massive dealumination up to an Si/AI ratio of 1300 mesopores were created. It was one of the authors’ conclusions that the mesopores are important for an efficient mass transport of the bulky molecules and that dealumination, i. e., dilution of the active sites, slows down both coke formation and undesired alkylations, e. g. in para-meta positions. 2-Methylnaphthalene which is an intermediate in the synthesis of vitamin K can be formed by isomerization of 1-methylnaphthalene on acid catalysts. Usually, however, the isomerization is accompanied by undesired side reactions, especially by transalkylation and coking, and this was found to be particularly true for zeolite H Y [70]. In HZSM-5, on the other hand, the isomerization turned out to be severely hampered by diffusional limitations. As shown in Fig. 2, certain zeolites with an intermediate pore width, characterized by Spaciousness Indices from 3 to ca. 16, gave excellent results: Initially, the equilibrium
6\”
100
>:
I
I
HZSM-12
(SI = 3)
75-
Si/AI = 35
- \m: O S
.
100
-
I
I
H-EU-1
(SI = 5) Si/AI = 20
75 - ; 05
1
-
-I
w
P
-
X
-
25-
100
I
I
I
I
H-Mordenite
1001
(SI = 7) Si/Al = 6.7-
O
,,-x,
y,*, A y,, n
- 1
-
251
I
I
I
H-Beta
(SI = 16)
--
I
STREAM, h Fig. 2. Isomerization of I-methylnaphthalene in acid zeolites with intermediate pore widths at 300 “C after [70] (SI: Spaciousness Index; Tr: Transalkylation products).
26
conversion of ca. 70 % was achieved. The bimolecular transalkylation into naphthalene and dimethylnaphthalenes which requires a very bulky transition state and/or intermediate, was almost completely suppressed. With HZSM-12, H-EU-1 and H-Beta the deactivation was moderate. Only with H-mordenite, there was a rapid deactivation due to coke formation which has probably to do with the relatively high aluminum content rather than with the unidimensional pore system; otherwise one would expect the same rapid deactivation for HZSM-12 which has unidimensional pores as well. 2,6-Dimethylnaphthalene is a particularly valuable compound. It can easily be oxidized into naphthalene-2,6-dicarboxylicacid which is a starting material for high quality polyesters, polyamides and liquid crystals. The problem in its synthesis is that it is usually formed together with 9 other isomers which are difficult to separate. In principle, zeolite catalysis offers the possibility of enhanced selectivities for the 2,6-isomer, since its molecular dimensions are smaller than those of most other isomers. Indeed, Matsuda et al. [71,72] found that the disproportionation of 2-methylnaphthalene in H Z S M J leads to a dimethylnaphthalene fraction which contains the 2,6-isomer (and the 2,7-isomer which has the same molecular dimensions) in excess of thermodynamic equilibrium. The yields, however, were very small, presumably due to diffusional limitations. By dealumination of the external surface with (NH&SiF6 it was shown that the shape selective disproportionation occurs inside the ZSM-5 pores [73]. Another way to 2,6-dimethylnaphthalene is the shape selective alkylation of naphthalene or 2-methylnaphthalene with methanol. Fraenkel et al. [74] did observe such enhanced selectivities for 2,6- and 2,7-dimethylnaphthalene on HZSM-5, but not on H-mordenite or HY. These authors proposed [75,76] that the shape selective conversion of naphthalene derivatives occurs uf the cxternul surjiuce of zeolite ZSM-5, in so-called "half-cavities". This concept which was generalized by Derouane et al. [77,78] has, however, been questioned severely [70,79]. It is more likely that the shape selective alkylation of naphthalene or 2-methylnaphthalene on HZSM-5 occurs inside the pores and is slowed down by diffusional limitations. So far, most shape selectivity effects were observed in zeolite Z S M J . It is clear from the origin of these effects [66,67] that with bulkier reactant and/or product molecules, shape selectivity can occur in zeolites with 12-membered ring pores as well. An example is the suppression of transalkylation of methylnaphthalenes in HZSM-12, H-EU-I, H-mordenite or H-Beta, as shown in Fig. 2. Even in the very open and spacious pore system of zeolite Y shape selectivity effects were observed, e. g. during the competitive hydrogenation of cyclohexene and cyclododecene in Rh/NaY [go], hydrocracking of hydropyrenes in NiW/ultrastable Y [81] or catalytic cracking of cis-decalin/trans-decalinmixtures in ultrastable Y [82]. Petroleum refining and commodity petrochemicals The application of zeolite catalysts in petroleum refining and the manufacture of commodity petrochemicals is sometimes considered as a mature technology. Even in these fields, however, there is room for improvements and new developments. Large efforts are
being undertaken by the industry to develop FCC catalysts which yield gasoline with enhanced octane numbers, especially motor octane numbers. Once the petroleum prices rise again all problems associated with catalytic cracking of residues will receive renewed attention, especially the resistance of the zeolite in FCC catalysts towards vanadium and nickel. Possibly, new zeolite based hydrocracking catalysts will be developed. For example, Idemitsu Kosan Co. claims a novel iron on Y-type zeolite catalyst for residue hydrocracking [83]. For the isomerization of light alkanes, zeolite catalysts with higher acid strength and hence higher activity are desirable, so that lower process temperatures can be applied, where the equilibrium is more favorable for the high-octane isomers. Making aromatics from liquefied petroleum gas (LPG) by BP/UOP's Cyclar [85] or Mobil's M2-Forming process [86] is ready for commercial application and the same is probably true for the production of distillates from light olefins by the MOGD process [87]. Acid pentads could have advantages [88] over the conventional organic cation exchange resins which are used as industrial catalysts for the production of methyl tert.-butyl ether. Another process which offers a large potential for zeolite catalysts is the manufacture of cumene from benzene and propene. Dewaxing by isomerization The cold flow properties, e. g. the pour point and cloud point, of petroleum fractions are largely determined by their content of normal paraffins. One method for the removal of these disturbing waxes is shape selective hydrocracking on ZSMd based catalysts [13]. In this process, the undesired paraffins in the boiling range of the middle distillate or lubricating oil feedstock are converted into gasoline and C, + C4 (LPG) hydrocarbons. Another approach is to transform the waxy paraffins into branched isomers which have considerably better cold flow properties. This way, the yield loss necessarily associated with dewaxing by selective hydrocracking could be avoided. The isomerization of long chain n-alkanes can indeed be achieved on bifunctional zeolite catalysts with a strong hydrogenation component, e. g., Pt/CaY or Pd/LaY in the presence of hydrogen [89,90]. Under these conditions, skeletal isomerization takes place at the acid sites via carbenium ions. These carbenium ions, however, can also undergo 0-scission, then the overall reaction is hydrocracking. Since skeletal rearrangement and 0-scission are consecutive reactions, the selectivity for long chain iso-paraffins is high at low conversions, but as the conversion is increased, hydrocracking becomes more and more severe. In other words, the yield of the desired iso-paraffins in dependence of the conversion of long chain n-paraffins passes through a maximum. Even with the best catalysts based on zeolite Y, this maximum yield of long chain iso-alkanes amounted to ca. 60 % (891. Significantly higher yields of long chain iso-alkanes can be attained on certain other zeolites, especially on bifunctional forms of Beta [91,92], ZSM-22 [43] and ZSM-23 [43]. Two examples are shown in Fig. 3: The maximum yields of iso-decanes from n-CloHzz or isotetradecanes from n-CI4H30 on Pd/H-Beta are above 70 %. In the future, the fundamentals of dewaxing by isomerization need further clarification. Presumably, the mechanistic reasons for the unusually high yields of iso-alkanes are different
28
for zeolite Beta on the one hand and medium pore zeolites such as ZSM-22 or ZSM-23 on the other hand. It has been argued [43] that in the latter zeolites, B-scission is slowed down because the highly branched precursors for the most favorable type of 13-scission (so-called type A O-scission) cannot form due to spatial constraints. No explanation for the interesting behavior of zeolite Beta has been proposed so far. Apparently, not all samples of zeolite Beta do give the high yields of long chain iso-alkanes [27]. It is particularly desirable to find out the key property of zeolite Beta which governs its catalytic performance in isomerization of n-alkanes. Furthermore, isomerization experiments with mixtures of homologous n-alkanes on bifunctional Beta catalysts are needed in the future. 100
-
n Decane
n-Tetradecane WIF = 6LOg-hlmol
WIF = 570g.hImol
$ 60
0 Xn-De
0 Xn-Te
>
0, tl! >
Ylso
A Ycr.
60
Ili
0
10 X
z
0 v) OL
W
20
2
0 V
210
220
230
210
250
260
270
280
190
200
210
220
230
210
250
260
270
TEMPERATURE , ' C
Fig. 3. Isomerization and hydrocracking of long chain n-alkanes on 0.27 wt.-% Pd/H-Beta after [92]. (The hydrogen partial pressure was 2 MPa). Isobutane/butene alkylation This refinery process converts the C4 by-product from catalytic crackers into high quality gasoline. Today, alkylation is carried out either with concentrated sulfuric acid or anhydrous hydrogen fluoride as catalysts. There is a very large incentive for replacing these processes by a new technology which is environmentally more acceptable and safer. In the sixties and seventies, the potential of acid faujasite catalysts in isobutane/alkene alkylation has been explored by several groups [93-1011. Typical results are depicted in Fig. 4 and Fig. 5: 011fliefiesh zeolite, the olefin is totally converted (Fig. 4), and a perfect alkylate is formed (Fig. 5 ) which consists fully of iso-alkanes. After a certain time on stream, however, the olefin conversion drops and the product quality deteriorates drastically: More and more C8-alkenes form and eventually, the zeolite no longer alkylates isobutane; rather, the butene
29
-
100
+a 0
0 I-
z
80
60
u)
U
W
5
40
0 0
TIME
ON
STREAM, rnin
Fig. 4. Conversion of isobutane with 1-butene (11:l) at 80 "C in the liquid phase on CeY-98 zeolite after [loo].
bp
I W
I
I
I
I
Fig. 5. Reaction of isobutane with I-butene on CeY-98 at 80 "C. Composition of the Cg product fraction after [loo].
oligomerizes. This dramatic loss of catalyst performance is due to the build-up of low temperature coke from the olefin. Concomitantly, the zeolite looses its hydrogen tranfer activity. The yields of alkylate which could be achieved on faujasite catalysts were much too low for an industrial application. If one arbitrarily considers the product as an alkylate as long as it contains 90 mol-% or more alkanes in the C,-fraction, then one can define an integrated yield
30
of alkylate (total mass of alkylate formed/mass of catalyst) for a given catalyst and set of reaction conditions. Typically, these yields were in the order of SO to SO0 mg/g [loo]. Zeolite HZSM-S was reported to be inactive in isobutane/butene alkylation a t temperatures up to 100 "C and this was attributed to the narrow pores [102]. In the meantime, a much broader variety of zeolitic materials and techniques for their manipulation are known. Given the high incentive for replacing the H$04 and HF catalysts we foresee a renewed interest in the use of zeolite catalysts for isobutane/butene alkylation. Organic intermediates and fine chemicals Although so far, the industrial application of zeolite catalysts has been essentially restricted to hydrocarbon chemistry, their potential in the synthesis of organic intermediates, especially of oxygen and nitrogen containing compounds, was recognized by several research groups as early as in the sixties. This early work was reviewed by Venuto and Landis [103]. With the advent of the modern molecular sieve materials, novel modification techniques and a better understanding of shape selective catalysis, a renewed interest arose in the utilization of zeolite catalysts for organic syntheses from about 1980 onward. O n e result was the introduction of at least one new commercial process, viz. the hydroxylation of phenol to hydroquinone/catechol mixtures on titanium silicalite by EN1 [22,23]. A few others, such as the conversion of methanol with ammonia with enhanced yields of methylamine and dimethylamine at the expense of the less valuable trimethylarnine on narrow pore zeolites like Rho or ZK-5 [104-1091 are considered to be ready for commercialization. In recent years, the potential of zeolite catalysts was explored in numerous organic reactions, both in the gas and in the liquid phase. Essential parts of this modern work have been discussed and evaluated in a number of excellent review articles [110-1141. The application of zeolite catalysts in the manufacture of organic intermediates and fine chemicals will continue to be a thrust area in the nineties. In the following paragraphs a few selected reactions will be addressed which, from the author's viewpoint, are examples for the specific behavior and versatility of molecular sieve catalysts. Gallezot et al. [ 1151 investigated the selective hydrogenation of cinnamaldehyde in the liquid phase at 60 "C under a hydrogen pressure of 40 bar. The first hydrogenation step can lead to either 3-phenylpropanal or cinnamyl alcohol (Fig. 6), depending on whether hydrogen adds to the carbon-carbon double bond or the carbonyl group. Ultimately, 3-phenylpropanol is formed. Two samples of Pt/Y zeolite were prepared by ion exchange of NaY with Pt(NH3)4?+ and subsequent decomposition of the ammine complex under different, but well defined conditions. The Pt loadings were 11 and 14 wt.-%, respectively. Transmission electron microscopy revealed that, in the first sample, the noble metal was homogeneously distributed inside the zeolite with a particle diameter of 1 to 2 nm. In the second sample, the platinum was still inside the zeolite pores. However, it formed much larger agglomerates with an average diameter of 5 nm, and these metal particles were preferentially located in a shell near the external surface.
31
a
cn = CH - C H ~ O H
clnnamsldehyde (CAL)
clnnemyl alcohol (COU
Catalyst
11 wt.-% PtlY (1-2 nm. insidel 14 wt.-% PtlY (5 nm, inside]
97 %
"Liquid phase, T = 6OoC. pnz = 40 bar After P. Gallerot el at.. Catal. Letters
2 169 11990)
Fig. 6. Selective hydrogenation of cinnamaldehyde to cinnamyl alcohol on platinum clusters encapsulated in zeolite Y after [ 1151.
The results summarized in Fig. 6 show that hydrogenation on platinum in Zeolite Y gave much better selectivities for cinnamyl alcohol than on a non-zeolitic reference catalyst like platinum on charcoal. This was interpreted by a special shape selectivity effect: Inside the faujasite pores, the cinnamaldehyde molecule can adsorb on the platinum clusters, sitting in the supercages, only in an end-on mode, i. e., via its carbonyl group, through the twelvemembered ring window. By contrast, the adsorption and activation of the carbon-carbon double bond in the center of the molecule is considered to be strongly hindered, but only as long as the supercage is completely filled with platinum; a small cluster of six or less metal atoms would leave enough space for cinnamaldehyde molecules to enter the supercage and adsorb laterally, i. e., via the carbon-carbon double bond, so that hydrogenation leads to 3-phenylpropanal. This is how the authors explain that on Pt/Y with the 1 to 2 nrn clusters the undesired hydrogenation of the carbon-carbon double bond occurs to a certain extent. Adamantane has been found to form readily by rearrangement of tetrahydrodicyclopentadiene (tricycIo[5.2.1.O2b]decane, see Fig. 7) which, in turn, is easily accessible from the Diels-Alder dimer of cyclopentadiene. Suitable catalysts for this interesting reaction are acid or bifunctional forms of zeolite Y, such as rare earth Y (REY), Pt-Re/REY or H Y [116,117]. For example, at temperatures around 250 "C, adamantane selectivities between 40 and 50 % could be achieved at conversions above 50 %. Likewise, mixtures of 1- and 2-methyladamantane were obtained on the same catalysts from tricyclo[6.2.1.0.2>7]undecane [ 1171. One of the conclusions from these studies was that the faujasite structure, due to the existence of almost spherical cages, is in general a promising catalyst for the formation of spherical molecules, and adamantane or its homologues might be just examples for spherical molecules which can be advantageously synthesized inside the cages of zeolites.
L
a) K. Honna et al.. ldemllsu Koaan Co., since 1985 b) G.C. Leu and W.F. Maler. LanOmulr 1987
endoTrlcyclo [5.2.1.0"]
exo-
Adamantane
decsne
Fig. 7. Formation of adamantane from tetrahydrodicyclopentadiene in zeolite Y after [116,117]. A more recent article [ 1181 reports on the formation of adamantane and its homologues from a simple alkene, viz. 1-hexene, at 240 "C in SAPO-34, a silicoalumophosphate with the chabazite structure. However, unlike in the above-mentioned studies with zeolite Y, the adamantanes did not appear in the gaseous effluent from the reactor; rather, they were built up and entrapped inside the chabazite cages (see Fig. 8), presumably again via tetrahydrodicyclopentadienes as intermediates. The adamantanes are, of course, too bulky to escape through the eigth-membered ring windows of the chabazite cage. The only way to detect them is the dissolution of the used molecular sieve, e. g., in hydrogen fluoride, followed by extraction with an appropriate solvent, e. g., CH,Cl,. It has been again concluded from these results [118] that the microporous catalyst with its cages of molecular dimensions can be considered as an ideal host for transition state guests which lead to spherical, polynuclear products. At higher reaction temperatures around 400 "C, naphthalene and its methyl derivatives were preferentially formed inside the cages (cf. Fig. 8) of SAPO-34, probably in related chemical pathways via dicyclopentadienes. It is too early to judge upon the general validity of these concepts. If zeolite cages act indeed as hosts which favor the formation of spherical products, then one could imagine a completely new direction of zeolite catalysis: A zeolite with cages of appropriate size and geometry could act as u templute which directs a chemical reaction towards the desired product, which may or may not have a spherical shape. If this product is of sufficiently high value (and/or the molecular sieve used as a reaction template sufficiently cheap) then it could even be economical to dissolve the zeolite so that the product becomes accessible through extraction. In the precise sense, the zeolite would then no longer play the role of a catalyst (with a high number of closed catalytic cycles or turnovers) but simply act as a selectivity directing reaction vessel with uppropriute moleculur dinzensions.
33
124.
After J.R. Anderson el. al.. J. Catal. 259 (1990) Flow reactor, I-hexene in N, dissolution of used catalyst in HF. extraction with CH,CI,, analysis by G C l M S
1-Hexene
-
b'2400c
1
Adamantane Methyladamantane Dimethyladamantane Ethyladamantane
=4OO0C
Naphthalene Methylnaphthalenes Dimethylnaphthalenes
Fig. 8. Formation of polycyclic hydrocarbons entrapped inside the chabazite-like cages of SAPO-34 from 1-hexene after [118]. An example for such a reaction which one could envisage in a very speculative manner is the synthesis of compounds with the dodecahedrane (C,,H,,, see Fig. 9) skeleton. Such compounds which strongly resemble a sphere became accessible in the eighties by multi-step syntheses, usually via precursors with the pagodane structure [ 119-1241. To our knowledge, no systematic and professional attempt has been undertaken so far, to utilize molecular sieves for
L H . Paquette et al., alnce ce. 1981 H. Prinzbach e l al., since ca. 1985
Fig. 9. A speculative example for the use of molecular sieves as reaction templates: Pagodanes to dodecahedranes. dodecahedrane synthesis. The reasons are obvious: Modern synthetic organic chemistry and professional zeolite chemistry are dealt with in different laboratories and separated from each other. It is a managerial challenge to bring inorganic host/guest chemistry and synthetic organic chemistry together.
34
Stereoselective catalysis Enantioselective catalysis in zeolites is often looked upon as one of the ultimate goals. It requires chiral centers next to the catalytically active sites. In principle, different methods can be used to induce chiral properties. According to Dessau [125], chiral hydrogenating catalysts result if a n acid zeolite loaded with a hydrogenation metal, e. g., Pt/HZSM-5, is neutralized with an optically active amine, such as S-(-)- a-methylbenzylamine (Fig. 10). O n such a catalyst, the hydrogenation of
2-Phenylbutene
y
5
2-Phenylbutane
Pt/(CH 3 - C x - N H , ) Z S M - 5 I H C,H,-
C - CH, I1
b-
C,H,-
0 +H2
Acetophenone
H
kCH, OH
1-Phenylethanol
Afler R.M. Dessau. US Patent 4 554 262, Mobil Oil Corp.. 1985
Fig. 10. Enantioselective hydrogenation of prochiral compounds after [ 1251. prochiral compounds, e. g., 2-phenylbutene or acetophenone, is claimed to give 2-phenylbutdne or 1-phenylethanol, respectively, in a n optically active form, i. e. with an excess of one cnantiomer. It is astonishing that, since the appearance of Dessau’s patent [ 1251, his approach has not heen dealt with more comprehensively in the scientific literature. Zeolite Beta, as it can be synthesized today, is an intergrown hybrid of two closely related structures [120,127]. One of these end members, the so-called polymorph A, forms an enantioniorphic pair while polymorph B is achiral. Since the structure of zeolite Beta has been determined, large efforts are being undertaken to develop synthesis procedures for its pure polymorph A. If, at the end, these endeavors will be successful and methods for the separation of the enantiomorphic pair will be available, then the old dream of enantioselective catalysis i n zeolites could become reality. Base catalysis Base catalysis in zeolites is still an underdeveloped field of science and technology, especially in comparison with acid catalysis. In the literature, essentially three approaches to create basic sites in zeolites can be discerned. Interestingly, these attempts were almost completely restricted to faujasites. It was recognized early [ 1281 that faujasites exchanged with large alkali cations, especially CsX and RbX, catalyze the side chain alkylation of toluene with methanol or formaldehyde, rather than ring alkylation to xylenes which occurs on acid zeolites. Later, side chain alkylation to styrene/ethylbenzene mixtures on faujasites exchanged with alkali cations has
35
been studied in detail by several groups [129-1351. It has been generally concluded that this is a base catalyzed reaction via carbanions, although the nature of the basic sites under reaction conditions remained a matter of debate [136,137]. Clusters from sodium metal can be generated in the pores of faujasites using various techniques. According to Jacobs et al. [138-1411 the preferred method from the viewpoint of catalysis is the impregnation of the dehydrated zeolite with NaN, followed by the thermal decomposition of sodium azide under carefully controlled conditions. In faujasites modified by this procedure, three types of sodium particles can be distinguished by ESR spectroscopy: Na43+ clusters located inside the sodalite cages, neutral Na, clusters inside the pore system and neutral Nay particles at the external surface. Basic framework oxygen anions next to intracrystalline Na, clusters seem to be the active sites [141] for base catalyzed reactions such as the isomerization of cis-2-butene [ 139,1401, the side chain alkylation in toluene, ethylbenzene or cumene with ethene [ 1411 or regioselective ring opening in unsymmetrical epoxides [142]. While basic zeolites prepared via the azide method appear to be very active catalysts, it remains to be seen how stable they are against potential poisons, such as water formed in many reactions or sulfur compounds present in traces in many feedstocks. The third method, introduced by Hathaway and Davis [143,144], consists of impregnating a faujasite, e. g. CsNaY, with cesium acetate followed by decomposition of the acetate at 450 "C in air or helium. It appears that the active site in such catalysts is cesium oxide rather than metallic cesium, cesium carbonate or cesium hydroxide [ 1451. These modified faujasites were tested in various base catalyzed reactions, such as the dehydrogenation of isopropanol to acetone [ 144-1461 and the alkylation of methane, ethane, acetone or toluene with methanol [146]. It will be interesting to observe whether this new method is accepted by other groups. If
Si 0, 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
A l PO1 Al 0
Zeolite IAcid 1
0
0
H'B
/ V S /o\; i / \ / \
o \ i 0
Sip0 ?
0
0
0
0
0
0
0
0
0
He
H@
/VSi /o\si /O\$/O I \ / \ / \ / \
0
0
0
0
0
0
0
0
OH 0
OH8
OH8 0
(Base 1
i ' o
i ' o
o/\o
d'o
o/\o
Fig. 11. Neutral, acid and hypothetical basic molecular sieves.
/ \
0
0
36
so, this could foster base catalysis in zeolites including, perhaps, shape selective base catalysis with its huge potential in the manufacture of organic intermediates and fine chemicals. A hypothetical approach towards a new generation of molecular sieve Briinsted bases might be seen in the preparation of microporous silicophosphates, either by direct synthesis or post-synthesis modification. As shown schematically in Fig. 11, and for obvious reasons, a molecular sieve S i 0 2 , such as silicalite 1 or 2, has no lattice charge. The same is true for the overuff lattice in AIP04s due to the strictly alternating arrangement of @lo4)and (PO4)+ tetrahedra. Alumosilicate zeolites are cation exchangers and, in their H + forms, strong Briinsted acids. By mere analogy and in a formal manner a porous silicophosphate would be expected to be an anion exchanger and, in its OH- form, a Briinsted base. In a triangular compositional diagram (Fig. 12), the alumosilicate zeolites and the hypothetical silicophosphates are just limiting cases which are represented by positions on the Si02-A102' and SO2-PO2+ connecting lines, respectively. There exist a whole field of cation exchangers, embracing both the SAPOs and alumosilicate zeolites, and a whole field of
A
Electroneutrality
.....................
AI02-/ Field of cation exchangers
0.5
-
:. Al-
: rich
Field of anion /e xc ha ng e rs ?
SiO, high : silica : ) .
ZEOLITES
Fig. 12. Schematic representation of the possible compositions of molecular sieve materials in the Si02/A102/P02 system, adapted from Flanigen et al. [%I.
37
hypothetical anion exchangers, and these two fields are separated from each other by the electroneutrality line which connects the AIPO4s and SiO2. It has, in fact, been reported [147] that the overall framework charge in a zeolite can be changed from negative to positive if a synthetic zeolite is treated with a melt of ammonium dihydrogen phosphate at 230 "C. For the resulting materials with anion exchanger properties the name phosphated aluminosilicates or PASOs [147] was coined. It remains to be seen whether the new procedure is reliably applicable to other zeolites and by other groups. If so, the door to a new class of microporous Bronsted bases would be open. It is sometimes difficult to distinguish unambiguously between acid and base catalysis on molecular sieve materials. Recently, Dessau [ 148,1491 proposed the conversion of HZSM-5 (SIIAI = 35) T = 225OC
I1
I1
0
Acetonylacetone
CH
I
X=25%
2,5-Dlmethylluran
CHS-C-CH,-CH,-C-CH, 0
HC
-.,A-
H,C
-C /C%
H,C,
/CH
I
NaZSM-5 (SI/Al = 13 000) (Na/AI = 300) T = 300%
II
X=51% S=95%
$
0 3-Methylcyclopentenone
After R.M. Dessau, Zeolites 10, 205 (1990)
Fig. 13. Conversion of acetonylacetone as a valuable test reaction for distinguishing between acid and base catalysis in zeolites after [149]. 2,5-hexanedione (acetonylacetone) as a test reaction which, from the products formed, allows a safe discrimination (Fig. 13). On acid zeolites, e. g. HZSM-5, 2,s-dimethylfuran is formed almost exclusively at medium conversions; base catalysis, on the other hand leads to 3-methylcyclopentenone, again in a very high selectivity a t conversions above 50 %. A broader application of this test reaction could be very helpful in future research on base catalysis in zeolites. Engineering aspects: Coupling of exothermic and endothermic reactions Lower alkenes, i. e., mainly ethene and propene along with some butenes, are currently produced on an industrial scale by steam cracking of petroleum or natural gas derived hydrocarbon fractions, such as light naphtha. Steam cracking is a non-catalytic pyrolysis at high temperatures around 850 "C in the presence of steam. The process is highly endothermic. An alternative route to lower alkenes is the methanol-to-olefin (MTO) process [13]. Ultimately, this process relies on coal or natural gas as raw materials from which methanol
38
Alter S. Nowak el .I.. since 1983
Fig. 14. Production of lower alkenes by coupled methanol/hydrocarbon cracking (CMHC) after [150-152]. X is the conversion and Y denotes product yields. can readily be manufactured via gasification to CO and H, followed by methanol synthesis. MTO is a catalytic process: Methanol is dehydrated on Z S M J type catalysts under such conditions that the yield of C,- to C,-alkenes is maximized. This is usually the case at temperatures in the order of 400 "C. MTO which is not yet a commercial process, is strongly exothermic. Nowak et al. [150-1521 advanced the idea to combine the exothermic MTO reaction and the endothermic hydrocarbon cracking with such a feed ratio that the overall process is nearly thermoneutral (Fig. 14). This is favorable from an engineering point of view since the problems associated with heat removal in MTO and heat generation in steam cracking are avoided. One problem in the coupled rnethunol/liyrlrocurbon cracking (CMHC) is to find an appropriate reaction temperature which is not excessively high for MTO, yet sufficiently high for hydrocarbon cracking. Since the latter reaction which in the non-catalytic steam cracking process proceeds via radicals, is now catalyzed by an acid zeolite (presumably with a mechanism via carbocations), much lower temperatures in the range of 600 to 700 "C seem to suffice. Typical conversions and product yields achieved in CMHC with methanol and n-butane are given in Fig. 14. For the most part, zeolite HZSM-5 was used as catalyst, recent studies [152] revealed, however, that molecular sieves with a lower acid strength, such as iron containing ZSM-5, might offer advantages in CMHC, especially a lower coking and deactivation rate. Zeolites: Hosts for a variety of guests; ship-in-the-bottle catalysts It is evident that in the channels or cages of zeolite molecular sieves, a broad variety of guests with catalytically attractive properties can be accommodated or encapsulated. Indeed, host/guest chemistry is playing an ever increasing role in the preparation of sophisticated
39
zeolite catalysts. Zeolite host/guest chemistry which is, at this time still at its infancy, can be expected to become a thrust area in the nineties. In a few review articles [113,153,154], the potential of zeolite host/guest chemistry for catalysis has been outlined and the different approaches were described. In complete analogy with a ship-in-the-bottle, transition metal complexes which are too bulky to escape through the twelve-membered ring window of faujasite, can be synthesized in its supercages from sufficiently small building blocks which do have access to the cages. The best known examples are phthalocyanine complexes of cobalt, nickel, copper and iron [ 155-1591. For their preparation, the transition metal ion is first exchanged into the zeolite whereupon the complex is synthesized by reaction with 1,2-dicyanobenzene at temperatures around 300 "C. Usually, part of the phthalocyanine complexes are formed at the external surface, but this part can be selectively removed by Soxhlet extraction. Iron phthalocyanine in NaY has also been synthesized from [HFe3(CO),,]- which was first oxidized and then reduced under controlled conditions whereupon 12-dicyanobenzene was admitted [ 1601. Metal phthalocyanines immobilized in zeolite Y have been described to exhibit interesting catalytic properties, e. g., in oxidation reactions with iodosobenzene [ 158,1591: Reactant shape selectivity occurred with cyclohexane/cyclododecane mixtures, n-octane was oxidized in a regioselective manner and stereoselective oxidation of methylcyclohexane and norbornane took place. Recently, an electron donor/acceptor complex (Na+)4 (FePc)@/NaY was prepared from iron phthalocyanine (FePc) encapsulated in NaY by reaction with sodium naphthalene, Na+ (CloH8)-, and this complex gave unusually high ratios of trans-2-butene/ cis-2-butene in the selective hydrogenation of butadiene [ 1601. Another guest which has been synthesized in zeolite Y as a host is cobalt salen [161]. About one Co2+ per two supercages was exchanged into NaY followed by sublimation of the free ligand salen [1,6-bis(2-hydroxyphenyl)-2,5-diaza-l,5-hexadiene]into the zeolite voids. Cobalt salen, a tetradentate chelate has dimensions greater than the window diameter of zeolite Y, hence it is a true ship-in-the bottle. The complex and its pyridine adduct were found to form adducts with dioxygen and can be considered as hemoglobin mimic [161]. Metal clusters entrapped in the intracrystalline voids are of utmost importance in zeolite catalysis. Since the pioneering work of Gallezot et al. [162,163] it is known that the size and location of such clusters depend, to a large extent, on the preparation conditions. The metal can be introduced by ion exchange with its cation or a cationic complex, especially an ammine complex, followed by a controlled thermal treatment including reduction. Metals which are difficult to introduce by ion exchange in aqueous suspension, such as molybdenum, vanadium and others, can often be easily incorporated by solid state ion exchange which has recently found much interest [ 164-1661. An alternative method for preparing highly dispersed metal clusters in molecular sieves is the adsorption and controlled decomposition of volatile metal compounds, mostly carbonyls [167]. In all these methods, metal atoms or small metal oligomers are probably formed as intermediates which migrate rapidly through the zeolite pores and coalesce to form larger agglomerates. Once these particles are larger than the windows of the cages the metal agglomerates find themselves entrapped, e. g., in a supercage
40
of faujasite. In other words, it is the ship-in-the-bottle principle which enables the preparation of highly dispersed metals in zeolites. Similarly, the encapsulation af metal carbonyls in zeolites has found considerable interest because many carbonyls can act as selective catalysts. An example is the ship-in-the-bottle formation of Pd,,(CO), clusters in NaY [168]. While, in the eighties, the potential role of zeolites as hosts for a variety of guests has clearly been recognized, host/guest chemistry has been essentially confined to faujasites. It is likely that, in the nineties, host/guest chemistry will be systematically extended to additional microporous materials, such as AlPO,-8, VPI-5 or the hexagonal counterpart of faujasite, Breck's Structure Six,which can now be synthesized in a pure form using a crown-ether based supramolecule, viz. 18-crown-6,as template [ 1691. Photochemical transformations of guest molecules in zeolites as hosts It is an attractive idea to combine the host/guest principle with photochemistry. Indeed, one can expect that the selectivities of certain photochemical reactions can be altered if they are carried out under the spatial constraints inside the pores of an appropriate zeolite. Examples for shape selective photochemical reactions have already been described in the literature, and one example is summarized in Fig. 15.
it;
Norrish Type II Reaction: C6H5
C6H5
H
R
H
H L
~~~~~
~
E/C In: R=
C,H,
R= C,,H2,
~~~~
Benzene
NaX
1.9 2.7
Alter V. Aemamurlhv el 01..
CBX
NaZSM-5
NaZSM-11
1.5
1.9
>>lo0
>>I00
0.62
2.7
6.8
>> 100
>> 100
0.48
Na-Beta
J. Cham. SOC, Chsm. Commun. 1213 (1989)
Fig. 15. Shape selective photolysis of alkanophenone guests in zeolites as hosts after [170]. The Norrish type I1 reaction, i. e., the photochemical decomposition of alkanophenones was studied in non-acidic forms of zeolites X, ZSM-5, ZSM-11 and Beta [170]. The intermediate 1,4-biradical can stabilize into two directions, viz. by cyclization (C) to a cyclobutanol or elimination (E) to acetophenone and an olefin. In a liquid solvent like benzene, E/C amounts to 1.9 for octanophenone and 2.7 for octadecanophenone, i. e., the elimination is slightly preferred. Essentially the same E/C ratios are observed in zeolite X with its spacious pore system. In the pentads, by contrast, the cyclization was almost completely suppressed, and
41
this was attributed to the restriction of the rotational motion of the central u bond in the 1,4-biradical. In other words and in the language of the zeolite community, this is a photochemical example for restricted transition state shape selectivity [66,67]. Interestingly, zeolite Beta with its intermediate pore width clearly favors cyclization, especially for octadecanophenone with its long alkyl chain, which is not yet understood. Similar photochemical studies with molecules adsorbed in zeolites were carried out by Turro et al. [171,172] and Zimmermann and Zuraw [173]. It has been found, for example, that the photochlorination of n-alkanes in pentad zeolites proceeds with a high selectivity for monochlorination at terminal methyl groups [172]. Conclusion: Zeolites and host/guest chemistry Shape selective photochemistry, in a rigorous sense, is beyond the scope of a review on zeolite catalysis. Nevertheless, the topic was briefly addressed because it clearly demonstrates that through host/guest chemistry, zeolites begin to spread into various new branches of science. Indeed, we foresee an ever increasing importance of the host/guest principle in zeolite chemistry and catalysis. Given the role of zeolites as inorganic hosts, the zeolite community is adviced to watch the current progress in tailor-made organic host molecules, i. e., suprari~olecitlrrr chetnistty [ 174-1801. Enzyme mimics in zeolites, ship-in-the-bottle catalysts and shape selective photochemistry in zeolites are just a few examples which show that zeolite chemistry and supramolecular chemistry are growing together. Acknowledgements The author expresses his gratitude to the following funding institutions which have sponsored his research on catalysis in zeolites: Deutsche Forschungsgemeinschaft, Bundesministerium fur Forschung und Technologie, Fonds der Chemischen Industrie and Max Buchner-Forschungsstiftung. References
J.S. Magee and J.J. Blazek, in J.A. Rabo (Ed.), Zeolite Chemistry and Catalysis; ACS Monograph 171, American Chemical Society, Washington, D.C., 1976, p. 615-679. C.E. Jahnig, H.Z. Martin and D.L. Campbell, in: B.H. Davis and W.P. Rettinger, Jr. (Eds.), Heterogeneous Catalysis- Selected American Histories: ACS Symposium Series, Vol. 222, American Chemical Society, Washin ton, D. C., 1983, pp. 273-292. C.J. Groenenboom, in: H.G. Karge and J. Weitkamp fEds.), Zeolites as Catalysts, Sorbents and Detergent Builders- Applications and Innovations; Studies in Surface Science and Catalysis, Vol. 46, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 99-113. A. Corma, in: P.A. Jacobs and R. van Santen (Eds.), Zeolites: Facts, Figures, Future; Studies in Surface Science and Catalysis, Vol. 49, Part A, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 49-67. A.P. Bolton, in: J.A. Rabo (Ed.), Zeolite Chemistry and Catalysis; ACS Monograph 171, American Chemical Society, Washington, D. C., 1976, p 714-779. J.W. Ward, in: G. Poncelet, P. Grange and P.A. Jacobs (Eds.f Preparation of Catalysts 111; Studies in Surface Science and Catalysis, Vol. 16, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1983, pp. 587-618.
42
[ 71
R.F. Sullivan and J.W. Scott, in: B.H. Davis and W.P. Hettinger, Jr. (Eds.), Heterogeneous Catalysis Selected American Histories; ACS Symposium Series, Vol. 222, American Chemical Society, Washington, D. C., 1983, p .293-313. J.W. Ward, in: B.E. Leach (Ed.), Applied Industrial Catalysis. Vof 3, Academic Press, Orlando, 1984, p 271-392. I.E. Maxwell, Catalysis Foday 1 (1987) 385-413. H.W. Kouwenhoven, in: W.M. Meier and J.B. Uytterhoeven (Eds.), Molecular Sieves; Adv. Chemistry Series, Vol. 121, American Chemical Society, Washington, D. C., 1973, pp. 529-539. W. Hiilderich and E. Gallei, Ger. Chem. Eng. 8 (1985) 337-345. N.Y. Chen and W.E. Garwood, Catal. Rev.-Sci. Eng. 28 (1986) 185-264. N.Y. Chen, W.E. Garwood and F.G. Dwyer, Sha e Selective Catalysis in Industrial Basel, 1989. Applications, 303 p Marcel Dekker, Inc., New C.D. Chang, Catal. kev.-Sci. Eng. 25 (1983) 1-118 K.H. Keim, J. Maziuk and A. Tiinnesmann, Erdo1,'Kohle- Erdgas, Petrochem. 37 (1984) 558-562. N.Y. Chen, W.E. Garwood and R.H. Heck, Ind. Eng. Chem. Res. 26 (1987) 706-711. W.O. Haag and N.Y. Chen, in: L.L. Hegedus (Ed.), Catalyst Design- Progress and Perspectives, John Wiley & Sons, New York, 1987, pp. 163-212. R.N. Bennett, G.J. Elkes and G.J. Wanless, Oil Gas J. 73 (No. 1, Jan. 6, 1975) 69-73. J.D. Hargrove, G.J. Elkes and A.H. Richardson, Oil Gas J. 77 (No. 3, Jan. 15, 1979) 103-105. S.Y. Yanik, E.J. Demmel, A.P. Humphries and R.J. Campagna, Oil Gas J. 83 (No. 19, May 13, 1985) 108-117. P.H. Schipper, F.G. Dwyer, P.T. Sparrell, S. Mizrahi and J.A. Herbst, in: M.L. Occelli (Ed.), Fluid Catalytic Cracking- Role in Modern Refining; ACS Symposium Series, Vol. 375, American Chemical Society, Washington, D. C., 1988, p . 64-86. G. Perego, G. Bellussi, C. Corno, M. Taramasso, F. Buonomo a n f A . Esposito, in: Y. Murakami, A. Iijima and J.W. Ward (Eds.), New Developments in Zeolite Science and Technology; Proc. 7th Intern. Zeolite Conference, Kodansha/Elsevier, Tokyo/Amsterdam, 1986, pp. 129-136. B. Notari, in: P.J. Grobet, W.J. Mortier, E.F. Vansant and G. Schulz-Ekloff (Eds.), Innovation in Zeolite Materials Science; Studies in Surface Science and Catalysis, Vol. 37, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1988, pp. 413-425. S.D. Burd, Jr. and J. Maziuk, Oil Gas J. 70 (No. 27, July 3, 1972) 52-60. Reference 13, pp. 158-165. K.Becker, H. John, K.-H. Steinberg, M. Weber and K.-H. Nestler, in: D. Ka116 and Kh.M. Minnchev (Eds.), Catalysis on Zeolites, AkadCmiai Kiad6, Budapest, 1988, pp. 515-542. J.A. Martens, J. Perez-Pariente and P.A. Jacobs, Proc. Intern. Symp. Zeolite Catalysis, Sibfok, Hungary, May 13-16, 1985, pp. 487-405. J.A. Martens, J. Perez-Pariente and P.A. Jacobs, in: R. Setton (Ed.), Chemical Reactions in Organic and Inorganic Constrained Systems; Nato AS1 Series, D. Reidel Publishing Co., Dordrecht, Boston, Lancaster, Tokyo, 1986, pp. 115-129. P. Ratnasamy, R.N. Bhat, S.K. Pokhriyal, S.G. Hedge and R. Kumar, J. Catal. 119 (1989) 65-70. E. Bourgeat-Lami, P. Massiani, F. di Renzo, F. Fajula and T. des Courieres, Catal. Letters 5 (1990) 265-272. B. Coq, J. Pardillos and F. Figueras, Appl. Catal. 62 (1990) 281-294. P. Massiani, B. Chauvin, F. Fajula and F. Figueras, Appl. Catal. 42 (1988) 105-120. B. Chauvin, V. Rajoafanova, B. Coq, F. Figueras, F. Fajula and T. des Courieres, in: P.A. Jacobs and R.A. van Santen (Eds.), Zeolites: Facts, Figures, Future; Studies in Surface Science and Catalysis, Vol. 49, Part B, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 1397-1406. B. Chauvin, P. Massiani, R. Dutartre, F. Figueras and F. Fajula, Zeolites 10 (1990) 174-182. J.A. Martens, J. Perez-Pariente, E. Sastre, A. Corma and P.A. Jacobs, Appl. Catal. 45 (1988) 85-101. E.G. Derouane and D.J. Vanderveken, Appl. Catal. 45 (1988) L15-L22. I. Manninger, X.X.Lun, P. TCtCnyi and Z. Pahl, Appl. Catal. 51 (1989) L7-Ll1.
-
cork,
[ 141 [ 151
[ 231
[ 201 [ 301
43
591
R. Kumar, G.N. Rao and P. Ratnasamy, in: P.A. Jacobs and R.A. van Santen (Eds.), Zeolites: Facts, Figures, Future; Studies in Surface Science and Catalysis, Vol. 49, Part B, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 1141-1150. S.G. Hegde, R. Kumar, R.N. Bhat and P. Ratnasamy, Zeolites 9 (1989) 231-237. G.N. Rao, R. Kumar and P. Ratnasamy, Appl. Catal. 49 (1989) 307-318. J. Weitkamp, S. Ernst and R. Kumar, Chem.-1ng.-Tech. 59 (1987) 135-137. R. Kumar and P. Ratnasamy, J. Catal. 116 (1989) 440-448. S. Ernst, G.T. Kokotailo, R. Kumar and J. Weitkamp, in: M.J. Phillips and M. Ternan (Eds.), Catalysis: Theory to Practice; Proc. 9th Intern. Congress on Catalysis, Vol. 1, The Chemical Institute of Canada, Ottawa, 1988, pp. 388-395. S. Ernst, J. Weitkamp, J.A. Martens and P.A. Jacobs, Appl. Catal. 48 (1989) 137-148. R. Kumar and P. Ratnasamy, J. Catal. 118 (1989) 68-78. S.L. Lawton and W.J. Rohrbaugh, Nature 247 (1990) 1319-1322. V.J. Frilette, W.O. Haag and R.M. Lago, J. Catal. 67 (1981) 218-222. R.A. Morrison, US Patent 4 686 316, assigned to Mobil Oil C o p , Au . 11, 1987. J.A. Martens, M. Tielen, P.A. Jacobs and J. Weitkamp, Zeolites 4 (1984) 98-107. J.A. Martens and P.A. Jacobs, Zeolites 6 (1986) 334-348. P.A. Jacobs and J.A. Martens, in: Y.Murakami, A. Iijima and J. W. Ward (Eds.), New Developments in Zeolite Science and Technology; Proc. 7th Intern. Zeolite Conference, Kodansha/Elsevier, Tokyo/Amsterdam, 1986, pp. 23-32. J. Weitkamp, S. Ernst and R. Kumar, Appl. Catal. 27 (1986) 207-210. J. Weitkamp, S. Ernst and C.Y. Chen, in: P.A. Jacobs and R.A. van Santen (Eds.), Zeolites: Facts, Figures, Future; Studies in Surface Science and Catalysis, Vol. 49, Part B, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 1115-1129. M. Tielen, M. Geelen and P.A. Jacobs, in: Proc. Intern. Symposium on Zeolite Catalysis, Si6fok, Hungary, May 13-16, 1985, pp. 1-18. B. Kraushaar and J.H.C. van Hooff, Catal. Letters 1 (1988) 81-84. B. Kraushaar-Czarnetzki and J.H.C. van Hooff, Catal. Letters 2 (1989) 43-48. A. Thangaraj, R. Kumar and P. Ratnasamy, Appl. Catal. 57 (1990) L1-L3. E.M. Flanigen, B.M. Lok, R.L. Patton and S.T. Wilson, in: Y. Murakami, A. Iijima and J.W. Ward (Eds.), New Developments in Zeolite Science and Technology; Proc. 7th Intern. Zeolite Conference, Kodansha/Elsevier, Tokyo/Amsterdam, 1986, pp. 103-112. E.M. Flanigen, R.L. Patton and S.T. Wilson, in: P.J. Grobet, W.J. Mortier, E.F. Vansant and G. Schulz-Ekloff (Eds.), Innovation in Zeolite Materials Science; Studies in Surface Science and Catalysis, Vol. 37, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1988, pp. 13-27. J.A. Rabo, R.J. Pellet, P.K. Coughlin and E.S. Shamshoum, in: H.G. Karge and J. Weitkamp (Eds.), Zeolites as Catalysts, Sorbents and Detergent Builders Applications and Innovations; Studies in Surface Science and Catalysis, Vol. 46, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 1-17. M.E. Davis, C. Saldarriaga, C. Montes, J. Garces and C. Crowder, Nature 331 (1988) 698-699. M.E. Davis, C. Montes, P.E. Hathaway, J.P. Arhancet, D.L. Hasha and J. M. Garces, J. Am. Chem. SOC.111(1989) 3919-3924. P.J. Grobet, J.A. Martens, I. Balakrishnan, M. Mertens and P.A. Jacobs, Appl. Catal. 56 (1989) L21-L27. E.G. Derouane, L. Maistriau, Z. Gabelica, A. Tuel, J.B. Nagy and R. von Ballmoos, AppI. Cdtal. 51 (1989) L13-L20. M.E. Davis, P.E. Hathaway and C. Montes, Zeolites 9 (1989) 436-439. R.M. Dessau, J.L. Schlenker and J.B. Hig ins, Zeolites 10 (1990) 522-524. S.M. Csicsery, Pure Appl. Chem. 58 (19865 841-856. J. Weitkamp, S. Ernst, H. Dauns and E. Gallei, Chem.-1ng.-Tech. 58 (1986) 623-632. W.O. Haag, D.H. Olson and P.B. Weisz, in: H. Griinewald (Ed.), Chemistry for the Future; Proc. 29th IUPAC Congress, Cologne, June 5-10, 1983, Pergamon Press, Oxford, 1984, pp. 327-336. G. S. Lee, J. J. Maj, S . C. Rocke and J. M. GarcCs, Catal. Letters 2 (1989) 243-248. J. Weitkamp and M. Neuber, in: Proc. Intern. Symposium on Chemistry of Microporous Crystals, Tokyo, Japan, June 26-29, 1990, accepted for publication. T. Matsuda, K. Yogo, T. Nagaura and E. Kikuchi, in: J. C. Jansen, L. Moscou and M. F. M. Post (Eds.), Zeolites for the Nineties; Recent Research Reports, 8th Intern. Zeolite Conference, Amsterdam, July 10-14, 1989, pp. 431-432.
-
T. Matsuda, K. Yogo, T. Nagaura and E. Kikuchi, J. Japan Petrol. Institute 33 (1990) 214-220. T. Matsuda, K. Yogo, Y. Mogi and E. Kikuchi, Chemistry Letters (1990) 1085-1088. D. Fraenkel, M. Cherniavsky and M. Levy, Proc. 8th Intern. Congr. on Catalysis, Vol. 4, Verlag Chemie, Weinheim, Deerfield Beach, Basel, 1984, pp. 545-554. D. Fraenkel, M. Cherniavsky and M. Levy, J. Catal. 101 (1986) 273-283. D. Fraenkel, J. Molecular Catal. 51 (1989) L1-L7. E.G. Derouane, J. Catal. 100 (1986) 541-544. E.G. Derouane, J.-M. Andre, A.A. Lucas, J. Catal. 110 (1988) 58-73. M. Neuber and J. Weitkamp, in: J.C. Jansen, L. Moscou and M.F. M. Post (Eds.), Zeolites for the Nineties; Recent Research Reports, 8th Intern. Zeolite Conference, Amsterdam, July 10-14, 1989, pp. 425-426. I. Yamaguchi, T. Joh and S. Takahashi, J. Chem. Soc., Chem. Commun. (1986) 1412-1413. H.W. Haynes, Jr., J.F. Parcher and N.E. Helmer, Ind. Eng. Chem. Process Des. Dev. 22 (1983) 401-409. H.B. Mostad, T.U. Riis and O.H. Ellestad, Appl. Catal. 58 (1990) 105-117. R. Iwamoto, K. Nita, S. Hidaka, K. Inarnura and A. Iino, in: K. Tanabe, H. Hattori, T. Yamaguchi and T. Tanaka (Eds.), Acid-Base Catalysis, Kondansha/VCH-Verlagsgesellschaft, Tokyo/Weinheim, 1989, pp.313-3 18. J.A. Maness, Jr. and K.M. Dooley, J. Catal. 117 (1989) 322-334. P.C. Doolan and P.R. Pujado, Hydrocarbon Process. 68 (No. 9, 1989) 72-76. Reference [13], pp. 205-207. Reference [13], pp. 195-198. PChu and G.H. Kiihl, Ind. Eng. Chem. Res. 26 (1987) 365-369. J. Weitkamp, Ind. Eng. Chem. Prod. Res. Dev. 21 (1982) 550-558. J. Weitkamp, W. Gerhardt and P.A. Jacobs, in: Proc. Intern. Symposium on Zeolite Catalysis, Sihfok, Hungary, May 13-16, 1985, pp. 261-270. J. Weitkamp, R. Kumar and S. Ernst, Chem.-1ng.-Tech. 61 (1989) 731-733. S. Ernst, R. Kumar and J. Weitkamp, in: B. Viswanathan and A. Meenakshisundaram (Eds.), Catalysis- Concepts and Applications; Proc. 9th National Symposium on Catalysis, Madras, India, Dec. 15-17, 1988, pp. OR 11-1 to OR 11-11. W.E. Ganvood and P.B. Venuto, J. Catal. 11 (1968) 175-177. F.W. Kirsch, J.D. Potts and D.S. Barmby, Preprints, Div. Petr. Chem., Am. Chem. SOC.13 (No. 1,1968) 153-164. F.W. Kirsch, J.L. Lauer and J.D. Potts, Preprints, Div. Petr. Chem., Am. Chem. Soc. 16 (NO.2, 1971) B-24-B-39. F.W. Kirsch, J.D. Potts and D.S. Barmby, J. Catal. 27 (1972) 142-150. R. Schijllner and 1 I.H6lzel, Journal f. Prakt. Chemie 317 (1975) 694-704. R. Schijllner and H. Holzel, Zeitschr. Chem. 15 (1975) 469-475. Kh.M. Minachev, E.S. Mortikov, S.M. Zen’kovskii, N.V. Mostovoi and N.F. Kononov, Preprints, Div. Petr. Chem., Am. Chem. Soc. 22 (1977) 1020-1024. J. Weitkamp, in: B. Imelik et al. (Eds.), Catalysis by Zeolites; Studies in Surface Science and Catalysis, Vol. 5, Elsevier, Amsterdam, Oxford, New York, 1980, pp. 65-75. J. Weitkamp, in: L.V.C. Rees (Ed.), Proc. 5th Intern. Zeolite Conference, Heyden, London, Philadelphia, Rheine, 1980, pp. 858-865. Y.F. Chu and A.W. Chester, Zeolites 6 (1986) 195-200. P.B. Venuto and P.S. Landis, Adv. Catal. 18 (1968) 259-371. F. Weigert, J. Catal. 103 (1987) 20-29. R.D. Shannon, M. Keane, Jr., L. Abrams, R.1-I. Stnley, T.E. Gier, D.R. Corbin and G.C. Sonnichsen. J. Catal. 113 (1988) 367-382. R.D. Shannon, M. Keane, Jr., L. Abkms, R.H. Staley, T.E. Gier, D.R. Corbin and G.C. Sonnichsen, J. Catal. 114 (1988) 8-16. R.D. Shannon, M. Keane, Jr., L. Abrams, R.H. Staley, T.E. Gier and G.C. Sonnichsen, J. Catal. 115 (1989) 79-85. H.E. Bergna, M. Keane, Jr., D.H. Ralston, G.C. Sonnichsen, L. Abrams and R.D. Shannon, J. Catal. 115 (1989) 148-158. L. Abrams, M. Keane, Jr., and G.C. Sonnichsen, J. Catal. 115 (1989) 410-419. W.Holderich, M. Hesse and F. NBumann, Angew. Chem. 100 (1988) 231-251. H. van Bekkurn and H.W. Kouwenhoven, Recl. Trav. Chim. Pays-Bas 108 (1989) 283-294. .
I
45
K. Smith, Bull. SOC.Chim. France (1989) 272-278. R.F. Parton, J.M. Jacobs, D.R. Huybrechts and P.A. Jacobs, in: H.G. Karge and J. Weitkamp (Eds.), Zeolites as Catalysts, Sorbents and Detergent Builders - Applications and Innovations; Studies in Surface Science and Catalysis, Vol. 46, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 163-192. W.F. Holderich, in: H.G. Karge and J. Weitkamp (Eds.), Zeolites as Catalysts, Sorbents and Detergent Builders Applications and Innovations; Studies in Surface Science and Catalysis, Vol. 46, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1989, pp. 193-209 P. Gallezot, A. Giroir-Fendler and D. Richard, Catal. Letters 5 (1990) 169-174. K. Honna, M. Sugimoto, N. Shimizu and K. Kurisaki, Chemistry Letters (1986) 315-318. G.C. Lau and W.F. Maier, Langmuir 3 (1987) 164-173. J.R. Anderson, Y.F. Chang and R.J. Western, J. Catal. 124 (1990) 259-267. L.A. Paquette, Y. Miyahara and C.W. Doecke, J. Am. Chem. SOC.108 (1986) 1716-1718. W.D. Fessner, B.A.R.C. Murty, J. Worth, D. Hunkler, H. Fritz, H. Prinzbach, W.D. Roth, P. von R. Schleyer, A.B. McEwen and W.F. Maier, Angew. Chem. 99 (1987) 484-486. P.R. Spurr, B.A.R.C. Murty, W.D. Fessner, H. Fritz and H. Prinzbach, Angew. Chem. 99 (1987) 468-488. J.P. Melder, H. Fritz and H. Prinzbach, Angew. Chem. 101 (1989) 309-312. J.P. Melder, R. Pinkos, H. Fritz and H. Prinzbach, Angew. Chem. 101 (1989) 314-319. R. Pinkos, J.P. Melder and H. Prinzbach, Angew. Chem. 102 (1990) 102-104. R.M. Dessau, US Patent 4 554 262, assigned to Mobil Oil Corp., Nov. 19, 1985. M.M.J. Treacy and J.M. Newsam, Nature 332 (1988) 249-251. J.M. Newsam, M.M.J. Treacy, W.T. Koetsier and C.B. de Gruyter, Proc. Royal SOC. London A 420 (1988) 375-405. T. Yashima, K. Sato, T. Hayasaka and N. Hara, J. Catal. 26 (1972) 303-312. M.L. Unland and J.J. Freeman, J. Phys. Chem. 82 (1978) 1036-1040. J.J. Freeman and M.L. Unland, J. Catal. 54 (1978) 183-196. M.D. Sefcik, J. Am. Chem. SOC.101 (1979) 2164-2170. M.L. Unland and G.E. Barker, in: W.R. Moser (Ed.), Catalysis of Organic Reactions; Chemical Industries, Vol. 5, Marcel Dekker, New York, Basel, 1981, pp. 51-71. H. Itoh, T. Hattori, K. Suzuki and Y. Murakami, J. Catal. 79 (1983) 21-33. C. Lacroix, A. Deluzarche, A. Kiennemann and A. Boyer, J. Chimie Physique 81 (1984) 481-490. C. Lacroix, A. Deluzarche, A. Kiennemann and A. Boyer, Zeolites 4 (1984) 109-111. D. Barthomeuf, J. Phys. Chem. 88 (1984) 42-45. J.M. Garces, G.E. Vrieland, S.I. Bates and F.M. Scheidt, in: B. Imelik et al. (Eds.), Catalysis by Acids and Bases; Studies in Surface Science and Catalysis, Vol. 20, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1985, pp. 67-74. L.R.M. Martens, P.J. Grobet and P.A. Jacobs, Nature 315 (1985) 568-570. L.R.M. Martens, P.J. Grobet, W.J.M. Vermeiren and P.A. Jacobs, in: Y. Murakami, A. Iijima and J.W. Ward (Eds.), New Developments in Zeolite Science and Technology; Proc. 7th Intern. Zeolite Conference, Kodansha/Elsevier, Tokyo/Amsterdam, 1986, pp. 935-941. L.R.M. Martens, W.J.M. Verrneiren, P.J. Grobet and P.A. Jacobs, in: B. Delmon, P. Grange, P.A. Jacobs and G. Poncelet (Eds.), Preparation of Catalysts IV; Studies in Surface Science and Catalysis, Vol. 31, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1987, pp. 531-542. L.R.M. Martens, W.J. Vermeiren, D.R. Huybrechts, P.J. Grobet and P.A. Jacobs, in: M.J. Phillips and M. Ternan (Eds.), Catalysis: Theory to Practice; Proc. 9th Intern. Congress on Catalysis, Vol. 1, The Chemical Institute of Canada, Ottawa, 1988, pp. 420-428. M. Onaka, K. Sugita and Y. Izumi, J. Org. Chem. 54 (1989) 1116-1123. P.E. Hathaway and M.E. Davis, in: J.C. Jansen, L. Moscou and M.F.M. Post (Eds.), Zeolites for the Nineties; Recent Research Reports, 8th Intern. Zeolite Conference, Amsterdam, July 10-14, 1989, pp. 455-456. P.E. Hathaway and M.E. Davis, J. Catal. 116 1989 263-278. P.E. Hathaway and M.E. Davis, J. Catal. 116 119891279-284.
-
P.E. Hathaway and M.E. Davis, J. Catal. 119 (1989) 497-507. A. Dyer, S.A. Malik, A. Araya and T.J. McConville, in: P.A. Williams and J. Hudson (Eds.), Recent Developments in Ion Exchange; Elsevier Applied Science, London, New York, 1987, pp. 257-263. R.M. Dessau, in: J.C. Jansen, L. Moscou and M.F.M. Post (Eds.), Zeolites for the Nineties; Recent Research Reports, 8th Intern. Zeolite Conference, Amsterdam, July 10-14, 1989, p .457-458. R.M. Dessau, Zeohes 10 (1990) 205-206. S. Nowak, H. Giinschel, A. Martin, K. Anders and B. Lucke, in: M.J. Phillips and M. Ternan (Eds.), Catalysis: Theory to Practice; Proc. 9th Intern. Congress on Catalysis, Vol. 4, The Chemical Institute of Canada, Ottawa, 1988, pp. 1735-1742. A. Martin, S. Nowak, B. Liicke and H. Giinschel, Appl. Catal. 50 (1989) 149-155. A. Martin, S. Nowak, B. Liicke, W. Wieker and B. Fahlke, Appl. Catal. 57 (1990) 203-214. N. Herron, Chemtech (1989) 542-548. G.A. Ozin and C. Gil, Chem. Rev. 89 (1989) 1749-1764. G. Meyer, D. Wiihrle, M. Mohl and G. Schulz-Ekloff, Zeolites 4 (1984), 30-34. E.S. Shpiro, G.V. Antoshin, O.P. Tkachenko, S.V. Gudkov, B.V. Romanovskyand Kh. M. Minachev, in: P.A. Jacobs, N.T. Jiiger, P. Jiru, V.R. Kazansky and G. Schulz-Ekloff (Eds.), Structure and Reactivity of Modified Zeolites: Studies in Surface Science and Catalysis, Vol. 18, Elsevier, Amsterdam, Oxford, New York, Tokyo, 1984, pp. 31-39. B.V. Romanovsky, in: Proc. Intern. Symposium on Zeolite Catalysis, Sibfok, Hungary, May 13-16, 1985, pp. 215-222. N. Herron, G.D. Stucky and C.A. Tolman, J. Chem. SOC.,Chem. Commun. (1986) 1521-1522. C.A. Tolman and N. Herron, Preprints, Div. Petr. Chem., Am. Chem. SOC.32 (1987) 798-805. T. Kimura, A. Fukuoka and M. Ichikawa, Catal. Letters 4 (1990) 279-286. N. Herron, Inorg. Chem. 25 (1986) 4714-4717. P. Gallezot, Catal. Rev.-Sci. Eng. 20 (1979) 121-154. P. Gallezot, in: B. Imelik et al. (Eds.), Catalysis by Zeolites; Studies in Surface Science and Catalysis, Vol. 5, Elsevier, Amsterdam, Oxford, New York, 1980, pp. 227-234. P.E. Dai and J.H. Lunsford, J. Catal. 64 (1980) 173-183. A.V. Kucherov and A.A. Slinkin, Zeolites 7 (1987) 38-42. H.G. Karge, H.K. Beyer and G. Borbdy, Catalysis Today 3 (1988) 41-52. R.F. Howe, J. Ming, W. She-Tin and Z. Jian-Him, Catal. Today 6 (1989) 113-122. L.L. Sheu, H. Kniizinger and W.M.H. Sachtler, Catal. Letters 2 (1989) 129-138. F. Delprato, L. Delmotte, J.L. Guth and L. Huve, Zeolites 10 (1990) 546-552. V. Ramamurthy, D.R. Corbin and D.F. Eaton, J. Chern. SOC.,Chem. Commun. (1989) 1213-1214. S.C. Stinson, Chern. Eng. News 66 (No. 26, June 27, 1988) 27-30. N.J. Turro, J.R. Fehlner, D.P. Hessler, K.M. Welsh, W. Ruderman, D. Firnberg and A.M. Braun, J. Org. Chern. 53 (1988) 3731-3735. H.E. Zimmerrnann and M.J. Zuraw, J. Am. Chem. Soc. 111 (1989) 7974-7989. I. Tabushi and Y. Kuroda, Adv. Catal. 32 (1983) 417-466. F. Viigtle and E. Weber (Eds.), Host Guest Complex Chemistry, Macrocycles, 421 pp., Springer-Verlag, Berlin, Heidelber , New York, Tokyo, 1985. J. Franke and F. Vogtle, Angew. Chem. 97 t1985) 224-225 F. Ebmeyer and F. Vogtle, Angew. Chem. 101 (1989) 95-96. E. Weber and F. Vogtle, Chemie in unserer Zeit 23 (1989) 210-212. F.H. Kohnke, J.P. Mathias and J.F. Stoddart, Angew. Chem. Adv. Mater. 101 (1989) 1129-1134. M.E. Tanner, C.B. Knobler and D.C. Cram, J. Am. Chem. Soc. 112 (1990) 1659-1660.
47
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
ZEOLITES AS CATALYSTS FOR ALKANE OXIDATIONS R.F.
P art on, D.R.C.
Huybrechts, Ph. Buskens and P.A. Jacobs
K.U. Leuven, Dept. B i o t e c h n i s c h e Wetenschappen, Oppervlaktechemie, K a r d i n a a l M e r c i e r l a a n 92, 8-3030 Belgium
Laborat orium voor Heverlee (Leuven),
SUMMARY The o x i d a t i o n o f u n a c t i v a t e d alkanes was s t u d i e d on z e o l i t e encaged p h t h a l l o c y a n i n e s w i t h t e r t i a r y b u t y l hydroperoxide and on t i t a n i u m s i l i c a l i t e w i t h aqueous hydrogen p e r o x i d e as o x i d a n t . On b o t h t y p e s o f c a t a l y s t s secondary and/or t e r t i a r y a l c o h o l s and ketones a r e formed. The o x y f u n c t i o n a l i z a t i o n occurs i n a r e g i o s e l e c t i v e way on t h e o u t e r carbon atoms o f t h e hydrocarbon c h a i n , t h u s r e f l e c t i n g t h e shape s e l e c t i v e properties o f the catalysts. INTRODUCTION The ma jo r
industrial
applications
of
zeolites
in
refinery
and
pet roc hemic a l o p e r a t i o n s a r e r e l a t e d t o r e a c t i o n s where t h e z e o l i t e i s used as
an
acid
isomerization,
or
bifunctional
alkylation,
catalyst,
etc.
e.g.
cracking,
[l]. Th eref ore,
hydrocracking,
up t o now most o f t h e
fundamental work i n t h e f i e l d o f z e o l i t e s has been c o n c e n t r a t e d on t h e i r a c i d o r metal s t a b i l i z i n g p r o p e r t i e s . On t h e c o n t r a r y ,
relatively l i t t l e
e f f o r t s have been done t o e x p l o i t t h e u n i q ue f e a t u r e s o f z e o l i t e c a t a l y s t s , such as shape s e l e c t i v i t y , i n o t h e r t y p e s o f r e a c t i o n s [ 3-51. One o f t h e l e a s t studied z e o l i t e catalyzed reactions i s t h e o x y f u n c t i o n a l i z a t i o n o f alkanes i n t h e l i q u i d phase. N e v e r t h e l e s s , t h e g r e a t abundance o f alkanes makes them t o one o f t h e g r e a t e s t r e s o u rces f o r t h e energy and f o r t h e chemical
industry.
The extreme
low r e a c t i v i t y o f
alkanes makes
their
a c t i v a t i o n and f u n c t i o n a l i z a t i o n a c h a l l e n g e f o r many chemical s c i e n t i s t s . D u r i n g t h e l a s t 10 y e a r s many e f f o r t s have been done i n t h i s area u s i n g homogeneous c a t a l y s t s [ Z ] .
Therefore, m a j or c o n t r i b u t i o n s o f r e s e a r c h e r s i n
the f i e l d o f z e o l i t e catalyzed reactions f o r t h e f u n c t i o n a l i z a t i o n
of
alkanes cannot h o l d o f f . I n t h i s work a l k a nes a r e o x i d i z e d by a p e r o x i d i c oxygen
source
(ROOH)
to
the
c o r r e s p o n ding
maintenance o f t h e carbon c h a i n s t r u c t u r e ,
alcohols
and
ketones
with
according t o t h e f o l l o w i n g
genera l r e a c t i o n scheme: CnH(2n+2)
+
ROOH
-
ROH
+
H20
+
CnH(2,,+qOH
( + CnHZnO)
48
The behav i o r o f two new t y p e s o f c a t a l y s t w i l l be discussed, namely p h t h a l l o c y a n i n e c o n t a i n i n g l a r g e - p o r e z e o l i t e s and t i t a n i u m s i l i c a l i t e 1 (TS-1). The f o r m e r one uses t - b u t y l h y d r o p e r o x i d e as o x i d a n t , aqueous hydrogen p e r o x i d e i s t h e oxygen s o urce f o r t h e l a t t e r one.
whereas
ALKANE OXIDATIONS ON ZEOLITE ENCAGED PHTHALLOCYANINES
Introduction M e t a l l o - p o r p h y r i n e s and p h t h a l l o c y a n i n e s a r e a c t i v e c a t a l y s t s f o r t h e s e l e c t i v e e p o x i d a t i o n o f alkenes and h y d r o x y l a t i o n o f alkanes and many o t h e r o x i d a t i o n r e a c t i o n s [6-81. They r e p r e s e n t models f o r t h e a c t i v e s i t e o f cytochrome P-450.
However,
t h e y l a c k a p r o p e r mimic o f t h e p r o t e i n
function.
Metallo-phthallocyanines encaged i n z e o l i t e s Y have been proposed as complete enzyme mimics [9-131.
Z e o l i t e s can r e p l a c e t h e p r o t e i n p o r t i o n o f
n a t u r a l enzymes and m o d i f y t h e r e a c t i v i t y i n t h e same way as enzymes do by imposing s t e r i c c o n s t r a i n t s on t h e environment o f t h e a c t i v e complex. I n t h e p r e s e n t work, t h e c o n s t r u c t i o n o f a mimic o f cytochrome P-450 is
at t emp t e d
supercages
by
in
situ
o f zeolite Y
synthesis
of
iron-phthallocyanines
and i n t h e channels
o f VPI-5.
in
the
I t s catalytic
a c t i v i t y and s e l e c t i v i t y i s t e s t e d i n t h e o x y f u n c t i o n a l i z a t i o n o f n-a1 kanes w i t h t e r t i a r y b u t y l hydroperoxide. ExDerimental Materials A sample o f NaY w i t h a Si/A1 r a t i o o f 2.46 was purchased f r o m Ventron. VPI-5 was s y n t h e s i z e d a c c o r d i n g t o a r e c e n t l i t e r a t u r e procedure [14]. Dicyanobenzene
(DCB)
hy dro pero x ide (t.BHP) hexane,
n-heptane,
(+98%), f e r r o c e n e
(FeCp2)
(+98%),
1,2-
t e r t i a i r y butyl
(70% i n w a t e r ) , acetone, dichloromet hane (CH2C12), n n-octane,
n-nonane
and
n-decane
(all
+99%)
were
purchased f rom A l d r i c h . H e - p h t h a l l o c y a n i n e (H2Pc) (+98%) was purchased f rom A l d r i c h and F e - p h t h a l l o c y a n i n e (FePc) (+98%) f r o m Strem Chemicals. F erric enium - Y i s p r e p a r e d by adding 5 g o f a i r - d r y NaY t o 50 m l o f a s o l u t i o n o f f e r r o c e n e i n acetone c o n t a i n i n g 84 mg (=1 complex p e r u n i t c e l l ) o f FeCp2, f o l l o w e d b y a i r d r y i n g a t 343 K. The l a t t e r s o l i d i s mixed w i t h 5 g o f DCB and h e a t e d under He atmosphere a t 423 K f o r 4 hours. T h i s khaki -coloured
sol i d
is
succesively
p y r i d i n e and a g a i n w i t h acetone,
soxhlet
extracted
with
acetone,
u n t i l a c o l o u r l e s s e x t r a c t i s obt ained.
The f i n a l c a t a l y s t i s a i r d r i e d a t 343 K. FePcVPI-5 i s s y n t h e s i z e d a t 523 i n s t e a d o f 423 K a c c o r d i n g t o t h e same procedure.
49
Characteri sat i o n I.R.
c h a r a c t e r i s a t i o n o f t h e samples i s c a r r i e d o u t u s i n g t h e K6r
t ec hnique.
spectroscopy
U.V.
was
used
for
the
semi-quantitative
d e t e r m i n a t i o n o f t h e amount o f i n t r a c r y s t a l l i n e p h t h a l l o c y a n i n e s ,
after
d i s s o l u t i o n o f t h e z e o l i t e i n c o n c e n t r a t e d s u l f u r i c a c i d (0.1 g o f c a t a l y s t i n 10 m l o f c on c e n t r a t e d H2S04 f o r 4 h ) . The i r o n c o n t e n t o f t h e samples i s determined by chemical a n a l y s i s . X - r a y powder d i f f r a c t i o n i s used t o ensure good c r y s t a l l i n i t y o f t h e z e o l i t e a f t e r t h e s y n t h e s i s and p u r i f i c a t i o n procedures. R eac t io n Drocedure O x y f u n c t i o n a l i z a t i o n r e a c t i o n s o f n - a lkanes a r e c a r r i e d o u t a t room temperature and atmospheric p r e s s u r e w i t h t.BHP as oxidans and acetone as s o l v e n t . Product a n a l y s i s was done w i t h GC on a 50 m CP S i l - 8 6 c a p i l l a r y column f rom Chrompack. R e s u l t s and d i s c u s s i o n C h a r a c t e r i s a t i o n o f encaqed FePc I n spite o f d i f f e r e n t conditions,
such as t h e r e s p e c t i v e s y n t h e s i s
c o n d i t i o n s mentioned above, a r e a p p l i e d f o r FePc and FeCp2 impregnat ion on NaY and VPI-5
is
the application
o f the
st andard s o x h l e t
extraction
procedure s t i l l s u f f i c i e n t t o remove a l l i r o n f rom t h e s o l i d s . When i n s i t u synthesis
of
FePc
is
made
in
both
molecular
sieve
structures,
the
e x t r a c t i o n procedure removes o n l y p a r t o f t h e i r o n . Thus a l l r e s i d u a l i r o n p r e s e n t i s ' a s s o c i a t e d w i t h encaged FePc.
Indeed,
after extraction the
c h a r a c t e r i s t i c l i n e s o f Cp2 (814 c m - l ) and DC6 (965 c m - l ) a r e absent as shown by I . R . d a t a [ 1 5 ] . Chemical a n a l y s i s combined w i t h U.V.
spectroscopy o f samples d i s s o l v e d
i n c onc ent rat e d s u l f u r i c a c i d a l l o w t o d et ermine t h e amount o f FePc and H2Pc pre s ent
i n t h e molecular sieves.
The NaY samples
used f o r
the
c a t a l y t i c experiments have about 1 FePc f o r e v e r y 77 supercages. I n case o f VPI-5 samples t h e amount o f i n t r a c r y s t a l l i n e FePc corresponds t o 0.021 f o r each u n i t c e l l
.
M o l e c u l a r g r a p h i c s a n a l y s i s o f FePc l o c a t e d i n t h e supercages o f NaY z e o l i t e s shows a s a d d l e - t y p e d e f o r m a t i o n o f t h e p l a n a r FePc molecule which det e rmines t h e f r e e a p e r t u r e s t o t h e a c t i v e s i t e [15]. I t c o n t r a s t s w i t h FePc encaged i n VPI-5 where no d e f o r m a t i o n o f t h e c h e l a t e occurs. C a t a l y t i c a c t i v i t v w i t h FePcY Zeolite-enclosed
FePc
are
suitable
catalysts
for
the
selective
o x i d a t i o n o f p a r a f f i n e s t o ketones and a l c o h o l s a t ambient t emperat ure and pre s s ure w i t h t.BHP as oxygen atom donor. O x i d a t i o n r e a c t i o n s a r e performed
50
b o t h i n a i r and i n n i t r o g e n atmosphere w i t h o u t s i g n i f i c a n t c o n v e r s i o n and p r o d u c t d i s t r i b u t i o n .
Fe3+-Y,
ferricenium-Y,
as w e l l as H2Pc impregnated on NaY a r e i n a c t i v e ,
changes i n
f errocenium-Y
i n d i c a t i n g t h a t t h e low
amounts o f FePc p r e s e n t i n t h e m o l e c u l a r s i e v e s a r e t h e a c t i v e s i t e s . F i g . 1 shows t h e c o n v e r s i o n o f a s e r i e s o f n - a l k a n e s w i t h d i f f e r e n t c h a i n l e n g t h on FePcY and FePcVPI-5 c a t a l y s t s a t a c o n s t a n t r a t e o f t.BHP addition.
I n FePcY
initial
reaction stoichiometry.
conversions
approach
limit
the
imposed
by
Upon f u r t h e r a d d i t i o n o f oxidans t h e c o n v e r s i o n
curve g r a d u a l l y deviates from t h e t h e o r e t i c a l l i n e ,
pointing t o catalyst
d e a c t i v a t i o n . Turnover numbers (N) p e r FePc molecule amount t o 6,000
and
exceed t h o s e o f homogeneous FePc (N r a n g i n g f rom 20 t o 30) [15] as w e l l as t h o s e r e p o r t e d i n t h e l i t e r a t u r e on FePcY c a t a l y s t s w i t h iodosobenzene as ox idans (N=5.6) dichloromethane,
[9, 121. Because o f t h e l o w a c t i v t y o f FePc, d i s s o l v e d i n it
is
not
shown
in
Fig.
1.
Initial
activities
are
presumably h i g h on homogeneous FePc c a t a l y s t s b u t t h e complexes a r e r a p i d l y d e s t r o y e d as shown by b l e a c h i n g o f t h e s o l u t i o n . FePcY t u r n s y e l l o w - b r o w n a f t e r r e l a t i v e h i g h t u r n o v e r numbers i n c o n t r a s t w i t h
FePcVPI-5 which
remains a c t i v e even a f t e r t u r n o v e r numbers over 2,000. The l a t t e r c a t a l y s t does n o t undergo any c o l o u r change a t a l l and can be reused. O bviously, t h e stability
and t h e r e f o r e
the
overall
conversion
is
improved when
the
complexes a r e encaged i n t h e m o l e c u l a r s i e v e s . It i s c l e a r t h a t t h e z e o l i t e p r o t e c t s t h e p h t h a l l o c y a n i n e s t r u c t u r e a g a i n s t o x i d a t i o n compared t o t h e homogeneous case, j u s t as t h e p r o t e i n m a n t l e does i n t h e enzyme cytochrome P-450.
However,
even a f t e r
correction f o r differences
i n experimental
c o n d i t i o n s , FePcVPI-5 remains l e s s a c t i v e t han FePcY. I n t h e f o r m e r system, f r e e t.BHP
accumulates g r a d u a l l y w h i l e w i t h FePcY a l l
t.BHP
added i s
consumed a t once. The l i n e a r i n c r e a s e i n c o n v e r s i o n observed w i t h FePcVPI5, suggests t h e absence o f c a t a l y s t d e a c t i v a t i o n , a t l e a s t up t o t u r n o v e r numbers o f 2,000. A t t h i s st a g e o f t h e work, t w o p o s s i b l e e x p l a n a t i o n s can be advanced t o r a t i o n a l i z e t h e h i g h e r a c t i v i t y o f FePcY and t h e s u p e r i o r s t a b i l i t y o f FePcVPI-5. As a r e s u l t o f t h e t u b u l a r n a t u r e o f t h e pores o f VPI-5 and ev ent u al d i f f u s i o n l i m i t a t i o n s o f t h e r e a c t i o n , o n l y t h e FePc complexes a t the external r i m o f the crystals are active i n i t i a l l y but are gradually consumed d u r i n g r e a c t i o n . Consequently, FePc complexes 1 ocat ed more towards t h e c e n t r e o f t h e c r y s t a l s become a c t i v e . Secondly, that
the
s ad d l e - c o n f o r m a t i o n
of
environment o f t h e a c t i v e i r o n ,
FePc
in
NaY
i t can be s p e c u l a t e d
changes
the
electronic
thus increasing not o n l y i t s c a t a l y t i c
a c t i v i t y b u t a l s o i t s v u l n e r a b i l i t y towards s e l f - o x i d a t i o n . The y i e l d o f a l c o h o l s and ketones on p e r o x i d e b a s i s on FePcY and FePcVPI-5,
r e s p e c t i v e l y t h e FePc c a t a l y s t s
i s 75% and l e s s t han lo%,
51
r e s p e c t i v e l y . The m a j o r s i d e r e a c t i o n i s t h e decomposition o f t h e o r g a n i c peroxide t o
molecul a r oxygen and t. b u t a n o l .
F i g . 1 a l s o shows a f a s t e r d e a c t i v a t i o n r e a c t i o n f o r an alkane when t h e c h a i n l e n g t h i n c r e a s e s f r o m 6 t o 10 carbon atoms. The reason f o r t h i s behav iour i s n o t c l e a r . 15 .~
5,
h
FePcY
3
v
10
0
octane FePcY
u
decane FePcY
C 0
. I
rA h
w
b
C
0
u
~
5
1 0 0 % ketone formation
octane FePcVPI- 5
0 0.0
0.5
1 .o
1.5
Ratio of t.BHP to alkane (mol/mol) F i g . 1. O x i d a t i o n c o n v e r s i o n o f n - a l k a nes w i t h t . B H P t o ketones and a l c o h o l s a t 298 K and 0 . 1 MPa. Wi t h FePcY 0 . 5 g o f c a t a l y s t (lFePc/77 supercages), 200 mmol o f n-alkane, 100 m l o f acetone and a t.BHP i n j e c t i o n r a t e o f 21.9 mmo1.h-' i s used. W i t h FePcVPI-5 0.1 g o f c a t a l y s t (0.021 p e r u n i t c e l l ) , 50 mmol o f n-C8, 50 m l o f acetone and a t.BHP i n j e c t i o n r a t e o f 8.76 mmo1.h-l i s used. The s t r a i g h t l i n e r e p r e s e n t s t h e maximum p o s s i b l e c onv ers io n when o n l y ketones a r e formed and t.BHP c o n v e r s i o n i s complete. Fig.
2A shows f o r n - o c t a n e o x i d a t i o n on FePcY,
FePcVPI-5 and FePc
c a t a l y s t s t h e d i s t r i b u t i o n o f ketones and a l c o h o l s . A l l t hese c a t a l y s t s e x h i b i t a h i g h s e l e c t i v i t y towards t h e f o r m a t i o n o f ketones. Encaged FePc complexes a r e more s e l e c t i v e f o r ketones t han t h e homogenous ones. S i m i l a r behav ior i s a l s o e s t a b l i s h e d by Herron e t a1 [ 9 ] and t h e y a t t r i b u t e i t t o d i f f u s i o n e f f e c t s i n t h e m o l e c u l a r s i e v e s.
The ketone t o a l c o h o l r a t i o
which ranges f r o m about 4 t o 9, i s t y p i c a l f o r f r e e r a d i c a l c h a i n o x i d a t i o n r e a c t i o n s and s i m i l a r t o what i s r e p o r t e d by Fontecave and Mansuy [ 16] f o r t h e system MnTPP(C1 )/02/ascorbate. They a t t r i b u t e t h e f o r m a t i o n o f ketones t o t h e involvement o f f r e e r a d i c a l s . A l s o t h e oxygen f o r m a t i o n i n d i c a t e s that
r e a c t i o n s o f r a d i c a l a r n a t u r e a r e i n v o l v e d i n t h e mechanism.
It
suggests t h a t t h e a c t i v e s i t e i s n o t an "oxenoi'd" species as i n enzymes [17] and t h e i r mimics [6-81, b u t t h a t t h e r e a c t i o n r a t h e r occurs v i a a f r e e r a d i c a l c h a i n mechanism. T h i s c h a i n mechanism seems t o be d i s t u r b e d by t h e z e o l i t e framework where t h e r e a c t i o n y i e l d on a p e r o x i d e b a s i s exceeds t h a t
52
o f f r e e FePc by a f a c t o r of
10. However,
Haber e t a l .
[ 18]
using the
oxidation of cyclohexane with hydrogen peroxide over c h l orotetratolylporphyrinatochromium( 111), o b t a i n e d under comparable c o n d i t i o n s k et o ne t o a l c o h o l r a t i o s between 1.4 and 4, and claimed t h a t t h e r e a c t i o n s were n o t r a d i c a l a r i n n a t u r e . Another e x p l a n a t i o n i s t h a t d i f f e r e n t k i n d o f r e a c t i o n s a r e i n v o l v e d i n t h e f o r m a t i o n o f p r o d u c t s . Indeed,
i t i s shown
[19] t h a t t h e i r o n - c a t a l y z e d t.BHP o x i d a t i o n o f cyclohexane proceeds w i t h different
parallel
r e a c t i o n pathways.
formation o f t.butoxy
However,
the total
s u b s t i t u e n t s on t h e a l k y l chain’,
lack o f
the
i n our results,
c o n t r a s t s w i t h a f r e e r a d i c a l mechanism.
A ketone
C2lC3
alcohol
100,
C2IC4
I
FePcY
FcPcVPI- 5
Catalyst
FePc
FcPcY
PePcVPI- 5
PCPC
Catalyst
2. S e l e c t i v i t y f o r k e t o n e and a l c o h o l formation (A) and r e g i o s e l e c t i v i t y (B) ;n t h e o x i d a t i o n o f n-oct ane a t l o w c o n v e r s i o n s o v e r FePcY, FePcVPI-5 and FePc c a t a l y s t s . C o n d i t i o n s a r e t h o s e o f F i g . 1. On homogeneous FePc no r e g i o s e l e c t i v i t y i s observed as t h e C2/C3 and C2/(C4tC5)
r a t i o s a r e a l l around 1. F i g . 28 i n d i c a t e s t h a t r e g i o s e l e c t i v i t y
e x i s t s f o r b o t h m o l e c u l a r s i e v e s , p o s s i b l y due t o t h e encaged n a t u r e o f t h e complex. However, l o w e r v a l u e s o f t h e C2/C3 and C2/C4 r a t i o s a r e o b t a i n e d i n VPI-5 compared t o z e o l i t e Y, p o i n t i n g t o t h e e x i s t e n c e o f shape s e l e c t i v i t y . The m o l e c u l a r g r a p h i c s a n a l y s i s enables q u a n t i f i c a t i o n o f t h e f r e e pore a p e r t u r e s , which f o r b o t h m o l e c u l a r s i e v e s a r e about 0.6 nm [ 15] . I t shows t h a t t h e d i f f e r e n c e i n s e l e c t i v i t y can h a r d l y be caused by
d i f f e r e n c e s i n t h e z e o l i t i c environment. The enhanced c o n s t r a i n t observed f o r FePcY s hou l d t h e n be r e l a t e d t o t h e s a d d l e - t y p e d e f o r m a t i o n o f t h e complex. R e g i o s e l e c t i v i t y i n t h e a l c o h o l f r a c t i o n i s i n s i g n i f i c a n t .
53 TABLE 1 R e g i o s e l e c t i v i t y i n t h e ketone f r a c t i o n f o r t h e o x i d a t i o n o f C6 t o C10 n a1 kanes on FePcY c a t a l y s t a expressed as s t a n d a r i z e d molar r a t i o s b o f C2/C3 and C2/ (C4tC5)
.
Substrate
Regi osel e c t i v i t y C Ketone f r a c t i o n C2/C3 C2/(C4tC5)
n - hexane n - heptane n -o c t a ne n-nonane n-decane
0.92 1.15 1.29 1.38 1.56
a; b; c;
1.51 1.95 2.80 2.59
R eac t io n c o n d i t i o n s : i n a m i c r o r e a c t o r o f 3 m l w i t h 2.4 mmol o f t.BHP, 6 mmol o f p a r a f f i n , 0.1 g o f FePcY and 1.5 m l o f acetone and a t 298 K, 0.1 MPa, 5 % c o n v e r s i o n . C o r r e c t i o n s a r e made so as t o have an equal number o f C2 and (C4+C5) p o s i t i o n s i n t h e n-alkane chain, i r r e s p e c t i v e o f i t s chain length. R a t i o o f o x y g e n a t i o n a t C p o s i t i o n 2 over oxygenat ion a t i n n e r C positions. T a ble 1 shows t h a t f o r n-hexane as s u b s t r a t e no r e g i o s e l e c t i v i t y i s
observed.
However,
f o r l o n g e r c h a i n s t h e z e o l i t e enclosed FePc e x e r t s a
shape s e l e c t i v e e f f e c t o f i n c r e a s i n g i n t e n s i t y on t h e i n s e r t i o n o f oxygen. Carbon atoms a t p o s i t i o n 0 - 1 a r e p r e f e r e n t i a l l y o x i d i z e d o v e r carbon atoms a t p o s i t i o n w-2, which i n t u r n a r e o x i d i z e d f a s t e r t h a n t h o s e a t p o s i t i o n 4 and 5. Thus t h e r e g i o s e l e c t i v i t y i n c r e a s es w i t h c h a i n l e n g t h , o b v i o u s l y , because l o n g e r a l k y l c h a i n s a r e more l i a b l e t o s t e r i c c o n s t r a i n t s e x e r t e d by t h e z e o l i t e framework a t t h e l e v e l o f t h e a c t i v e s i t e . Conclusions I r o n - p h t h a l l o c y a n i n e s a r e encaged i n m o l e c u l a r s i e v e s by an i n s i t u s y n t h e s i s f rom ferrocenium-Y and f e r r i c e n i u m-VPI -5.
The a c t i v i t y o f i r o n -
p h t h a l l o c y a n i n e s i n z e o l i t e Y i s h i g h e r t h a n i n t h e VPI-5 c a t a l y s t which i n t u r n exceeds t h o s e o f t h e f r e e complexes as shown by t u r n o v e r numbers o f respectively,
6,000,
2,000 and 25. However,
iron-phthallocyanine-VPI-5 i s
more s t a b l e t h a n t h e i r o n - p h t h a l l o c y a n i n e - Y c a t a l y s t , which t u r n s i n a c t i v e a f t e r t u r n o v e r numbers h i g h e r t h a n 6,000, and t h e f r e e complex s u f f e r s f rom v e r y f a s t o x i d a t i v e d e s t r u c t i o n . Shape s e l e c t i v i t y i n t h e r e g i o s e l e c t i v e o x i d a t i o n o f alkanes i s h i g h e r on FePcY t h a n on FePcVPI-5, which i n t u r n i s higher
t h an
explained framework
by
on
the
non-selective
a combined e f f e c t
of
phthallocyanine.
the
molecular
homogeneous
of
sieve
steric and
catalyst.
constraint the
This
can
imposed by
deformation
of
be the the
The shape s e l e c t i v e o x i d a t i o n o f t h e secondary carbon
atoms o f normal alkanes i n c r e a s e s w i t h t h e c h a i n l e n g t h .
54
ALKANE OXIDATIONS ON TITANIUM SILICALITE Introduction T i t a n i u m s i l i c a l i t e - 1 o r TS-1 i s a t i t a n i u m c o n t a i n i n g d e r i v a t i v e o f s i l i c a l i t e - 1 , w h i c h i n a s e l e c t i v e way c a t a l y z e s t h e o x i d a t i o n o f o r g a n i c s u b s t r a t e s w i t h d i l u t e d hydrogen p e r o x i d e as o x i d a n t [20-231. The r e a c t i o n s
i n t h e p a t e n t 1 i t e r a t u r e i n c l u d e t h e (mono)epoxidat ion o f ( d i ) o l e f i n s [24-251, t h e c o n v e r s i o n o f o l e f i n s t o g l y c o l monomethyl e t h e r s
described
i n t h e presence o f methanol [26],
t h e h y d r o x y l a t i o n o f aromat ics [27], t h e
o x i d a t i o n o f p r i m a r y o r secondary a l c o h o l s t o aldehydes o r ket ones [28], t h e ammoxidation o f ketones i n t h e presence o f NH3 [29], o f v in y lbenz ene s t o 8 - p h e n y l aldehydes [30].
and t h e c o n v e r s i o n
Very r e c e n t l y , t h e o x i d a t i o n
o f u n a c t i v a t e d a l k a n e s on TS-1 was d e s c r i b e d i n d e p e n d e n t l y b y Tatsumi e t a l . [31] and by us [32]. ExDerimental TS-1 was p r e p a r e d by t h e h y d r o t h e r m a l procedure d e s c r i b e d i n example 2 o f US 4,410,501.
I t s i n f r a r e d spectrum i s c h a r a c t e r i z e d by an I R band a t
960 cm-l, as r e p o r t e d f o r T S - 1 [20-211, and no e x t r a z e o l i t i c c r y s t a l l i n e o r amorphous phases c o u l d be d e t e c t e d by XRD o r SEM. The o x i d a t i o n r e a c t i o n s were conducted a t 373K under v i g o r o u s s t i r r i n g i n a s t a i n l e s s s t e e l b a t c h r e a c t o r p r e v i o u s l y f l u s h e d w i t h N2. A f t e r c o m p l e t i o n o f t h e r e a c t i o n , t h e two l i q u i d phases (H202 and alkane) were homogenized w i t h an excess o f acetone. The o x i d a t i o n p r o d u c t s were analyzed by GC on a 50 m CP S i l - 8 8 c a p i l l a r y column f rom Chrompack, u s i n g t o l u e n e as internal
s t a nd a r d .
W i t h i n t h e accuracy o f t h e GC a n a l y s i s ,
yields
of
a l c o h o l s and ketones add up t o 100% on a carbon b a s i s . H202 conversions were det e rmined by c e r i m e t r i c t i t r a t i o n [33]. R e s u l t s and d i s c u s s i o n F i g . 3 shows t h e o x i d a t i o n p r o d u c t s o b t a i n e d i n t h e o x i d a t i o n o f n hexane w i t h 35% aqueous H202 as a f u n c t i o n o f r e a c t i o n t i m e . The o x i d a t i o n p r o d u c t s a r e d e r i v e d f r o m n-hexane by o x i d a t i o n a t t h e secondary carbon p o s i t i o n s , w h i l e o x i d a t i o n o f p r i m a r y C - H bonds i s below d e t e c t i o n 1 i m i t s . Wi t h i n c r e a s i n g n-hexane conversion, t h e s e l e c t i v i t y f o r ketones i n c r e a s e s a t t h e expense o f t h a t f o r a l c o h o l s , i n d i c a t i n g t h a t t h e o x i d a t i o n oc c u r s i n two c o n s e c u t i v e s t e p s,
i.e.
hexane i s o x i d i z e d t o a
m i x t u r e o f 2- and 3-hexanols, w h i c h i s t h e n f u r t h e r o x i d i z e d t o 2- and 3 hexanones. The absence o f i s o m e r i s a t i o n o f t h e p r o d u c t s was c o n f i r m e d by t h e f o r m a t i o n o f o n l y 2-hexanone o r 3-hexanone when r e s p e c t i v e l y 2-hexanol o r 3-hexanol were used as s u b s t r a t e s . Theref ore, t h e o v e r a l l r e a c t i o n scheme f o r n-hexane o x i d a t i o n on TS-1 can be r e p r e s e n t e d as f o l l o w s :
55
H202 2-hexanol
H202
n-hexane
,
1
2-hexanone
1
3-hexanone
H202
3-hexanol
I r
I
3-hexanone
/ /
lZ@#2-hexanone
/
EB 2-hexanol total
n 0
1
0.5
2
3
Reaction time (h) F i g . 3. n-Hexane o x i d a t i o n by H202 on TS-1 as a f u n c t i o n o f r e a c t i o n t i m e . R eac t io n c o n d i t i o n s : 500 mg o f TS-1, 115 mmoles o f n-hexane, 240 mmoles o f H202 (35% i n H20), 45 m l o f acetone; 373K, 700 RPM s t i r r i n g r a t e . T a ble 2 shows t h e H202 y i e l d and s e l e c t i v i t i e s i n t h e o x i d a t i o n o f d i f f e r e n t n-alkanes and hexane isomers by 35% aqueous H202 [31]. TABLE 2 Alkane o x i d a t i o n s by H202 on TS - l a Substrate
H202 y i e l d b
(%I n-pentane n - hexane n -o c t a ne n-decane 2-methylpentane 3-methyl pentane 2,2-dimethylbutane a; b; c;
68 70 65 56 59 58 50
Regiosel e c t i v i t y C Total products Ketone f r a c t i o n C2/C3 C2/(C4tC5) C2/C3 C2/(C4tC5) 2.12 1.13 1 1.12 5.67 0.78
1.17 0.60
2.45 1.78 2.77 3.28 2.70
3.59 2.56
>loo
OD
R eac t io n c o n d i t i o n s : 400 mg o f TS-1, 310 mmoles o f alkane, 210 mmoles o f H202 (35% i n H20), 60 m l o f acetone, 3 hours a t 373K and 1000 RPM. H202 y i e l d = f r a c t i o n o f H202 used f o r alkane o x i d a t i o n on t o t a l H202 c onv ers io n ; H202 c o n v e r s i o n s a r e h i g h e r t han 90% i n a l l r e a c t i o n s . R a t i o o f o x y g e n a t i o n a t C p o s i t i o n 2 over oxygenat ion a t i n n e r C positions.
56
F o r t h e d i f f e r e n t n - a l k a n e s as w e l l as o f t h e f o u r hexane isomers t h e e f f i c i e n c y o f H202 use f o r o x i d a t i o n i s h i g h on TS-1, b u t decreases as t h e dimensions
of
the
substrate
p r o d u c t s i s almost
increase.
statistically
The t o t a l
amount
of
oxygenated
d i s t r i b u t e d over t h e d i f f e r e n t carbon
p o s i t i o n s , w i t h o n l y a s l i g h t s e l e c t i v i t y f o r o x i d a t i o n a t t h e w-1 p o s i t i o n o f t h e carbon c h a i n . W i t h i n t h e ketone f r a c t i o n however, a v e r y pronounced s e l e c t i v i t y f o r 2-ketones i s observed. T h i s means t h a t t h e f i r s t o x i d a t i o n step,
i.e.
the formation o f
alcohols,
i s only
l i t t l e regioselective,
whereas i n t h e second o x i d a t i o n s t e p , t h e 2 - i s o m e r s o f t h e a l c o h o l m i x t u r e a r e s e l e c t i v e l y f u r t h e r o x i d i z e d t o 2 - k e t o n e s . T h i s s e l e c t i v i t y i s imposed by t h e shape s e l e c t i v e p r o p e r t i e s o f TS-1. Furthermore, i t i s seen t h a t f o r branched
a1 kanes,
secondary ones.
tertiary
C-H
bonds
are
selectively
oxidized
over
The a l k a n e o x i d a t i o n on TS-1 t h u s f o l l o w s t h e ' n o r m a l '
r e a c t i v i t y o r d e r : t e r t i a r y C-H > secondary C-H >>> p r i m a r y C-H. Furthermore
the
H202
yield
decrease
with
increasing
substrate
dimensions. T h i s o f f e r s s t r o n g e v i d e n c e f o r an i n t r a c r y s t a l l i n e r e a c t i o n . Indeed, because o f t h e g e o m e t r i c a l c o n s t r a i n t s imposed by t h e p o r e geometry of
TS-1
on
the
catalyzed
reactions,
the
oxidation
rate
of
organic
s u b s t r a t e s decreases as t h e i r dimensions i n c r e a s e , and t h e r e f o r e t h e main side reaction,
i.e.
t h e r m a l and c a t a l y t i c d e c o m p o s i t i o n o f H202 t o 02,
becomes more c o m p e t i t i v e . F i g . 4 shows t h e dependance o f t h e TS-1 c a t a l y z e d n-hexane o x i d a t i o n on t h e amount o f s o l v e n t i n t h e r e a c t i o n m i x t u r e . 100
3-hexanone
be
80
@%%l 2-hexanone
W
.-
3-hexanol
60
v1
2-bexanol
h
2
40
C
0
0
total
20
0 0
10
20
30
45
60
90
Acetone (ml) F i g . 4 . n-Hexane o x i d a t i o n by H202 on TS-1 as a f u n c t i o n o f acetone c o n c e n t r a t i o n . R e a c t i o n c o n d i t i o n s : 500 mg o f TS-1, 115 mmoles o f n-hexane, 240 mmoles o f H202 (35% i n H20), I h r a t 373K and 700 RPM. F i g . 4 shows t h a t t h e hexane c o n v e r s i o n e x h i b i t s an optimum a g a i n s t t h e acetone c o n c e n t r a t i o n .
I n absence o r a t v e r y l o w c o n c e n t r a t i o n s o f
57
acetone, t h e s o l u b i l i t y o f t h e o r g a n i c s u b s t r a t e i n t h e aqueous H202 phase, and t h e r e f o r e t h e d r i v i n g f o r c e f o r d i f f u s i o n o f t h e s u b s t r a t e t o t h i s phase i s v e r y l o w . As a r e s u l t t h e r e a c t i o n r a t e may become d i f f u s i o n l i m i t e d . When t h e amount o f acetone i s increased, b o t h t h e hexane s o l u b i l i t y and t h u s t h e d r i v i n g f o r c e f o r d i f f u s i o n increase, which ' r e s u l t s i n an in c re as e d r e a c t i o n r a t e . When t h e acetone c o n c e n t r a t i o n i s s t i l l f u r t h e r inc re as e d however, b o t h H 2 0 2 and n-hexane become more d i l u t e d i n the reaction solution,
t h u s c a u s i n g a decrease i n r e a c t i o n r a t e .
The
compromise between h i g h hexane s o l u b i l i t i e s and d i f f u s i o n r a t e s and l o w d i l u t i o n s e x p l a i n s t h e optimum o f t h e acetone c o n c e n t r a t i o n . The t r a n s i t i o n of a diffusionally
t o a chemically
controlled
system w i t h
increasing
s o l v e n t c o n c e n t r a t i o n , i s c o n f i r m e d by t h e r e s u l t s shown i n F i g . 5, f o r nhexane o x i d a t i o n as a f u n c t i o n o f c a t a l y s t c o n c e n t r a t i o n b o t h i n t h e presence and i n t h e absence o f acetone. hexane
c onv ersi o n
increases
in
a
I n t h e presence o f acetone t h e
proportional
way
with
the
catalyst
c o n c e n t r a t i o n up t o conversions o f about 60%. I n t h e absence o f a p o l a r solvent
deviations
c onv ers io ns ,
of
this
indicating
linearity
that
the
are
reaction
already is
no
observed longer
at
low
chemically
c o n t r o l 1ed.
m
100
80
0
h
ae
0
d
W
60
0
C
0,'
0
U
.I
vl
L,
a2 P
.'
40
0
2-bexanol
O
El
$
0
20 0
++1 0
. 2 5 . 5 0 .75 1.00
TS-1
0
. 2 5 . 5 0 .75 1.00
(g)
F i g . 5. n-Hexane o x i d a t i o n by H20 as a f u n c t i o n o f TS-1 c o n c e n t r a t i o n . R eac t io n c o n d i t i o n s : 115 mmoles n-hexane, 2 4 0 mmoles o f H 2 0 2 (35% i n H 2 0 ) , 1 h r a t 373K and 700 RPM, a; 4 5 m l acetone added, b; no s o l v e n t added.
03
As i s seen i n F i g . 4 , t h e acetone c o n c e n t r a t i o n a l s o s t r o n g l y a f f e c t s
the
k et o ne/ a lc o h o l
ratio
of
the
oxidation
product s.
The
decreasing
s e l e c t i v i t y f o r ketones w i t h i n c r e a s i n g acetone c o n c e n t r a t i o n i s p r o b a b l y p a r t l y connected w i t h t h e v a r i a t i o n o f t h e conversion, b u t i s a l s o caused by t h e v a r i a t i o n o f hexane c o n c e n t r a t i o n i n t h e aqueous H202 phase. Indeed,
58
a t l o w acetone c o n c e n t r a t i o n s , t h e hexane s o l u b i l i t y and t h u s t h e s u b s t r a t e concentration i s low,
and t h e r e f o r e a hexanol m o l e c u l e w h i c h has been
formed on t h e c a t a l y s t w i l l be e a s i l y c o n v e r t e d t o t h e k e t o n e . A t h i g h e r acetone c o n e n t r a t i o n s and h i g h e r hexane s o l u b i l i t i e s however, c o m p e t i t i o n between hexane and hexanol f o r o x i d a t i o n w i l l become s i g n i f i c a n t , and o n l y a f r a c t i o n o f t h e hexanol m i x t u r e ,
will
m a i n l y 2-hexanols,
be f u r t h e r
oxidized. A l t h o u g h no e x t e n s i v e m e c h a n i s t i c s t u d y o f t h e a l k a n e o x i d a t i o n on titanium
silicalite
has
it
been made,
is
believed
that
the
reaction
proceeds v i a t h e a c t i v a t i o n o f H202 on t h e t i t a n i u m s i t e s w i t h f o r m a t i o n o f T i - p e r o x o s p e c i e s , which a r e t h e a c t u a l oxygen donors. such
species
subsequent
was
shown
vacuum
by
drying
adsorption [32,34].
disappearance o f t h e 960 cm-’ (Ti=O) group [26],
of
This
I R band,
H202
on
The f o r m a t i o n o f Ti-silicalite
treatment
results
in
and the
ascribed t o t h e surface t i t a n y l
and t h e f o r m a t i o n o f a 425 nm a b s o r p t i o n band i n t h e
v i s i b l e spectrum. The l a t t e r a b s o r p t i o n i s c h a r a c t e r i s t i c f o r T i complexes
w i t h peroxo l i g a n d s [31].
H e a t i n g o f t h e H202 t r e a t e d t i t a n i u m s i l i c a l i t e
l e a d s t o t h e disappearance o f t h e 425 nm and t h e reappearance o f t h e 960 c m - l a b s o r p t i o n bands i n t h e V I S and I R s p e c t r a , due t o d e c o m p o s i t i o n o f the
Ti
peroxo
complexes
with
reformation
of
the
titanyl
group.
The
f o r m a t i o n o f hydroxyhydroperoxo o r peroxo complexes by r e a c t i o n o f hydrogen p e r o x i d e w i t h t i t a n y l groups on TS-1 i s r e p r e s e n t e d i n scheme 1.
H202
0 II
OH
OOH
\ /
Ti
Ti
-
0-0 \
<
+ H20
Ti
Scheme 1: F o r m a t i o n o f s u r f a c e peroxo s p e c i e s by r e a c t i o n o f T i - s i l i c a l i t e w i t h H202. C o n s i d e r i n g t h e i n e r t n e s s o f s a t u r a t e d hydrocarbons, p r o b a b l y r e q u i r e s a h o m o l y t i c mechanism,
i t was shown above t h a t t h e p r i m a r y o x i d a t i o n p r o d u c t s a r e
Furthermore, alcohols,
t h e i r oxidation
involving radical intermediates.
ketones b e i n g formed i n a c o n s e c u t i v e o x i d a t i o n .
A tentative
mechanism f o r a l k a n e h y d r o x y l a t i o n on TS-1 i s r e p r e s e n t e d i n Scheme 2, analogy
with
the
mechanism
proposed
for
the
stoichiometric
*
-
in
alkane
o x i d a t i o n s by V-peroxocomplexes [ 3 1 ] . /
-0 \ /
Ti
Scheme 2: s i 1 i c a l it e .
d
-0* \
Ti
H-C-
Hydroxylation o f
\ \
0-0-H Ti
a l k a n e s by
/
C\
surface
0 II
Ti
/
HO-C-
\
peroxo species
on T i -
59
I n this
scheme,
an open d i r a d i c a l
f orm o f t h e T i
peroxo complex
a b s t r a c t s a hydrogen f r o m t h e alkane t o g i v e a carbon r a d i c a l . The a l c o h o l i s formed by r e c o m b i n a t i o n o f t h e carbon r a d i c a l w i t h t h e h y d r o x y l r a d i c a l coming f rom Ti-0-OH, w i t h r e f o r m a t i o n o f t h e t i t a n y l group. Conclusions T i t a n i u m s i l i c a l i t e can be used as
a selective catalyst
for
the
o x i d a t i o n o f u n a c t i v a t e d a1 kanes by aqueous hydrogen p e r o x i d e . Secondary and/or
t e r t i a r y a l c o h o l s a r e t h e p r i m a r y o x i d a t i o n product s,
t h e f ormer
ones a r e f u r t h e r o x i d i z e d t o ketones i n a c o n s e c u t i v e r e a c t i o n . The shape s e l e c t i v i t e p r o p e r t i e s o f t h e c a t a l y s t favour a r e g i o s e l e c t i v e o x i d a t i o n a t t h e 0-1 p o s i t i o n o f t h e carbon c h a i n which i s most pronounced i n t h e f o r m a t i o n o f ketones. A d d i t i o n o f a p o l a r s o l v e n t t o t h e r e a c t i o n system improves t h e mutual s o l u b i l i t y o f t h e aqueous H202 phase and t h e alkane phase, and r e s u l t s i n an i n c r e a s e o f t h e r e a c t i o n r a t e and a decrease o f t h e k et o ne/ a lc o h o l
r a t i o o f t h e p r o d u c t s. A homolyt ic,
non-free r a d i c a l
c h a i n r e a c t i o n mechanism, i n v o l v i n g T i peroxo compounds i s proposed. GENERAL CONCLUSIONS
Iron-phthallocyanines
encaged
i n zeolite Y
o r VPI-5 and t i t a n i u m
s i l i c a l i t e were used as c a t a l y s t s f o r t h e o x i d a t i o n o f alkanes w i t h t - b u t y l hy dro pero x ide o r aqueous hydrogen p e r o x i d e b e i n g used i n t h e f ormer and t h e l a t t e r case, r e s p e c t i v e l y . Both t y p e s o f c a t a l y s t c a t a l y z e t h e f o r m a t i o n o f a m i x t u r e o f secondary and/or controlled free
radical
t e r t i a r y a l c o h o l s and ketones. However,
chain
r e a c t i o n mechanism i s proposed f o r
a the
p h t h a l l o c y a n i n e based c a t a l y s t s , w h i l e a c o n s e c u t i v e h o m o l y t i c , b u t n o t f r e e r a d i c a l c h a i n mechanism i s p r o b a b l y i n v o l v e d i n t h e t i t a n i u m s i l i c a l i t e c a t a l y z e d r e a c t i o n . A l t h o u g h t h e t h r e e c a t a l y s t s have d i f f e r e n t p ore systems, v e r y s i m i l a r shape s e l e c t i v e e f f e c t s , r e s u l t i n g i n t h e r e g i o s e l e c t i v e o x i d a t i o n a t t h e o u t e r p o s i t i o n s o f t h e carbon chain, a r e observed. T h i s suggests t h a t t h e s t e r i c c o n t r o l a t t h e a c t i v e s i t e i t s e l f i s comparable i n t h e t h r e e cases. ACKNOWLEDGMENTS
RFP and DRCH acknowledge a f e l l o w s h i p S c i e n t i f i c Research.
f rom t h e
B e l g i a n Fund o f
REFERENCES 1. J.W. Ward, A p p l . I n d . C a t a l . , 3 (1984) 271-392. 2. C . L . H i l l , A c t i v a t i o n and F u n c t i o n a l i z a t i o n o f Alkanes, John W i l e y & Sons, N.Y., ( 1 9 8 9 ) . R.F. Parton, J.M. Jacobs, D.R. Huybrechts and P.A. Jacobs, Stud. S u r f . 3. S c i. , 46 (1989) 163-193.
60
4. 5. 6. 7. 8. 9. 10. 11.
12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.
H. van Bekkum and H.W. Kouwenhoven, Recl. Trav. Chim. Pays-Bas, 108 (1989) 283-294. W.F. H o l d e r i c h , Stud. S u r f . S c i . , 49 (1989) 69-93. 6. Meunier, B u l l . SOC. Chim. France, 4 (1986) 578-594. D. Mansuy, Pure Appl. Chem., 59 (1987) 759-770. I . Tabushi, Coord. Chem. Rev., 86 (1988) 1-42. N. Herron, G.D. Stucky and C.A. Tolman, J. Chem. SOC. Chem. Commun., (1986) 1521- 1522. G. Meyer, D. Wohrle, M. Mohl and G. S c h u l z - E k l o f f , Z e o l i t e s , 4 (1984) 30-34. B.V. Romanovsky, Proceed. 8 t h I n t . Congr. C a t a l ., V e r l a g Chemie, Weinheim, 4 (1984) 657-667. N. Herron, J. Coord. Chem., 19 (1988) 25-38. T. Kimura, A. Fukuoka and M . I c h i k a w a , Shokubai, 30 (1988) 444-447. P.J. Grobet, J.A. Martens, I . B a l akrishnan, M. Mertens and P.A. Jacobs, Appl. C a t a l . , 56 (1989) L21-L27. R.F. Parton, L. U y t t e r h o e v e n and P.A. Jacobs, Second I n t . Symp. Heterog. C a t a l . & F i n e Chemicals, P o i t i e r s , (1990). M. Fontecave and D. Mansuy, Tetrahedron, 40 (1984) 4297-4311. Cytochrome P-450: S t r u c t u r e , Mechanism and B i o c h e m i s t r y , Ed. P.R. O r t i z de M o n t e l l a n o , Plenum Press: New York, (1986). J. Haber, R. Iwanejko and T. Mlodnicka, J . Molec. C a t a l . , 55 (1989) 268- 275. R.A. L e i s i n g , R.E. Norman and L. Que, Jr., I n o r g . Chem., 29 (1990) 2555-2557. M. Taramasso, G. Perego and B. N o t a r i , U.S. Pat . 4,410,501 (1983). G. Perego, G. B e l u s s i , C . Corno, M. Taramasso, F. Buonomo and A. E s p os it o , i n "Proc. on t h e 7 t h I n t . Z e o l i t e Conf., Tokyo, August 17-22 (1986) 129-136. B. N o t a r i , Stud. S u r f . S c i . C a t a l . , 37 (1987) 413-425. U. Romano, A. E s p o s i t o , F . Maspero, C . N e r i and M.G. C l e r i c i , Stud. S u r f . S c i . C a t a l . ,55 (1990) 33-41. C. N e r i , A. E s p o s i t o , B. A n f o s s i and F . Buonomo, Eur. Pat . 0 100 119 (1984). F. Maspero and U. Romano, Eur. P a t . 0 190 609 (1986). C. N e r i , 6. A n f o s s i and F . Buonomo, Eur. Pat. 0 100 118 (1984). A. E s p os it o , M. Taramasso and C. N e r i , DE 3 1 35 559 (1982). A. E s pos it o , C . N e r i and F . Buonomo, Eur. Pat . 0 102 655 (1984). P. R o f f i a , M . Padovan, E. M o r e t t i and G. De A l b e r t i , Eur. Pat. 0 208 311 (1987). C . N e r i and F . Buonomo, Eur. Pat. 0 102 097 (1986). T. Tatsumi, M. Nakamura, S. Nagashi and H. Tominaga, J. Chem. SOC., Chem. Commun. (1990) 476. D.R.C. Huybrechts, L. De B r u y c k e r and P.A. Jacobs, Nat ure, 345 (1990) 240-242. A.I. Vogel, "A t e x t b o o k o f q u a n t i t a t i v e i n o r g a n i c analyses", Ed. Longmans, 3 t h Ed. (1961) 325. D.R.C. Huybrechts, I. Vaesen, H.X. L i and P.A. Jacobs, i n p r e p a r a t i o n .
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeotes 0 1991 Elsevier Science Publishers B.V., Amsterdam
61
SORPTlON AND SEPARATION OF BINARY MIXTURES OF CH4, N2 AND COz I N ZEOLITES LOVAT V.C. REES Imperial College of Science. Technology & Medicine, London. SW7 2AY, U.K. SUMMARY The sorption of methane, nitrogen and carbon dioxide and their binary mixtures has been studied i n both silicalite-1 and Na-Y zeolites. The above sorbates have increasing polarity and the effect of change of polarity on their sorption behaviour in the uncharged silicalite and highly charged Na-Y frameworks has been determined. Separation factors calculated from the initial slopes of the single component isotherms are compared with those obtained by experiment for both sorbents. Three isotherm models have been tested for their ability to predict the sorption of nitrogedmethane mixtures in Na-Y. 1NTRODUCTION The separation of gas mixtures by pressure swing adsorption (PSA) is becoming more widely used as the cost of energy increases. The design and optimisation of PSA separation units requires a detailed knowledge of equilibrium adsorbate mixture data over the temperature and pressure range likely to be used in such units. There is also a need to select the adsorbent with the optimum properties for the specific adsorbate mixtures to be separated. This paper describes some interesting results which compare the sorptions of binary mixtures of CH4, N2 and COz in two zeolites with completely different surfaces. The first is silicalite-1, which is the pure s’ilica end member of the zeolite ZSM-5, and Na-Y a more open, highly charged zeolite containing a high concentration of aluminium atoms in the framework and sodium ions to neutralize the negative charge introduced in the framework by the substitution of Si by Al atoms. EXPERIMENTAL The sorption of the single components and their binary mixtures was measured in a modified version of the isosteric method developed by Bulow et a1(1,2) and is fully described in a previous paper(3) The important feature of this method in the case of mixtures, is the constant composition of the sorbed phase over the temperature range scanned.
62
The silicalite sample, supplied by Laporte Inorganics (Widnes, UK), consisted of clean. elongated slabs 15-25 p m in length containing less than 0.01 % of aluminium. The zeolite was calcined in air at 550°C to remove the organic template and outgassed at all other times at 350°C. The Na-Y zeolite was also supplied by Laporte Inorganics and had the unit cell formula Nas6[(A10~)~6(Si0~)136].250Hz0. No detectable amorphous material was present in the sample. 27Al magic angle spinning NMR confirmed that no dealumination of the sample occurred whilst activating the sample at 100°C to remove most of the zeolitic water followed by heating to 350'C for 14 hours. The 29Si MAS NMR spectrum was deconvoluted to give a Si/AI ratio of 2.5 f 0.2 confirming the chemical analysis above. RESULTS AND DISCUSSION The quality of the isosteres obtained by the isosteric method are clearly demonstrated in Fig. 1 and 2 for the sorption of nitrogen in silicalite1 and Na-Y respectively. Isosteres of similar quality were obtained for methane and carbon dioxide sorbed in these two zeolites. The temperature range covered by the silicalite-l isosteres is 273-343K and the Na-Y isosteres is 110-250K for nitrogen and methane and 210-270K for the carbon dioxide. The isosteric apparatus was designed specifically to cope with mixtures and in Fig.3 the quality of the isosteres obtained with a mixture of 75 mole % N2 and 25 mole % COz sorbed in Na-Y is shown to be equal to that obtained with single component sorbates. The isotherms obtained from the isosteres in Fig.1-3 are shown in Fig.4-6 respectively and, once again, clearly show the quality of the single component and binary mixture data which can be obtained from the isosteric method. In Fig. 7 isotherms for the sorption of methane, nitrogen and carbon dioxide in silicalite-1 and Na-Y at 273K may be compared. Carbon dioxide is very strongly sorbed in Na-Y compared with silicalite-1 demonstrating the strong interaction of the large electric fields in Na-Y with the large quadrupole moment of carbon dioxide. Nitrogen has a much smaller quadrupole moment than that of carbon dioxide. Although the small quadrupole moment of nitrogen does lead to its enhanced sorption in the highly charged Na-Y supercages over the uncharged silicalite-l pores this enhancement is only -16% compared with the 50 fold enhancement found with carbon dioxide. It is interesting to note that the non-polar methane molecule is more strongly sorbed by some 75% in silicalite-1 than in Na-Y
63 12
11
.
10
5
9
I
no
-n
E
~.
7
3.0
3.2
3.1
3.3
3.4
3.5
3.6
I
low1
Fig. I Pure nitrogen silicalite- 1 isosteres. P,=l Pa. Sorbate loadings (in mmol/g): (a) 0.012: (b) 0.047; (c) 0.069; (d) 0.097: (e) 0.139; (f) 0.157: (8) 0.195.
14
12
10
B
6
4
2
D 3.0
4.0
5.0
6.0
7.0
E.0
lOC/T
Fig. 2 Pure nitrogen Na-Y isosteres. P,=lPa. Sorbate loadings (in mmol/g): (a) 0.0702: (b) 0.121: (c) 0.237; (d) 0.449 ; (e) 0.762: (f) 0.925; (g) 1.449; (h) 4.522; (i) 7.633.
64
Fig. 3 75 mol% N2/25 mol% C02 Na-Y isosteres. P,=lPa. Sorbate loadings (in mmol/g): (a) 0.0265; (b) 0.125; (c) 0.235; (d) 0.416;
Fig.4 Nz-silicalite- 1 isot heriris
Fig.5 N 2 - N a Y i sot her 111s
65 1.4
1.2
Fig.6
1.0
. 2
0.n
b
0.6
P
75 mol% N2/25 mol% C 0 2 - NaY isotherms
0.4
0.2
0.0
0
20
40
60
80
100
120
140
160
1
e.0 7.0 6.0
5.0 4.0
3.0
2.0 1.0
0.0
I
Pressure/kPe
Pressure/kPe
Fig. 7 Comparison of the sorption of CH4, N2 and C02 at 27313 in silicalite-1 and NaY zeolites 30
I 7
I
25
Fig.8 Isosteric heat of sorption. qT. of C02/N2 mixtures in silicalite- I as a function of mole fraction, Yco?.of C 0 2 in gas phase.
5
0.0
0.2
0.4
ycor
0.6
0.9
I
1.0
66
demonstrating the enhanced repulsion-dispersion interaction of this sorbate in the snialler channels and intersections of silicalite-1 compared with the corresponding interaction in the much larger supercages of Na-Y. The Henry's Law constants, KH. have been determined from the initial slopes of the sorption isotherms of methane, nitrogen and carbon dioxide in silicalite-l and Na-Y and these constants are listed in Tables 1 and 2 respectively. These tables also include the separation factors a;,? = K H ( I ) / K H ( ~ )for gases ( 1 ) and ( 2 ) calculated from these constants and the corresponding experimentally determined separation factors al,? = (Xl/Y,) (Y?/X?) where Xi and Yi are the mole fractions of component i i n the sorbed and gas phases respectively. Good agreement between these a' and a separation factors can be seen. TABLE 1 Henry's Law Constants. K H and Separation Factors a' and a for Sorption of CO?. N: and CHJ in Silicalite. Temp KH/( 10-6mol.kg.'.Pa-' a' a (K)
273
I I I
I
85.0
N? 4.97
10.3
2.54
CQ
CHj 17.0
4.31
CO:/N2 17.04 13.29 9.44 5.65 4.04
COI/CH4 C H 4 / N 2 IC@/N2 4.99 3.43 I -20 4.38 3.04 3.69 2.69 2.81 2.01 -4 1.69 2.38
C02/CHj CHJ/N2 3.5 3.5 3.0 2.8 2.7 2.7 2.2 2.5 2.0
TABLE 2 Henry's Law Constants, KH and Separation Factors a' and a for Sorption of CO,, N2 and CHJ in Na-Y Zeolites a'
Temp1
1 50 200 247 270
2 . 5 9 ~10-3 0.363 8.37~10-~ 1 . 9 9 ~ 1 0 - ~1 . 5 0 ~ 1 0 - ~ 3.95~ 5.79~
4 . 0 7 10-3 ~ 1 . 3 6 ~ 1 0 - ~ 4337 2 . 5 0 ~ 1 0 - ~ 1327 9.75~ 693
a
I .57 2657 795 405
1.63 1.67 1.68
1.52
The experimentally determined a factors for C02/N2 and C O ~ / C H J mixtures sorbed in Na-Y are not given in Table 2. Because of the very strong sorption o f carbon dioxide in Na-Y at 2 0 0 - 2 7 0 K the concentrations of carbon dioxide in the gas phase at equilibrium were too small to be measured by the on-line mass-spectrometer and the a factors could not, therefore. be calculated. These low concentrations of carbon dioxide are fully consistent with the very large a' separation factors in Table 2 calculated from the KH constants.
67
The separation factors in Table I for silicalite indicate that this sorbent would be an excellent sorbent to use for PSA separations of binary mixtures of carbon dioxide, nitrogen and methane. Na-Y could be used to separate methanehitrogen mixtures but the large separation factors when carbon dioxide is one component of the mixture would lead to desorption problems if the PSA separation was carried out in room temperature. Na-Y could be used for such separations but the temperature would need to be raised well above 300K. From the slopes of the isosteres the isosteric heat of sorption, q, can be calculated by the Clausius-Clapeyron equation: d(lnP)/d( 1/T) = -q/R. The initial values of q at low coverages were found to be 15.3, 7.6 and 23.6 kJ/mol for methane, nitrogen and carbon dioxide in silicalite-l and 18.1, 16.9 and 34.4 kJ/mol for these sorbates respectively in Na-Y. The small difference in q for methane in the two zeolites reflects the non-polar nature of the sorbate. The heat of sorption of methane in Na-Y will contain a polarization energy contribution from the high concentration of sodium cations present in the supercages which will tend to be balanced by an additional dispersion-repulsion contribution to q from the closer fit of the methane molecule in the smaller channels of silicalite-I. The 9.3 and 10.8 W/mol difference in q for nitrogen and carbon dioxide respectively in the two zeolites arise mainly from the additional quadrupole moment/electric field gradient contribution to q for these two sorbates in Nay. Heats of sorption of mixtures The isosteric heat of sorption of mixtures, q,, in silicalite-1 were found to be controlled by the composition of the gas phase i.e. q, = Y,q, +Y2q2 where q is the isosteric heat of sorption of the pure component, Y the gas phase mole fraction and subscripts 1 and 2 represent the two components of the mixture. Fig. 8 shows the linear variation of q, with Yco2 for nitrogedcarbon dioxide mixtures in silicalite-I. Thus the heat of sorption, q,, is not controlled by the composition of the sorbed phase but by the gas phase composition. The above equation for q, was found to be invalid for the ionic zeolite Na-Y. The isosteric heats of sorption, q,, of nitrogen/carbon dioxide and methane/carbon dioxide mixtures do not fall between the heats qN20r qcQnd 9C O , respectively as would be the case if the above equation was valid. Because of the very large separation factors involved the gas phase for these systems is virtually pure nitrogen or methane respectively and the isosteric heats of the mixtures should. therefore, be that of nitrogen or
68
methane i.e. - 17kJ/mol. Although the heat of sorption of nitrogen/carbon dioxide mixtures is initially - 16kJ/mol this heat decreases with increasing coverage to 13kJ/mol indicating that the carbon dioxide molecules are not only preferentially sorbed on the high energy sites but that they contribute an additional repulsion contribution to the heat of sorption of the nitrogen molecules because the heat of sorption of pure nitrogen is never found to decrease below 17kJ/mol. Carbon dioxide molecules similarly affect the heat of sorption of methane i n methanelcarbon dioxide mixtures. The value of qT decreases to -14kJ/mol at higher loadings and is thus less than the minimum value of 16kl/m0l found for the sorption of pure methane in Na-Y. The isotherms and isosteric heat of sorption of nitrogen and methane in Na-Y are very similar. The isosteric heat for nitrogedmethane mixtures, qT, was found to be controlled by the component present in excess in the sorbed phase. Because of the small differences in q and qCH, it was N2
difficult to really test the relationship between qT and the isosteric heats of the pure single components. Model Dredictions of mixture sorDtion in Na-Y The object of much of this work is to develop theoretical models which are capable of predicting the sorption behaviour of binary mixtures from knowledge only of the single component sorption data. In a previous paper (3) the Ideal Adsorbed Solution (IAS) theory was used to make such predictions of the sorption behaviour of nitrogedcarbon dioxide mixtures in silicalite-1. The calculated isotherms were found to be in close agreement with the experimental isotherms for three different binary mixture compositions.
The root mean square deviations, [ ~ ( v ~ . ,-, ~ , , ~ ~ ~ ) ~ / n ] ” ’
where n is the number of experimental points, between the experimental. v,,, and predicted, vpred.amounts sorbed were found to be of the order of 5%. Thus the IAS model is capable of accurate predictions for this specific system. In Fig. 9-11 the isotherms predicted by the I A S theory for three different nitrogen/carbon dioxide mixture compositions in the sorbed phase of Na-Y may be compared with the experimentally determined isotherms. Isotherm temperatures of 220, 250 and 270 K were covered for each mixtures to indicate the effect of temperature on the predictions. The IAS theory predicts a) too large amounts sorbed at high nitrogen compositions of the sorbed phase b) smaller amounts sorbed than found experimentally when the sorbed phase contained equal amounts of the two components and
69 40
+ 3 5 .
experimental
-- IAS Theory
25
M
75
100
125
150
175
200
Ressurc/kPa
Fig. 9
Isotherms for 75 mol% N2/25 mol% C02 sorbed in Na-Y
30
.-,
+
experimental
-
IAS Theory
20
: 10 u
0
Fig. 10 Isotherms for 50 mol% N2/50 mol% C02 sorbed in Na-Y Fig. 1 1 Isotherms for 25 mol% N2/75 mol% C02 sorbed in Na-Y 40
30
-
.
=
20
0
10
0
+
experimental
-- IAS Theory
70
c) very much smaller amounts sorbed than found experimentally at high carbon dioxide compositions of the sorbed phase. These differences were the same at all three temperatures studied. Although the agreement was poor between experiment and IAS theory predictions at the three different sorbed phase compositions studied the agreement was best for the sorbed phase containing equal quantities of the two components. The root mean square (rms) deviations calculated for these IAS predictions were -7, 5 and 28 molecules per unit cell for the 75%N2, 50%N2 and 25%N2 sorbed phase compositions respectively. The IAS theory was similarly unable to predict the sorption behaviour of methanelcarbon dioxide mixtures in Na-Y over the same composition and temperature ranges covered for the above nitrogenlcarbon dioxide mixtures. However, the IAS predictions were found to be reasonably accurate for a sorbed phase containing equal amounts of methane and carbon dioxide at 220 and 250K. The root mean square deviations calculated for these two isotherms were found to be less than 1 molecule per unit cell at 220 and 250K respectively. The isotherms obtained with mixtures of nitrogen and methane are similar to the pure nitrogen and pure methane isotherms at the same temperature. The mixture isotherms were less curved when the sorbed phase contained excess nitrogen consistent with the less curved nature of the pure nitrogen isotherms compared to the pure methane isotherms. Na-Y showed a small, but reasonable selectivity for methane. This selectivity is clearly shown in the McCabe-Thiele diagram in Fig.12 over the whole range of sorbed phase compositions. This diagram also clearly demonstrates the increasing selectivity towards methane as the temperature increases. The corresponding McCabe-Thiele diagram for nitrogenlcarbon dioxide mixtures in silicalite-1(3) showed the opposite temperature effect. As the temperature was raised the selectivity towards carbon dioxide decreased. The mixture data obtained for the sorption of nitrogenhethane mixtures in Na-Y were suitable for testing three different predictive models ( i ) IAS (ii) Competitive Langmuir(4) and (iii) Ruthven's(5) Statistical Thermodynamical Model. These three models were tested for three sorbed phase mixture compositions of 75%/25%, 50%/50% and 25%/75% nitrogenhethane and at three different sorption temperatures of 150, 200 and 250K. Fig.13-15 compare the predictions of the three models against the experimental data for these three sorbed phase mixture compositions at the most sensitive temperature of 150K. The Competitive Langmuir model is clearly seen to give the best fit to the experimental data. At the higher
71 1.0
-
-
0.6
Ne
0
a
0.6
-
0.4
-
n I?)
E
2 Y
a CH,
0.0
0.0
0.2
0.6
0.4
1
0.8
M o l e Fraction Sorbed Phase (X)
Fig. 12
loo 90
10
2o 0
McCabe-Thiele diagram for N2/CH4 mixtures sorbed in NaY
f
i
4 0
40
80
120
160
2 I0
Pressure/kPa
Fig. 13 Isotherms for 75 mol% N2/25 mol% CH4 sorbed at 150K in NaY (+) experimental versus predictions by Competitive Langmuir, TAS and Ruthven's Statistical Thermodynamical models.
72
.... ....
-
90 80
................................. Langmuir
t
.+
IAS
0 0
40
80
120
160
0
Pressure/kPa
Fig. 14 Isotherms for 50 mol% N2/50 mol% CH4 sorbed at 150K in NaY (+) experimental versus predictions by Competitive Langmuir, IAS and Ruthven's Statistical Thermodynamical models.
,
100
-
90
-;
70
-
50 60
-
\
m m
40
-
30
-
PI
9
.'; . . . . . . . . . . . . . . .L.a n.g.m.u .i r . . . . . . .
;cF -
80
\
.'
f'
:
- - - - - _ _ _ _ _ _ _ _ Ru t h u e n
c
20
0
4
0
.~
40
80
120
160
200
Pressure/kPa
Fig. 15 Isotherms for 25 mol% N2/75 mol% CH4 sorbed at 150K in NaY (+) experimental versus predictions by Competitive Langmuir, IAS and Ruthven's Statistical Thermodynamical models.
73
temperatures the IAS and the Competitive Langmuir model were both equally good at predicting the experimental isotherms. The root mean square deviations calculated for all three models for all three sorbed phase compositions and for all three temperatures are summarized i n Table 3. In this Table the very poor predictive ability of the Statistical Mechanical model compared with the other two models can be clearly seen. However, this is not surprising since this model is not really designed to cope with the large number of molecules sorbed per supercage in these mixtures. In order to cope with the large number of molecules sorbed per unit cell the unit cells were divided into smaller volumes which contain only 4 molecules at saturation. The model then assumes that there is little transference from one sub-volume to another and this is a gross simplification of the actual experimental situation. TABLE 3 Comparison of the predictive models for nitrogenlmethane mixtures sorbed Na-Y ________~
Sorhed Phase Composition mole
Temperature K
%NZ
Thermodynamic
mole %CH4
~~~
Root Mean Square Deviation* Statistical Competitive Ideal Adsorhed Langmuir
Solution
75/25
1 so 200 2so
22.0 23.7 24.7
4.7 2.3 2.1
12.7 2.9 1.4
50150
150 200 250
19.7 15.0 11.2
3.2 2.1 1 .o
12.2 2.9 0.3
150
18.1 18.3 9.2
8.4 4.0 1.6
3.1 2.0 1.2
25/75
200 250
t RMS deviations are in molecules per unit cell
74
REFERENCES
1 2 3 4 5
M. Bulow and P. Lorenz, Fundamentals of Adsorption 11: A. Liapus (Ed), Engineering Foundations. New York. USA, 1987, pl19. M. Bulow and P. Lorenz in preparation P. Graham, A.D. Hughes and L.V.C. Rees, Gas Sep. Purif. 3 (1989) 56. E.C. Markham and A.F. Benton. J. Amer. Chem. SOC. 5 3 (1931) 497 D.M. Ruthven, K.F. Loughlin and K.A. Holbrow Chem. Eng. Sci., 28 (1973) 701
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
75
USE OF ZSM ZEOLITES IN THE LIQUID PHASE SEPARATION OF ALCOHOLS R. SCHOLLNER and W.-D. EINICKE Department of Chemistry, Karl M a r x University, Leipzig, German Democratic Republic
Talstr.
35,
7010
SUMMARY
A study was made on the adsorption of alcohol-water mixtures on hydrophobic siliceous molecular sieves. The aim of this paper is to show the concept from the selection of the most effective structure types over the influence of the Si/Al ratios of the adsorbent.~from template synthesis and the dealumination of Pentasils from the “inorganic way’, the selection of a hydrophobic binder to the use of the zeolites in innovative processes.
INTRODUCTION The novel h l g h silica molecular sieves with distinct hydrophobic properties synthesized during recent years make a selective separation of organic compounds from aqueous solutions possible (ref. 1). As polnted out by several authors, an industrial potential is the adsorptive separation of alcohols from ethanol (ref. 2) and butanol-isopropanol (ref. 3) fermentation. I t seems that adsorption processes are alternative methods for the energy consuming conventional three stage process : strlpplng-rectificationvacuum rectlfication (ref. 4). The aim of this paper 1s to examine the thermodynamics of ethanol-water mixtures on hydrophobic sillceous adsorbents for an optimization of the adsorbents properties. The analysis of the excess isotherms was carried out by the thermodynamics of energetically homogeneous and heterogeneous adsorbents. Further investigations deal with the selection of suitable binders for hydrophobic molecular sieves, the characterization of the desorption step of ethanol from the adsorbents and the use of the molecular sieve pellets in innovative processes. THEORY To describe the liquid phase adsorption of blnary alcohol-water mixtures the adsorption excess was used (ref. 5) : r2e= n2e/ rn = no
(
x20- x21
/
m
(1)
The adsorption excess can bedetermined experimentally, when n
0
76
molecules of a mixture of known composition were contacted with the mass m of the adsorbens and the mole fraction of the solution changes from x20 to the equilibrium mole fraction x21. The representation of the adsorption excess versus the equilibrium mole fraction gives the so-called adsorption excess Isotherm, classified by Schay and NagY (ref. 6). For the thermodynamical description of the excess isotherms two main approaches exist (ref. 7). From the technological point of view one disadvantage of the method of excess functions of Gibbs (ref. 8 ) 1s that there is no information concerning the content and the distribution of the two components inside a porous adsorbent. For this reason the model of the adsorbed phase (ref. 9) was used, where in the case of the zeolites the free adsorption volume corresponds to the space of the adsorbed phase. From the analysis of the adsorption excess isotherms the limiting value of adsorption n2,0s, the equilibrium diagram x2' vs. x 2 1 , the individual isotherms n i S , the differences of the interaction energies o t both components including their distribution and the activity coefficients of the adsorbed phase are available. The limiting values of adsorption determined by the extrapolation method of schay (ref. lo), the isopiestical uptake from the gas phase and the pyknometric method of Dubinln (ref. 11) are very similar and demonstrate that the ethanol molecules can occupy the whole free adsorption volume of the adsorbents. The composition of the adsorbed phase x2' was determined according to : X2S
=
"2,o
S
r x21 +
r2
1-r I r2e "210 r + where r is the ratio of the limiting values of adsorption of both components r = "1,o S '"2,o S ' The equilibrium constant for the exchange process between the bulk and the adsorbed phase is given by :
where y, are the activity coefficients of the components in the adsorbed and the bulk phase respectivily. The activity coefficients of the bulk phase were calculated by UNIFAC (ref. 12). The value K connected with the free standard enthalpy of the exchange and the difference of the interfacial enthalpies can determined by
77
the integration of the Gibbs adsorption equation 1
with the upper integration boundary x2l=l. Taking into account the energetical heterogeneity, wing equation for the overall adsorption isotherm was 13) : F(UZ1) d U21
X
the folloused (ref.
(5)
‘21,min where x 2 S (x21,U21) is the local isotherm, U21 the difference of the interaction energies of both components and F(UZ1) their distribution. The functlon F(UZ1) was calculated by a numerical regularization method on the basis of a singular value decompositlon (ref. 14) without any assumption concerning the shape of the distribution function. For comparison for all systems the same regularization parameters were used. EXPERIMENTAL The molecular sieves were calcinated to remove organic compounds, exchanged with 0.1 M aqueous NaN03 solution and washed before use. For the determination of the adsorption excess about 0 . 2 g activated adsorbens was contacted with about 1 ml of alcohol-water mixture of known composition. The equilibrium mole fractlon was determined refractometrically after 48 hrs. The DTG measurements were carried by means of a microbalance (Setaram, Lyon) with a controlled linear heating rate of 10 K/ min and a dried argon stream a t a flow rate of 3 l/h. RESULTS AND DISCUSSION The efficiency of selective separation of alcohol-water mixtures on pentasil zeolites is shown in Fig. 1 by the equilibrium diagrams on NaZSM-5 (Si/A1=19) at 293 K. The alcohols were preferentially adsorbed in the whole concentration range, which is due to the dominance of the nonspecific disperslon interaction.
78
25
5.0
7.5
Fig. 1 Equilibrium diagrams in wt-% for ethanol , 1-propanol 0 and 1-butanol-water mixtures on NaZSM-5 (Si/Al= 19) at 293 K
weight fractiodo/0)
In the technologically interesting concentration range ethanol was concentrated from 10 t o 90 percent by weight. The equilibrium constant K is increasing linearly with a higher carbon number of alcohol. The limiting values of adsorption are independent of the carbon number of the alcohol with about 130 mg/ g. In comparison with other hydrophobic adsorbents (charcoals, copolymers, dealuminated Y-zeolites) the low capacities of the pentasils seems to be a disadvantage of these zeolites. For this reason we investigated other siliceous molecular sieves with distinct hydrophobic properties to get information about the influence of the growth and the geometry of the channels and the cavities of the adsorbents on the liquid phase adsorption of ethanol-water mixtures. The characterization of the sodium forms of the molecular sieves is given in Table 1. TABLE 1 Characterization of the hydrophobic adsorbents used ~~
Zeolite ZSM-5 Mordenite US-EX Hectorite
Pore System
Chemical composition
mg EtOH/g
small channels Na4. (AloZ) 4. (sio2)91. 128.5 channelstcavities Na2m5(A102) 2.5(si02)40.0 100.0 cav 1 ties 235.4 Na4. 8 (A102) 4.8 (' i02) 158.9 "large channels* Na0,7Mg5, 3 ~ i o . 7 ~ i 8 0 2 0 ( ~130.2 ~)
79
The excess isotherms for the adsorption of ethanol-water tures on the adsorbents at 293 K are demonstrated in Fig. 2.
mix-
Fig. 2 Excess isotherms for the adsorbents 0 ZSM-5, 0 mordenl te, us-Ex and A hectorite at 293 K
It can be seen that the shape of the excess isotherms is quite different. While the isotherm for the zeolite NaZSM-5 is of t y p 1 1 1 of the Schay-Nagy classificationl the other adsorbents show type V behaviour which is connected with an adsorption azeotrope in the low ethanol region. The equilibrium diagrams calculated by eqn. (2) are illustrated in Flg. 3. It seems that the larger the pores of the adsorbens the lower the efficiency of the ethanol-water separation. From our expierences we would expect that similar Si/Al ratios and respectivily similar ratios of the nonspecific and specific adsorption interaction leads to the same excess isotherm type. Therefore the differences only can be caused by the behaviour of the adsorbed phase. In the bulk phase the water molecules are associated in clathrates (ref. 15). The first ethanol molecules of the mixture can OCCUPY the cavities of the clathrate structure. A f t e r saturation of the clathrate cavities a pseudo two phase system exists with filled clathrates and a random ethanol-water mixture. With regard to the high ethanol region, the degree of association increases is due to the decreasing number of hydroxyl groups.
80
Equllibrium diagrams for the systems mentioned in fig. 2 Fig. 3
mole fraction x i
In contrast the small channels of the ZSM-5 zeolite with a diameter of about 0.6 nm do not allow the formation of the clathrate structures. As reported earlier (ref. 16) the activity coefficients show in contrast to the bulk phase negative deviations from Raoults law. Therefore the hydrophobic Part of the ethanol molecule should be adsorbed on the oxygen walls of the channels and the OH-group of the molecules can interact with water and/or ethanol on the basis of hydrogen bonding. In the case of the zeolites mordenlte and US-Ex and the layer silicate the activity coefficients are similar to the bulk phase. This behaviour confirms the possibility of the formation of clathrate structures in the adsorbed phase. For all adsorbents with large channels or cavities the water clathrates and the sil condioxid framework compete with the first ethanol molecules in the adsorbed phase. That means that hydrophobicity is an essentia but not a sufficient condition for an effective adsorptive ethanol-water separation. Because the zeolites with small channels are the most effective ones, the considerations below only deal with the pentasil zeolites. As pointed out by several authors (refs. 17, 18), the hydrophobicity depends strongly on the aluminium content of the unit cell of the pentasil zeolites. Therefore we investigated a series of zeolites from template synthesis with different Si/Al-ratios given in Table 2. I
81
TABLE 2 Characterization of the NaZSM-5 zeolites investigated Si/Al
chemical composition
85 50 40 19 13.5
Nal. 1 (A102)1.1 (‘‘O2) 94.5 Nal. 9 (A’02) 1.9 (s102) 94.1 Na2. 3 (A102)2. 3 (si02)93. 7 Na4,8 (A102) 4.8 (“O2) 91.2 Nag.6(Alo2)6.6(Si02)8g,4
adsorbed mg g-l zeolite ethanol water 117.0 112.8 115.6 128.4 132.2
32.1 33.5 33.1 36.5 42.0
The systems are of interest for an analysis in terms of the thermodynamics of adsorption from solution on energetically heterogeneous adsorbents. The idealized aluminium-free end member of the pentasils, the so-called silicalite, can only interact with the molecules of the mixture on the basis of dispersion forces. W i t h the installation of aluminium into framework positions the possibility of specific interaction occurs and the adsorbents become more energetically heterogeneous. The distribution functions of the adsorption energy differences are shown in Fig. 4.
Si IAI.13
c
Si/AI= 40
Si/Al= 50
SilAI.85
0.1
02 -10 0 10
Fig. 4
Kl: mMa Si/AI=19
-10
0 10
-
10 0 10
.
U/kJmol”
-10 0 10
Distribution functions for the energy differences (from left to right increasing Si/Al-ratio)
The numerical calculations demonstrate, that the greater the aluminium content of the adsorbent the higher the peak of the low energy difference. This fact led us to the conclusion that this peak corresponds to the difference of the interactions of both moThe position lecules with the field of the dipoles (A104)-- Na+ and the height of the other peak is almost independent of the Si/ A1 ratio. Therefore this peak was assigned to the difference of the London-type interactions of the molecules with the zeolitic
.
82
framework. It is interesting to note that in dependence on the adsorbate-adsorbate interaction the peaks shift toward higher or lower U21 values. For Fig. 4 the calculations of the interaction within the adsorbed phase were carried out by the Ising model in the Bragg-William approximation (ref. 19) with the assumption that the interaction based on hydrogen bonding and a coordination number c=3. Accordingly with the model of an energetically homogeneous adsorbens i t becomes clear that the lower the aluminium content of the pentasils the higher the separation efficiency for ethanolwater mixtures. The demand for the use of silicalite in the adsorptive ethanol separation was also underlined by the desorption experiments (ref. 20). While in the DTG curve for the ethanol saturated silicalite only one desorption effect at 380 K appears, the curves for the aluminium containing samples show two maxima. The area of the high temperature effect (510-590 K) increases with increasing alum nium content. That means that an economic thermal regeneration of the adsorbens after the adsorption step is only possible for si 1cate. Two disadvantages of the pentasils from template synthesis are the high production costs and the problems of air pollution during the burning of the pentasils to remove all organic compounds. Therefore several authors investigated the dealurnination of pentasil zeolites synthetisized with Si/A1 ratios varying from 13 to 20 by the inorganic way without template (ref. 21). For the dealumination of zeolites the acid extraction, the modification with C0Cl2, SiC14 and (NH4)2SiF6 and the hydrothermal treatment were proposed (ref. 22). To obtain pentasils with higher Si/A1 ratios we used the hydrothermal dealumination of the parent pentasil HS 30 at water steam pressures of 4, 1 3 , 40 and 80 kPa (ref. 23). As suggested by Breck (ref. 24) the nonframework aluminium species built during the dealumination exist in a cationic form. The dealuminated samples were modified by three procedures : A : cation exchange with 0.1 M aqueous Nacl at 350 K 8 : cation exchange with 0.1 M aqueous NaOH at 350 K C: extraction with 0.1 M aqueous HC1 and recationization with 0.1 M aqueous NaOH at 350 K The framework Si/A1 ratios of the pentasils are shown in Table 3. It can be seen that the dealumination of the strongly steamed samples can be partially reversed by treatment with NaOH. A fur-
83
TABLE 3 (Si/A1IF ratios after modification determined by 27Al MAS NMR (S
water pressure/kPa 0 4
13 40 80
i /All
A
B
C
15 20 25 53 64
15 21 24 27 25
15 18 26 36 46
ther effect of the NaOH modification is the decrease of the micropore volume of the pentasils which Is due to the partial dissolution of the zeolitic framework. The formation of a secondary pore system was detected by nitrogen adsorption at 77 K as shown in Fig. 5.
tI 20
Fig.
5
40 60 p(kPa1
80
Nitrogen adsorption (open symbols) and desorption ( filled symbols) isotherms for the samples modified by procedure B by water steam pressures 0 0 , V 4, 0 13, A 40 and 0 8 0 kPa
From the hysteresis between adsorption and desorption isotherms the following micropore and secondary pore volumes (Table 4) were calculated by the "t-method" of de Boer (ref. 25). In the case of the acid treated dealuminated pentasils i t can be seen from Table 3 that the extraction of the non-framework aluminium was incomplete. During the following recationization with NaOH, a part of these species was reinserted into framework positions. It is interesting to note that the modification with HC1
84
TABLE 4 characterization of the Pore volume of the pentasiles modified procedure B water pressure k Pa 0
4 13 40 80
"micro cm 3 g-l
"sec cm 3 g-l
"sum cm3 9-1
0.166 0.154 0.154 0.146 0.129
0.060 0,081 0 101 0.122 0.194
0,226 0.235 0.255 0.268 0.323
m
by
surf. m 2 g-l 70 115 140 145 230
must lead to a reconstruction of the dealuminated framework because the realumination is not connected with the formation of the secondary pore system. The adsorption behaviour of all modified samples in the ethanol-water separation is shown by the equilibrium constants in Fig. 6. For comparison the values for the template synthesis products (ref. 26) are also given.
6 Equilibrium constants of the dealuminated pentasils in comparison t o the products from template synthesis ( OHS 3 0 , A A e A B , AC and 0 (ref. 26)) Flg.
I
:
:
;
:
:
:
1 2 3F4 5 6 Aluminium per unit cell
For the NaCl recationized samples i t can be seen from F i g . 6 that the equilibrium constants do not achieve the values of the template synthesis. The differences must be assigned to the influ-
85
ence of the non-framework aluminium which prefers water adsorption. The extraction of these species with HC1 leads to higher equilibrium constants. But the formation of SiOH groups detected by 1H NMR (ref. 27) as adsorption centres for specific interactions prevent a better separation of the ethanol-water mixture. In the case of the NaOH treated pentasils the formation of the secondary pore system gives no contribution to a higher concentration of ethanol. I t seems that association of the water molecules in the secondary pore system is the main reason for the decrease of the equilibrium constants. The evidence compares well with the results of the DTG investigations on the same samples. The portion of ethanol desorbed at higher temperatures increases in comparison with the template synthesis products with the same Si/Al ratio (ref. 20). Therefore the use of the dealuminated and modified pentasils produced with our procedure seems to be doubtful for ethanol-water separation The use of the pentasil zeolites in the industrial scale-up makes i t nessesary to find a suitable binder f o r pelletizatlon. The information concerning a hydrophobic binder in the literature is limited. Several authors used Si02 (ref. 28) and especially for catalytic investigations A1203 (ref. 2 9 ) . Both binders detoriat the efficiency of the ethanol-water separation. Therefore we used hydrophobic cellulosic triacetate. For comparison zeolite NaZSM-5
.
Fig. 7 Adsorption excess isotherms for ethanol-water mixtures on 0 NaZSM-5, pelletized with Vsio2, A ~ 1 and~ O C0 A
~
86
(Si/Al=19) was pelletized with 30 Per cent by weight of Si02, A1203 and cellulosic acetate (CA). The excess isotherms for ethanol-water mixtures are shown in Fig. 7. According to the Schay-Nagy isotherm classification the system sthanol-water-NaZSM-5+A1203 belongs to the type I V , which is connected with an adsorption azeotrop. From the excess isotherms the equilibrium constants, the limiting values of adsorption and the ethanol enrichment (E) in the adsorbed phase from solutions of 10 percent by weight were determined and are given in Table 5. TABLE 5 Adsorption characteristics of the pelletized pentasil adsorbens
ns/mmole g-l
NaZSM-5 powder NaZSM-S+CA NaZSM- 5+S i O2 Na Z SM- 5+A1 2O
3.34 2.30 3.37
4.16
K
28.7 27.8 13.8 5.2
E/Wt -% 90 90
50
30
I t can be seen that the zeolite NaZSM-5 pelletlzied with A1203 shows the highest capacity for ethanol-water mixtures, followed by the zeolite Powder and the NaZSM-5+Sio2 system. The CA containing pellets give the lowest limiting value of adsorption. In comparison with the powder, the capacity was decreased during the pelletization by about 31 percent. That means that the binder is inert concerning the adsorption of ethanol-water mixtures. From the equilibrium constants and the ethanol enrichment in the adsorbed phase i t becomes clear that the cellulosic acetate maintalnes the hydrophobicity of the zeolite. In the case of the binder Si02 the silanol groups and the possibility of associate formation in the mesopore system of the binder decrease the hydrophoblcity of the system. Also A1203 decreases the efflclency of the ethanol-water separation considerably. Additionally the associate formation a migration of aluminium species from the binder into framework positions during the calcination was detected (ref. 30). However, i t seems that cellulosic acetate is a suitable binder for hydrophobic siliceous molecular sieves in the adsorptive ethanol-water separation. The binder content can be reduced to about 5
87
percent by weight. The hydrophobic Pellets were sucessfully applicated in the - removal of ethanol from the gas and the aqueous phase - separation of alcohols from ethanol and butanol fermentation - in-situ fermentation of ethanol
-
dealcoholization of beer and wine
ACKNOWLEDGEMENTS The authors are indebted to Dr. W. Reschetilowski for helpful contributions and the group of Prof. Br%uer for numerical calculations. REFERENCES E.M. Flanigen, J.M. Bennett, R.W. Grose, J.P. cohen, R.L. 1 Patton, R.M. Kirchner and J.V.C. Smith, Nature (London), 271 (1978) 512 N.B. Milestone, Report NO. C.D. 2355, Methods of Separating 2 and Concentrating of Ethanol and Other Alcohols, New Zealand, March 1985 I.S. Maddox, Biotechnol. Lett.. 4 (1982) 23 3 A. Serra, M. Poch and C. Sola, Rev. Agroquim. Tecnol. Ali4 ment, 27 (1987) 372 and 508 D.H. Everett, Trans. Faraday Sac. 61 (1965) 2478 5 L.G. Nagy and G. Schay, Magyar Kem. Foliorat. 68 (1960) 31 6 7 A.A. Lopatkin, Pure Appl. Chem. 61 (1989) 1981 J.W. Gibbs, Collected Works, Longmans Green, New York, 8 London, Toronto 1928 A.I. Rusanov, Fazovye ravnovesija i poverkhnostnye yavlenia, 9 Khimia, Moscow 1967, p. 13 10 0. Foti, L.G. Nagy and G. Schay, Acta chim. Aced. Sci. Hung. 00 (1974) 25 11 M.M. Dubinin, A.A. Fomkin, 1.1. seliverstova and V.V. Serpinski, Adsorption of Hydrocarbons on Zeolites, Berlin 1979, SUPPI.. V01.t P. 1 12 A. Fredenslund, J. Gmehling and P. Rasmussen, Vapor-Liquid Equilibria Using UNIFAC, Elsevier, Amsterdam, 1977 13 A . Dabrowski and M. Jaroniec, Adv. Coll. Interf. Sci. 27 (1987) 211 14 M. v.Szombathely, Ph. D. Thesis, Leipzig, 1988 15 8 . Rozenfeld, K. Jerie, A. Baranowski, J. Glinskl and S. Ernst, Positron Annhilation, (Ed. P.C.Jainl R.M.Singru and K.P. Gopinethan), world Sci. Pub. Co., Singapore 1985, P. 239 16 W.-D. Einicke, U. Messow and R. SchBllner : J. Coll. Interf. sci. 122 (1988) 280 17 G. Debras, A. Gourgue, J.B. Nagy and G. de CliPPelier, Zeolites 5 (1985) 377 18 H. Nakamoto and H. Takahashi, Zeolites 2 (1982) 67 19 T.L. Hill, Statistical Mechanics, Mc Graw-Hill, New York, 1956 20 W.-D. Einicke, W . Reschetilowski, H. siegel, J. BBhm and R. Schdllner, J. Chem Soc., Faraday Trans. 88 (in Press) 21 DD-WP 207186 (1984) 22 W.-D. Einicke, W. Reschetilowski, M . Heuchel, M. v.Szombathely, M. Jusek, H.-R. Poosch, P. Brluer, W.
88
Schwieger and K.-H. Bergk, Chem. Techn. 42 (1990) 215 E. Brunner, H. Ernst, D. Freude, M . Hunger, C.B. Krause, D. Prager, W. Reschetilowski, W. schwieger and K.-H. Bergk, zeolites 9 (1989) 282 2 4 D.W. Breck and G.W. Skeels, Proc. 4th Conf. Zeolites (J. Katzer,Ed.) ACS SYmP. ser. 40, Washington D.C., 1978, p. 271 25 B.C. Lippens and J.H. de Boer, J. Catal. 4 (1965) 319 2 6 W.-D. Einicke, M. Heuchel, M. v.szombathely, P. Brtluer, R. SchBllner and 0. Rademacher, J. Chem. Soc., Farad. Trans. I 85 (1989) 4277 27 W. Reschetilowski, W.-D. Einicke, M. Jusek, R. SchBllner, D. Freude, M. Hunger and J. Klinowsi, Appl. Catal. 56 (1989) L 15 28 H.R. Burkholder, G.E. Fanslow and D.D. Bluhm, Ind. Eng. Chem. Fundam. 25 (1986) 414 2 9 A.S.T. Chiang and Y.J. Yang, J. Chin. I. ch. E. 18 (1987) 63 3 0 D.S. Shihabi, W.E. Garwood, P. Chu, J.N. Miale, R.M. Lago, C.T.W. Chu and C.D. Chang, J. Catal. 93 (1985) 471 23
I
G. Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
89
BASIC PRINCIPLES AND RECENT RESULTS OF 'A MAGIC-ANGLE-SPINNING AND PULSED FIELD GRADIENT NUCLEAR MAGNETIC RESONANCE STUDIES ON ZEOLITES
H. PFEIFER, D. FREUDE AND J. m G E R Sektion Physik der Karl-Marx-Universitat Leipzig, LinnkstraBe DDR-7010 Leipzig
5,
ABSTRACT In the first part it is shown both experimentally and theoretically that the H' NMR chemical shift of surface hydroxyl groups as measured for strongly dehydrated zeolites by the MAS technique is a suitable parameter to characterize their strength of acidity. Moreover, this method allows a direct and unambiguous determination of the concentration and accessibility of the various OH groups. In contrast, for hydrated zeolites a superposition of several effects (e.g. Plank's mechanism, formation of hydroxonium ions, adsorption of water molecules on Lewis acid sites) may occur which requires a more detailed analysis of the spectra and additional information for an unambiguous interpretation. In the second part recent results of pulsed field gradient (PFG) NMR studies of self-diffusion in zeolites are presented which include xenon self-diffusion, the anisotropy of diffusion in zeolites, NMR tracer exchange experiments and a critical comparison of PFG NMR diffusivities with results of other techniques. H' MAS NMR STUDIES ON DEHYDRATED ZEOLITES The elementary step of catalysis by Bronsted acidic sites is the proton transfer from the surface hydroxyl group ZOH to the adsorbed molecule M: ZOH
+
M
=
zo- +
MH+
(1)
Hence, the catalytic activity of a catalyst will be determined by at least three independent parameters for each sort of surface hydroxyl groups (Bronsted acidic sites): The strength of acidity as defined by the rate constant of the proton transfer to the adsorbed molecule, (ii) the concentration and (iii) the accessibility of the acidic sites. (i)
90
Strenuth of Bronsted Aciditv According to eq. (1) the rate constant of the proton transfer depends both on the properties of the acidic site ZOH and of the particular molecule M. In order to define a parameter which characterizes the protonating ability of the ZOH group but does not depend on the particular molecule we decompose reaction (1) into two processes: ZOH H+
+M
=
20-
=
MH+
+
H+
which leads to the definition of the strengths of gas phase acidity and basicity as the standard Gibbs free energy change of processes (2) and ( 3 ) , respectively (ref. 1). In the following we shall use "strength of acidity" as a synonym for the standard Gibbs free energy change of process (2) since it is exactly this parameter which describes quantitatively the protonating ability of the Bronsted acidic site. To compare the strength of acidity with the deprotonation energy of the ZOH group h E D P l a quantity which follows from quantum chemical calculations, one must take into consideration that the standard Gibbs free energy hGDpo is the sum of the deprotonation energy A EDp, of the zero-point energy change AEDpo and of the Gibbs free energy change b GDPtherm which results from the conversion of the three vibrational degrees of freedom of the proton as part of the ZOH group into its three translational degrees of freedom after leaving this group. Assuming that the zero-point energy change AEDpo is a constant and that the contribution of b GDptherm is negligible (ref. 2) , the deprotonation energy &EDPI i.e. the energy difference of 20and ZOH, can be taken as a suitable measure for the strength of acidity. With regard to the position of the H ' MAS NMR signal of the ZOH group which is generally described by the so-called chemical shift i.e. the shift of the resonance relative to a standard (tetramethylsilane) in ppm, the following qualitative argument may be taken as a hint, that there is a direct relation between J H and the strength of acidity: A higher value of 6H corresponds per definition to a reduced shielding of the external magnetic field and hence to a larger value for the net atomic charge of the hydrogen. On the other hand an enhancement of the net atomic
91
charge will lead to a reduction of the deprotonation energy. In agreement with this suggestion experimental values for 6H and for the absolute gas phase acidities aGDPo measured for hydroxyl groups of various organic compounds in the gaseous state which are plotted in Fig. 1 show in fact a good correlation (cf. ref. 3)
I
PhOH
Fig. 1. Dependence of the chemical shift JH for the OH proton of various molecules on the absolute gas phase in the acidities A G gaseous statef)’The 6 values were taken from ref.H;2, the AGDpo-values were measured by Lias, Bartmess et al. and are presented in ref. 73.
$’PPm
2
1
0
1550
1500
1450
1400
Another experimental proof of our suggestion that 6H is a suitable and sensitive measure for the strength of acidity is given by Fig. 2. In this figure Sanderson‘s intermediate electronegativity Sm, computed for zeolites of composition HA102(Si02)n according to the equation sm = (SH S A ~So 2n+2 SSi)1/ (3n+4)
(4)
with SH = 3.55; SA1 = 2.22; So = 5.21; Ssi = 2.84 (cf. ref. 4), is plotted together with experimental values of 6H in dependence on the silicon-to-aluminium ratio n. The nearly identical functional dependence of both quantities provides ample evidence for the usefulness of JH as a measure for the strength of acidity. In this connection it should be mentioned that due to the use of CH4 as an inner standard (ref. 3 ) the absolute values of 6H plotted in Fig. 2 are reduced by about 0.3 ppm with respect to ref. 5 and that the accuracy of the absolute values of JH has been enhanced considerably so that the present error does not exceed ? 0.05 ppm. Moreover it was possible to show by a thorough investigation that the residual linewidths which limit the accuracy for a determination of the values of 6H for bridging OH groups on zeolites are not controlled by artefacts or imperfections rather than by the distribution width of the strength of acidity (ref. 6).
92
nFig. 2. Values for the intermediate electronegativity Sm and the chemical shift bH of the (accessible) bridging OH groups in dependence on the silicon-to-aluminium ratio n for various zeolites (HX: zeolite H-X; HY: zeolite H-Y; HM: mordenite; HE: erionite; HZ: H-ZSM-5). For bridging OH groups of a zeolite H-Y with a silicon-toaluminium ratio n = 2.6 the residual linewidth which can be reduced in the MAS NMR spectra after partial deuteration of the sample or by using multiple pulse sequences to reduce the dipolar proton proton interaction is of the order of 0.3 ppm (ref. 7). Taking into consideration the empirical formula (ref. 3 ) f1cm-l
= 3906
-
74.5 GH/ppm
(5)
which connects GH with the stretching vibration frequency f of the OH groups, the residual MAS NMR linewidth of 0.3 ppm corresponds to a residual linewidth of the IR band of the bridging OH groups of ca. 20 cm-l. This value agrees quite well with experimental results published recently in literature (ref. 8 ) . A s a third argument which supports our suggestion that bH can serve as a measure for the strength of acidity we refer to nonempirical quantum chemical calculations (ref. 2) although due to the approximations inherent in such a treatment it is not such a conclusive proof as the above mentioned experimental results. In Fig. 3 calculated shielding constants uH (connected to the chemical shifts SH by the equation uH = Gbare proton - GH) taken from ref. 2 are plotted in dependence on calculated deprotonation energies hEDp. Once more one observes a good correlation between both quantities.
93
20
-
+
"iO A E D ~ /kJrnd-'
-
Fig. 3 . Dependence of calculated values for the chemical shielding aH on calculated values for the deprotonation energy (ref. 2 ) . Summarized, Figs. 1-3 demonstrate clearly that bH is a suitable and sensitive spectroscopic quantity to measure the strength of acidity as defined by eq. ( 2 ) . In principle however, one should take into account that in these figures isolated hydroxyls of similar type are compared so that it is not clear whether there exists a generally valid correlation between 6H and EDp. Further experiments and calculations are necessary to answer this question which is of basic interest. In the H' MAS NMR spectra of dehydrated zeolites containing only oxygen, silicon and aluminium in the framework, in general five lines can be separated which have been denoted (refs. 3 and 9) as a, b, c, d, and e: Line a at 1.8 to 2.3 ppm is caused by non-acidic (silanol) OH groups. In the case of carefully dehydrated (200 OC, 10 K/h, lo-' Pa, 2 4 h) silica this line appears at somewhat lower values (1.65 ppm for aerosil 200/Degussa). A distinction between single and geminal OH groups however is not possible since from the single line shown in Fig. 4 (dehydrated sample) we must conclude
I 6Hsing1e -
6 Hgeminall 5 0.1
ppm
which is in agreement with infrared studies lfsingle - fgeminall but at variance with a former interpretation of Lippmaa et al. (ref. 10).
(6)
(7)
94
on air
Si 1.65 ppm 3.0ppm 2.0ppm
10
-61
5
0 H ppm
-5
10
-61
5
0
-5
H ppm
Fig. 4. H' MAS NMR spectra of carefully dehydrated (200 O C , 10 K/h, lo-' Pa, 24 h) silica (aerosil 200 1 Degussa) and after keeping it 30 s on air (ref. 11). The adsorption of water which interacts with SiOH groups gives rise to a shift of ca. 0.4 ppm to higher values of JH as can be seen from Fig. 4 (ref. 11). The H' MAS NMR signal of silanol groups in molecular sieves of SAPO-type may also appear at lower values of JH (cf. Fig. 5 ) .
S A PO-511
S A PO-512
f b)
'HMAS NMR
H
Si O A I Ibl SiOH
Fig. 5. H' MAS NMR (Bruker MSL 300) and infrared stretching vibration spectra (Digilab FTS-20) of two differently synthesized specimens of SAPO-5 (ref. 13).
3630 f/rm-'L-
-
3630 t/cm-1
95
Line b at 3.8 to 4.4 ppm is ascribed to acidic O H groups which are known to be of bridging type (SiOHA1). The value of JH increases with increasing silicon-to-aluminium ratio of the zeolite (cf. Fig. 2). Line c at 4.8 to 5.6 ppm is also ascribed to acidic OH groups of the bridging type but influenced by an additional electrostatic interaction of the hydroxyl proton presumably with neighbouring oxygen atoms of the six-membered ring (corresponding to the shift of the so-called LF-band in infrared spectroscopy). The fact that for the bridging OH groups pointing into the large cavities (HFband) and into the small cavities (LF-band) separate lines appear in the 'H MAS NMR spectra (lines b and c, respectively) excludes the possibility of a fast proton exchange among the four oxygens around an aluminium atom of the zeolite framework. Line d at 6.5 to 7.0 ppm is due to residual ammonium ions. Line e at 2.5 to 3.6 ppm represents hydroxyl groups associated with extra-framework aluminium species. Due to the limited space available for these OH groups their JH-value will be determined at least to a certain degree by an additional electrostatic interaction of the hydroxyl proton with other oxygen atoms. In accordance with this suggestion, for isolated AlOH groups the chemical shift is much lower and in the interval -0.5 ppm 5 bH 5 1 ppm (cf. Fig. 5 and ref. 12). For P O H groups the value of 6H is between 1.5 ppm and 4 ppm depending among others on interactions mentioned for line c.
Concentration of Bronsted Acidic Sites With respect to a measurement of the concentration of hydroxyl groups (Bronsted acidic and non-acidic sites) nuclear magnetic resonance spectroscopy has an extremely important advantage compared with infrared spectroscopy since the area of an H' MAS NMR signal is directly proportional to the concentration of the hydrogen nuclei contributing to this signal irrespective of their bonding state so that any compound with a known concentration of hydrogen atoms can be used as a reference (mostly water). In Fig. 5 H' MAS NMR (Bruker MSL 300) and infrared stretching vibration spectra (Digilab FTS-20) are shown for two differently synthesized specimens of SAPO-5 (ref. 13). While the positions of the various signals in the NMR and IR spectra correspond to each other quite well and are in agreement with IR results published recently (ref. 14), there are dramatic differences in the
96
relative intensities. Hence, even the relative intensity of an OH stretching vibration band cannot be taken as a measure for the concentration of the respective hydroxyl group in contrast to the NMR signal. Accessibilitv of Bronsted Acidic Sites The accessibility of hydroxyl groups can be easily determined through a study of the ' H MAS NMR spectra after loading the adsorbent with a suitable molecule which however must be fully deuterated in order to avoid an unwanted additional H ' NMR signal. Using deuterated pyridine the concentrations of accessible and non-accessible silanol groups of silica could be determined (ref. 15). With the same probe molecule it was also possible to show unambiguously that line (b) in the H' MAS NMR spectra of zeolites H-Y is due to OH groups which are easily accessible to pyridine (ref. 16) and that line (c) is caused by OH groups pointing into the small cavities (LF-band in infrared spectroscopy). Through this direct method it was possible to correct the formerly (ref. 17) postulated wrong correlation (LFband and line (b)). In Fig. 6 results are shown for a shallow-bed (400 OC) pretreated SAPO-5 unloaded, and after keeping it loaded with deuterated n-hexane for 1 hour at 5 0 O C . There is no doubt that the 3.9 ppm signal is caused by bridging OH groups which are easily accessible to n-hexane in contrast to the 4 . 9 ppm signal.
' MAS NMR spectra of a shallow-bed ( 4 0 0 OC) pretreated Fig. 6. H SAPO-5. a) unloaded, (b) after keeping the sample loaded with deuterated n-hexane for 1 hour at 50 O C .
97
So the higher value of JH must be due to the above mentioned
additional electrostatic interaction and we ascribe the signals at 4 . 9 ppm and 3.9 ppm to bridging OH groups located in the 6and 12-membered rings of the SAP0 - 5 structure. H'
NMR STUDIES ON HYDRATED ZEOLITES H' MAS N M R studies of hydrated zeolites may be complicated by a superposition of three effects: (i) Plank's mechanism, i.e. the adsorption and dissociation of water molecules on extra-framework multivalent cations like Ca2+ with a formation of bridging OH groups. The H' MAS NMR signal of the latter hydroxyls appears at JH values of ca. 0 ppm and 2 . 8 ppm for calcium ions in the large and small cavities of zeolites Y, respectively (ref. 18). It seems to be noteworthy that for these OH groups eq. ( 5 ) is not fulfilled and that a simple interpretation cannot be given at present for this experimental finding. (ii) Formation of hydroxonium ions at Bronsted acid sites which, however, take part in a fast proton exchange with physically adsorbed water molecules and OH groups. In general, it should be necessary to include in this exchange also water molecules hydrogen-bonded to Bronsted acid sites. These species for which the chemical shift JH is unknown could not be observed in the infrared spectra of rehydrated zeolites (ref. 19). It is possible to determine quantitatively the concentration of hydroxonium ions from the position of the H' NMR signal. In shallow-bed (400 OC) treated zeolites H-Y the probability to find a water molecule in the state of a hydroxonium ion is ca. 0 . 2 0.3 for a rehydration corresponding to one water molecule per bridging OH group (ref. MAS
-
20).
(iii) Adsorption of water molecules on Lewis acidic sites giving rise to a narrow line at JH = 6 , 5 ppm. In the case of hydrothermally pretreated zeolites H-Y (540 OC, 20 h, 4 kPa water vapour pressure) a concentration of (2 ? 0.5) Lewis acid sites of this type per unit cell could be found. Surprisingly the MAS sideband pattern of the signal at 6.5 ppm could only be explained quantitatively if these Lewis acid sites are not connected with extra-framework aluminium ions. Therefore, we assume that the signal at 6.5 ppm is caused by water molecules adsorbed on threefold coordinated and positively charged silicon atoms of the zeolite framework (ref. 20), i.e. on sites which where proposed by
98
Kazansky (ref. 21) in order to explain an infrared band at 4035 cm-’ for H2 adsorbed on a zeolite H-Y activated at 400 O C under deep-bed conditions. Further experimental and theoretical work seems necessary to prove whether threefold coordinated and positively charged silicon atoms in the zeolite framework really exist. PULSED FIELD GRADIENT (PFG) NMR SELF-DIFFUSION STUDIES Xenon Self-Diffusion In the last few years, xenon has turned out to be an efficient adsorbate for probing the pore structure and the internal surface of adsorbents by means of NMR spectroscopy (refs. 22-24). The advantage of xenon in comparison to other adsorbates is brought about by the large chemical shifts of ‘”Xe NMR as a consequence of the large electron shell, and by the fact that xenon as a noble gas leaves the adsorbent structure essentially unaffected. In particular, in zeolite research I2’Xe NMR has been successfully applied for probing pore and channel dimensions (ref. 25), cation distributions (ref. 26), reaction sites (ref. 27) and structural peculiarities (ref. 28). Following the pioneering work of J. Fraissard and coworkers, it is now applied in various laboratories (refs. 29-31). The advent of 12’Xe NMR in zeolite research has been accompanied by some controversy in the interpretation of the obtained spectra. In particular, on observing separate lines in the NMR spectra and attributing them to certain states or regions of the xenon atoms within the adsorbate-adsorbent systems, one has to imply that the rate of exchange between these states or regions is less than the difference in the Larmor frequencies of these lines (ref. 32). If the spatial separation of the respective regions is known, an estimate of the diffusivities of the xenon atoms within the sample becomes possible (ref. 33). It is obvious that in such cases the validity of the interpretation of the 12’Xe NMR spectra would be substantially supported if the diffusivities could be measured directly. As yet, owing to their large gyromagnetic ratios, pulsed field gradient NMR (PFG NMR) self-diffusion measurements of adsorbed molecules have been exclusively carried out with ‘H and ‘’F nuclei. However, experimental progress in the PFG NMR technique (ref. 34) recently enabled the application of 12’Xe NMR to a direct measurement of xenon self-diffusion in zeolites (ref. 35).
99
P1 250
.
Fig. 7 . Mean squ re displacements
of
Fig. 7 provides an example of the measurement of the mean square displacements of the xenon atoms in the three different zeolite types NaX, NaCaA and ZSM-5 in dependence on the observation timeA. For comparison, the data for methane in ZSM-5 are included. According to Einstein’s relation
each of the observed linear relations yields a value for the self-diffusion coefficient D. Fig. 8a shows the results of I2’Xe measurements in a representation of the thus determined selfdiffusion coefficients in dependence on the sorbate concentrations. The sorbate concentrations are given in atoms per 24 Si(A1)-atoms, corresponding to one channel intersection in the case of ZSM-5 and to one large cavity in the case of the X and A type zeolites. The error bars indicate the uncertainty in the self-diffusion coefficients resulting from the scattering of the experimental data of the mean square displacements. A comparison of the diffusivities of xenon and of methane presented in Figs. 8a and 8b, respectively, shows similar tendencies: For both adsorbates, the mobility in zeolite NaX is found to be higher
i
100
than in the other adsorbents. This corresponds to the fact that the diameters of the windows between the large cavities in the Xtype structure (0.75 nm) are considerably larger than the corresponding diameters in zeolite NaCaA (0.4 0 . 5 nm) or the channel diameters in zeolite ZSM-5 (0.55 nm).
...
a
1 0
I
0
.
.
2
3
2
1
.
. 4
.
.
6
4
.
.
8
atoms per 24 si (A1 )atoms
-
molecules per 24 Si ( A l l atoms
Fig. 8 . Concentration dependence of the intracrystalline selfdiffusion coefficient D. of xenon (Fig. 8a) and methane (Fig. 8b) in zeolite NaX (I),Na6$%CaA ( 0 ) , Na45%CaA ( 0 ) and ZSM-5 ( x ) at 293 K.
In zeolite A , the degree of calcium exchange determines the relative number of Ilopenll windows, i.e. the number of windows unblocked by cations, which thus are easily penetrable for the adsorbate molecules (refs. 36, 37). A numeric calculation given in ref. 36 yields that a reduction of the calcium content from 63% to 45% should reduce the diffusivity by a factor of about 2.
101
The experimental results collected in Table 1 (3rd and 6th column) are in satisfactory agreement with this prediction. Considering the concentration dependence of the diffusivities, for xenon and methane again identical tendencies are observed: While in zeolites NaX and ZSM-5 the diffusivities are found to decrease monotonously with increasing concentration, for zeolite NaCaA the diffusivities either remain constant or increase slightly with increasing concentration. It has been demonstrated in ref. 3 8 that concentration dependences of the first type are due to a reduction of the jump length (in NaX) or of the jump rate (in ZSM-5), while the latter behaviour results if mass transfer is controlled by the passage through an llactivatedll state as represented by the windows in the A type structure. Obviously, the similar size and shape of xenon and methane lead to coinciding dependences. Probinu Surface Barriers by 12’Xe PFG NMR Besides the possibility of a direct comparison with the llconventionalll I2’Xe NMR line shift analysis of the same system, the application of xenon in NMR diffusion studies may offer promising prospects in providing a more sensitive tool for probing the existence of surface barriers. These prospects are based on the expectation that (i) due to their larger kinetic diameter (ca. 0.49 nm, ref. 39) in comparison to methane (ca. 0.41 nm, ref. 39) the xenon atoms should be more affected by any narrowing in the zeolite pore systems, and that (ii) the reduced magnetic interaction of 12’Xe with the environment should allow a significant enhancement of the observation time. Fig. 9 represents typical plots of the signal intensity (spin echo amplitude) versus the width of the gradient pulses as observed in 12’Xe NMR pulsed field gradient experiments (ref. 40). Obviously, for zeolites NaX and ZSM-5 the obtained NMR data may be satisfactorily represented by the relation (ref. 34)
with Y denoting the attenuation of the NMR signal (spin echo attenuation) under the influence of the field gradient pulses of intensity g, width 6 and separationA.
102
I
a I
’
.
’
’
’
a2 0
0 OD4
L
.
’
0.2
o
(1 ’
55 ms
b
-6,,mSI
‘60 ms 02
1
zeolites NaX (Fig. 9a, crystallite diameter ca. 50 pm), ZSM-5 (Fig. 9b, crystallit5 dimensions 100 x 30 x 30 pm ) , and Na45%CaA
stands for the mean square displacement of the molecules during A , and T i s the gyromagnetic ratio of the considered nuclei. Thus, for the self-diffusion measurements of xenon in zeolites NaX and ZSM-5, the complete information turned out to be contained in the mean square displacements. In Figs. 10a and lob, these mean square displacements are represented in dependence on the observation time b For the A-type zeolites, the observed echo attenuations (cf. Fig. 9c) turned out to be represented by a superposition of two exponentials. One has to conclude, therefore, that the xenon atoms contributing to the NMR signal may be subdivided into two groups having different transport properties.
.
103
b
0
C
.
I
20
40
60
80
Alms
Fig. 10. Dependence of the mean square displacement according to eq. !9) on the observation time A for xenon in NaX (Fig. 10a) and ZSM-5 (Fig. lo%), and and of i, according to eq. (10) for xenon i NaCaA (Fig. lOc, Ca5+ content as indicated in Table 1).
Following the conception of the NMR tracer desorption technique (ref. 34), these two subgroups result very easily by distinguishing between those diffusants which remain in the interior of the individual crystallites over the whole observation time (i = intracrystalline diffusion) and those which may leave their crystallites and which are able, therefore, to cover large diffusion paths through the intercrystalline space (1.r. = long range diffusion). The echo attenuation is given therefore by
104
with i and l.r. representing the mean square displacements of the two subgroups, respectively, and with r(A) denoting the relative amount of the latter subgroup, i.e. the relative amount of atoms leaving the individual crystallites during the observation timed. For a discussion of the transport properties of the zeolite crystallites, an echo attenuation of the form of eq. (10) contains two types of information: the mean square displacement i of the atoms in the intracrystalline space as obtained from the slope of the second, slower decaying part of the 1n"f-vs.- J2 plot, and the quantity r ( h ) , resulting from the relative contributions of the two constituents. Both quantities are represented in Figs.10~and 11, respectively, in dependence on the observation time. It is obvious, that the function r(A) contains the same information as a conventional tracer desorption curve (ref. 3 4 ) . In previous NMR self-diffusion studies with methane, in general echo attenuations of the type of Fig. 9c have been observed (ref. 34). Such a behaviour has again been found on investigating the methane diffusivities in the zeolite specimens applied in this study.
r [A] 1.0-
Fig. 11. Relative amount r(A) of xenon atoms which have left their crystallites during the observation time A ("NMR tracer desorption curves11)for z olite NaCaA at content as 293 K, with indicated in Table 1. The magnitude of r ( 0 ) is determined by analysing the In y -vs- J~ plots as presented in Fig. 9c according to eq. (10): Extrapolating the second, less steeply decaying part to the ordinate yields the quantity 1-r( A )
Gas+
l3Pm
o
20
LO
60
eo
A,ms
.
For a comparison of the NMR tracer desorption curves with each other as well as with other transport related properties it is useful to introduce a mean intracrystalline life time
This definition coincides with that of the first statistical moment of an uptake curve (ref. 41). If the tracer exchange is controlled by intracrystalline diffusion it can be shown that
105
for crystallites of spherical shape ri is equal to
with cR2> and Di denoting the mean square radius of the crystallites and the intracrystalline self-diffusion coefficient, respectively (refs. 34, 41). Numerical values of the relation between Di and 7 7 for crystallites with the shape of parallelepipeds are given in ref. 53. According to Einstein's relation (eq. ( 8 ) ) which is valid for self-diffusion in homogeneous systems, the intracrystalline selfdiffusion coefficient Di may be easily determined from the slope of the representation of the mean square displacement versus the observation time A , provided that the mean square displacements cr2(b)> are still smaller than the mean square radii of the crystallites. TABLE 1 Values for the intracrystalline self-diffusion coefficients Di and the intracrystalline mean life time ri of xenon and methane in various specimens of zeolites NaCaA, NaX and ZSM-5 at 293 K and a sorbate concentration of about 3 methane molecules or xenon atoms per 24 Si(A1) atoms. Adsorbents and the mean c r y s t a l l i t e diameters i n lun
Xeno% D.
T.
/ms
Ims Y c o n t
45 X 63 X 80
x
Wax
.
13 20 5
1.0t0.3 1.5t0.4 1.5t0.4
3.0tl.0 4.0t1.0 0.4t0.2
80t2O 45t10 252 8
1.0t0.3 1.5t0.4 1.520.4
3.0tl .O 6t2 4.0tl .O 4t2 0.4t0.2 lt0.5
50
5.0t1.5 5.0t1.5
8.0t2.O
1.5t0.5
15t10 5t3
20t5 20t5
2tl 5tl 0.4t0.2 2t1
O.Pt0.3 0.Pt0.3
11t3 160t40
,40 >70
4tl 4t1
20 ZSM-S(a) ZSM-5(b )
25 * 100~30x30
3tl 4t1 45t10 ~ 3 0
mean c r y s t a l l i t e dimension
Table 1 gives a summary of the self-diffusion coefficients thus determined for sufficiently short observation times from the
106
presentations of Fig. 10 and from corresponding results for methane. Furthermore included are values for the intracrystalline mean life time ri and the quantity riD calculated from Di and < ~ 2 >via eq. (12). For those adsorbents (ZSM-5 and NaX) which did not reveal a two-phase behaviour in the In Y -vs- 62 plots, the values for the intracrystalline mean life time were estimated by realizing that the rms displacements can only exceed the mean crystallite radii by a noticeable value if during the observation time h a substantial part of the adsorbate molecules or atoms have left their crystallites, i.e. if the observation time A is at least of the order of the magnitude of the intracrystalline mean life time r i . Thus, for NaX the intracrystalline mean life times have been put equal to those observation times, for which the mean square displacements coincided with the mean square radii. For both ZSM5 specimens even for the largest observation times the mean square displacements remained smaller than the crystallite dimensions, so that ri must be assumed to be larger than the maximum observation time A , Direct information about the existence of surface barriers is provided by a comparison between the values for the intracrystalline mean life time ri and the quantity riD which represents the minimum possible life time calculated via eq. (12) from the intracrystalline diffusivities. Table 1 shows that for methane these two quantities are in reasonable agreement. One has to conclude therefore that molecular exchange is mainly controlled by intracrystalline diffusion and that, consequently, possible surface barriers can only be of minor importance. This result is in agreement with the finding of previous studies (ref. 3 3 ) that on applying methane as a probe molecule, in zeolite NaCaA surface barriers do only occur after coking or hydrothermal pretreatment. It appears from Table 1, that xenon reveals a completely different situation: In all NaCaA specimens as well as in the considered ZSM-5 specimen with the smaller crystallites, xenon desorption is significantly retarded by influences different from intracrystalline diffusion, i.e. by an additional transport resistance on the external surfaces or in a surface layer of the crystallites. This result may be related to the larger kinetic diameter of the xenon atoms, making them suitable for probing the surface permeability of adsorbents with limiting
107
...
free diameters of this order of magnitude (0.4 0.5 nm for NaCaA, ca. 0.55 nm for ZSM-5). It is not unexpected that for zeolite NaX, also with xenon no surface resistances are observed, since the increase of the diameter of the probe particle in comparison to methane is still insignificant compared with the limiting free diameters of NaX (ca. 0.75 nm). The experimental finding that the existence of a surface barrier may depend on the sort of the diffusing particles reflects an essential feature of the surface barrier: Though being determined by structural or energetic peculiarities at the crystallite surface (cf. ref. 43), the surface barrier can only be defined in connection with a certain probe molecule. This statement becomes self-evident on considering the reciprocal of the surface barrier, the surface permeability which - as the diffusity - must be a function of the diffusant. It is possible that the low literature data for intracrystalline diffusion of xenon in zeolite NaCaA deduced from uptake measurements (lXlO-I4 m2/s (ref. 44), 1.2x10-'' m2/s (ref. 4 5 ) or 8x10-'I m2/s (ref. 46), determined for a loading of about 1 atom per cavity and at temperatures of about 300 K) are caused by the significant surface barriers for this system. Diffusion Anisotrovv Among the problems connected with new zeolite structures, a possible anisotropy of self-diffusion deserves special consideration. Since the PFG NMR method allows the observation of molecular migration with respect to the direction of the applied field gradient (ref. 3 4 ) which may be easily changed by applying suitable gradient coil arrangements, favourable conditions exist for the investigation of diffusion anisotropy. Literature provides various examples of such diffusion studies with liquid crystals and membranes where molecular diffusion proceeds in a well oriented sample of macroscopic dimensions. A completely different situation exists if one considers molecular diffusion in crystalline catalysts and adsorbents. Being of the order of a few micrometers, the diameters of the individual crystallites are too small to allow the conventional method of studying diffusion anisotropy. However, also in this case PFG NMR may provide information about the extent of diffusion anisotropy. This information is contained in the shape of the N M R spin echo attenuation versus the gradient intensity, which in the case of isotropic diffusion yields a simple
108
exponential dependence (cf. eq. (9)), but which in the case of anisotropic diffusion may be shown to obey the relation (refs. 34, 47) ~ ( c3,,Y)= exp {-C(D,~ cos2y sin’>
+ D sin’ 7 sin2nt + DZZcos2a ) 1 ,
where the quantities DXX, Dyy and D,, denote the principal values of the diffusion tensor. The orientation of the field gradient with respect to the principal axes system is represented by the angles &and Y The quantity
.
is determined by the parameters of the pulse sequence, explained in connection with eq. (9). In the case of powder samples, i.e. of randomly oriented crystals, averaging over all possible directions leads to the expression
c Y(C)>
=
1
J1exp{471 -1
c
with I ( J ) = J2;xp{-c 0
sin22
As an example, Fig. 1 2 shows the result of a numerical
calculation of eq. (15) for an axial symmetric diffusion tensor (ref. 47). It s obvious, that for D,,>D,, the sensitivity of the PFG NMR method with respect to deviations from isotropic diffusion is h gher than in the reverse case. Fig. 13 compares the echo attenuation obtained in PFG NMR measurements of methane diffusion in ZSM-5 (ref. 47) with the theoretical dependence calculated from eq. (15) with the tensor elements DXX (= 1.22 lo-’ m’s-l), Dyy (= 14.9 10-9m2s-1) and D,, (= 1.22 10-9m2~-1) as determined in recent Molecular Dynamics calculations (ref. 48). It turns out that the calculated pattern is already beyond the range compatible with the experimental behaviour, so that one has to conclude that the influence of anisotropy is overestimated in the Molecular Dynamics calculations.
109
Fig. 12. Spin echo attenuation due to anisotropic diffusion in randomly oriented crystallites with axial symmetry for D , , > , D L and D , , c DL as a function of the product of the field gradient intensity parameter C defined by eq. (14) and the quantity =(Dxx+Dyy+Dzz)/3.
Fig. 13. Spin echo attenuation of the proton NMR signal of methane adsorbed in an assembly of ZSM-5 crystallites at 23 OC ( 0 ) I and comparison with the theoretical curve- b ai ed from eq. (15) with bhs tfnsor elements Dxx = 3 59x10 8m s , D = 14.9~10-m sand D,, = 1.22x10-9m2s-1 (ref. 48) ( - - - I rxgpectively.
5 -7
I
Limits of ADwlication The versatility of the PFG NMR method to study diffusion processes in zeolites and its successful application to numerous problems in zeolite science and technology (cf. ref. 42) implies the temptation to apply it also under such conditions where
110
extreme caution is recommended. This is in particular true for heterogeneous or, in general, multi-phase systems, where the contribution of the various phases to the spin echo is controlled by their nuclear magnetic relaxation properties (ref. 52) or, with other words, where with increasing observation time A the contribution of the phases with the largest relaxation times will become more and more dominant. So it may happen that by formally applying the PFG NMR method one determines the diffusivity of a trifling amount of very mobile molecules (occurring e.g. in tracks within the zeolite crystallites or on their outer surfaces) rather than that of the phase represented by most of the adsorbed molecules. It is most likely that the surprisingly high values for the diffusivity of benzene in silicalite which were recently determined using PFG NMR (ref. 49) are due to this effect. In order to exclude the possibility of such artifacts in NMR self-diffusion measurements, a modification of the traditional rf pulse sequence has been proposed (ref. 50). It allows an unambiguous check whether the obtained diffusivities are in fact unaffected by the influence of a trifling amount of very mobile molecules. NMR Tracer Exchanse The lowest diffusivities accessible by PFG NMR in zeolites are of the order of m2s-' (refs. 34, 42). However, for systems with diffusivities below this limit an alternative NMR method for studying mass transfer may be applied (ref. 52). Under such conditions and for crystallite diameters k l 0 pm, the rates of molecular adsorption or desorption become small enough so that it is possible to study these processes by an analysis of the intensity of the NMR signal. As an example, Fig. 14 shows the arrangement of such an experiment for the investigation of the rate of molecular exchange for benzene in H-ZSM-5 with the surrounding atmosphere (refs. 52, 53). The exchange experiment is started by crushing the glass vial using the glass cylinder as a striker: The hydrogen form of benzene initially adsorbed on the zeolite will then be exchanged by the deuterated compound stemming from the gas phase.
111 gaseous C6D6 adsorption vessel granulated NaX +C6D6
20 mm glass vial H - Z S M - 5 .CgH(
Fig. 14. Experimental arrangement for the NMR tracer exchange measurements.
D Fig. 15. Comparison of the self-diffusion coefficients obtained from NMR tracer exchange measurements
\
\
\
i\ f \
\
( 0 293 K;
A 386 K)
with values calculated from sorption uptake experiments ( o 303 K; 0 363 K; A 393 K) using eq. (17). Both experimental series have been carried out with the same silicalite specimens.
f
10Eo
1.0
Loading I molecules per 0.25 U.C. This exchange process may be directly recorded via the intensity of the H' NMR signal. It is shown in Fig. 15 that the selfdiffusivities determined from these experiments are in satisfactory agreement with the results of uptake experiments, when the relation between the coefficient of diffusion (Dd) which
112
follows from the uptake experiment and the coefficient of selfdiffusion (D) is assumed to be (ref. 54)
with c(p) denoting the sorbate concentration in equilibrium with the sorbate pressure p. Clearly, the experimental arrangement as shown in Fig. 14 may be easily modified to allow both adsorption or desorption experiments (in this case, the sorbate pressure in the surrounding atmosphere must be much greater or much less than over the adsorbent within the glass vial) and counter diffusion experiments. Taking advantage of high resolution NMR spectroscopy (ref. 5 5 ) , such experiments (e.g. the observation of the exchange between aliphatic and aromatic compounds) may even be carried out without the necessity of applying deuterated compounds.
TABLE 2 Comparison of PFG NMR diffusivities with the results of quasielastic neutron scattering and Molecular Dynamics simulations as well as with the results of macroscopic non-equilibrium measurements (uptake, zero length column, permeation). Data refer to the limit of small sorbate concentrations, the unit is m2s-'. Adsorbent
Tewperature/K
Neutron Scattering
NaCaA
NaX
nethane ethane
ethylene benzene
300
250
300
H-ZSM-5
458
293
~ x I O -4~~ 1 0 . ~ r e f . 67 r e f . 35
5x10-" r e f . 59
S X ~ O - ~ r e f . 68
3xlO-l' r e f . 62
7xlO-l' r e f . 64
=5x10-9 2x10-9 r e f . 48 ref.-$O 4x10 r e f . 71 3xlO-l' r e f . 65
Uptake
Permeation
250
-~ los9 6 ~ 1 0 . ~ 1~. ~ X I D 8x1D-10 r e f . 61 r e f . 63 ef. 37 r e f . 37
Molecular Dynamics
Zero length colum
methane xenon
3 . ~ 1 0 ~ ~ ~ r e f . 40
8~10-l~ r e f . 46
8 ~ 1 0 ~ ' ~ r e f . 66 10.10 334 K r e f . 69
113
ComDarison of PFG NMR Diffusivities with the Results of other Technisues The comparison of the diffusivities determined in PFG NMFt measurements with the results of other techniques has concerned zeolite workers over more than a decade (refs. 56, 57). While there is still some controversy whether in general the results of uptake (i.e. of non-equilibrium) measurements after applying eq. (17) should or should not agree with the NMR self-diffusivities (cf. ref. 52), in the last few years by other microscopic techniques (neutron scattering, 2H NMF2) as well as by Molecular Dynamics calculations a large amount of diffusivity data has been provided which are in remarkable agreement with the results of PFG NMR measurements. Some of these results are compared in Table 2.
REFERENCES 1
J.E. Bartmess, J.A. Scott, R.T. Mc Iver, J. Am. Chem. SOC.
2 3 4
J. Sauer, J. Mol. Cat. 54 (1989) 312.
101 (1979) 6046.
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
H. Pfeifer, Colloids and Surfaces 36 (1989) 169. R.T. Sanderson: Chemical Bonds and Bond Energy, Academic Press 1976. D. Freude, M. Hunger, H. Pfeifer, W. Schwieger, Chem. Phys. Lett. 128 (1986) 62. E. Brunner, J. Chem. SOC., Faraday Trans. (in press). E. Brunner, D. Freude, B.C. Gerstein, H. Pfeifer, J. Magn. Res. (in press). J. Dwyer, J. Dewing, N.E. Thompson, P.J. O'Malley, K. Karim, J. Chem. SOC., Chem. Comm. (1989) 843. H. Ernst, Z. Phys. Chenie, Leipzig, 269 (1988) 1073. E. Lippmaa, A.V. Samoson, V.V. Brey, Y.I. Gorlov, Doklady Akad. Nauk 259 (1981) 403. E. Brunner, unpublished. U.M. Mastikhin, I.L. Mudrakovsky, A.V. NOSOV, Bruker Report 2 (1989) 18. M. Hunger: Thesis (Prom. B) in preD. S. Halik, J.A. Lercher, H. Maver, Chem. SOC., Faraday - . Trans. 1; 84 (1988) 4457. H. Rosenberger, H. Ernst, G. Sche er! I. Junger, R. Sonnenberger, 2. Phys. Chemie, Leipzig 263 (1982) 864. D. Freude, M. Hunger, H. Pfeifer, W. Schwieger, Chem. Phys. Lett. 128 (1986) 62. H. Pfeifer, D. Freude, M. Hunger, Zeolites 5 (1985) 274. M. Hunger, D. Freude, H. Pfeifer, D. Prager, W. Reschetilowski: Chem. Phvs. Le t. 163 (19891 221. A. Jentys, G. Warecka, M. Derewinski, J.A.' Lercher, J. Phys. Chem. 93 (1989) 4837. M. Hunger, D. Freude, H. Pfeifer, J. Chem. SOC., Faraday Trans. in press. V.B. Kazansky, Catalysis Today 3 (1988) 367. J. Fraissard, T.Ito, in: Proc. 5th Internat. Conf. Zeolites, Naples 1980, Heyden, London 1980, p, 510.
>.
114 23 24 25 26 27 28 29 30 31 32
J. Rippmeester, J. Am. Chem.. SOC. 104 (1982) 289. J. Fraissard, T. Ito, Zeolites 8 (1988) 350. J. Demarquay, J. Fraissard, J. Chem. Phys. Lett. 136 (1987) 314. T. Ito, Fraissard, J. Chem. SOC., Faraday Trans. 1, 83 (1987) 451. J. Fraissard, T. Ito, L.C. de Menorval, in: Proceedings of the 8th Internat. Congress on Catalysis Berlin, 1984, p. 111.
.
T. Ito, L.C. de Menorval, E. Guerrier, J. Fraissard, Chem. Phys. Lett. 111 (1984) 3271. E.W. Scharpf, R.W. Grecely, B.C. Gates, C. Dybowski, J. Phys. Chem. 9 0 (1987) 9. R. Shoemaker, T. Apple, J. Phys. Chem. 9 1 (1987) 4024 B.F. Chmelka, R. Ryoo, S.-B. Liu, L.C. de Menorval, J. Am. Chem. SOC. 110 (1988) 4465. H. Winkler, D. Michel, Adv. Coll. Interf. Sci. 23 (1985) 149.
33 34
J. Karger, H. Pfeifer, J. Fraissard, Z. Phys. Chemie, Leipzig 195 (1987) 268. J. Karger, H. Pfeifer, and W. Heink, Adv. Magn. Res. 12 (1988) 1.
35 36 37
W. Heink, J. Karger, H. Pfeifer, and F. Stallmach, J. Am. Chem. SOC. 112 (1990) 2175. D.M. Ruthven, Canad. J. Chem. 52 (1974) 3523. J. Caro, J. Karger, G. Finger, H. Pfeifer, and R. Schollner,
Z. Phys. Chem., Leipzig, 257 (1976) 903. 38
H. Pfeifer, J. Karger, A. Germanus, W. Schirmer M. Bulow, and J. Caro, Ads. Science Techn. 2 (1985) 229.
39
Landoldt-Bornstein IIZahlenwerte und Funktionenll, Band I, Teil 1, p. 325, 370, Springer, Berlin, Gottingen, Heidelberg, 1950. J. Karger, H. Pfeifer, F. Stallmach, and H. Spindler, Zeolites 10 (1990) 288. R.M. Barrer IIZeolites and Clay Minerals as Sorbents and Molecular Sievest1,Academic Press, London, 1978 J. Karger, H. Pfeifer, Zeolites 7 (1987) 90. M. Kocirik, P. Struve, K. Fiedler, M. Billow, J. Chem. SOC. Faraday Trans. 1, 89 (1988) 3001. D.M. Ruthven, R.I. Derrah, J. Chem. SOC., Faraday Trans. I
40 41 42 43 44
7 1 (1974) 2031. 45 46 47
M.M. Dubinin, A.K. Gorlow, A.M. Voloshchuk, in: Proc. 5th Internat. onf. Zeolites, Naples, 1980, Heyden, London, 1980, p. 554. M. Bulow, U. Hartel, U. Muller, K.K. Unger, Ber. Bunsenges. Phys. Chem. 94 (1990). B. Zibrowius, J. Caro, J. Karger, Z. Phys. Chem., Leipzig, 269 (1988) 1001.
48 49
P. Demontis, E.S.Fois, G.B. Suffritti, S . Quartieri, J. Phys. Chem. 94 (1990) 4329. M. Morgan, T. Cosgrove, R. Richardson, Coll. Surf. 36 (1989) 209.
50 51 52 53 54
W. Heink, J. Karger, H. Pfeifer, Z. Phys. Chem., Leipzig, (in press). C. Forste, W. Heink, J. Karger, H. Pfeifer, N.N. Feoktistova, S.P. Zhdanov, Zeolites 9 (1989) 299. C. Forste, J. Karger, H. Pfeifer, Proc. 8th Int. Zeolite Conf., Amsterdam, Elsevier, 1989, p. 907. C. Forste, J. Karger, H. Pfeifer, L. Riekert, M. Bulow, A . Zikanova, J. Chem. S O C . Faraday Trans. 1, 86 (1990) 881. D.M. Ruthven, IIPrinciples of Adsorption and of adsorption Processes" Wiley, New York, 1984
-
115 55
H. Pfeifer, W. Meiler, D. Deininger, Ann. Rep. NMR Spectr.
56 57
D.M. Ruthven, Am. Chem. SOC. Symp. Ser. 40 (1977) 320. J. Karaer. J. Caro, J. Chem. SOC., Faraday Trans. 1, 73
58 59
J. Karger, D.M. Ruthven, Zeolites 9 (1989) 267. E.Cohen de Lara, R. Kahn, F. Mezei, J. Chem. S O C . , Faraday Trans. 1, 79 (1983) 1911. Z . Xu, personal communication. A. Germanus, J. Karger, H. Pfeifer, Zeolites 4 (1984) 188. C.J. Wright, C. Riekel, Mol.Phys.36 (1978) 695. J.Karger, D.M.Ruthven, J.Chem.Soc., Faraday Trans. 1,
15 (1983) 292. (1977)-1363.
60 61 62 63
77 (1981) 1485. 64 65 66 67 68 69 70
71 72 73
H. Jobic, M. Bee, J. Karger, H. Pfeifer, J. Caro, J. Chem. S O C . , Chem. Commun. (1990) 341. M. Bulow, W. Mieth, P. Struve, P.Lorenz, J. Chem. SOC. Faraday Trans. 1, 79 (1983) 2457. M. Eic, M. Goddard, D.M.Ruthven, Zeolites 8 (1988) 327. J. Caro, M.Bulow, W.Schirmer, J.Karger, W.Heink, H.Pfeifer, S.P.Zhdanov, J. Chem. SOC. Faraday Trans. 1, 85 (1989) 4201 H. Jobic, M. BBe, J. Caro, J. Karger, M. Billow, J. Chem. SOC. Faraday Trans. 1, 85 (1989) 4201. D.T. Hayhurst, A.R. Paravar, Zeolites 8 (1988) 27. S.D. Pichett, A.K. Nowak, J.M. Thomas, B.K. Peterson, J.F.P. Swift, A.K. Cheetham, C.J.J. den Ouden, B.Smit, M.F.M. Post, J. Phys. Chem. 94 (1990) 1233. R.L. June, A.T. Bell, D.N. Theodorou, J. Phys. Chem., in press. J.P. Chauvel, N.S. True, Chem. Phys. 9 5 (1985) 435. R.W. Taft, I.A. Koppel, R.D. Topsom, F. Anvia, J. Am. Chem. SOC. 112 (1990) 2047.
This Page Intentionally Left Blank
G . Ohlrnann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
117
SPECTRAL STUDY OF LEWIS ACIDITY OF ZEOLITES AND OF ITS ROLE IN CATALYSIS
V .B .KAZANSKY N.D.Zelinsky Institute of Organic Chemistry, Moscow, USSR SUMMARY
The Lewis acidity of cationic forms of zeolites and of those modified by mild steaming or by Zn and Ga ions was studied using molecular hydrogen and light paraffins as molecular probes. The extent of their perturbation was controlled by IR diffuse reflectance spectroscopy. It was concluded, that Lewis sites should be considered not as low coordinated cations, but rather as acid-base pairs in which the neighboring basic oxygen is equally important. The role of extra lattice aluminum or modifying Zn and Ga ions in cracking of paraffins or their aromatization is discussed. INTRODUCTION The catalytic properties of zeolites are mainly connected with
their aci-
dity. Therefore the central problem of catalysis on zeolites is the nature of their acid active sites. The recent information on the Broensted acidity of zeolites is rather complete, since it could be easily followed by such direct spectroscopic methods as IR or high resolution solid state NMR. On the contrary the knowledge of the nature and the properties of Lewis acid sites is quite not as full. This is because they could be investigated only by indirect way by means of molecular probes with spectral control of perturbation of adsorbed molecules. For this purpose most often ammonia or pyridine adsorption is used. These
molecules have however very little in common with hydrocarbons, whose catalytic transformations on zeolites are of
the greatest
interest. In other
words, if adsorbed ammonia is protonated by a Broensted acid site or is strongly bonded by a Lewis site, this does not yet mean, that the adsorbed hydrocarbons would behave in the similar way. Another, more principal difference between strong bases and hydrocarbons consists in the chemical nature of their interaction with Lewis acid sites. In the case of hydrocarbons or hydrogen the elementary step, which is most often discussed, is the heterolytic dissociation: 6- 6 H--H
-Me--0- + H
2
--.c
-Me--O-
+
H- H+ ----L
-Me--O-
(1)
Unlike this ammonia or pyridine adsorption results in coordinative binding:
118
-Me--0- + NH3
c_)
(2)
Y3 -Me--0-
Thus, the strong bases are the tests only for Lewis sites, whereas the molecu- lar hydrogen is rather the test for the acid-base ion pairs. Therefore the conclusions on the nature, the strength and the number of Lewis sites obtained from adsorption of strong bases could not be
so
simply transferred
to
adsorption of hydrocarbons. Of course the best molecular probes for Lewis acidity themselves. Therefore in the previous studies of aluminum
are
the reactants
oxide and H-zeo-
lites we used the adsorption of light paraffins or the low temperature adsorption of molecular hydrogen (refs. 1.2). Below the application of
the similar
approach for the investigation of Lewis acidity of zeolites modified by various methods is reported. CATIONIC FORMS OF ZEOLITES
Zodlurn-Errns The univalent cations or multivalent cations introduced by ion exchange represent the simplest examples of Lewis acid sites in zeolites. Therefore already in the very first papers on their catalytic application the probable role of such active sites was postulated and discussed (ref.3). According to these works the cations could activate adsorbed molecules due to their electrostatic polarization. The strength of electric field created by cations and of the degree of such polarization have been however never estimated experimentally. Some ideas on this point could be drawn out from our
re-
sults on low temperature molecular hydrogen adsorption (ref.2,4). IR spectra of molecular hydrogen adsorbed at 77
K and the pressure of 2-5
Torr on sodium forms of different zeolites are presented in Fig.1. They that low temperature adsorption of hydrogen
show,
results in the appearance of se-
veral bands in the region of H-H bond stretching vibrations. This indicates the existence of various types of adsorption sites, which seem
to be
sodium
ions in different positions of the zeolite framework. They are able to polarize adsorbed hydrogen molecules in various extent, resulting in the appearance of several absorption bands. When the spectra of adsorbed hydrogen contain more than one line, then the decrease of hydrogen pressure or evacuation at 77 K result
in the following
change of their relative Intensities: the high frequency bands at w
=
4120-1 4150 cm-l disappear , whereas those at lower frequencies of w = 4080-4110 cm
remain unchanged. This could be explained by the different strength of hydro-
119 gen binding by cations located in various positions of the zeolite framework.
It correlates with the values of low frequency shifts of the H-H stretching vibrations.
Fig.1 Diffuse reflectance IR spectra of molecular hydrogen adsorbed on sodium forms of zeolites at 77 K at various pressures. P=5 Torr:(b)-NaM,(c)-NaZSM-5, (e)-NaA, (f)-NaX, (g)-NaY. P=100 Torr: (a)-NaM, (d)-NaA, (h)-NaY.
At low pressure the spectra of pentasils and mordenites are the simplest (Fig.1. spectra (a)-(c) 1. For mordenite they consist of a single band at 4108 -1 cm , which is shifted by 55 cm-l towards lower frequencies as compared with the stretching vibrations of free hydrogen in gas phase at 4163 cm-l. This is consistent with conclusion of ref. 5, that in mordenite Na cations are preferentially located at one main type of sites at the centers of eight membered rings. At higher pressure of 100 Torr (spectrum (a) 1 an additional shoulder -1 of lower intensity appears near 4120 cm . It may be connected with sodium cations distributed among other available sites of localization. For NaZSM-5 zeolite there is no published data on Na cations distribution. Our results show, that there is only one type of sodium in its framework (Fig.1, spectrum (c) 1. The corresponding stretching frequency shift of hydrogen is very close to that of hydrogen adsorbed by mordenite. Therefore it
is
highly probable, that in NaZSM-5 the sodium ions are also localized at
the
similar positions. According to X-ray analysis, in NaA 8 of the overall number of
12 sodium
ions are located near the centers of the six membered rings, which form the sodalite cages. The 3 of remaining 4 Na cations are situated near the centers of 8-rings and one is statistically located over 12 fold equipoints (ref.6).
120
This is also consistent with the spectra (d) and (el of Fig.1, which show, that the majority of Na ions bind hydrogen more strongly and exhibit the
lar-
ger low frequency shifts of H-H stretching vibrations, while the rest of sodium ions could hold and perturb molecular hydrogen only at higher pressures. The spectra of hydrogen adsorbed on sodium forms of faujasites are more complicated (Fig.1, spectra (f)-(h) 1. According to our results (ref.4) they could be ascribed to H molecules perturbed by sodium ions located in SI; SI, 2 and in SIIpositions. Thus, low temperature molecular hydrogen adsorption allows to follow both the different sites of sodium ions localization in
the
zeolite frameworks and the distribution of ions among these sites. The important feature of IR spectra of hydrogen adsorbed by sodium forms of zeolites is the dependence of the line shifts on the type of
zeolite frame-
works. In this respect hydrogen behaves quite differently from adsorbed carbon monoxide (ref.7) which for the same cations exhibits very
close shifts for
different frameworks. Thus, the perturbation of adsorbed hydrogen depends not only on the nature of cations, but also on the type of
the zeolite crystal
lattice. This is in consistence with the by-point adsorption of molecular hydrogen, which was earlier supported by quantum chemical calculations for aluminum oxide (ref.8). If this conclusion is also correct for zeolites, then for their sodium forms the shifts should be dependent on the basic properties of oxygen. This is really observed experimentally. Indeed, the least basic oxygen is that
one
in high silica zeolites, which contain the lowest amount of aluminum in their framework. This results in the smallest low frequency shifts of H2
stretching
vibrations. On the contrary the largest shifts were observed for NaA zeolite, which contains the most basic oxygen. However, even in this case the value of the low frequency shift is only 90 cm-l i.e. is about twice less, than for hydrogen adsorbed by acid-base pairs of aluminum oxide sodium ions are rather weak perturbing sites of adsorbed
that
(ref.1). Thus,
hydrogen. This is
consistent with the fact, that contrary to aluminum oxide we never observed dissociative hydrogen adsorption on sodium forms of zeolites.
Bllralent-catlonlr-~orms_oy_qeoliteq The IR spectra of hydrogen, adsorbed at 77 K on alkaline earth forms of mordenite are presented in Fig.2. They contain two bands for Mg
and Ca
exchanged samples, but only single bands for the samples exchanged with Sr and Ba ions. Such difference could be easily explained by various hydrolysis extent of these cations.
121
J
Q
i C
d Fig.2. IR diffuse reflectance spectra of molecular hydrogen adsorbed at 77 K on alkaline earth forms of mordenite dehydrated at 720 K. (a)-MgNaM, (blCaNaM,(c)-SrNaM,(d)-BaNaM. The degree of cationic exchange is about 60 %. Hydrogen pressure is 30 Torr. Indeed,the second dissociation constants of (MgOHl' and
(CaOH)'
rather low (3'10-3 and 4'10-2 respectively). On the contrary for (BaOH)' cations they are much higher
(
ions are (SrOH)'
and
about 0.15-0.21. Therefore after dehy-
dration at elevated temperature magnesium and calcium ions exist in the zeolite framework in two forms: as bivalent Me2'
cations and as hydrolyzed MeOH'
species. On the contrary Sr and Ba cations are completely dehydroxylated resulting in doubly charged ions. Such an interpretation is supported by IR spectra of these samples in OH stretching region. For dehydrated Mg and Ca forms they contain both the bands from structural OH groups and those from basic hydroxyls of MeOH' On the contrary removal of water from S r ' 2
fragments.
+2
and Ba containing samples results
in complete dehydration and only the bands of silanol groups are observed. The comparison of results on hydrogen adsorption on sodium and earth forms of mordenite shows, that for Na'
and MeOH'cations
alkaline
the shifts are
very close. This also supports the above interpretation based on hydrolysis of M g ' 2
and Ca'2
cations, which behave in hydrolyzed form similar to univalent
ions. More surprising are the small shifts exhibited by nonhydrolyzed bivalent Sr"
and Ba"
cations, which are only slightly larger, than for sodium. In our
opinion, this is consistent with the above idea of bypoint adsorption of
122 hydrogen. If it is really of the bypoint nature, then the charge and
the polarizing
ability of lattice oxygen is also important. In other words the charges of cations and their dimensions are not the only factors affecting polarization of adsorbed hydrogen. In addition, for bivalent cationic forms the higher positive charge results in the partial electron density transfer from basic
oxygen
to cations. This decreases both positive charges of cations and those of
oxy-
gen in ion pairs and results in smaller shifts of stretching vibrations of adfrom
sorbed molecular hydrogen in comparison with which one could expect
the
formal electric charges. DEHYDROXYLATED HYDROGEN FORMS OF ZEOLITES Historically, one of the first mechanisms of dehydroxylation of H-faujasites was suggested by Uytterhoeven, Crystner and Hall (ref. 91. They postulated the formation of trigonally coordinated A1 and Si atoms, when
two hydroxyl
groups are removed from the zeolite framework:
-
H 0 \
2
/ \
Si /
\ /
\
/
A1 \
+
Si
d
/
\ /
/
A1 \
\
0 / \
Si
+ /
/
A1
\ /
(31
\
Later this mechanism was found to be incorrect, since according to the numerous solid state high resolution NMR data
the dehydroxylation of
hydrogen
forms of faujasites results instead of formation of lattice Lewis sites in dealumination (ref.101. Nevertheless our experimental results on low temperature molecular hydrogen adsorptioii clearly show, that f o r zeolites dehydroxylation still could occur according
high silica containing to
scheme
(31 (refs.
2,111.
IR spectra of H2 adsorbed at 77 K on HZSM-5 zeolite preheated in vacuo at different temperatures are presented in Fig.3. At moderately high
temperatu-
res, when dehydroxylation has not yet started (770K) , two bands with
maxima
at 4125 and 4105 cm-l are observed (fig.3 (a) 1. With the increase of the pretreatment temperature and of the extent of surface dehydroxylation their intensity decreases. On this ground we assign these bands to hydrogen interacting with surface hydroxyl groups. Since the position of the high frequency band is the same as when hydrogen is adsorbed on silica, this band could be attributed to the molecules, perturbed by silanol groups. Then the band with a lower frequency of 4105 cm-1 should be ascribed to hydrogen molecules interacting with the acidic bridged
hydroxyls with
stretching
123 vibrational frequency of 3610 cm-1
.
Fig. 3. IR spectra of hydrogen, adsorbed at 77 K on HZSM-5 zeolites dehydroxylated in vacuo at various temperatures: (a) - 770 K. (b) - 970 K. (c) 1220 K. (d) - HY zeolite preheated at 770 K under "deep bed" conditions. If the pretreatment temperature of HZSM-5 zeolite is increased up to 910 K
its partial dehydroxylation takes place. This results in the decrease of
the
bands intensity of the bridged hydroxyl groups. Simultaneously two new bands from adsorbed hydrogen with maxima at 4010 and 4035 cm-l and values of low frequency shifts of 125 and 150 cm-l
the very
are appearing
large
(Fig.3b).
They should be attributed to adsorbed hydrogen interacting with two different kinds of Lewis acidic sites, which are formed after removal of hydroxyl groups in accordance with scheme ( 3 ) . There are also the following additional arguments in favor of ment of these bands to hydrogen molecules adsorbed on
the assign-
trigonally coordinated
Si and A1 atoms of the zeolite crystal lattice. Firstly, if the dehydroxylation of H-high silica zeolites is carried out under still more
severe condi-
tions and the dealumination of their crystal lattice really takes place, one more low frequency band cm-'appears
of adsorbed hydrogen with
the maximum at
4060
[Fig.3c). Simultaneously the intensity of the band of adsorbed hy-
drogen with the maximum at 4010 cm-l decreases. This is easy
to explain,
if
this band is assigned to molecular hydrogen interacting with trigonaly coordinated aluminum atoms, the number of which decreases upon dealumination. Then the band at 4035 cm-l should be attributed to trigonal silicon atoms.
124 Such an interpretation is also confirmed by a similar, but much more distinct picture, which is observed after a high temperature pretreatment of HY zeolite under more severe "deep bed" conditions, when dealumination really
is
the main process. For instance. according to (ref.12) the degree of framework dealumination of Y zeolite preheated at 770 K in a "deep bed" is as high as 85 -1 %. As seen from the Fig 3(d) in this case the band at 4060 cm ,which was previously attributed t o hydrogen interacting with extra lattice aluminum, really predominates and the band at 4010 cm-l from trigonal lattice aluminum atoms is practically absent. The following simple suggestion may help in understanding the difference in the dehydroxylation of hydrogen forms of
faujasites and of high silica
zeolites: The removal of aluminum from the crystal lattaicedoes not only disturb the balance of the electric charges in their structure, but also changes the valence angles and the hybridization of the orbitals in low coordinated Si and A1 atoms being formed. Each of these sites possesses therefore a certain excess energy and is a source of some strains in the zeolite structure. When stoichiometric dehydroxylation of faujasites occurs, the number of defects is very high and is comparable with the total A1 and Si
such
content. For ratio of 2
instance, when full dehydroxylation of a zeolite with a Si to A1
takes place, half of the A1 atoms and a quarter of Si atoms should be
conver-
ted from the tetrahedral coordination into trigonal coordination. Then the total amount of low coordinated aluminum and silicon atoms will be about
30 %.
Such a perturbation of the zeolite crystal lattice is apparently too strong. Therefore the stabilization of the crystal structure takes place resulting in its dealumination and partial decomposition. Unlike this, the percentage of lattice Lewis acidic sites which are formed when high silica H-zeolites are dehydroxylated, does not exceed
several per
cent. In other words, Lewis sites appear in their crystal lattice as a relatively small number of structural defects. Therefore high silica zeolites remain stable without dealumination or decomposition up to much higher temperatures.
ON THE PROBABLE ROLE OF LEWIS ACID SITES IN CRACKING OF PARAFFINS At the moment there is no generally accepted opinion about
the role of
extra lattice aluminum in catalytic transformations of hydrocarbons on zeolites. On the other hand there is also no doubt, that their mild steaming or dealumination results in considerable increase of catalytic activity. Probably the best evidence of this effect was presented by Lago, Haag et. al. (ref. 13). They showed, that the mild steaming of HZSM-5 at 773 K resulted
125 in
a considerable, about one order of magnitude, increase of the activity in
cracking of n-hexane. The more severe treatment decreased the activity, which after prolonged steaming was even lower than that of the starting material. The authors explained this effect by the appearance after mild tion of some new superacidic Broensted active sites. However
dealumina-
they failed
in
their direct experimental observation. Therefore this explanation is only of a speculative character. The effect of the mild steaming was reproduced by us in ref.14, where we also investigated the Broensted and Lewis acidity of H E M - 5 zeolite at different stages of its dealumination. It was actually found, that
the steaming
created several new types of Broensted acid sites, connected with extra lattice aluminum ions. No one of them exhibited however any strong acid properties as evidenced by OH stretching shifts created by adsorption of benzene. Therefore the enhancement of catalytic activity after mild steaming is hardly connected with an appearance of any new superacidic Broensted active sites. On the other hand, the low temperature hydrogen adsorption indicated that the mild dealumination created the extra lattice Lewis sites. Their number was increasing with the time of water vapor treatment. This result allowed to explain the enhancement of the catalytic activity in the following more
traditi-
onal and simple way: According to ref.15 the cracking of paraffins proceeds via chain mechanism, where both Lewis and Broensted acid active sites are involved: R H + L -
+
:C=C:
+ Hsurf + Rads
+
H2
+ +
,‘c=c:
--c
(41
Rads
+ 1
:C=C:
+
R
‘ads
+
The Lewis sites play here the role of the dehydrogenating active centers, resulting in formation of olefins. The latter are then transformed by ted acid sites into adsorbed carbenium ions, which are
involved
Broensin chain
propagation via /3 scission and hydrogen transfer reactions. Since the steaming creates extra lattice Lewis active sites, this favors the dehydrogenation of paraffins and the generation of adsorbed carbenium ions at the initial stages of the treatment. At the same time the dealumination also
destroys the Broensted acid sites. These two factors do influence the cata-
lytic activity in opposite directions. Therefore after mild steaming the reac-
126 tion rate is increased. It reaches then some maximal value after moderate time of steaming and goes down for the higher extent of
the dealumination, where
the steam treatment results in a considerable decrease of Broensted acidity. This explains the extreme character of the activity dependence upon the time of high temperature water vapor treatment. To check this explanation the special experiments were carried out in which the cracking of n-hexane doped by small amounts of
1-hexene was
studied on
steam treated and on untreated HZSM-5 samples. As one should expect, the catalytic activity of the nonsteamed zeolite and of that one
treated
in optimal
hydrothermal conditions changed after 1-hexane addition in such a way
that
they reached almost the same values. In other words the activity of the steamed zeolite became two times lower, whereas for the untreated sample 2-3 times higher.
Dehydrogenating activity of Lewis sites was also confirmed by
the
increased H2 yield (by 1.5-2 times) in the cracking of n-hexane on mildly dealuminated zeolites as compared with initial H-form. Thus, the role of Lewis sites in catalytic cracking was independently proved by purely chemical exper iments . Recently some other examples of modification effects in developing the strong acidity of hydrogen forms of zeolites were reported (refs. 16-17). For instance. Lunsford et.al. (ref.16) found, that dealuminated faujasites and HZSM-20 zeolites prepared such, that they had a minimum amount of extra framework aluminum were relatively inactive in cracking of hexane, which was used as a test of strong acidity. Upon exchanging La+3 ions into the zeolites, the cracking activity first increased with respect to lanthanum loading and then went through a maximum. The authors explained these results by strong Broensted acidity. It was
suggested, that
the model of
lanthanum ions formed
[La(OH)2Lal+4 or La(OH)+2 species in 6 cages, which were responsible for
the
withdrawal of electrons from the framework hydroxyl groups, thus making protons more acidic. In our belief this explanation sounds rather illogical, since hydrolyzed lanthanum species definitely are of the basic character. Therefore they hardly would increase the acidity of structural hydroxyl groups, but rather neutralize them. It looks, that the more probable explanation is that one similar to the already discussed above for extralattice aluminum
ions.
According to it the role of dehydrogenating active sites is played by the lanthanum-oxygen acid-base pairs, which are the source of olefins, resulting
in
the chain carbenium ion mechanism of cracking. THE ACID-BASE PAIRS IN HIGH SILICA ZEOLITES MODIFIED BY Zn AND Ga IONS. Another example, where Lewis acid-base pairs play the role of dehydrogena-
127 ting active sites, represent the high silica zeolites modified by oxides. They are known as active catalysts for aromatization of
Zn and Ga light paraf-
fins (ref.18) and therefore attract the growing attention of a number of
in-
vestigators. There are rather strong indications that these catalysts are of bifunctional nature combining the dehydrogenation of light paraffins on modifying oxides with the consequent aromatization of olefins on super acid Broensted sites of high silica zeolite. Therefore in our work (ref.19) the dehydrogenating
cen-
ters of these catalysts were studied in frames of the similar approach of
the
low temperature molecular hydrogen adsorption. In addition the distribution of modifying oxides was also investigated by means of XPS and by ESR of adsorbed bulky stable free radicals, which could not penetrate inside the zeolites micropores. The most interesting results were obtained for the samples modified by Zn+2 ions. It was found, that both the ion exchange and the impregnation with zinc nitrate decreased the intensity of the band of bridged acidic hydroxyl groups with the stretching frequency of 3610 cm-l. In addition a shoulder from basic ZnOH'
groups at 3660 cm-' was observed.
On the other hand, the modification by Zn also resulted
in appearance of
several different Lewis sites. This is evident from the Fig.4, where
the
IR
spectra of molecular hydrogen adsorbed at 77 K on the sample modified
by
Zn
are represented.
z,'L +
HI
Fig.4 IR diffuse reflectance spectra of molecular hydrogen adsorbed at 77 K on HZSM-5 zeolite modified by 3 weight % of ZnO and preheated in vacuo at 770 K. (a)-hydrogen adsorption on nonmodified sample.(bl-adsorption at 77 K without any additional treatment.(c)- adsorption at 77 K after keeping the sample in hydrogen at 300 K for 10 minutes.
128 The important point is, that the largest shift of adsorbed hydrogen considerably exceeded that, which was earlier observed for hydrogen adsorbed on the surface of the bulk ZnO (ref.20). This means, that the adsorption sites on the surface of zinc oxide and in high silica zeolites modified by Zn+2 ions are of the different nature. This difference could be explained by the different nature of oxygen, which surrounds Zn+2ions in both of these cases.
In the modified
zeolites this
could be the oxygen of the zeolite framework, which certainly is different from that of ZnO. Such interpretation is confirmed by the following experimental results:
It was found, that the sites, which most strongly perturb adsorbed hydrogen at low temperature do dissociatively adsorb it at room temperature. This is evident from the spectrum (c) of the Fig.4, which shows, that after keeping of modified ZnZSM-5 sample at room temperature in hydrogen and cooling it back to
77 K the low frequency band of adsorbed molecular hydrogen was disappearing. Instead the growth of the band from the bridged acidic hydroxyl
groups with
the stretching frequency of 3610 cm-l was observed. This could be explained by the heterolytic dissociative adsorption of hydrogen, which blocks the Lewis acid-base pairs and makes them unaccessible for the adsorption of molecular hydrogen:
-Zn--0-
t
H2
-
+ H
H
-Zn--0-
(5)
Since the stretching frequency of the resulting OH groups is the same as in the hydrogen form of high silica zeolite, the corresponding oxygen atoms of the acid-base pairs are those of the zeolite framework. The acid-base pairs in zeolites modified by Zn+2 do also perturb light paraffins. This is evident from Fig.5, where the IR diffuse reflectance spectra of methane and ethane adsorbed at room temperature on ZnZSM-5 as well as on HZSM-5 zeolite dealuminated by mild hydrothermal treatment are presented.
addition to the lines from the physically adsorbed molecules marked by
In
aste-
riks, the broad bands at 2750-2860 cm-' strongly shifted towards low frequencies are observed. They resemble the corresponding IR bands from light paraffins specifically adsorbed on Lewis acid-base pairs of aluminum oxide (ref.1). The shifts observed for zinc-containing acid-base pairs are somewhat larger as compared with extralattice aluminum ions in the steamed zeolite. After evacuation these bands are removed more difficult than those from the physically adsorbed molecules. Like i n the case of hydrogen at high temperatu-
129 re paraffins are able to be adsorbed dissociatively. This is evident from disif it is adsorbed at 77 K
on
the samples preheated in hydrocarbons at elevated temperature. This effect
is
appearance of the bands of molecular hydrogen
completely reversible, since after evacuation at high temperature the low temperature adsorption of hydrogen again resulted in appearance of the corresponding IR bands. successful. For
the
sample prepared by impregnation we did not observe by IR spectroscopy the
in-
The spectral study of GaZSM-5 zeolites was Just not fluence of Ga ions neither on the intensity of
so
the bridged acidic hydroxyl
groups nor on the formation of any new Lewis acid sites. This result was
ra-
ther surprising, since the samples modified by gallium were also active in aromatization of light paraffins.
Fig.5 IR diffuse reflectance spectra of methane (a,b) and ethane (c) adsorbed at 300K and P=10 Torr on ZnO/HZSM-5 zeolite (a,c) and on the mildly dealuminated HZSM-5 zeolite (b). The explanation of this discrepancy was found, when we studied the chemical composition of modified samples by XPS .It was found, that the external surface of the zeolite micrograins was markedly enriched in Ga, whereas Zn
ions
were distributed homogeneously inside the zeolite micropores. Therefore the lack of IR observation of Lewis acid sites for gallium doped samples should be explained by their low concentration inside the channels.
130 We observed these low coordinated gallium ions on the outer surface of zeolite micrograins by more sensitive ESR technique. As a molecular probe the adsorption of 2.2.6.6 - tetramethylpiperidine 1-oxyle stable free radical was used, which due to its large dimensions could not penetrate inside the zeolite micropores. The ESR spectra of
this free
radical
exhibited
splitting from low coordinated Ga ions, which proved their
the hyperfine
existence on
the
external surface of micrograins of modified zeolite (ref. 1 3 ) . CONCLUSIONS The increase of activity in transformations of paraffins after ming or modification of zeolites by various oxides is usually
mild
stea-
connected with
formation of some new super acidic Broensted active sites (ref.13,16,21). The present paper supports an alternative explanation,according to which the main effect of modification consists in creation of Lewis acid
sites.
In
the steam treated samples their role is played by extra lattice aluminum. is shown, that such active centers could dehydrogenate paraffins and
way assist the easier formation of adsorbed carbenium ions. This
in
It this
results
in
higher reaction rate of cracking. In the case of catalytic transformations of light paraffins
to aromatics
the role of dehydrogenating Lewis sites is played by modifying Zn and Ga ions. They differ in their location, since the former are homogeneously inside the zeolite micro grains, whereas the latter are mainly
distributed localized on
their outer surface.
It was also shown, that the Lewis active sites should be considered as acid-base pairs, where both low coordinated cations and the neighboring basic oxygen of the zeolite crystal lattice are equally important. The best
molecu-
lar probes for their study are molecular hydrogen and paraffins, i e. the substrates, which are involved in catalytic reactions with participation of
such
active sites. REFERENCES V.B.Kazansky, V.Yu. Borovkov, A.V.Zaitsev, IR spectroscopic study of hydrogen and light paraffins interaction with Lewis acidic sites on 7 and -r)-aluminas,Proceedings of 9 th International Congress on Catalysis, Calgary 1988, Ed. by M.J.Phillips and M.Teman, Pub. by Chem. Inst. of Canada 3 (19881. pp.1426-33. J.A.Rabo, P.E.Pickert, D.N.Starnires, J.E.Boyle, Molecular sieve catalysis in hydrocarbon reactions, Proc.2nd Int. Congr. Catalysis, Paris 1960. V.B.Kazansky, V.Yu.Borovkov, L.M.Kustov, IR diffuse reflectance study of oxide catalysts. Use of the molecular hydrogen adsorption as a test for surface active sites, Proc of 8 th 1nt.Congr. on Catalysis, Berlin 1984, Dechema Verlag Chemie, 3 (19841,pp. 3-13.
131
4 L.M. Kustov, V.B. Kazansky, IR - spectroscopic study of an interaction of cations in zeolites with simple probe-molecules. I. Adsorption of molecular hydrogen on alkaline forms of zeolites as a test for localization sites, J . Chem. SOC., Faraday Transactions (1991) (in press). 5 W.J. Mortier, Compilation of extraframework sites in zeolites, Butterworth, Guildford, (1982). 6 R.Y. Yanagida, A.A. Amoro, K. Seff, A redetermination of the crystal structure of dehydrated zeolite 4A, J. Phys. Chem., 77(3) (1973), p p . 805-9. 7 C.L. Angell, P.C. Shaffer, Infrared spectroscopic investigation of zeolites and adsorbed molecules. 11. Adsorbed carbon monoxide, J. Phys. Chem., 70(5) (1966), pp. 1413-17. 8 I . N . Senchenya, V.B. Kazansky, Nonempirical quantum chemical study of molecular hydrogen adsorption and heterolytic dissociation on aluminum oxide, Kinetika i Kataliz, 29(5) (1988), pp. 1331-37. 9 L.B. Uytterhoeven, L.G. Crystner, W.K. Hall, Studies of the hydrogen held by solids, J . Phys. Chem., 69(6) (1965), p p . 2117-2126. 10 J.M. Thomas, J . Klinowski, The study of alumina silicates and related catalysts by high resolution solid state NMR spectroscopy, Adv. Catal. 33 (1985), Ed. D.D. Eley, H. Pines, P.W. Weisz, p p . 200-361. 11 V.B. Kazansky, On the nature of Lewis acidic sites in high silica zeolites and the mechanism of their dehydroxylation, Catalysis Today, 3 (1988), pp. 367-72. 12 D. Freude, T. Froelich, G. Pfeifer, G . Scheler, NMR studies of aluminium in zeolites, Zeolites, 3 (1983), pp. 171-177. 13 M. Lago, W.O. Haag, R.G. Mikovsky, D.H. Olson, S.D. Hellring, K.D. Schmitt and J.T. Kerr, The nature of the catalytic sites in HZSM-5 - activity enhancement, Proc. 7th International Zeolite Conference, Tokyo, Kodansha, 1986, p p . 677-84. 14 V.L. Zholobenko, E. Loeffler, L.M. Kustov, U. Lohse, Ch. Peuker, V . B . Kazansky, G. dhlmann, Zeolites, (1990), in press. 15 P.A. Jacobs, Carboniogenic activity of zeolites, Elsevier, Amsterdam, 1977. 16 R. Carvajal, Po-jen Chu, J.H. Lunsford, The role of polyvalent cations in developing strong acidity: a study of lanthanum-exchangedzeolites, J . Catal., in press. 17 V.L. Zholobenko, L.M. Kustov, V.Yu. Borovkov, V.B. Kazansky, The role of Lewis acidic sites in initial stages of the n-hexane cracking catalyzed by high silica containing zeolites, Proc. 6th Int. Symp. on Heterog. Catal., 1987, Sofia, V o l . 2 , p p . 240-45. 18 T. Mole, J.R. Anderson, G . Greer, The reaction of propane over ZSM-5-H and ZSM-5-Znzeolite catalysts, Appl. Catal., 17(1) (1985), p p . 141-154. 19 V.B. Kazansky, L.M. Kustov, A . Y u . Khodakov, On the nature of active sites for dehydrogenation of saturated hydrocarbons in HZSM-5 zeolites modified by zinc and gallium oxides, Proc. 8th Int. Zeolite Conf., Amsterdam, 1989, Zeolites: facts, figures, future, Ed. P.A. Jacobs, R.A. Van Santen, Elsevier, 1989, pp. 1173-83. 20 R.J. Kokes, Hydrogenation and related reactions over metal oxides, Proc. 5th Int. Congr. Catal., Miami Beach, 1972, Catalysis, Ed. J.W. Hightower, North Holland Pub. Co., 1973. 21 R.A. Beyerline, G.B. McVicker, L.M. Jacullo, I.J. Zilmiak, Influence of framework and nonframework aluminum on the acidity of high-silica protonexchanged FAU-framework zeolite, J. Phys. Chem., 92(7) (1988), p p . 1667-70.
This Page Intentionally Left Blank
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
133
COMPAKA'I'I V E MEASU KKMEN'FS ON ACIDITY OF ZEO1,ITES
Hellmut G. Karge Fritz-Haber-Institut der Max-Planck-Gesellschaft Berlin, 1000 Berlin 33, FRG
A €3STKACT
Methods of identification of various acidic sites in zeolites (Brensted sites, different types of Lewis sites, cations) are briefly reviewed. Similarly, techniques for determination of the density of acid sites are discussed and illustrated with selected examples. In particular, a good agreement between the results obtained by IR (using pyridine as a probe molecule) and ESR (employing NO) in determining Lewis site densities of, for instance, progressively dehydroxylated H-Y is demonstrated. Emphasis is laid on the discussion of techniques, such as titration, in aprotic solvents, IR with and without probe molecules, 'H MAS NMR, temperature-programmed desorption and microcalorimetric measurements of heats of adsorption of basic probe molecules, which are suitable for characterising the strength of acidic sites. Quantitative evaluation of TPD spectra using a kinetics model as well a s the combination of different methods is described in more detail, and results obtained by various techniques on selected examples of acidic zeolites are compared.
Most of the important reactions carried out over zeolite catalysts are acid-catalysed even though there is a n ever increasing interest in zeolite catalysis on basic sites [ l - 2I. The term "acidity of zeolites", however, is by no means unequivocal. "Acidity" could designate (i) the nature of acidic sites (e.g. Brcdnsted sites 131, various forms of Lewis centres such a s "true" Lewis sites [4-5], cations), (ii) the number or density of the respective sites or, finally (iii) their strength. Although there exists a n overwhelming evidence that a t least for the large majority of acid-catalysed reactions the acidic Brplnsted centres of zeolites (acidic hydroxyls) are the relevant active sites [6-10!, there is still a debate on the role of the Lewis centres [ll-191 in zeolite catalysis. In any event, there i s no severe difficulty to discriminate between the various types of acidic sites. A survey of the most relevant methods of distinguishing the various types of acidic sites is presented in the first section. In fact, this can be rather brief since a number of excellent earlier and also more recent review articles are available which cover this subject [8, 20-231. Similarly, the determination of the relative or absolute density or number of acidic sites will not be dealt with in great detail. Most of the techniques suitable for identifying the nature of the acidic sites can be developed in such a way that they provide quantitative information. Thus, in the second section some of the important methods for quantitative evaluation of acidic sites in zeolites will be presented and for a few instances the results ofdifferent techniques will be compared.
134
A more difficult problem concerning the acidity of zeolites is the evaluation of the acidity strength. Indeed, more recently some theoretical concepts have been developed to rationalise the acidity strength of Bransted acid sites in zeolites [24-321. They relate acidity strength to long range collective phenomena and/or more localised effects of structure and chemical composition. Hence, theoretical ideas such a s "activity" or "efficiency" (regarding the zeolite as a crystalline liquid, see Refs. [241, [27]), Sanderson electronegativity (see Refs. [25],[261, [281, [301), topological densities (see Ref. [31]), quantum mechanical effects of bond lengths and angles (see Ref. [291) and deprotonation energies (see Refs. [29], [32]) are involved. Detailed discussions of these concepts were presented by Dwyer [221 and Rabo and Gajda [231. There is, however, an urgent demand to have a t hand reliable techniques which enable us to characterise and measure the acidity strength of zeolite materials, not only to verify or falsify theoretical concepts, even though this is certainly a n important point. Progress in experimental techniques will extend the data base for further improvement of theoretical approaches to the phenomenon of acidity strength. A practical need is felt to determine the acidity strength experimentally in a reliable manner, since in many instances i t has become quite obvious that not only the number but also the strength of acidic sites is particularly decisive for the behaviour of zeolitic materials in sorption and catalysis (see, e.g. Refs. [331, [341). Quite a number of methods have been suggested, developed and employed to characterise the strength of acidic sites in zeolites. In the third part of this contribution a n attempt is made to evaluate some of the more important ones and to illustrate them by pertinent examples taken from the literature or own measurements. Special emphasis is laid on the comparison of results obtained for one and the same acidic zeolite sample by the use of different techniques for the determination of density and strength of acidic sites.
1. THE NATURE OF ACIDIC SITES IN ZEOLITES AND TECHNIQUES FOR THEIR DISCRIMINATION Table 1 presents a list of techniques proposed for discrimination between the various acidic sites encountered in zeolites. Pertinent references are also given; however, here and in the following text they are by no means exhaustive, due to the extreme broadness of the field. Possible acidic sites are the acidic OH groups, in particular the so-called bridging H ones in the = Si - 0 - A1 = configurations, i.e. potential Bransted acid centres (proton donors). Furthermore, coordinatively unsaturated sites occur (electron pair acceptors), e.g. "true" Lewis sites [4-5, 351 such a s A10+ or charged A1,0,"+ clusters or cations, Me"+, or acceptor sites on tiny oxide particles inside or outside the pore structure.
135
TABLE 1
NATURE OF ACIDIC SITES IN ZEOLITES Methods of Identification
Examples
probes
P O . 131
AI-OH; P-OH
'H MAS NMR, 15N NMR
[39-421
?AI (framework)
IR
AIO +,[AI,0y](3X*21
27AlMAS, 15N NMR
Al-oxides (extra-framework)
ESR
Bransted Bridging OH;
0
Lewis
+
Ref.
IR,IR
Other oxides Cations, Men
+
I
+
probes
+
probes
[201 [39.411 [65-671
IR + probes
Regarding the methods to distinguish between the various types, i t has been claimed several times that Hammet indicators are non-specific whereas aryl indicators should indicate only Bransted but not Lewis sites [36].This, however, was disproven [37-38,76-771. It was shown by Karge [37]via U V M S transmission spectroscopy that, for instance, triphenyl methanol, @,C(OH), reacts with a surface being free from Bransted sites as well, giving the typical spectrum of @,C+ carbenium ions. The most powerful and widely employed techniques for discrimination of acidic sites are the IR and NMR spectroscopy used with or without probe molecules [a,20,39431.Acidic OH groups may be detected by the position of the corresponding OH stretching vibration in the range of ca. 3520 to 3650 cm-l [8,9, 20, 441 or by their typical chemical shift in 'H MAS NMR [41,421. Adsorption of probe molecules on acidic OH'S gives rise to characteristic features in the spectrum which also reveal the presence of Bransted sites. Thus, after interaction with pyridine, typical bands stemming from ring deformation modes occur a t 1632 cm-' ( w 8a) or ca. 1540 cm-l ( v 19b). Reaction with ammonia results in the appearance of the v,(NH4+) band around 1445 cm-l. Similarly, adsorption of 15N-pyridineon acid Bransted sites causes a chemical shift of S = 171-174 ppm vs. solid NII,NO, [39-401. 27Al MAS NMR allows for the determination of the number of A1 atoms tetrahedrally coordinated in the framework, Al'. Since tetrahedrally coordinated A1 corresponds (in zeolites not modified via dealumination) to the presence of Bransted acid sites, this provides another means to indicate those centres [431. True Lewis sites (i.e. Al-containing species on extra-framework sites, AINF, [4-5, 35, 451)as well as cations can be identified via IR spectroscopy with the help of suitable probe molecules such as NH,, pyridine, acetonitrile or CO. Ammonia molecules coordinatively bound to true Lewis sites and I or cations give rise to a broad band around
136
1620 cm-l. The u(19b) band of pyridine attached to Lewis sites is usually more intense
and sharp bands occur around 1450 cm-I for true Lewis sites and in the range from 1440
to 1551 cm-l for cations depending on their Coulomb field [20, 46-47]. Similarly, the band position of adsorbed acetonitrile and CO is affected by the nature (e/r) of the respective cation [ 37,48-491. In the case of IR spectroscopy usually only one type of Al-containing extra-framework species is indicated by the probe, i.e. with pyridine just one band appears a t 1452 cm-l. Only in a few instances a doublet (1456,1451 cm-l) was observed [45,50]. In contrast, 27A1MAS NMR frequently reveals the presence of different types of extra-framework Al. In fact, the interpretation seems to be difficult; however, besides octahedrally coordinated extra-framework A1 also penta- and even tetrahedrally coordinated species are being discussed [41,43,51-531.
2. DENSITY (NUMBER) OF ACID SITES IN ZEOLITES Tables 2 and 3 present lists of methods which were applied to determine the den-
sity and / or number of Bransted and Lewis sites in zeolites, respectively.
'ABLE 2
DENSITY (NUMBER) OF BRQNSTED ACIDIC SITES Method
Remarks
Ref. -
Titration
Restricted applicability
[69,74-771
IR Spectroscopy
Absorbance of bands of acidic OHs andlor probes, e.g., PyH , NH4 , H2 (extinction coefficients required for absolute numbers)
[20,13-14.91
Area of respective signal
[41-42)
2 7 ~MAS 1 NMR
Area of respective signal
I431
29SiMASNMR
Combined with chemical analysis
[911
Ion Exchange
Exchange capacity
[601
Test Reactions
Restricted applicability
(34.62-641
Adsorption and Desorption of Probe Molecules
Restricted applicability (no other adsorption sites involved; or in combination with, e.g., IR)
[83-84.891 [94-1041
+
'HMASNMR
+
Those methods which are reliable in discriminating the various types of sites (e.g. IR spectroscopy, '13 NMR, 29Si and 27Al NMR spectroscopy) can be in most cases employed for quantitative measurements as well.
137
In this context, i t should be mentioned that the NMR techniques provide the absolute numbers of sites, n (sites), under investigation in a very straight-forward manner, uiz. by the area of the respective signals. In contrast, the bands of IR absorbance, which are typical of certain acidic sites, must be evaluated with the help of the appropriate extinction coefficients according to
n (sites)=
1;'vdv I
where E, d, v1 and w2 are the extinction coefficient, thickness of the sample, beginning and end of the particular band, respectively. The extinction coefficient must be determined in an independent experiment (see, e.g. Refs. [54-561 ) and may depend on the coverage of sites. However, in many cases i t is sufficient to use relative densities to obtain the relevant information about the relationship between numbers of sites and, e.g. catalytic or adsorption properties [9-10,201. Figure 1 illustrates the application of IR spectroscopy and 'H NMR spectroscopy on the determination of the number of Bronsted sites, respectively [3,57-581.
3670 3580 cm-' W
bridgin acidic 08s
v] = 49.2 0Hh.c.
= 48.8 0Hh.c.
U
z
a c
-
t-
5 wl
z
a
'I
z
7 4,s 2
I-
! E I
I(1)
IR-SPECTROSCOPY
J.E. Uytterhoeven. 1. G. Chrirtner and W. K. Hall 1. Phyr.Chem.69(1965)2117 (2) J.E. Uytterhoeven. P. Jacobs, K. Makay and R. Schoonheydt 1. Phyr. Chem. 72 (1968) 1768
I ~ H M A S N M R SPECTROSCOPY] H.Pfeifer, D. Freude and M. Hunger Zeolites 5 (1985) 274
Fig. 1 Determination of bridging acid OH groups in HY (90) by IR and 'H hIAS NMR. Most of the other techniques indicated in Tables 2 and 3 suffer from certain restrictions. With respect to the titration method, this will be elucidated in the third section. Deter-
138
mination of the ion exchange capacity, e.g. with NH4+, gives usually a good estimate of the upper limit for the total number of acidic Brernsted sites, however, often even very weak sites are involved, without any relevance to catalysis. Adsorption and desorption of probe molecules requires, inter alia, t h a t only one type of sites exists in the zeolite, otherwise the assignment of the amounts of adsorbed or desorbed probe molecules might be erroneous. In such cases, where more than one type of site is involved, the combination with IR and/or NMR measurements is advisable [59-611. Finally, the evaluation of sites via the rate of suitable test reactions is well established for a series of homologous zeolite catalysts [34,62-641. Comparison of quite different types of zeolites could be misleading. This field, i.e. application of test reactions for determination of density of active sites is not yet satisfactorily explored. As already pointed out (see also Table 3), it is possible to identify and quantitatively determine acidic Lewis sites by IR spectrosopy with the aid of suitable probe molecules such as pyridine, ammonia, carbon monoxide and hydrogen. The direct measurement of Al-containing extra-framework species (true Lewis sites) can be achieved via 27Al MAS NMR,provided the Al-content is not too low [43,53]. TABLE 3
DENSITY (NUMBER) OF LEWIS ACIDIC SITES
Ref.
Method
Remarks
0
IR Spectroscopy
Absorbance of bands of probes, e.g., Py L, NH3 -,L, CO L, H2 -,L (extinction coefficients required for absolute numbers)
[20,13-14,91
0
27AIMAS NMR Spectroscopy
Requirements:A1 content not too low, all Al observable (area of respective signal)
1431
0
29Si MAS NMR
Combined with chemical analysis
0
ESR Spectroscopy NO adsorption
[91l [65-681
-.
-
However, ESR spectroscopy using NO as a probe, provides also a valuable tool not only for identification but also for quantitative evaluation of Lewis sites in zeolites. This was first shown by Lunsford et al. [65-671, see also Ref. [591. Figure 2 shows a n ESR spectrum obtained after adsorption of NO on a n H-ZSM-5 sample [68]. It is almost identical with the ESR spectrum of NO adsorbed on alumina [66-671. Double integration provides the spin density as a measure of the density of Lewis sites. In a series of experiments, these densities were determined as a function of pretreatment temperature. As expected, the density of Lewis sites increases with pretreatment temperature as a result of increasing dehydroxylation 1681. This is indicated in Figure 3 by the corresponding increase of the spin density of NO adsorbed. In a parallel
139
set of experiments the Lewis site density was determined through IR measurements of the intensity of the 1452 cm-’ band after pyridine adsorption. Figure 3 demonstrates that the results of both techniques are in very good agreement.
r
I : : )
p(N0) = 50Pa
lo’* Adsorbent : H - ZSM - 5 (2)
3.
Part
Tad (Py) Tdel (Py) Tad(NO)
z
Fig. 2 ESR spectra of NO adsorbed on y-Al,O, and H-ZSM-5 (2) or CAZ 49
CAZ 49
I
1
= 10-5pa = 475 K (2h)
= 475 K (1 h) = 298K ( l h )
- 0.6
a W
n m
- 0.3
0
I
600
I 1 I I 700 800 900 1000 A C T I V A T I 0 N T E M P E R A T U R E [K]
10.0 1100
Fig. 3 Density of Lewis sites of H-ZSM-5 (2) or CAZ 49, measured through NO (ESR) and pyridine (IR) adsorption
140
Table 4 displays comparative data of measurements of site densities obtained via various methods. The relative densities (%) for Bransted sites were obtained via MAS NMR (see AIF ) and IR (see A[OH] and A[PyH+]), those for Lewis sites via MAS NMR (see AINF1, IR (see A[Py -B L] and ESR (see spin densities D[NO]). With the exception of the PyH+ data the agreement is satisfactory. The determination of Brflnsted sites via the density of pyridinium ions, (A[PyH+]),seems to be disturbed by the presence of a relatively high amount of extra-framework aluminium (AINF)which may give rise to (weak) Al-OH groups reactive towards pyridine or, another possibility, may affect the extinction coefficients of the PyH+ species. The presence of Al-OH in H-ZSM-5 (21, i.e. CAZ 49, was indicated by an IR band at 3680 cm-'; this band was missing in the case of H-ZSM-5 (1).This would explain the high value of A(PyH+) of CAZ 49 compared to CAZ 36.
TABLE 4
COMPARISON OF (RELATIVE) SITE DENSITIES MEASURED VIA IR, NMRAND ESR
1 [%I
Brcansted Sites
[%I
[%I
I
I [%I
I
Lewissites
["/.I
[%I
H-MD-1
100
100
100
100
100
_-
H-MD-2
65
63
84
95
84
__
100
100
100
100
100
H-ZSM-5(1)
I
I
I
100
3. CHARACTERISATION OF THE STRENGTH OF ACIDIC SITES IN ZEOLI'I'ES Titration of acidic sites Tables 5 and 6 list a number of techniques frequently used for determination of the strengths of Bransted and Lewis sites. The titration technique was developed to characterise the acidity strength of solid inorganic materials such as oxides and salts (see Tanabe, 1691 references therein) and is still widely used in this area of research [70,71]. It was also employed in studying the acidity of the classic silica-alumina cracking catalysts [37,72]. The particular procedure is described in detail in the literature (see, for instance Refs. [34,36,69,72-731). In zeolite chemistry, pioneering and careful work was carried out by Hirschler L361, Beaumont et al. 1741, Kladnig 1751, Kittelmann [761 and Unger e t al. [771. In fact, Beaumont et al. [74] obtained very interesting results when applying the titration technique on faujasite-type X and Y zeolites. These authors found a distinct
141
correlation between the degree of exchange of N a + against NH4+ (or H') and the efficiency of thus created acidic hydroxyls. This enabled them to define a n efficiency coefficient in analogy to the activity coefficient of electrolytes. In the case of H,Na-X (Si/Al =1.23 or Al / [A1 Si]=0.45) this coefficient was u,=0.16 whereas for Y-type zeolite, H,Na-Y (Si/Al = 2.43 or A1 / [A1 Si] = 0.29),a, = 0.6 was obtained. This was the first evidence for a (collective)effect of the Si/Al ratio on the acidity of zeolites (compare also Ref. [241).
+
+
TABLE 5
STRENGTH OF BRQNSTED ACIDIC SITES Method
Remarks
Ref.
0
Titration
Restricted applicability
[ 36,34,74-771
0
TPD
Probe molecules, e.g., NH3! pyridine (eventually combined with spectroscopy)
(83-84,89-901
0
Microcalorimetry
Adsorbates, e.g., NH3, pyridine, weaker bases (eventually combined with spectroscopy)
[94-104)
0
IR Spectroscopy
Shift of OH; with and without probe molecules, e.g. benzene, CO
[28,1081 [lo91
UV-VIS Spectroscopy
aromatics as probe molecules
[1101
0
'HMASNMR
Chemical shift of the 'H signal
[41-42.1111
0
Test Reactions
Various reactions requiring different acidic strength
[1141
TABLE 6
STRENGTH OF LEWIS ACIDIC SITES Method
Remarks
0
TPD
Probe molecules, e.g., NHJ, pyridine or weaker bases (eventually combined with spectroscopy)
0
Microcalorimetry Adsorbates, e.g., NH3, pyridine or weaker bases (eventually combined with spectroscopy) IR Spectroscopy
0
Adsorption of bands of, e.g. pyridine or CO coordinated t o cations
ESR Spectroscopy Desorption of NO
Ref. (83-84,89-901
[94-1041
[20,14,91
[a1
142
Even though the method was subject to frequent criticism [78-791, the agreement between the total number of acidic sites (H, 2 -3.0) and the same number evaluated, for instance, from the stoichiometry seems to provide some confidence in its reliability. However, i t turned out that the procedure was obviously not applicable to other zeolite structures such as mordenites [75-77,80411. The reason is not, as has been often claimed, that the bulky indicator molecules cannot penetrate into the channels or pores of structures other than faujasite [23,82]; mordenite for instance, is a large-port zeolite with twelve-membered rings as pore openings very similar to faujasite. Moreover, it i s questionable whether slight colour changes of indicator molecules inside the pore structure would be detectable by the eyes of the experimentalist. However, the main reason for the failure of the titration method in the case of many zeolites seems to result from a very strong interaction of the solvent molecules with the acidic sites. I
I
b : 1 rnin. after injection of excess butylamine (16 m mo1.g-1) c : 7 h after injection
3680 Y
U
z
U II-
-
I z VI
4 L*
+
I
,000
I
3500 W A V E N U M B E R [cm']
3001
Fig. 4 Interaction of solvent (nheptane) with H F (3650 cm-I ) OH groups of H,NaY and in-situ titration with butylamine
In the pore system of mordenite the replacement of the solvent molecule by the titrator base (e.g., butyl m i n e ) is thus severely impeded. This is demonstrated by Figures 4 and 5 which contrast the different behaviour of H,Na-Y and H,Na-MOR in a titration experiment followed in-situ by IR. Thus, one has to state that, regardless of other difficulties, the method fails when applied to acidic zeolites such as mordenite, erionite etc. This is in agreement with the above-mentioned findings [75-77,80-811.
143
Va
v-
a
afterdehydrationat 725 Kand 1O'Pa
b after admiwon at solvent. CCI,
c
after 1st mje(tion of n butylamine ( 0 25 m mol q 11, 8 h
d
-a
after dehydration at 1 2 5 K and 10 5 Pa
b after admiwon of ~ o l v e n l CCI. .
b after admission of solvent CCI.
c
c
after 2nd mjeamn of n-butylamme
atterlnd mjert~onof n-butylamlnc
1025mmol (I-) t 2 h
I
I
3800
WAV
2800
E N U M B E R [cm-'1
Fig. 5 Interaction of solvent (CCI,) with acidic OH groups of H,Na-MOR and in-situ titration with butylamine It seems to work, however, with faujasite-type zeolites. In fact, in the case of a homologous series of Y-type zeolites with systematically varied acidity strength a spectrum of the acidity strength distribution was obtained as is shown in Figure 6.
-
Y'
0
€ E
-
very strong H, c-8 2 strong -82
2-
3
Y
VI Y
+ m
2
0
U
a LL
0 '
* -
I-
VI
z u
o
HNaY
BeNaY
MgNaY CaNaY
SrNaY
BaNaY
LaNaY
Fig. 6 Spectra of acidity strength of a homologous series of Y-type zeolites
144
Figure 7 demonstrates that only the fraction of the strongest sites (Ho 5 -8.2)correlates with the rate of conversion of the test reaction, viz. disproportionation of ethylbenzene
MI.
D E N S I T Y O F V E R Y S T R O N G A C I D SlTES(H,<-8.2)
[mmol.g-1]
Fig. 7 Rate of ethylbenzene disproportionation vs. density of strong sites in Y-zeolites
Temperature-programmed desorption of probe molecules Temperature-programmed desorption of basic probe molecules is a widely used method to characterise the strength of acidic sites in zeolites [83-841. Usually relatively strong bases are employed such as ammonia or pyridine. To minimize the effect of the adsorbate base on the strength distribution of the remaining (non-covered) acidic sites the use of weaker bases is sometimes preferred (see, e.g. Ref. [85]).The amount of desorbing species is usually monitored via GC or MS. It is worth noting that the rate of desorption of the probe molecules might be affected by diffusion restrictions. Furthermore, the measurements may depend on experimental conditions such a s geometry of the arrangement, heating rate, reliable temperature control of the sample vs. oven etc. For example, two essentially differing geometric arrangements caused differences in the results by 10-15%[861. Therefore, a warning is appropriate with respect to comparison of absolute TPD data obtained under different conditions. The factors indicated above must be carefully taken into account. However, a ranking of the acidity strength of members of a series of acidic zeolites investigated under identical conditions is usually feasible. In most cases of TPD characterisation of acidic sites in zeolites, the peak temperatures of the TPD spectra are adopted as an indication of the (relative) strength of the sites. However, the spectra are usually relatively broad and, moreover, TPD by itself does not provide any information about the nature of the sites from which the probe
145
desorbs. Therefore, several efforts were made by, for instance, V. V. Yushchenko e t a!. [85], Dima and Rees [87], Hashimoto et al. [88] and Karge and Dondur [891 to evaluate in a more quantitative manner the TPD spectra and/or to combine TPD with other methods, such as IJi spectroscopy, to identify the types of sites releasing the probe molecules. Figure 8 displays, as an example, TPD spectra of ammonia desorbing from two different H-ZSM-5 samples, i.e. H-ZSM-B(l) or CAZ 36 and H-ZSM-5(2) or CAZ 49. I
I
I
I
T E M P E R A T U R E [K]
Fig. 8 TPD of ammonia from two different H-ZSM-5 samples, CAZ 36 and CAZ 49
It suggests a significant difference in acidity strength in that sample CAZ 49 exhibits a peak temperature higher by 36 K than that of sample CAZ 36. A kinetics model, developed by Dondur and Karge [89-901was applied for evaluation. The results are shown in Figures 9 and 10 and in Table 7 (vide infra). One recognizes that (i) the analysis of the spectra according to the above-mentioned model reveals the presence of three types of sites each of them possessing its own distribution of strengths and (ii) the differences indicated by the most frequent activation energy of desorption, Edes,for the respective sites are less than lo%, the accuracy of the measurements being in the order of 3%. These relatively low differences in acidity strength were confirmed by calorimetric measurements (vide infra). Nevertheless, the samples exhibited remarkably different catalytic behaviour, e.g. in MTG reaction and coke formation [911, which seemed at least to some extent to be correlated with both differences in number and strength of the acidic sites. The assignment of the sub-peaks obtained by evaluation of the original TPD spectra (shown in Figures 9 and 10) was achieved by concomitant IR investigation of the effect of dehydroxylation on the intensity changes of those sub-peaks. A more elegant technique is the desorption of the probe molecules from a sample placed in an IR cell. In
146
w n.
.lo
I
’
1
I
104
I
0.4 -
I
I
I
I
CATALYST : CAZ 36
1
mz
ao
ul W
n A C T I V A T I 0 N E N E R G Y 0 F D E 5 0 R P T I 0 N [k 1 m o 1’1 3
Fig. 9 Calculated distributions of the activation energies of desorption of NH, from various sites of €I-ZSM 5 (1) or CAZ 36.
ul W
n A C T I V A T I 0 N E N E R G Y 0 F 0 E S 0 R P T I 0 N [k J . m o I ’1
Fig. 10 Calculated distribution of the activation energies of desorption of NH, from various sites of H-ZSM-5 (2) or CAZ 49 this case the change of the partial pressure of the desorbing species (monitored by MS or G C ) and the change in intensity of the corresponding adsorbate IR bands, both a s a function of temperature are simultaneously measured [80,86,921. The peak maximum in a TPD spectrum of ammonia desorbing from Bransted sites of a zeolite should, for
147
instance, correspond to the maximum change in the intensity of the ammonium ion band at 1445 cm I . An example for such simultaneousTPD-MS and IRmeasurements is shown in Figure I I .
Tart. = 670 K
1450 cm-l
01
6 - S I T E S (11)
I
I
I
Fig. 11 Combined IR and MS during TPD of NH, from hydrogen mordenite
400 500 600 700 800 T E M P E R A T U R E 0 F D E S 0 R P T l O N [K]
I
1
I
I
I
I
I
I
I
I
-
r I
.
I-
A
0
r
0
-
Heating rate : 5 K . min-1
0.000
I
I 90
I
I 120
I
I
I
150 T E M P E R A T U R E [K]
I
180
I
I
210
Fig. 12 TPD of N O from two different H-ZSM-5 samples, CAZ 36 and CAL 49, monitored by ESK
-
148
Valuable information giving a t least a n estimate of the strength of sites occurring on a particular zeolite saniple may be provided by discontinuous desorption of bases. Most frequently this is followed by IR, monitoring the change of the absorbance of typical bands due to pyridine adsorption after heating the sample to fixed desorption temperatures (see, e.g. Ref. [93I). Temperature-programmed desorption of suitable probe molecules from Lewis acid sites can be used in a n analogous manner asdescribed for Bransted sites [89-901. An interesting new approach is the temperature-programmed desorption of N O from Lewis sites, monitored by ESR, because, this probe appears to be very specific to these types of sites (vide supra, Refs. [65-681).An example is illustrated by Figure 12 1681.
Microcalorimetric measurements There are only a few research groups applying microcalorimetry to characterise and compare acidity strength via the differential heat of adsorption, even though this technique provides in a very clear-cut manner quantitative data for scaling the strength of acidity. However, the technique requires very careful experimenting and is frequently somewhat time-consuming. Important contributions came from the groups of Auroux e t al. [59,94-971, Klyachko et al. [98-1011, Thamm et al. [102-1031, and Khvoshchev et al. [ 1041.An example taken from a study of Klyachko et al. [981is shown in Figure 13.
1 : Na - MOR 2 : H - MOR 3 : H-MOR 4:H-MOR
(Si/AL= 5,O) (Si / A l = 6.3) (Si/AI=lO,O) o o o o (Si/Al=23,5)
Klyachko et al. Acta Phys. Chem. 24 (1978) 183
100
50
1
5 3 C O V E R A G E [mmol 9-11
7
Fig. 13 Differential heat of adsorption of NH, on mordenites (after Klyachko et a]., ReT 1981)
149
In the study by Kapustin e t al. [ l o l l interesting results were obtained with HZSM-5 samples of systematically varied Si/Al ratios, employing TPD of ammonia for the sake of comparison. The heats of adsorption of NH, derived from microcalorimetric measurements and TPD were in satisfactory agreement. However, microcalorimetry yielded a distribution of the strengths of strong acidic sites whereas TPD provided only a mean value. Similar to the situation in TPD investigations (vide supra) stronger bases a r e frequently employed as probes, the use of weaker ones, however, seems to be preferable in many cases 1961. When strong bases are the adsorbates, care must be taken to ensure sufficient mobility to achieve adsorption equilibrium. Thus, for adsorption of, e.g., ammonia and pyridine higher adsorption temperatures are required which, in turn, necessitates the maintainance of a good isothermicity of the whole set-up. Moreover, the same problem arises in microcalorimetry a s with TPD investigations: only concomitant experiments identifying the sites of adsorption enable u s to ascribe, in a unambiguous way, the measured heats of adsorption to the types of sites which interact with the probe molecules under release ofthat energy [100,105]. Results of such studies, i.e. microcalorimetric measurements of NH, adsorption onto H-ZSM-5 samples (paralleled by IR measurements of the relative amount of BrGnsted and Lewis sites before and after dehydroxylation), are incorporated into Table 7 [1051.
IR spectroscopic measurements IR spectroscopy is sometimes employed a s a sole method for determination of acidity strengths of sites, in other words, not as a technique complementary to other ones (vide supra). Thus, it was claimed that the band position (wavenumber) of the OH stretching band of acidic OH groups is a measure of their strength. This was first discussed by Barthomeuf [24] when comparing different zeolite structures in the hydrogen form; see also Refs. [26-28,1061. However, one has to be aware of the fact that other factors, such as the environment might affect this band position as well, i.e. without any significant relationship to the acidity strength. Most prominent examples are positions of the high and low frequency bands of OH groups in the faujasite structure. Their accessibility for more bulky probe molecules differs remarkably but not, in fact, their strength [107]. I t seems to be even more dificult to rely on the small shifts of the OH bands caused by slight but systematic variations of the acidic properties within a homologous series of zeolite samples. An example is presented in Figure 14 (see Ref. [47]) for a series of mordenites (see also Ref. [621). Even though TPD measurements seem to provide support [62,801,the author hesitated to interpret the small but systematic shift in the OH band positions as being indicative of decreasing strength of Brflnsted sites in the sequence from H-MOR through to Ba-MOR. As mentioned earlier, the OH stretching band of acidic OH groups
150
covers, without any doubt, a whole spectrum of sites with a more or less broad distribution of acidic strengths (vide supra). The fraction of sites falling into certain ranges of strengths with corresponding OH modes may vary and, therefore, their superimposition may lead to shifts in the overall position of the envelope OH band.
I
3600
3580
3615
3605
3616
3618
3800-3400
cm
3620
3620 3612
W a v e n umber
Fig. 14 Wavenumbers ol'the stretching bands of acidic OH groups of a homologous series of niordenites 1471. An estimate of the strength of the acidic sites can be obtained from the shift of the OH bands of the acidic hydroxyls upon interaction with suitable adsorbates such a s benzene [28,1081 or CO [1091. The resulting shifted bands, however, caused by hydrogen or II bonding, are usually relatively broad and do not reveal more subtle differences in acidity strength. Sometimes the observed shifts may provoke an overestimation or the differences in acidity. This seems to be so with the results obtained with two II-ZSM-5 samples (CAZ 36, CAI, 49) a s the comparison with TPD and microcalorimetric data reveals (see Table 7). Similar experiments were carried out in the U V M S region by Naccache e t al. [110] using aromatic probe molecules. The strength of Lewis sites, i.e. more specifically cations as electron pair acceptor sites, can be reliably scaled by the shift of the bands of, for instance, pyridine, nitriles or CO coordinatively attached to them. Examples are provided in the work by Ward [20,46], Angel1 e t al. 148-491 and Karge [37,47].
151
TABLE 7 COMPARISON OF MEASUREMENTS OF ACIDIC STRENGTH Strong Brensted Sites
Zeolite
0;: [kJ.rnol-'l
Lewis Sites
TZIK
Edc, [kJ.mol-']
H-MOR-D (1)
12
700
115
355
137
H-MOR-D (2)
17
690
115
360
140
H-ZSM-5 (1)
33
619
104
90'
337
130
H-ZSM-5 (2)
25
655
110
115'
3 59
137
dv(0H)
Ed,* [kJ-mol-']
'H MAS NMR Spectroscopy Pfeifer and co-workers [41-42,1111 have demonstrated that the chemical shift in 'H MAS NMR spectra of acidic zeolites might be a suitable measure of the strength of the related 'H-containing sites. Thus, in the case of HY a shift (in ppm vs. TMS) of 6 = 1.8-2.2 indicates the weak or non-acidic silanol groups (corresponding to the IR band around 3740cm-'), 6=3.8-4.4 and S=4.8-5.6 are related to the two types of bridging acidic OH groups with IR bands a t about 3640 and 3550 cm-l in faujasite-type acidic zeolites, indicating a different environment but not a different acidity strength. A shift of S = 2.6-3.6 ppm is, finally, ascribed to weak acidic OH groups associated with extraframework aluminium-containingspecies. Also interesting relationships were found between the chemical 'H shifts and dealumination of faujasite-type or mordenite-type zeolites via acid leaching or hydrothermal treatment (steaming) [41,1111. Basically, however, the situation with 'H-MAS-NMR measurements of the strength of acidity is the same as encountered with IR. The signals observed are certainly covering a whole spectrum of sites with various strengths. 'H MAS NMR also does not, hitherto, provide a direct means to determine the particular distribution of these acidity strengths. Test reactions A few studies attempted to employ test reactions not only to characterise the density of acidic centres but also their strength [112-1143. The basic idea was that different acid-catalysed model reactions may require acidic sites of different strengths which would enable us, conversely, to characterise the strengths of catalytically active acid sites in zeolites by the rates of a suitable set of model reactions. Thus, i t has been
152
pointed out, for instance, by Weitkamp [112] that isobutane alkylation by butene over rare earth faujasite-type zeolites requires stronger sites than butene oligomerisation and this, in turn, occurs on sites stronger than those necessary for butene isomerisation. Similarly, Karge e t al. [62,113] found, with regard to the required strength of sites, the sequence dealkylation of ethylbenzene > disproportionation of ethylbenzene > dehydration of cyclohexanol. In a more recent study, Guisnet and co-workers [1141 used the following set of test reactions: isomerisation and cracking of: ( l ) , n-hexane ( a t 673 K); (2), 2-methyl pentane ( a t 673 K); (3), 2,4 dimethyl pentane ( a t 623 K); (4), 1,2,4 trimethyl pentane (at 625 K); isomerisation and disproportionation of: (51,o-xylene (at 623 K);(61, 1,2,4-trimethyl benzene ( a t 625 K) and (7), 3,3-dimethyl 1-butene ( a t 473 K).The reaction rates were extrapolated to 625 K. Their results suggested t h a t the required strength of the acidic sites, catalysing these reactions, decreases in the above sequence from (1)to (7). I t appears that this approach could be further developed and may provide another possibility of scaling the acidity strength of sites quantitatively through the activation energies of a set of suitable model reactions.
CONCLUSIONS While well-established techniques are available for identification of acidic centres in zeolites and quantitative determination of their density, improved methods are still required for characterisation and reliable measurement of acid strength and strengths distribution. For this purpose, temperature-programmed desorption (TPD) of suitable probe molecules seems to be appropriate provided that possible experimental sources of error (limitations in mass and heat transfer, effects of geometrical arrangements, etc.) are avoided. In this case, evaluation of the TPD data on the basis of a valid kinetics model may provide quantitative results such a s number of types of sites involved, fractional population of these sites, activation energies of desorption (Ed,,), and distribution of Edes. However, in order to identify the nature of the sites from which the probe molecules desorb complementary experiments (such a s IR or 'H MAS NMR measurements) are necessary. TPD of NO, monitored by ESR, seems to offer a valuable tool for characterisation of Lewis sites. Microcalorimetric measurements of the heat of adsorption released by sorption of probe molecules onto acidic sites is a very attractive way to quantitatively determine their strengths. However, analogous to the case of TPD, complementary measurements are required to assign the calorific data to the types of active centres involved. Correlating of the rates of a suitable set of test reactions to the strength of the catalysing acidic sites is an attractive alternative. This method deserves further development since it could provide a relative simple and fast technique which does not require sophisticated equipment.
153
ACKNOWLEDGEMENT Contributions from colleagues, in particular from Dr. Vera Dondur, Prof. Dr. Richard Fiedorow, Dr. Linda Jozefowicz, Dr. Jurgen Ladebeck, Dr. Jurgen Schweckendiek a n d Dip].-Phys. Frank Witzel are gratefully acknowledged. The author is indebted to Dipl. Phys. Uwe Hartel, Mrs. Erika Popovic and Mr. Walter Wachsmann for excellent experimental assistance. REFERENCES 1
2
3 4 5 6 7 8 9 10 11 12 13 14
15 16 17 18 19 20
D. Barthomeuf and A . de Mallmann, in P.J. Grobet, W.J. Mortier, E.F. Vansant and G. Schulz-Eckloff (Editors), "Innovation i n Zeolite Materials Science", Proc. Int. Symp., Nieuw oort, Belgium, Sept. 13-17, 1987, Elsevier, Amsterdam, 1988; Stud. Surface Sci. 37 (1988) 365-374. L.R. Martens, W.J. Vermeisen, D.R. Huybrechts, P.J. Grobet and P.A. Jacobs, i n M.J. Philips and M. Ternan (Editors), "Catalysis: Theory to Practice", Proc. 9th Int. Congress on Catalysis, Calgary, Canada, J u n e 26-July 1, 1988, The Chemical Institute of Canada, Ottawa, 1988, pp. 420-428. J.B. Uytterhoeven, L.C. Christner and W.K. Hall, J. Phys. Chem. 69 (1965) 21172126. G.H. Kuhl, in J.B. Uytterhoeven (Editor), Proc. 3rd Int. Conf. on Molecular Sieves, Zurich, Switzerland, Sept. 3-7, 1973, Recent Progress Reports, Univ. Leuwen Press, Paper No. 127, pp. 227-229. G.H. Kuhl, J. Phys. Chem. Solids38 (1977) 1259-1263. P. Venuto, Adv. Catalysis 18 (1968) 259-371. M.L. Poutsma, in J.A. Rabo (Editor), "Zeolite Chemistry and Catalysis", ACS Monograph 171, Am. Chem. Soc., Washington, D.C., 1976, pp. 437-551 (and references therein). P.A. Jacobs, "Carboniogenic Activity of Zeolites", Elsevier, Amsterdam, 1977. H. Karge, 2. Phys. Chem. [NF] 76 (1971) 133-153 (and references therein). K. Becker, H. Karge and W.-D. Streubel, J. Catalysis28 (1973) 403-411. N. Pesl and €I. Noller, in P. Fejes (Editor), Proc. Int. Symp. Zeolites, Szeged, Hungary, Sept. 11-14, 1978, Acta Universitatis Szegediensis, Nova Series, Acta Physica et Chemica 24 (1978) 267-272. L.M. Kustuv, V.Yu. Borovkov and V.B. Kazansky, Kin. Katal. (Engl. Translat.) 25 (1984)471-477; (Engl. Translat.: 393-398). V.B. Kazansky, V.Yu. Borovkov and L.M. Kustov, in Dechema (Editor), Proc. 8th Congress on Catalysis, Berlin (West), July 2-7, 1984, Verlag Chemie, Weinheim, 1984, pp. 3-14. L.M. Kustov, V.Yu. Borovkov and V.B. Kazansky, in P. J i r u et al. (Editors) "Structure and Reactivity of Modified Zeolites', Proc. Int. Symp. Prague, Czechoslovakia, July 9-13, 1984, Elsevier Amsterdam, 1984; Stud. Surface Sci. Catalysis, 18 (1984) 241,-247. V.B.Kazansky, in G. Ohlmann (Editor), Catalysis on Zeolites, Proc. Int. XX Annual Catalysis Meeting of the GDR, Reinhardsbrunn, GDR, May 26-29, 1987, Elsevier, Amsterdam, 1988; Catalysis Today 3 (1988) 367-372. L.M. Kustov, I.V. Mishin, V.Yu. Borovkov and V.B. Kazansky, Kinet. Katal. 25 (1984) 724-728; (Engl. Translat.: 614-617). Kh.M. Minachev, D.A. Kondrat'ev, A.A. Dergachev, A.Yu. Borodkin a n d I.V. Mishin, Izv. Akad. Nauk SSSR, Ser. Khim. (1983) 1723-1728; (Engl. Translat.: 1562-1566). V.L. Zhobolenko, L.M. Kustov, V.Yu. Borovkov and V.B. Kazansky, Kinet. Katal. 28 (1987) 965-969; (Engl. Translat.: 847-850). S. Beran. P. JirG and L. Kubelkova, J Mol. Catal. 16 (1982) 299-304. J. Ward, in J.A. Rabo (Editor), "Zeolite Chemistry and Catalysis", ACS Monograph 171, Am. Chem. SOC., Washington, D.C., 1976, pp. 118-284 (and references therein).
catalysis
154
21 22
23
24 25 26 27 28 29 30
31 32 33
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
D. Barthomeuf, in J.K. Katzer (Editor), ”Molecular Sieves - U”, Proc. 4 t h Int. Conf. Molecular Sieves, Chicago, USA, April 18-22,1977,ACS Symp. Series 40, Am, Chem. Soc., Washington, D.C., 1977,pp.453-472. J. Dwyer, in P.J. Grobet, W.J. Mortier, E.F. Vansant a n d G. Schulz-Ekloff (Editors), ”Innovation in Zeolite Materials”, Proc. Int. Symp., Nieuwpoort, Belgium, Sept. 13-17,1987, Elsevier, Amsterdam 1988; Stud. Surface Sci. Catalysis 37 (1988)333-354. J.A. Rabo and G.J. Gajda in D. Barthomeuf, E.G. Derouane and W. Hoelderich (Editors), ”Guidelines for Mastering the Properties of Molecular Sieves - Relationship between the Physicochemical Properties of Zeolitic Systems a n d Their Low Dimensionality”, NATO AS1 Series B: Physics, Volume 221,Plenum Publishing Corp., in press. D. Barthomeuf, in P. Fejes (Editor), Proc. Int. Symp. Zeolites, Szeged, Hungary, Sept. 11-14,1978;Acta Universitatis Szegediensis, Acta Physica et Chemica, Nova Series 24 (1978)71-75. W.J. Mortier, J. Catal. 55 (1978)138-145. P.A. Jacobs, W.J. Mortier and J.B. Uytterhoeven, J. Inorg. Chem. 40 (1978)19191923. D. Barthomeuf, J.Phys. Chem. 83 (1979)249-256. P.A. Jacobs, Cat. Rev. Sci. Eng. 24 (1982)415-440. W.J. Mortier, J. Sauer, J.A. Lercher and H. Noller, J. Phys. Chem. 88 (1984) 905 912. W.J. Mortier in Y. Murakami, A. Iijima and J.W. Ward (Editors), ”New Developments in Zeolite Science and Technology”, Proc. 7th Int. Zeolite Conf., Tokyo, J a p a n , August 17-22,1986,Kodansha, Tokyo and Elsevier, Amsterdam, 1986,pp. 423-428. D. Barthomeuf, Mat. Chem. Phys. 17 (1987)49-71. J. Sauer in J. Klinowski and P.J. Barrie (Editors), Proc. Annual Meeting of the British Zeolite Ass., Cambridge, U.K., April 17-19,1989;Elsevier, Amsterdam, Studies Surf. Sci. Catalysis 52,1989,pp. 73-90. S.H. Abbas, T.K. A1 Dawood, J. Dw er, F.R. Fitch, A. Geor opoulos, F.J. Machado and S.M. Smyth, in B. Imelik e t ay. (Editors), ”Catalysis y Zeolites”, Proc. Int. Symp., Ecully (Lyon), France, 1980,Studies Surface Sci. Catalysis 5 (1980)pp. 127-134. H.G. Karge, K. Hatada, Y. Zhang and R. Fiedorow, Zeolites 3 (1983)13-21. P.A. Jacobs and K.H. Beyer, J. Phys. Chem. 83 (1979)1174-1177. A.E. Hirschler, J. Catal. 2 (1963)428-439. H.G. Karge, Surface Sci. 40 (1973)157-168. L.N. Izmailova and E.I. K o h v , Kinet. Katal. 18 (1977)488-492;(Engl. Translat.: 409-412). J.A. Ripmeester, J. Am. Chem. SOC.105 (1983),2925-2927. J.H. Lunsford, W.P. Rothwell and W. Shen, J. Am. Chem. SOC.107 (1985)1540 1547. D. Freude, M. Hunger and 11. Pfeifer, Z. Phys. Chem. [NF] 152 (1987)429-440. H. Ernst, L). Freude, M. Hunger, H. Pfeifer and B. Seiffert, Z. Phys. Chem. (Leipzig) 268 (1987)304-314. H. Pfeifer, Colloids Surf. 36 (1989)169-177. K. Hatada, Y.Ono and Y. Ushiki, Z. Phys. Chem. [NF],117 (1979)37-42. H.G. Karge, Z. Phys. Chem. [ N F ] 122 (1980)103-116. J.W. Ward, J. Colloid Interface Sci., 28 (1968)269-279. H.G. Karge, in J. Katzer (Editor), ”Molecular Sieves - II”, Proc. 4 t h Int. Conf. Molecular Sieves, Chicago, USA, April 18-22,1977;ACS Symp. Series 40, Am. Chem. SOC.,Washin n, D.C., 1977,pp.584-595. C.L. Angell and M.$tOHowell, J. Phys. Chem. 73 (1969)2551-2554. C.1,. Angell and P.C. $chaffer, J. Phys. Chem. 70 (1966)1413-1417. F.K. Cannings, J. Phys. Chem. 72 (1968)4691-4693. J.P. Giison, G.C. Edwards, A.W. Peters, K. Rajakopalan, R. Worspechev, T.G. Roberie and M.P. Shatlock, J.C.S. Chem. Commun. (1987)91-92. E. Brunner, H. Ernst, D. Freude, M. Hun er and H. Pfeifer, in P.J. Grobet, W.J. Mortier, E.F. Vansant and G. Schulz-Etloff (Editors), ”Innovation in Zeolite Materials”, h J C . Int. Symp., Nieuw oort, Belgium, Sept. 13-17,1987,Elsevier, Amsterdam 1988;Stud. Surface Sci. t a t a l y s i s 37 (1988)pp. 155-165.
%
155
53
54 55 56 57 58 59 60 61 62 63 64
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
82 83 84 a5 86 87 a8
89
P.J. Crobet, €I. Geerts, M. Tielen, J.A. Martens and P.A. Jacobs, i n H.G. K a r g e a n d J. Weitkamp (Editors) ”Zeolites a s Adsorbents, Catalysts a n d Detergent Builders - Innovation and Application”, Proc. Int. Symposium on Zeolites, Wurzburg, FRG, Sept. 4-6, 1988, Elsevier, Amsterdam, 1989; Stud. Surface Sci. Catalysis, 37 (1989)pp.721-734. l’h. K. Hughes and H.M. White, J . Phys. Chem. 71 (1967) 2192-2201. A. Bielanski a n d J. Uatka, Bull. Acad. Polon. Sci., 22 (1974) 341-350. A. Bielanski and J. Datka. J. Catal. 32 (1974) 183-189. J.B. Uytterhoeven, P. Jacobs, K. Makay and R. Schoonheydt, J. Phys. Chem. 72 I 1968) 1768-1775. H . Pfeifer, I). Preude and M. Hunger, Zeolites 5 (1985) 274-286. J.C. Vedrine, A. Auroux, V. Bolis, P. Dejaifve, C. Naccache,P. Wierzchowski, E.G. Derouane, J.B. Nagy, J.-P Gilson, J.H.C. v a n Hooff, J.P. v a n den Berg a n d J. Wolthuizen, J. Catal. 59 (1979)248-262. U . Lohse, B. Parlitz and V. Patzelovti, J. Phys. Chem. 93 (1989) 3677-3683. R.B. Shannon, R.H. Staley, A.J. Vega, K.X. Fischer, W.H. B a u r and A. Auroux, J. Phys. Chem. 93 (1989)2019-2027. H.G. Karge, J. Ladebeck, Z. Sarbak and K. Hatada, Zeolites 2 (1982) 94-102. W.O. Haag, in D. Olson and A. Bisio (Editors), Proc. 6th Zeolite Conf., Reno, Nevada, USA, J u l y 10-15,1983,Butterworths, London 1984, pp. 466-478. H.G. Karge, Y. Wada, J. Weitkamp, S. Ernst, U. Girrbach a n d H.K. Beyer i n S. Kaliaguine and A. Mahay (Editors), ”Catalysis on the Energy Scene”, Proc. 9th Canadian Symp. on Catalysis, Quebec, Canada, Sept. 30-Oct. 3, 1984; Stud. Surface Sci. Catalysis, 19 (1984) pp. 101-111. J . H . Lunsford, J. Phvs. Chem. 72 (1968)4163-4168. J.H. Lunsford; J. Caial. 14 (1969) 379-385. J.H. Lunsford, L.W. Zingery and M.P. Rosynek, J. Catal. 38 (1975) 179:188. E’. Witzel. H.C. Karee. A. Gutsze and U. Hartel. to be published in Zeolites. Z. Tanabe, ”Solid Aiids and Bases”, Kodansha,i‘okyd, Academic Press, New York and London, 1970. Z. Tanabe, in J. Philips and M. Ternan (Editors), ”Catalysis-Theory to Practice” Proc. 9th Int. Congress Catalysis, Calgary, Canada, J u n e 26 - J u l y 1, 1988, T h e Chemical Institute of Canada, Ottawa, 1988, pp. 85-113. S.L. Soled, G.B. McVicker, L.L. Murrell, L.G. Sherman, N.C. Dispenziere, S.L. Hsu, D. Waldman, J. Catalysis 111 (1988) 286-295. H.A. Benesi and B.H. Winquist, Advances in Catalysis 27 (1978)97-182. H.A. Benesi, J. Phys. Chem. 61 (1957)970-973. R. Beaumont and D. Barthomeuf, J. Catal. 26 (1972) 218-225. W. Kladnig, J. Phys. Chem. 80 (1976)262-269. K.U.R. Kittelmann, PhD Thesis, Technische Hochschule Darmstadt, 1979. K. Unger, K.U. Kittelmann and W.K. Kreis, J. Chem. Tech. Biotechnol. 31 (1981) 435-469. M. Deeba and W.K. Hall, J. Catal. 60 (1979)417-429. M. Deeba and W.K. Hall, Z. Phys. Chemie [NF] 144 (1985) 85-103. J. Schweckendiek, PhD Thesis, Fritz-Haber-Institut der Max-Planck-Gesellschaft, Berlin, und Universitat Bremen, 1982.A.K. Gosh and G. Curthoys, in S. Kaliaguine a n d A. Mahay (Editors), ”Catalysis on the Energy Scene”, Proc. 9 t h Canadian Symp. on Catalysis, Quebec, C a n a d a , Sept. 30 - Oct. 3, 1984; Elsevier, Amsterdam, 1984; Stud. Surface Sci. Catalysis 19 (1984) 147-154. M.W. Anderson and J. Klinowski, Zeolites6 (1986) 150-153. J.L. Falconer and J.A. Schwarz, Cat. Rev. Sci. Eng. 25 (1983) 141-227. I,. Forni and E. Magni, J. Catal. 112 (1988)437-443. V.V. Yushchenko, C.J. Vanegas and B.V. Romanovskii, Reakt. Kinet. Catal. Lett.,40 (1989) 235-240. H.G. Karge and Z. Zheng, publication in preparation. E. Dima and L.V.C. Rees, Zeolites 7 (1987)219-227. K. Hashimoto, T. Masuda and T. Mori, in Y. Murakami, A. Iijima a n d J.W. W a r d (Editors), ”New Developments i n Zeolite Science and Technology”, Proc. 7th Int. Zeolite Conf., Tokyo, J a p a n , Aug. 17-22, 1986; Kodansha, Tokyo and Elsevier, Amsterdam, 1986, pp. 503-510. H.G. Karge and V. Dondur, J. Phys. Chem. 94 (1990) 765-772.
156
90 V. Dondur and H.G. Karge, Surface Sci., 189/190 (1987)873-879. 91 H.G. Karge, M. kaniecki, M. Zidek, G. Onyestyak, A. Kiss, P. Kleinschmit and M. Siray, in P.A. Jacobs and R.A. van Santen (Editors) "Zeolites: Facts, Figures, Future", Proc. 8th Int. Zeolite Conf., Amsterdam, The Netherlands, July 10-14, 1989; Elsevier, Amsterdam, 1989; Stud. Surface Sci. Catalysis, 46 (1989) pp. 13271337. 92 H.G. Karge and J. Schweckendiek, in D. Shopov e t al. (Editors), Proc. 5th Int. Symp. on Heterogeneous Catalysis, Varna, Bulgaria, Oct. 3-6, 1983; Publishing House of the Bulgarian Academy of Sciences, Sofia, 1983, pp. 429-439. 93 J . Take, €1. Yoshioka and M. Misono, in M.J. Philips and M. Ternan (Editors), "Catalysis: Theory to Practice", Proc. 9th Int. Congress on Catalysis, Calgary, Canada, J u n e 26-July 1, 1988, The Chemical Institute of Canada, Ottawa, 1988 Vol. 1, pp. 372-379. 94 A. Auroux, P. Wierzchowski and P.C. Gravelle, Thermochimica Acta 32 (1979) 165-170. 95 A. Auroux, P.C. Gravelle and J.C. Vedrine, in L.V.C. Rees (Editor),Proc. 5th Int. Zeolite Conf., Naples, Italy, J u n e 2-6,1980; Heyden, London, 1980, pp. 433-439. 96 A. Auroux, in P.J. Grobet, W.J. Mortier, E.F. Vansant and G. Schulz-Eckloff (Editors), "Innovation in Zeolite Materials Science", Proc. Int. Symp., Nieuwoort, Belgium, Sept. 13-17, 1987, Elsevier, Amsterdam, 1988; Studies in Surf. gci. and Catal sis37 (1988) pp. 385-391. 97 A. Auroux, Y.8. J i n and J.C. Vedrine, Applied Catalysis 36 (1988)323-330. 9 8 A.L. Klyachko, T.R. Brueva, I.V. Mishin, G.I. Kapustin and A.M. Rubinstein, in P. Fejes (Editor), Proc. Symp. on Zeolites, Szeged, Hungary, Sept. 11-14, 1978; Acta Universitatis Szegediensis, Acta Physica et Chemica 24 (1978) pp. 183-188. 99 A.L. Klyachko, G.I. Kapustin, T.R. Brueva and A.M. Rubinstein, Zeolites 7 (1987) 119-122. 100 I . Bankus, J . Valyon, (3.1. Kapustin, D. Kallo, A.L. Klyachko and T.R. Hrueva, Zeolites8(1988) 189-195. 101 G.I. Kapustin, T.R. Brueva, A.L. Klyachko, S. Beran and B. Wichterlova, Applied Catalvis 42 (1988)239-246. 102 11. l'hamm, J. Phys. Chem. 91 (1987) 8-11. 103 H. Thamm. H . 4 . Jerschkewitz and H. Stach. Zeolites 8 (1988) 151-153. 104 S.S. Khvoshchev, V.E. Skazyvaev and E.A. Vasileva, in L.V.C. Rees (Editor), Proc. 5th Int. Zeolite Conf., Naples, Italy, J u n e 2-6, 1980; Heyden, London, 1980, pp. 476-482. 105 I,. Jozefowicz, H.G. Karge and F. Asmussen, to be published in J. Phys. Chem. 106 M.F.M. Post, T. Huizinga, C.A. Emeis, J.M. Nanne and W.H.J. Stork, in H.G. Kargc and J. Weitkamp (Editors) "Zeolites a s Adsorbents, Catalysts and Detergent Builders - Innovation and Application", Proc. Int. Symposium on Zeolites, Wurzburg, FRG, Sept. 4-6, 1988, Elsevier, Amsterdam, 1989; Stud. Surface Sci. Catalysis, 37 (1989) pp. 363-375. 107 P.A. Jacobs and J.B. IJytterhoeven, J. Chem. SOC.,Faraday Trans. I, 69 (1973) 359-371. 108 PA. Jacobs, J.A. Martens, J. Weitkamp and H.K. Beyer, Faraday Discuss. 72 (1981) 353-369. 109 L. Kuhelkova, S. Beran and J.A. Lercher, Zeolites9 (1989)539-543. 110 C. Naccache, C. Fang Ren and G.Coudurier, in P.A. Jacobs and R.A. van Santen (Editors) "Zeolites: Facts, Figures, Future", Proc. 8 t h Int. Zeolite Conf., Amsterdam, The Netherlands, July 10-14, 1989; Elsevier, Amsterdam, 1989; Stud. Surface Sci. Catalysis, 46 (1989) pp. 661-668. 111 H. Pfeifer, 1). Freude and M. Hunger, Zeolites 5 (19859 275-286. 112 J. Weitkamp, in L.V.C. Rees (Editor), Proc. 5th Int. Zeolite Conf., Naples, Italy, J u n e 2-6,1980; Heyden, London, 1980, pp. 858-865. 113 H.G. Karge, H. Kosters and Y. Wada, in D. Olson and A. Bisio (Editors), Proc. 6th Int. Zeolite Conf., Reno, Nevada, USA, July 10-15, 1983; Butterworths, London, 1984, pp. 308-315. 114 G. Bourdillon, C. Gueguen and M. Guisnet, Appl. Catalysis61 (1990) 123-139.
G. Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
157
ACDITY and BASICITY in ZEOLITES
D. BARTHOMEUF
Laboratoire de RCactivitC de Surface et Structure, URA 1106, C.N.R.S., UniversitC P. et M. Curie, Paris, France
ABSTRAm The paper presents correlations between the acidity and the basicity of zeolites. Basic sites are of the Broensted or Lewis type. The last ones which occur the more frequently consist of the oxygen atoms of the framework. The charge on the oxygen, which gives information on the basic strength, can be calculated and the basicity can be experimentally measured using for instance pyrrole as a probe molecule. Several parameters are identified as affecting the basicity. They are the electronegativity of the framework atoms, the bonds angles and bond lengths, the structure ionicity, the crystallographic sitting of the oxygen, the location of Al (pairs, Al gradients). The properties of basic zeolites in adsorption and catalysis, their resistance to poisons, are presented in relation with their basicity.
INTRODUCTION For many years the attention on the main chemical properties of zeolites has been focussed on their acidity which gives rise to a large number of applications in acidic catalysis. Little has been published on the zeolites basic properties and on their applications in catalysis and adsorption (1-5). It is well admitted that Broensted acids and bases are conjugated and that weak acids have strong conjugate bases and reciprocally strong acids have weak conjugated bases (6). The existence of protons in zeolites has to be associated with that of basic sites. The paper will consider the factors which characterize the number and strength of both sites. It will not consider the case of cationic framework which are not known presently, but it will focus on the family of (Si-AI) zeolites giving reactions typical of basic centers. One of the first reaction tested in that sense in the side chain alkylation of toluene with methanol (1,2) which is favoured on basic zeolites over the ring alkylation catalyzed by acids. The catalysts are X type faujasites i.e. materials with a high A l content and exchanged with Rb and Cs in the alkaline cation series. Dehydrogenation of isopropanol is also favoured on the same type of catalysts (7).
158
NATURE OF ACID AND BASIC SITES The study of protonic and non protonic acidity of zeolites has been the subject of very important research. The Lewis acidity arises from Al or other cationic species, alkaline, alkaline earth or other cations. In a similar way the basic sites may be considered of the Broensted or Lewis type. The first one consists of basic OH groups and the second of the framework oxygen atoms. The charge
on these oxygen determines their strength. Some major differences have to be noted specifically for zeolites in the characteristics of the acidity and of the basicity. Considering first protons, their theoretical number is equal to the number of AlOb tetrahedra whose charge has to be neutralized, i.e. to the A1 content. It is also accepted that their strength increases as the A1 content decreases. The protons are mobile, jumping from oxygen to oxygen and may approach the reactant to form a cxbocation. The number of Lewis acid sites (extraframework Al or compensating cation) depends on the zeolite pretreatment for the Al or on the degree of cation exchange for the other cations. For basic zeolites, most of the samples used do not contain much Broensted basicity, but mainly Lewis basicity. The number of potential basic sites is equal to that of oxygen atoms in the framework. The difference from one zeolite to the other is determined only by the charge on the oxygen, i.e. by the basic strength and by the degree of heterogeneity of the charge distribution, some oxygen being possibly more basic than others due to different bond angles and bond lengths. In addition and by contrast with the protons these framework oxygen are not mobile and the reactant has to adjust itself to the structure geometry which may induce structural selectivities in adsorption and catalysis.
STRENGTH OF BASICITY AND ACIDITY A large number of studies, experimental (8) or theoretical (9) have been devoted to the characterization of the acid sites, number, strength and location. The calculation based on the principle of the equalization of the Sanderson electronegativity of the atoms constituting a compound (10) gives an approach of the average charge on the proton (1 1) or on the oxygen (1 2,13). The intermediate electronegativity of a substance. Sint. gives the mean electronegativity of
the atoms of the compound, which is the same for all of them. It suggests to rank the zeolites according to their Sinl value. The results depend in fact only o n the chemical fomiula and not on structural parameters. It will be shown funher that some advantage may be taken from this. Acidity The known increase in protonic acid strength with decrease in A1 content is exemplified in figure
1, curve a, giving the charge on the proton calculated from Sin1 (10,ll). The acid strength can also be decreased by exchange of a part of the protons by metallic cations, the stronger sites being exchanged first. It is also well accepted that these results obtained by considering Sinl values give only trends. Each zeolite structure and topology will superimpose other specific parameters.
159
Basicity The paper is focussed mainly on Lewis basicity since the oxygen atoms exist in all the zeolites frameworks. The Broensted basicity is less often present and in addition some OH may be amphoteric which makes the study less clear (14). The Lewis base strength can be calculated or determined experimentally. i) Theoretical approach Several methods can be used. Ab initio quantum chemical calculations have been applied to small clusters (15.16). The use of the Sanderson electronegativity equalization principle (10) gives an overage oxygen charge value. Curve b in figure 1 reports as a function of the A1 content the average charge on oxygen for a protonic zeolite. The oxygen charge decreases as the zeolite contains less Al, i.e. as the charge on the proton rises. This is in line with the concept of conjugated acid-base pairs (12). One may at that point wonder how the material keeps its electrical neutrality with a constant number of theoretical basic framework oxygen. It has to be kept in mind that zeolites are oxyacids. Their formula can be written as T On (OH), by analogy with other oxyacids such as CI On(OH), (12,17). In zeolites m expresses the Al atomic fraction i n the framework and consequently the average number of protons per TO2 tetrahedron (n + m = 2). The fomiula for a theoretical fully protonic X type zeolite (AI/AI + Si = 0.45) is T01.55 (OH)o.45. For a ZSM-5 material with AI/AI + Si = 0.04 (Si/AI = 24) it is T01.96 (0H)o.w. It is known in oxyacids that the acid strength of the proton increases with n (18), i.e. with the number of oxygen free of protons. This is in agreement in zeolites with the higher acid strengths at the lower Al contents, i.e. lower m. The framework negative charge which compensates the positive charge of all the protons is shared between a larger number of oxygen free of protons at low content, i.e., at high n. The charge bom by each individual oxygen then decreases and the electrical neutrality is maintained. The calculation of the average oxygen charges of zeolites fully saturated with alkaline cations using the same Sanderson approach is reported in figure 1 (curves c to g) as a function of the Al content. The change in oxygen charge as a function of cations electronegativity is given in figure 2 for different Al contents which correspond to usual zeolites compositions (X, Y,L, mordenite, beta and ZSM-5). Both figures 1 and 2 show that the higher basicity strengths are obtained at high Al content and low cation electronegativity. The lowest oxygen charge is obtained for the H forms. Besides the chemical composition other parameters may affect the charges and their distribution in the framework. It has been shown for a long time that the charge on the oxygen is larger as the TOT or OTO angles are narrower and as the TO bonds are longer (19). Then the charge on oxygen depends on the structure and in a given structure on the oxygen crystallographic site considered (4 types in faujasite, 10 in mordenite, 26 in ZSM-5). The introduction in a framework of elements other than Si and Al will change not only the electronegativity but also the bond angles and bond lengths and very likely the charge on the oxygen. For instance the replacement of Si by Ge in an X zeolite increases the measured oxygen basicity (20) while the simultaneous rise in Sint value would suggest a lowering of basicity. One may expect that the presence of other elements like B, P or of all the elements known to exist in S A W S , MeAPSO's and EIAPSO's materials also generates very
160
-&1
62 n
E
aJ 0.4 m
0.4
J
7
m ru
X %
0 0
r 0
3
0.3
0.3
aJ
%
n
m L
0 1
r m
u
0.2
0.2
ln lD 7J
0.1 0
0.1
0.2
0.3
Al/Al + S i
Fig, 1. Calculated charge on oxygen and on proton as a function of the A1 content. (a) (b) Protonic zeolite and zeolites exchanged with (c) Li, (d) Na, (e) K, (f) Rb, (9) Cs specific basicities. As to the structure effect, using the Electronegativity Equalization Method @EM) (21) which differs from the Sanderson approach (10) in that it takes into account structural parameters, it has been shown that the ionicity of the bonds in a zeolite depends on its crystallographic structure. Usual zeolites are ranked in the order of ionicity FAU < LTL < MFI < MOR (21) which is different from their Sin, order ( FAU < LTL < MOR < MFI (figures 1 and 2)). This means that the ionic charge on oxygen and proton in mordenite will be increased compared to the ones in a less ionic structure, even of similar composition. In addition and particularly for zeolites with a low A1 content the inhomogeneity in the A1 distribution (possible paired sites, gradient of A1 concentration ...) will also make the true charge of some specific oxygen atoms different from the average calculated one. One may expect that the experimental measurement of basicity will reflect these influences of the structure and of Al location. ii) Experimental measurement of basicity Several methods have been proposed to measure the basicity of solid materials (22,23). Some of them have been applied to the field of zeolites. The use of Co;? as adsorbate is not well suited since infrared studies showed that physical and chemical adsorption occur simultaneously (24) and that various and not well defined carbonated species may be generated (25). The shift in the NH infrared wavenumber of pyrrole adsorbed on basic zeolites may be used to evaluate the basic strength (20). It showed that NaX are more basic than NaY (20) which is in line with figures 1,2. Applied to various zeolite structures exchanged with alkaline cations, it is seen (Table I) that some samples do not show any basicity like ZSM-5 for instance, even in the Cs form (12). For the faujasite structure pyrrole adsorption detected basicity for all the X samples from LiX to CsX. In theyseries LiY was not basic and NaY close to the limit detected. These results suggest to draw in
161 Rb Na Cs K Li
-
-61
3.0
c
g 0.4
- - - - - -3.5 -------
3
0 C
0 a,
+ .-C v)
0.3
1 m U
,4.5
9 4 0.2
I
0
1
1
2
8
E l ectronegativ i ty
Fig. 2. Calculated Sint and charge on oxygen as a function of cation electronegativity for usual A1 contents of zeolites (a) X, (b) Y, (c) L, (d) MOR, (e) Beta, (0 ZSM-5. TABLE I Shift of NH vibration of pyrrole adsorbed on zeolites and calculated average charge on oxygen Zeolite csx NaX KY NaY KL Na-MOR Na-Beta CS ZSM-5 Na ZSM-5
240 180 70 30-40 30 30 30 0 0
- 0.461 - 0.413
- 0.383 - 0.352
- 0.356 - 0.278 - 0.240 - 0.236
- 0.225
(a) Shift of NH from the liquid (from 12) (b) Charge on oxygen calculated from Sanderson electronegativity (10-12) figures 1 and 2 the dashed lines which define the range of oxygen charges above which zeolites are basic and below which the charge on the oxygen is not high enough to generate a significant basicity. Such a range will be confirmed later in catalytic studies. The limit is around 3.5 for Sint and -0.36 for the oxygen charge. It follows that none of the hydrogen form has strongly basic oxygen. This is in general agreement with acid catalysis results. In the alkaline cation materials not only ZSM-5 but also beta and mordenite should not show any basicity. Nevertheless pyrrole adsorption studies could evidence an interaction of pyrrole with the basic sites of these two zeolites (Table I). The oxygen charge in Table I is calculated for the actual composition of the materials
162
studied. It takes into account the fact that the cation exchange is not always 100 % for the major cation. For X, Y and L zeolites the shifts of NH wavenumber upon pyrrole adsorption are in line with the expected charge on oxygen while the NH shift obtained for mordenite and beta demonstrate the existence of basic sites. These two zeolites are then examples of a strong influence of the structure and/or of the Al distribution on the charge on oxygen which is not taken into account in the Sanderson electronegativity calculation. In mordenite, cations are in location of low symmetry and a rather large range of TOT angles exists (26). This structure induces a higher ionicity of bonds than other structures do (21). In addition Al pairs have been proposed to exist in niordenites (27). The existence of small bond angles, of a high bond ionicity and of Al pairs may explain the presence of basic sites. The structure of beta zeolite has been determined (28) but little is published for this zeolite on bond angles, bond lengths, bond ionicity and A l location. The behaviour of beta zeolite in catalysis (29) suggests that it is less strongly acidic than its chemical composition would suggest. Both this too weak acidity and simultaneously the present too high basicity show a very strong influence on the framework charges of parameters related to the structure and very likely on A1 location at this very low Al content (Al/Al+Si in the range of 0.07 to 0.09). In the case of faujasite another probe has been also used to evaluate the basicity. Benzene just fits the twelve ring window where it can interact weakly through the -CH with the basic framework oxygen (30-33). I n the alkaline cation series the infrared shift of yCH and the amount adsorbed in the 12-R window increases from the Li to the Cs form reaching a significant shift for the more basic Cs materials (31). The LiY sample adsorbs benzene only very weakly on the oxygen showing a weak zeolite basicity. These results are in agreement with the pyrrole adsorption results and with the calculated results of figures 1 and 2 showing the limit basicity range. In beta zeolite a significant interaction of benzene with the 12 R window is seen (34) which is surprising if one takes into account only its Al content (i.e. Sinl) but which is in line with the pyrrole results. iii) Correlations aciditv-basicitv The opposite change in the charge of proton and oxygen seen in figure 1 should also exist for the acidic strength of metal cation versus the oxygen basicity when no protons are present. It was shown by the interaction of pyridine with the alkaline cations acting as Lewis sites that the Lewis acidity decreases from Li to Cs in the alkaline cation series in faujasites (12,35). It was also shown that the acid strength for a same cation is higher in Y than X (12) which is parallel to the change in H+ charge with the Al content. One may conclude that generally speaking the acid strength of a metal cation should be high at low Al content, i.e. in highly siliceous zeolites or dealuminated materials, while the framework basicity should be low. This agrees with the existence of conjugated acid-base pairs. At this point the advantage of considering Sin1 for predicting trends in acid-base propenies of zeolites has to be considered. The cases of mordenite, beta and GeX clearly show that the predictions based on chemical composition alone are not confirmed by experimental results. A major conclusion is that a strong influence of other parameters, (structure, topology, bond ionicity, Al distribution) becomes very important for those materials. This opens a field of research for
163
Characterizing what in zeolites properties (acido-basicity, ion exchange, influence on supported metals ...) is relevant mainly to chemical composition or to other parameters. The characterization of these parameters will lead to a better understanding and then to prediction of the properties. Another consequence is the guide line given by these results for obtaining highly basic zeolites. Following the trend of increasing basicity at high Al content, i n the faujasite family the highest possible value is obtained in X. In mordenite and beta for which AVAI+Si ratios are 0.17 and 0.07 respectively one may expect quite higher basicity at high Al levels by a combined effect of Sin*, structure, topology and Al location. Once a given structure is ranked by comparison with faujasite taken as a reference, it would be possible, for this structure to prepare a large series of samples with predetermined Lewis acidity and basicity by choosing the right Al content and the right cation. The family of samples could then be used to direct the properties in adsorption and catalysis. USE OF BASIC ZEOLITES IN ADSORPTION In adsorption, non protonic zeolites are commonly used i n order to prevent any catalytic transformation of the adsorbates on acid sites. The Na or Ca forms of A, X, Y, mordenite ... have been used extensively in fundamental studies or in applications (36,37). Besides separations of mixtures based on differences in the kinetics of diffusion of the adsorbates in the zeolite pores a number of separations rely on subtle differences between the adsorbates when they are in the adsorbed state in the zeolite cavities and channels. Two examples will be given. The first one concerns the separation of the four C8 aromatics. They have very close boiling points and selective adsorption on zeolite is well suited in that it avoids very difficult distillation or crystallisation separations. The preference of the zeolite for a given isomer strongly depends on the solid. In the faujasite series KY prefers p-xylene (38-39). NaY m-xylene (40) and RbX and CsX ethylbenzene (41). Ethylbenzene is rejected from CaY or CaX (42). A correlation was observed between the Y zeolite acidity and the p-xylene selectivity (38). In the X and Y zeolites it was shown that the order of preference for the isomers may be related to the adsorbent intermediate electronegativity (5) i.e. to its basic character. The relative x basicity (43) of the aromatics increases from ethylbenzene to p-xylene and to m-xylene (Table 11). The order of increased K basicity parallels that of increased electronegativity of the zeolite selective for each isomer (Table 11). The Cg aromatics interact only with the cations through their 7c electrons and due to steric hindrance not with the framework oxygen as benzene does. The results of Table I1 suggest that the interactions aromatic-cation are governed by a balance between the aromatic basicity and the Lewis acid character of the cation in the zeolite, itself related to the zeolite electronegativity. For beta zeolite, Figure 3 reports selectivity values in the separation of ethylbenzene from p-xylene in a mixture of the four Cg aromatics (45). The selectivity factor a12expresses the ratio of the concentrations of compound 1 in the zeolite and i n the liquid phase to the corresponding concentrations of compound 2. Values of a12 higher than one indicate the preferred adsorption of compound 1 in the zeolite. Figure 3 shows that any form of beta zeolite tested (protonic or saturated with alkaline cations) is selective for ethylbenzene over p-xylene (and over the o - and m - forms
164
also) (45) as the basic faujasites NaX to CsX in Table 11. TABLE I1 K basicity of c8 aromatics and Sin1 values for zeolites selective for each isomer
Zeolite
Selectivity for EB, PX or MX@)in a mixture of the four C8 aromatics
Sint
Cs5oNa3d( Rb53Na33X RbyjMggNa14X KX NaX K%Na2Y NaY
3.03 3.08 3.18 3.15 3.25 3.38 3.54
-
-
-
EB
PX
Mx
(1.5)@)
(1.6)
(2.0)
Ref
41c 41, 41, 41c 41, 4 h 4ob
X X
X X
X X X
(a) EB : Ethylbenzene, PX : paraxylene, Mx : metaxylene (b) K basicity (from 43 )
Rb Cs K
H
Na
2 X
a m
--. W
0 1
0 9
4.0
4.1
'int
Fig. 3. Separation coefficient between ethylbenzene and p-xylene as a fonction of the intermediate electronegativity of beta zeolite exchanged with (o)Cs, (0) Rb, (n) K, @) Na, (A) H. The values of Sint for beta lie in a range - 4 to 4.2 - which is usually obtained in acidic zeolites (Figure I). As previously for pyrrole (Table I) or benzene adsorption, beta zeolite looks like much more basic that one would expect from the Sin1 calculation. This confirms the strong influence of the structure or/and Al location on the charges born by the atoms in this zeolite. This is also in line with the moderate acidity of this material (29). The second example of separation on basic zeolites is given by the changes in the separation
165
coefficient a between methyl-2 and methyl-1 naphtalenes (46). The authors observed a decrease in the C I ~ M N I I M N coefficient as Sin, increases upon alkaline and alkaline earth ion exchange. The curves are distinct for X and Y zeolites.
USE OF BASIC ZEOLITES IN CATALYSIS Basic catalvsis For a long time reactions involving zeolites containing metallic cations but no protons have been proposed (47). Many of them may be relevant to basic catalysis, like reactions involving H2S. For the transformation of furan to thiophene (47) it has been shown that H2S dissociation occurs to generate mainly HS- and H+ species which are adsorbed in NaX on the Na+ cation and framework oxygen respectively (48). The activity of X zeolites exchanged with the different alkaline cations in the reactions of CH3OH
CH3SH and C2HsOH
a C2HsSH is the
highest for the zeolites with lowest Sint, i.e. for the most basic (49). Reactions like isopropanol dehydration, alkylation with methanol of toluene or of other aromatics and reactions involving heteroatoms (N,O ...) have been reviewed elsewhere ( 3 3 . It has to be pointed out that for the toluene alkylation with methanol the alkylation of the ring or of the chain is obtained on acidic or basic zeolites respectively. A correlation was shown between the extent of alkylation in each case and the electrostatic potential of the cation, the lower the electrostactic potential, the higher the side chain alkylation (50). A similar relationship was latter obtained considering the zeolite intermediate electronegativity and the yield of products. Interestingly the limit of acidic or basic zeolites defined by the selectivity of this reaction is at a value of Sinl close to 3.5 which is in agreement with the limit range already reported in figures 1 and 2. Recently it was shown that the condensation of benzaldehyde with derivatives of malonic esters has an increased activity for the fmjasites giving the highest NH shift in adsorbed pyrrole i.e. having the strongest basicity from Table I. Similarly GeX zeolite is very active in condensation reactions (53) which is in line with the strong basicity of this zeolite (20). Supported metals A few works are related to the study of metal supported basic zeolites. It is well known in the field of acidic zeolites that the electronic properties of the metal are modified through an electron transfer from the zeolite acid sites to the platinum particles (54). One may expect for basic zeolites an effect in the opposite direction. The influence of such a metal-support interaction explains that the extent of N i reduction is increased for the more basic NaX compared to NaY and that the presence of Ce3+ cations , able to give Ce4+ ions and one electron favors this reduction in CeNaX (55). The catalytic properties of the supported metallic particles also depend on the acidic or basic
character of the zeolites. The hydrogenation of CO to olefins and paraffin over RuY zeolites increases for the formation of unsatured hydrocarbons as the zeolite basicity rises from Ru supported on Nay, to KY and CsNaY (56). In the case of PtL zeolites in the alkaline cation form the activity for the dehydrocyclization of n-hexane to benzene increases from Li to Cs form. This was related to an increased electron
166
transfer from the zeolite to the Pt particles (57). Such an explanation appears to be also valid for the alkaline earth forms (58). Figure 4 reports the change in benzene yield in n-hexane aromatization as a function of oxygen charge and Sin1 evaluated from (57,59). The more active catalysts (Rb and Cs
3.65 1
3.60
3.55
'int
'.
I
I
A
-0.34
oxygen c h a r g e
Fig. 4. Benzene yield in n-hexane dehydrocyclization at 733 K (ref 59) as a function of Sinl and oxygen charge of PtL zeolite exchanged with (0)H, ( 0 ) Li, (A) Na, 0) K, g)Rb, (A) Cs. forms) for the production of benzene have the lowest Sinlvalues. This reaction is monofunctionnal, the active sites being the Pt atoms (59). The results confirm that a high zeolite basicity increases the aromatization activity of the Pt sites. The change in the electronic state of Pt was also shown from a decrease in the amount of CO adsorbed on Pt and in the benzene hydrogenation activity in L materials (58). In a series of Pt faujasites of the X or Y type exchanged with H, Na or Cs cations, the wavenumber of CO adsorbed on Pt particles decreases simultaneously with the benzene hydrogenation activity from the acidic to basic zeolites in the order PtHY > PtHNaY > PtNaY > PtNaX > PtCsX (44a,60,61). Simultaneously the apparent activation energy increases from 48 to 120 KJ.mole-1 which precludes any explanation based on diffusion due to the steric hindrance of the big Cs ions. The results show that the intrinsic electronic properties of the platinum are gradually changed following the decrease in acidity and increase in basicity of the zeolite. Figure 5 gives a scheme summarizing the continuous changes in properties from the strongly acidic to the highly basic zeolites. Choosing the right Al content and the right cation is a way to adjust for a given zeolite the properties to the expected behaviour of the solid in adsorption or in catalysis. Resistance to sulfur The experimental results presented here indicate that the zeolite basicity may direct the properties of supported metal . Changes i n the metal particles properties may be induced not only by the
167
zeolite but also by the species adsorbed on the metal. For instance sulfur atoms formed by decomposition of sulfur compounds are electronegative and favor an electron transfer from the Pt particles to the sulfur (62). In the case of Pt supported on acidic zeolites, the Pt-S bond is loose since less electrons are available in the metal due to the Pt-acid interaction (62). One may expect that
4.0 I
0.30 1
4
4
1
3.5
3.0
1
I
I
-6
acidity
basic i ty
increase
increase
(oxygen)
strength
increase
decrease
-
decrease
increase
e N-hexane a r o m a t i za t ion
increase
decrease
*
increase
4
4
'int
0.40
benzene hydrogenation
resistance t o S
Fig. 5. Change in properties of zeolites with their electronegativity. on the opposite, due to the electron transfer from the basic zeolites to the Pt particles, more electrons are available in the metal to form a strong bond Pt-S. Sulfur should be a stronger poison for m a s i c zeolites than for Pdacidic zeolites. This is in fact observed, the PtKL or PtBaL zeolites active in the n-hexane dehydrocyclization are very sensitive to sulfur (63). CONCLUSION The basicity of zeolites, like acidity, varies i n a large range from weak to strong. Several parameters which direct this basicity have teen identified. They give trends and guide lines for the increase in basicity. The easiest of these parameters to act on are the A1 content and the electronegativity of the cation. For instance in the case of zeolites like mordenite or beta which show some basicity despite their low Al level, an increase in the A1 content should give a stronger basicity. In the search of new selectivities in adsorption and catalysis, the basic zeolites offer a large range of opportunities. REFERENCES 1
2
Y.N. Sidorenko, P.N. Galich, V.S. Gutyrya, V.G. Il'In, I.E. Neimark, Dokl. Akad. Nauk SSSR, 173 (1967) 132. T. Yashima, K. Sato, T. Hayasaka, N. Hara, J. Catal., 26 (1972) 303.
168
3 4 5 6 7 8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23
24 25 26 27 28
29 30 31
32 33 34 35 36 37 38 39 40 41
Y. Ono, in "Catalysis by zeolites" (B. Imelik et al. eds), Stud. Surf. Sci. Catal., Elsevier, Amsterdam, 5 (1980) 19. M.L. Unland, G.E. Barker, in "Catalysis of Organic Reactions" (W.R. Moser ed.), Chemical Industries Series 5, Dekker, New York, 1981, p. 51. D. Barthomeuf, G. Coudurier, J. Vedrine, Mat. Chem. and Phys., 18 (1988) 553. W.B. Jensen, "The Lewis Acis-base concepts. An overview", Wiley, 1980, p. 51. T. Yashima. H. Suzuki, N. Hara, J. Catal., 33 (1974) 486. P.A. Jacobs, "Carboniogenic Activity of Zeolites", Elsevier, Amsterdam, 1977 ; D. Barthomeuf, Mat. Chem. Phys., 17 (1987) 49 ; K.V. Topchieva, Huo Shi Thuong, Vestn. Moskov. Univ. Khim., 15 (1974) 239. A. Goursot, F. Fajula, C. D a d , J. Weber, J. Phys. Chem, 92 (1988) 4456 ; I.N. Senchenya, V.B. Kazansky, S . Beran, J. Phys. Chem., 90 (1986) 4857 ; S. Beran, J. Mol. Catal., 26 (1984) 31 ; A. Goursot, F. Fajula, F. Figueras, C. D a d , J. Weber, Helv. Chim. Acta, 73 (1990) 112 ; S. Beran, J. Phys. Chem., 94 (1990) 335. R.T. Sanderson, "Chemical Bands and Bond Energy" Academic Press, 1976. W.J. Mortier, J. Catal., 55 (1978) 138. D. Barthomeuf, J. Phys. Chem., 48 (1984) 42. A. de Mallmann, D. Barthomeuf, in "Innovation in Zeolite Materials Science", (P.J. Grobet et al., eds), Stud. Surf. Sci. Catal., Elsevier, Amsterdam, 37 (1988) 365. W. Przystajko, R. Fiedorov, I.G. Dalla Lana, Zeolites, 7 (1987) 477 ; A. Bielanski, J. Datka, J. Catal., 32 (1974) 183. E.G. Derouane, J.G. Fripiat, J. Phys. Chem., 91 (1987) 145. P.J. OMalley, J. Dwyer, Chem. Phys. Lett., 143 (1988) 97. D. Barthomeuf, J. Phys. Chem., 83 (1979) 249. R.P. Bell, "The proton in chemistry", Chapman and Hall, London, (1973). p. 92. G.V. Gibbs, E.P. Meagher, J.V. Smith, J.J. Pluth, ACS Symp. Ser., 40 (1977) 19. P.O. Scokart, P.G. Rouxhet, Bull. Soc. Belg., 90 (1981) 983. K.A. Van Genechten, W.J. Mortier, Zeolites, 8 (1988) 273. K. Tanabe, "Solid Acids and Bases", Kodanska, Tokyo, Academic Press, New York, (1970). K. Tanabe, M. Misono, Y. Ono, H. Hattori, New Solid Acids and Bases, Kodanska, Elsevier, Stud. Surf. Sci. Catal., 51 (1989). Y. Delaval, R. Seloudoux, E. Cohen De Lara, J. Chem. Soc. Faraday Trans. I, 82 (1986) 365. C. Mirodatos, P. Pichat, D. Barthomeuf, J. Phys. Chem., 80 (1976) 1335. J.L. Schlenker, J.J. Pluth, J.V. Smith, Mat. Res. Bull., 14 (1979) 751. P. Bodart, J.B. Nagy, G. Debras, Z. Gabelica, P.A. Jacobs, J. Phys. Chem., 90 (1986) 5183. M.M.J. Treacy, J.M. Newsam, Nature (London), 332 (1988) 249 ; J.M. Newsam, M.M.J. Treacy, W.T. Koetsier, C.B. De Gruyter, Proceed. R. SOC., London, Ser. A, 420 (1988) 375 ; J.B. Higgins, R.B. La Pierre, J.L. Schlenker, A.C. Rohrman, J.D. Wood, G.T. Kerr, W.J. Rohrbaugh, Zeolites, 8 (1988) 446 . J.A. Martens, J. Perez-Pariente, P.A. Jacobs, Proceed. Int. Symp. Zeol. Catalysis, Siofok, Acta Phys. Chem. Szegediensis, 1985, p. 487. A. de Mallmann, D. Barthomeuf, Proceed. 7th Int. Zeolite Conf., Tokyo (Y. Murakami, A. Iijima, J.W. Ward , eds), Kodanska, Elsevier, 1986 p. 609. A. de Mallmann, D. Barthomeuf, Zeolites, 8 (1988) 292. H. Lechert, K.P. Wittern, Ber. Bunsenges Phys. Chem., 82 (1978) 1054. A.N. Fitch, H. Jobic, A.J. Renouprez, J. Chem. SOC., Chem. Comm., (1985) 284 and J. Phys. Chem., 90 (1986) 1311. S. Dzwigaj, A. de Mallmann, D. Barthomeuf, J. Chem. SOC.Faraday Trans., 86 (1990) 431. J.W. Ward, J. Phys. Chem., 72 (1968) 421 1. R.M. Barrer, "Zeolites and Clay Minerals as Sorbents and Molecular Sieves", Academic Press, London, 1976. D.M. Ruthven, "Principles of Adsorption and Adsorption Processes", John Wiley and Sons, New York, 1984. Maomi Seko, Oil and Gas J., July 2 (1979) 81. L.O. Stine, D.B. Broughton, (1969), US 3.636.121. to UOP ; R. Baerden Jr, R.J. De Foe, 1972, U.S. 3.686.343, to Esso. a) R.W. Neuzil, (1982), U.S. 4.326.092, to UOP b) D. Barthomeuf, (1986), U.S. 4.593.150 to Exxon. a) R.W. Neuzil, D.H. Rosback, (1976), U.S. 3.943.182 to UOP
169
42 43 44 45 46 47 48 49
50 51 52 53 54 55
56 57 58 59
60 61 62
63
b) P.R. Geissler, (1979), U.S. 4.175.099 to Exxon c) D. Barthomeuf, (1986), U.S. 4.593.149 and 4.613.725 to Exxon A.J. De Rosset, (1975), U.S. 3.917.734 to UOP. G.A. Olah, Acc. Chem. Res., 4 (1971) 240. a) A. de Mallmann, Thesis, Paris (1989) b) A. de Mallmann, D. Barthomeuf, Industrial and Engineering Chemistry Research, (1990), in press. D. Barthomeuf, (1986), U.S. 4.584.424 to Exxon D. Barthomeuf, L.G. Daniel, (1988), U.S. 4.751.346 to Exxon. V. Solinas, R. Monaci, E. Rombi, M. Morbidelli, in Zeolites as Catalysts, Sorbents and Detergent Builders, (H.G. Karge, J. Weitkamp, eds.), Stud. Sci. Catal., Elsevier, Amsterdam, 46 (1989) 595. P.A. Venuto, P.S. Landis, Adv. Catal., 18 (1968) 259. H.C. Karge, J. Rasko, 1. Colloid, Interf. Sci., 64 (1978) 522 ; H.G. Karge, M. Ziolek, M. Laniecki, Zeolites, 7 (1987) 197. M. Ziolek, D. Szuba, R. Leksowski, in "Innovation in Zeolite Materials Science", (P.J. Grobet et al., eds), Stud. Surf. Sci. Catal., Elsevier, Amsterdam, 37 (1988) 427. H. Itoh, T. Hattori, K. Suzuki, Y. Murakami, J. Catal., 70 (1983) 27. N. Giordano, L. Pino, S. Cavallero, P. Vitarelli, B.S. Rao, Zeolites, 7 (1987) 131. A. Coma, V. Fornes, R.M. Martin, H. Garcia, J. Pnmo, Appl. Catal., 59 (1990) 237. A. Corma, R.M. Martin-Aranda, F. Sanchez, J. Catal., submitted. P. Gallezot, Catal. Rev. Sci. Eng., 20 (1979) 121. M. Briend-Faure, J. Jeanjean, M. Kermarec, D. Delafosse, J. Chem. SOC.Faraday Trans I, 74 (1978) 1538 ; S. Djemel, M.F. Guilleux, J. Jeanjean, J.F. Tempere, D. Delafosse, J. Chem. Soc., Faraday Trans I, 78 (1982) 835. I.R. Leith, J. Chem. SOC. Cheni. Commun., (1983) 93. C. Besoukhanova, J. Guidot, D. Barthomeuf, M. Breysse, J.R. Bernard, J. Chem. SOC., Faraday Trans I, 7 1981) 1595. G. Larsen, G.L. Haller, Catal. Lett., 3 (1989) 103. J.R. Bernard, Proceed. 5th hit. Zeol. Conf. Zeolites, (L.V.C. Rees ed.), Heyden, London, 1980, p. 86. A. de Mallmann, D. Barthomeuf, in "Zeolites as Catalysts, Sorbents and Detergent Builders" (H.G. Karge, J. Weitkamp, eds), Stud. Surf. Sci. Catal., Elsevier, Amsterdam, 46 (1989) 429. A. de Mallmann, D. Barthomeuf, J. Chim. Phys., 87 (1990) 535. P. Gallezot, J. Datka, J. Massardier, M. Primet, B. Imelik, Proceed. 6th Int. Cong. Catal., (G.C. Bond, P.B. Wells, F.C. Tompkins, eds), The Chemical Society, London, 2 (1977) 696. T.R. Hughes, W.C. Buss, P.W. Tamrn, R.L. Jacobson, Proceed. 7th Int. Zeol. Conf., (Y. Murakami, A. Iijima, J.W. Ward, eds), Kodanska, Elsevier, Tokyo, (1986), p. 725.
This Page Intentionally Left Blank
G . ohlmann c s t a!. (Editors 1, Cakdysis and Adsorption by Zeolites (01991 Elsevier Science Publishers R.V., Amsterdam
171
MATRIX vs ZEOLITE CONTRIBUTIONS TO THE ACIDITY OF FLUID CRACKING CATALYSTS
ROLAND VON BALLMOOS AND CHI-MI T. HAYWARD Engelhard Corporation, Menlo Park, CN28, Edison, NJ 08818, USA
SUMMARY
The matrix of fluid catalytic cracking (FCC) catalysts serves multiple functions: it gives the microspheres their physical strength and the attrition resistance necessary to withstand the stresses of rapid circulation in the FCC unit and protects the soft zeolitic particles from abrasion. The matrix also serves as a cracking component for gas oil molecules which are too large to enter the zeolitic pores. This second function is extremely important for achieving the high conversions and high gasoline selectivities of today's FCC catalysts. The proper choice of matrix chemistry and its balance with the zeolite component are the keys to reduced gas and coke makes at high catalyst activity while maximizing the liquid product yield. We investigated the influence of low, moderate, and strongly acidic matrices and their synergisitc contributions to the catalyts' activity. "Staged bed" MAT experiments demonstrated that the overall conversion and gasoline yield were drastically increased when gas oil molecules could interact with the matrix material first, before contacting the zeolite. Secondary cracking on the matrix of products initally generated inside the zeolite was negligible. The strength of the acid sites in the matrix influences the coke make. A matrix component with moderate acidity was found to result in the most beneficial selectivities. INTRODUCTION Modern fluid catalytic cracking (FCC) catalysts represent a highly sophisticated class of materials. In principle, they consist of approximately 15 to 35 wt-% of an active zeolite component in an amorphous or clay matrix with a minor amount of an oxide binder. Yet, on a microscopic scale the composition of the 60 to 9 0 micron size microspheres is considerably more complex: many different forms of zeolite Y, a
172
faujasite type material, with great variations in silicon-toaluminum ratios and rare earth levels are used depending on the selectivity demands of the refiners. Several recent reviews on FCC catalysts have been published dealing with either overall catalyst properties and performances [1,2,3] or investigating zeolite modifications and their effects on selectivity [4]. They describe in detail the effects of varying the silicon-to-aluminum ratio of zeolite Y and highlight differences resulting from ultrastabilization (dealurnination) by thermal, chemical, or thermo-chemical treatments. The beneficial influence of the zeolite's rare earth oxide (REO) content on stability and gasoline yields have been known since the early days of FCC cracking with zeolites. Ultrastabilized Y zeolites, however, generate higher octane gasolines at a yield debit. In order to achieve an optimal compromise, many of today's FCC zeolite components represent ultrastabilized rare earth exchanged faujasites. The equally significant variations in catalyst matrices and binder materials are usually given less attention in scientific investigations of cracking catalysts. Nevertheless, several reviews have been published recently [5,6]. This presentation focuses on the importance of the matrix contributions to FCC cracking, the matrix properties, and their influence on FCC catalyst selectivities. Modern FCC catalyst matrices can be separated into four classes: aluminas, silicas, silica-aluminas, and clays. The chemical and catalytic functions of these matrices may vary quite significantly, as demonstrated below, but they all share some common functions as FCC matrices. The matrix provides the suitable particle size and shape for fluidized bed operation, nominally 60-90 micron spheres. FCC zeolite particles are less than 2 microns in size and very soft. The matrix materials render the microspheres attrition resistant and reduce the loss of active zeolite components. Because the FCC unit is heat balanced without much external heat input (only for feed preheat to approximately 250 OC) the endothermic cracking reaction requires the microspheres to serve as a heat transfer medium to vaporize the feed and to crack the hydrocarbons. The heat is provided by burning coke off the catalyst in the regenerator. It is then transferred by the hot ( - 730 OC) microspheres to the riser section of the unit. The matrix furthermore serves as an adsorbent and scavenger for feed contaminants such as nickel, vanadium, sodium, iron, sulfur, and nitrogen. An additional function of the matrix is to provide easy access for feed molecules to the active sites of the zeolite. A major function of the matrix is the cracking of feed molecules which are too large to enter the zeolite's pores. The kinetic diameter of molecules with boiling points above approximately 450 OC (atmospheric equivalent) is larger than the
7.5 A pore of faujasite [7]. Depending on feed cut points a signifcant fraction of the molecules may fall into this category of compounds which have to be cracked on the matrix first. We have investigated this process in some detail by separating the functions of the matrix and the zeolite. Some aspects of this work have been published in a recent paper [a]. EXPERIMENTAL The experimental catalysts used in this study were commercial FCC microspheres with a clay based matrix which were combined with surface area-stabilized silicas doped with metal oxides to generate model matrix compounds with different intrinsic acidities. We used an Engelhard rare earth exchanged US-Y catalyst as the base case. The weakly acidic matrix material was prepared by using a commercially availbale silica support modified by a metal oxide. The moderatly acidic model matrix was based on the same,silica modified with a group IIA metal oxide. The strongly acidic matrix was generated by leaching aluminosilicate clay microspheres. In some experiments, aluminum oxide was deposited on the above silica to simulate a strongly acidic matrix with variable site density. The zeolitic FCC particles were steam deactivated separately to simulate typical FCC equilibrium catalysts for 4 hours in 100% steam at 788 OC. The model matrix components were either steamed at 732 OC for 8 hours in one atmosphere steam or calcined at 580 OC to provide the target acid strength. The catalysts' activity and selectivity patterns were then determined in a modified MAT unit (modified ASTM D-3907 test). Figure 1 shows a schematic process flow diagramm of the MAT test modified by Engelhard [9]. Figure 2 shows the catalyst bed arrangement for the two different types of experiments. In the "Catalyst on Top1@configuration, 3 g FCC catalyst were loaded into the reactor tube on top of 3 g appropriately sized matrix component. Because the reactor was of the down-flow type, the zeolite-containing FCC catalyst was contacted by the feed first. In the "Matrix on Top" configuration, the loading order was inverted and the matrix component was added on top of the FCC catalyst'bed. 1.2 g of a Mid Continent gas oil were fed in 48 s (catalyst-to-oil ratio of 5 ) at 490 OC. The liquid product syncrude was analyzed by GC simulated distillation (ASTM D-2887), the gas fraction was analyzed by GC, and the coke on catalyst was determined by a LECO coke analyzer. Activity is defined as weight percent of gas oil converted to products boiling below 216 OC and coke, or as a dimensionless activity based on second order kinetics, where Activity = Conversion/(lOO - Conversion).
w
ENGELHARDS MAT PROCESS FLOW SCHEME
CHROYATOORAPHK: 'ROGEN
DELIVERV ZONE
IL W l V E R V TUBE
CATALYST BED
GLASS REACTOR MEAT
ZONE FURNACE
I
t
CALIBRATION GAS INLET
GAS COLLECTION RESERVOIR
0 % COOUNT
Figure 1: Schematic flow diagram f o r MAT unit modified by Engelhard based on ASTM D3907 test.
P -1
Q
P S
0 a,
U a, a,
LL
I
a,
c,
.0 '
a, N
.-X L
c,
I'
a,
c,
0
.-a, N
176
Figure 2: Catalyst bed arrangement for "Staged Bed" MAT experiments used to demonstrate the importance of matrix cracking before zeolite cracking in FCC.
Synergistic Interaction Between Matrix and Zeolite Increases Cracking Activity
9 ri c r.
ID
.. W
MAT A c t i v i t y
2.5
-
I
2I
1.5
~~
10.5 -
0Matrix
Zeolite
Calculated
Measured
Catalyst In MAT .Test
J 9
--1
?
Activity = Conv./(lOO-Conv.) Conversion to Prod. < 216 C
-i
177
RESULTS AND DISCUSSION We are reporting three sets of experiments which demonstrate the importance and beneficial effects of an optimal balance between the zeolitic cracking component and the amorphous matrix material: A first experiment highlights the synergistic effects of matrix and zeolite cracking. A second set of data shows that the matrix component is most effective in cracking large molecules before they are further converted to gasoline inside the zeolite. A third experiment distinguishes between matrices of different overall activity, expressed as a combination of acid strength and acid site density. Zeolite
-
Matrix Synergism:
A rare earth exchanged zeolite catalyst with minimal matrix activity showed an experimental MAT conversion of 5 2 wt% (activity = 1.1). A moderate activity amorphous material, when tested by itself in the absence of a zeolitic cracking component, yielded 30 wt% MAT conversion (activity = 0 . 4 ) .
A third catalyst was prepared from a blend of the first two materials. It had the same zeolite activity as the first, combined with the identical amorphous matrix activity of the second. If the effects of the matrix and the zeolite were additive, the activity of Catalyst 3 would be calculated to be 1.1 + 0 . 4 = 1.5. In fact, when the MAT activity was determined experimentally it was 2.2, with a conversion of 69 wt%. These results are represented graphically in Figure 3 .
. The substantially greater than expected activity of Catalyst 3 indicates that significant interactions between the components occurs. The synergism can be explained by a mechanism of initial matrix cracking of large feedstock molecules to smaller fragments which are subsequently cracked into the gasoline and gas range inside the zeolite. Approximately 20 % of the MAT gas oil molecules are too large to enter the zeolite's pores. They have to be cracked on an easily accessible matrix first. In fact, the MAT data of Catalyst 2 showed 30% conversion, primarily of heavier molecules which are cracked more easily over a moderately acidic matrix. The matrix cracking does not produce converted products in the gasoline range, but rather it generates "intermediate feed molecules". The conversion of these harder-to-crack molecules requires the enhanced activity of the zeolite. A recently published study demonstrated the effectiveness of matrix-type "bottoms cracking additives1'to conventional FCC catalysts in increasing their activity [lo]. The results of our experiments confirmed that separate particle matrix components were in fact active in pre-cracking gas oil molecules. The hypothesis of "matrix first" cracking was further investigated by separating the matrix and zeolite components
178
TABLES TABLE I
-
Staged bed MAT tests
Moderate Matrix/REY
..................................................................
Description Conversion, wt%
Activity,conv./(lOO-conv.)
Matrix first 69.0 2.2
Yields,wt % Bottoms ,>602OF (352OC) Gasoline,Cg -421°F (216OC) Dry Gas rH2-C2 Coke Selectivities, (yield/activity) Dry Gas Coke
10.2 54.4 1.0 3.1 0.45 1.40
Zeolite first 51.0 1.0 24.5 41.9 0.6 2.2 0.60 2.20
..................................................................
TABLE I1
-
Staged bed MAT tests
Description Conversion I wt%
Activity,conv/(lOO-conv.)
Moderate Matrix/=-USY
Matrix first 75.05 3.01
Yields, wt % Bottoms, > 602OF (352OC) Gasoline,Cg -421°F (216OC) Dry Gas ,H2 -C2 Coke Selectivities, (yield/activity) Dry Gas Coke
Zeolite first 65.75 1.92
6.84 55.54 1.50 4.06
12.66 51.01 1.20 3.18
0.50 1.35
0.62 1.66
-_______________________________________------------
179
physically and by reversing the order in which the gas oil molecules contacted the active sites: Importance of Matrix Cracking Before Zeolitic Cracking: We designed "staged bed" MAT experiments to verify the concept that conversion to desirable products depends on the feed contacting the matrix first. For the matrix-on-top experiments the FCC catalyst was carefully poured into the bottom of the MAT reactor tube. The matrix component was poured on top of this bed carefully to avoid mixing of the two components. The feed was now forced to percollate through the Ilmatrix bed" first. In the zeolite-on-top experiments the order of the components was reversed. The spent catalyst beds were thoroughly mixed before the coke analysis to obtain a reliable average carbon residue number. We repeated the experiment three times: first using a lower activity grade, high rare earth catalyst combined with a moderately active matrix, secondly using a more active "octanebarrel" moderate rare earth catalyst with a moderatly active matrix, and thirdly a combination of the octane barrel catalyst with a more strongly acidic but not very active matrix (lower surface area). The activity and selectivity results are summarized in Tables I through 111, respectively. In all three cases the effect of matrix cracking of the fresh feed was clearly present. Very little conversion of zeolite products occurred when the matrix was downstream of the FCC catalyst. The MAT activity of the samples with the "matrix-ontop" was 1.5 to over 2 times higher, depending on the initial activity of the FCC component. Similar effects had been observed before by other investigators [ll]. Figure 4 shows the drastic conversion increase for the 9natrix-on-topIl sample to valuable products, accompanied by a reduction in bottoms compounds (boiling above 317 OC). The absolute conversion differences between the less active REY and the more active octane-barrel RE-USY catalyst introduce a bias to the selectivity numbers: the REY yields high gasoline (at lower octane), whereas in the other two comparisons the gasoline yield is relatively smaller, but at higher octane. The difference in conversion is largely due to the increased production of wet gas (c3 and c4 olefins and paraffins) in the experiments with the octane-barrel catalyst. Matrix Type Optimizes Cracking Selectivity: In a final set of experiments, we studied how the product selectivity and octane yield can be optimized by selecting a matrix with a balanced combination of acid strength and site density. These experiments were carried out in the traditional MAT configuration with fully blended components. REY was used as the active FCC catalyst. The matrix types varied in composition
Matrix Sets up Large Molecules for Selective Cracking in Zeolite Staged Bed MAT, wt-%
80 70 60 50 40
30 20 10
0
Conv.
Activ.
Gasol. Bottms Coke Dry Gas
Zeolite/Weak Matrix
0 Weak
Matrix/Zeolite
181
and, hence, intrinsic acid strength, and surface area. We generated MAT data at 70 wt% conversion, shown in Table IV, which are directly comparable. Significantly different selectivities resulted depending on the type of matrix used. Catalyst A, which had the lowest matrix activity, produced the highest yield of coke and gas and the lowest yield of the most valuable products, gasoline and LCO. The reason for this adverse selectivity was overcracking promoted by the few very strong acid sites on the matrix. Catalyst B, with its moderatly active matrix, had the lowest coke and gas yields and higher yields of the desirable products gasoline and LCO. Catalyst c , which had the highest matrix activity, produced yields that generally fell between those of Catalyst A and Catalyst B. These results suggest that an optimum level of matrix activity exists for each active zeolite component. If the activity of the matrix is too low the desired level of heavy molecule conversion cannot be achieved and the overall catalyst activity is negatively affected. If the matrix activity is too high, pr'imarily because the matrix acid sites are too strong, the catalyst selectivities are severely shifted to undesirable products, such as coke and gas.
Catalyst
A
Matrix Activity
LOW
Acidity Surface Area
Strong Very Low
B
C
Moderate
Moderate
Weak Moderate
Moderate High
MAT Yields @ 70 wt-% Conversion Dry Gas wt%
1.1
1.2
9.7
11.1
C5+ Gasoline wt%
1.3 11.8 53.0
55.9
54.2
Valuable Products Gasoline+LCO wt%
71.2
75.6
74.7
3.9
3.3
2.1
1.5
3.5 1.3
c3 -I-c4 wt%
Coke wt% ic4/~4=ratio
..................................................................
182
CONCLUSIONS Our experiments using REY and RE-USY catalysts with a minimal matrix activity in combination with model matrix compounds of increasing intrinsic acid strength have clearly demonstrated the following principles:
-
A strong synergism exists between matrix and zeolite cracking under FCC conditions. The resulting activity is higher than the sum of the individual contributions.
-
It is imperative that the larger feedstock molecules contact the matrix component first t o convert them to smaller fragments which can be cracked inside the zeolite to gasoline products.
-
Matrix activity is a combination of acid site strength and density. A moderately active matrix results in a properly balanced FCC catalyst with optimal gasoline and coke selectivities. The FCC catalyst matrix plays a key role in determining overall activity, selectivity, and octane trends. The FCC catalyst manufacturer can control the matrix activity and acidity, allowing him to tailormake catalysts to meet specific refining needs.
ACKNOWLEDGMENTS We acknowledge the expert technical assistance of Mr. Egberto Villanueva in preparing the Samples for this study. We are indebted to Engelhard's Petroleum Catalyst hraluation group for the many experiments, in particular those of the staged bed configurations. Engelhard's Research Services group provided the physical and chemical characterization of the samples. We also thank S.H. Brown, M. Deeba, G.S. Koermer, J.B. McLean, E.L. Moorehead, and D. Stockwell for many stimulating discussions. REFERENCES
,
31(3) , 215-354 (1989).
1.
J. Scherzer, Catal.Rev.-Sci.Eng.
2.
P.B. Venuto and E.T. Habib, IIFluid Catalytic Cracking with Zeolite Catalystsl8, Marcel Dekker, New York (1979).
3.
B.W. Wojciechowski and A . ?orma, IICatalytic Cracking Catalysts, Chemistry and Kinetics", Marcel Dekker, New York (1986).
4.
Reference [l] above, and references therein.
5.
A.G.
Oblad, Oil and Gas Journal, March 27, 84 (1972).
183 6.
C.J. Groenenboom, "Zeolites as Catalysts, Sorbents, and Detergent Builders@~,Stud.Surf.Sci.Catal. 46, Elsevier, Amsterdam, 99 (1989).
7.
A. Humphries and J.R. Wilcox, lIZeolite/Matrix Synergism in FCC Catalysis8@,paper presented at the 1988 NPRA Annual Meeting (March 1988).
8.
C.T. Hayward and W.S. Winkler, Hydrocarbon Processing, February, 55 (1990).
9.
Engelhard Corporation, @IMicro Activity Testing of FCC Catalystsll, The Catalyst Report, Engelhard Corporation, Edison, NJ, USA (1988).
10.
M.M. Mitchell, H.F. Moore, and T.L. Goolsby, llImprovementof FCC Catalyst Performance with Bottoms Cracking Additives', paper presented at the Spring AIChE Meeting, Orland0,FL (1990).
11.
R.E.. Ritter and G.W. Young, paper AM-84-57 presented at the March NPRA Meeting (1984).
This Page Intentionally Left Blank
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 1991 Elsevier Science Publishers B.V., Amsterdam ZEOSORB HS 30
-
185
A TEMPLATE-FREE SYNTHESIZED PENTASIL-TYPE
ZEOLITE K.-H.
'
BERGK1, W.
SCHWIEGER',
H.
FURTIG'
and U. HADICKE2
M a r t i n - L u t h e r - U n i v e r s i t y H a l l e - W i t t e n b e rg, Department o f C h e m i s t r y , S c h l o R b e r g 2, H a l l e 4020, Chemie-AG B i t t e r f e l d - W o l f e n ,
B i t t e r f e l d 4400,
BRD
BRD
SUvlMARY The s u b j e c t o f t h i s p a p e r i s t h e s y n t h e s i s o f p e n t a s i l - t y p e z e o l i t e s w i t h o u t t e m p l a t e s . We d e s c r i b e i n v e s t i g a t i o n s o f the various synthesis variables s u c h as t h e c o m p o s i t i o n o f the r e a c t i o n mixture, a p p l i c a t i o n s i n a t e c h n i c a l process and, based on t h e z e o l i t e "Zeosorb HS 30", i n v e s t i g a t i o n s o f t h e a f t e r - t r e a t m e n t f o r a s p e c i f i c adjustment o f t h e Al-content i n the l a t t i c e o f the zeolite which are important f o r t h e a p p l i c a b i l i t y o f the catalyst.
-
-
INTRODUCTION S i n c e 1966 t h e Chemie-AG B i t t e r f e l d - W o l f e n has been t h e o n l y p r o d u c e r o f m o l e c u l a r s i e v e s i n t h e GDR.
The development
s t a r t e d w i t h t h e s y n t h e s i s o f z e o l i t e 4A i n 1960. t y p e s o f m o l e c u l a r sieve, t h e z e o l i t e s 3A and 5A,
The r a n g e o f
w h i c h a r e produced f u r t h e r i n c l u d e s
t h e f a u j a s i t e s 13X,
1OX and Y,
morde-
n i t e and e r i o n i t e t y p e s and c u r r e n t l y a l s o t h e p e n t a s i l t y p e Zeosorb HS 30 ( T a b l e 1). The z e o l i t e s a r e used i n v a r i o u s a d s o r p t i v e and ca t al,,t
i c processes.
The range o f p e n t a s i l t y p e b y m o l e c u l a r s i e v e s was w i d e n e d b y u s i n g a s p e c i a l method.
F o r a l o n g time the synthesis of
t h e s e t y p e s was c o n n e c t e d w i t h t h e a p p l i c a t i o n o f s u b s t a n c e s deternining the structure, as d e s c r i b e d i n r e f s .
t h e s o - c a l l e d t e m p l a t e compounds,
1-3.
TABLE 1 Chemie-AG B i t t e r f e l d - W o l f e n as a m o l e c u l a r s i e v e p r o d u c e r TYP e A
Fauj a s i t e
E r i o n it e
~~~~
~~
Zeosorb 3A, 4A, 5A
Zeosorb 13X, lox, Y
delivery form:
as powder ( a i r ,
~
Mo r d e n i t e
Pent a s i l
Zeosorb VlOO
Zeosorb HS30
~
Erionite dried,
activated),
as g r a n u l e
186 The f i r s t work on t h e a p p l i c a t i o n o f o r g a n i c s u b s t a n c e s i n t h e p r o c e s s of z e o l i t e f o r m a t i o n was c a r r i e d o u t b y BARRER and DENNY ( r e f .
4) and KERR ( r e f .
5).
T a b l e 2 shows a l i s t o f
v a r i o u s o r g a n i c compounds, w i t h w h i c h p e n t e s i l z e o l i t e s c o u l d be pr oduc ed i n t h e y e a r s a f t e r 1966. We have grouped t h e s u b s t a n c e s i n t o categories,
such as o r g a n i c c a t i o n s ,
n e u t r a l molecules,
o r g a n i c a n i o n s and
and h b v e l i s t e d t h e names o f t h e a u t h o r s and
the year o f publication,
when t h e s e f a c t s w ere known.
s t a g e s a r e t h e w o rk s o f ARGAUER and LANDOLT ( r e f . e t el.
(ref.
KERR ( r e f .
2),
Important
l), RUBIN
5) and LOWE and AROYA ( r e f .
From t h i s v a r i e t y and g r o u p s o f s u b s tances,
6).
the r o l e o f template
compounds i n t h e s y n t h e s i s p r o c e s s c a n be deduced t o b e u n c l e a r and d i v e r s e , The a p p l i c a t i o n o f o r g a n i c compounds i n a process, w h i c h i s otherwise e n t i r e l y inorganic, process i t s e l f ,
ceuses some p r o b l e m s f o r t h e
a s w e l l a s f o r t h e economy and t h e q u a l i t y o f
t h e products.
-
-
Such pr oble m s a r e ,
f o r instance:
the cost o f t h e template; t h e p o l l u t i o n o f w e s te w a t e r ; s a f e t y problems i n t h e t e c h n i c a l process; t h e p o l l u t i o n o f t h e o r g a n i c component b y w aste gas d u r i n g thermal decomposition; t h e c ar bon d e p o s i t s i n t h e z e o l i t e caused b y a p o s s i b l e i n c o m p l e t e c o mb u s ti o n o f t h e o r g a n i c component. These d i s a d v a n t a g e s demand t e m p l a t e - f r e e s y n t h e s i s
fore,
.
There-
b e f o r e commencing r e s e a r c h and development w ork on t h e
synthesis of pentesil zeolites, Bitterfeld-Wolfen
i t was n e c e s s a r y f o r Chemie-AG
( a s a p r o d u c e r o f m o l e c u l a r s i e v e s ) and t h e
Research I n s t i t u t e a t H a l l e t o c o n s i d e r t h e b a s i c raw m a t e r i a l s and t h e t e c h n i c a l f a c i l i t i e s w h i c h were a v a i l a b l e , i n o r d e r t o make i n d u s t r i a l p r o d u c t i o n f e a s i b l e .
187 TABLE 2 The d i f f e r e n t t y p e s o f o r g n n i c compounds,
w h i c h c a n be used
f o r t h e syntheses o f P e n t e s i l z e o l i t e s Type
Example
Author
R eference/ year o f publication
Organic cations
R~N+, R ~ P +
A rg a u e r and L a n d o l t Dwyer and J e n k i n s Kerr
1, 1972 7, 1976 5, 1969
polymeric cationic conpounds
................................ Organic molecules
Amine, diamine alcohols, dioxane, 2- amino-py r i d i n e
Rubin e t e l . Bremer e t e l . R o l l me n n P o s t and Nanne Klotz Whit t am
1975 1981 3, 1978 9, 1979 10, 1983 11, 1982 2, 8,
................................ Organic anions
Alkylbenzols u l f onate, polymeric anionic compounds
Hugiwara e t a l .
12,
1981
Roscher e t a l . Lowe and Aro ya
13, 6,
1984 1983
SYNTHESIS OF PENTASIL TYPE ZEOLITE The f i r s t p o s i t i v e r e s u l t s ( T a b l e 3) were o b t a i n e d i n l a b o r a t o r y t e s t s which looked s p e c i f i c a l l y a t t h e aspect of a c o n t i n u o u s s i m p l i f i c a t i o n o f t h e t e m p l a t e compounds.
The
c l a s s i c a l t e m p l a t e compound TEA-3 ( t e t r a e t h y l a m m o n i u m i o d i d e ) , i t s f o r m a l p r e c u r s o r s TEA ( t r i e t h y l a m i n e ) and
alkyliodide
,
s i m p l e amines,
a t e c h n i c a l m i x t u r e o f amines and a l c o h o l and
NH3 were used.
By t h i s method,
s p r o d u c t c o n t a i n i n g 60 t o 80 %
p e n t e s i l c o u l d a l r e a d y be p ro d u c e d w i t h o u t u s i n g t e m p l a t e s . A f t e r optimizing, considerably.
t h i s c o n t e n t l e v e l c o u l d be i n c r e a s e d
188 TABLE 3
S y n t h e s i s o f P ,n ta s i l 175
OC
zeolites with different additions at
and r e s t e d r e a c t i o n m i x t u r e s
Number
CrystelliProduct Composition of t h e reaction mixture z a t i o n time composition Na 2 ~ ~l / 203/ s i O 2 / H20/Z t / h/ Ma.-%
Addition (2)
TPA-J
5,7
1
100 1580 9,6
142
60 ZSM-5 40 Q u a r t z
TEA + n-Butyl-bromid
;?
1
Technical B u t y l am i n mixture
8,3
1
93,5
3877 36,7
168
80 ZSM-5 20 Q u a r t z
N-P r o p y l amin
8,3
1
93,5
3877 36,7
96
70 ZSM-5 30 Q u a r t z
+
8.3
1
93,5
3877 36,7
48
NH3
28
750 16
30 amorphous
240
70 ZSM-5
20 sio2-x
C2H5-OH
80 ZSM-5
NH3
25,9
From t h e s e t e s t s ,
1
88,4
1820
55
24
8 0 ZSM-5 10 amorphous
t h e f i r s t p a t e n t a b l e r e s u l t s were o b t a i n e d ,
and t h e s e a r e l i s t e d i n T a b l e 4. C a t i o n i c and/or
a n i o n i c p o l y m e r compounds a s w e l l as a t e c h n i c a l
m i x t u r e o f amines a r e u s e d as t e m p l a t e s ( r e f s . template-free
13-14).
The f i r s t
synthesis w i t h o r without the eddition o f n u c l e i
o f c r y s t e l l i z a t i o n are described i n refs.
15-17.
F i g u r e 1 shows a f i e l d o f s y n t h e s i s f o r t h e t e m p l a t e - f r e e type. The phases w h i c h c a n be seen a r e shown b y p l o t t i n g t h e r e a c t i o n t i m e a g a i n s t v a r i a t i o n s o f t h e SiOdA1203 molar r e t i o i n t h e
reaction mixture.
I n a d d i t i o n t o t h e t a r g e t phase,
pentasil
w i t h a M F I - s t r u c t u r e , m o r d e n i t e , q u a r t z and l a y e r s i l i c a t e s can a l s o be seen. Each t i m e t h e S i O d A 1 2 0 3 m o l a r r a t i o was i n c r e a s e d a d i f f e r e n t sequence o f phases c o u l d be seen. A t a S i02/ A1 2 0 3 m o l a r r a t i o o f 20
-
amorphous phase i n t o m o r d e n i t e o c c u r s , o f 25
-
25 c o n v e r s i o n f r o m t h e a t a medium m o l a r r e t i o
55 c o n v e r s i o n f r o m t h e amorphous phase i n t o q u a r t z v i a
P e n t a s i l oc c u rs , e t an even h i g h e r r a t i o c o n v e r s i o n f r o m t h e amorphous phase i n t o q u a r t z v i a l a y e r - s i l i c e t e occurs.
189
TABLE 4
E a r l i e r p a t e n t s f o r P e n t a s i l s y n t h e s i s (1982) of Chemie-AG B i t t e r f eld- W o l f en Number o f t h e p a t e n t
Content
DD-WP 207 184
c e t i o n i c and a n i o n i c p o 1ym e r ic comp o u n d s
DD-WP 207 186
template-f ree synthesis
DD-WP 207 185
organic-free, ammonia
DD-WP 205 674
t e m p l a t e - f ree, w i t h . t h e a d d i t i o n o f a c t i v a t e d seed m a t e r i e l s
DD-VIP 206 551
template-f ree w i t h vigorous s t i r r i n g o r the addition o f grinding bodies
DD-WP 206 979
t e c h n i c a l amin m i x t u r e as t e m p l a t e
I
'
BS YO
7004
b u t b y use o f
4.-""
300
,
~
r r n d r n o r ~
k-i,f---/Yfj
mordenih
I
I
I
/
24
48
I
I
I
72
96
720
Y
I
tU
14.0 h
HS 20
Fig.
1. A t e m p l a t e - f r e e
zeolites.
synthesis f i e l d o f Pentasil type Pentosil type zeolites; SS : s h e e t s i l i c a t e s .
MFI:
190
ss
MF I
MOR
c---------r 5 um
/
c---------r 2 um
/
F i g . 2. A r a s t e r e l e c t r o n microscope image o f t h e phases observed i n t h e t e m p l a t e - f r e e s y n t h e s i s f i e l d . MFI: P e n t a s i l t y p e z e o l i t e s f SS : s h e e t s i l i c a t e e ; MOR: M o r d e n i t e s .
191 T h i s d e s c r i p t i o n shows t h a t when t h e SiO2/Al2O3 m o l a r r s t i o
i s i n c r e a s e d , c r y s t a l l i z a t i o n g e n e r a l l y s l o w s down. We have named t h e r ang e s o f s y n t h e s i s , w h i c h were most i m p o r t a n t i n o r d e r t o achieve t e c h n i c a l r e e l i s a t i o n .
On t h e b a s i s o f t h e s e
r e s u l t s t h e p r o d u c t i o n o f t h e p r o d u c t was s t e r t e d .
The f i r s t
co m m er c ial p r o d u c t t o be d e v e l o p e d was Zeosorb HS 30. further,
product8 with
I
SiOdA1203 m o l a l r a t i o o f 20 (HS 20) end
40 (HS 40) can b e m a n u f a c t u r e d and o f f e r ' e d .
F i g u r e 2 shows a p h o to g ra p h , t a k e n t h r o u s h an e l e c t r o n m i c r o scope,
o f t h e phases t o be seen i n t h e f i e l d o f s y n t h e s i s .
Each phase has i t s c h a r a c t e r i s t i c morphology,
so t h a t even
s l i g h t i m p u r i t i e s can be seen,
although these i m p u r i t i e s cannot
y e t be shown b y means o f X-ray
techniques.
THE TECHNICAL PROCESS OF SYNTHESIS Based on t h i s s t u d y o f t h e l i m i t a t i o n s o f t h e f i e l d o f s y n t h e s i s end t h e k i n e t i c s o f t h e c r y s t a l l i z a t i o n process, an e n t i r e l y i n o r g a n i c , t e c h n i c a l p r o c e s s was developed. I n F i g u r e 3 a scheme o f t h e c o u r s e o f t h e r e a c t i o n i s gi ven. Based on a sodium a l u m i n a t e s o l u t i o n as a s o u r c e o f A1203, a NaOH s o l u t i o n and water,
an a l k a l i n e p r e p a r a t i o n i s p r o d u c e d
i n a f i r s t m i x i n g p ro c e s s .
S i l i c a s o l used as a s o u r c e of S i 0 2 ,
i s added i n a subsequent m i x i n g p ro c ess. A f t e r homogenizing, t h e r e a c t i o n m i x t u r e i s t r a n s f e r r e d t o an a u t o c l a v e end c r y s t a l l i z e d under hydrothermal conditions. has ended and h a s been c o o l e d down,
After the reaction
t h e p r o d u c t is s e p a r a t e d
fro m t h e m o t h e r l i q u o r b y f i l t r a t i o n .
I t i s t h e n washed u n t i l
t h e o u t - f l o w i n g w a t e r has a pH v a l u e o f a p p r o x i m a t e l y 9 and d r i e d a t e t e m p e r a t u r e o f 110
OC.
i s a d r i e d powder o f sodium form.
The p r o d u c t
-
Zeosorb HS 3 0
-
T h i s i s i t e c o m m e r c i a l form.
I f r e q u i r e d , HS 3 0 - p r o d u c t s c a n a l s o be s u p p l i e d i on-exchanged ( l o a d e d w i t h d i f f e r e n t c a t i o n s ) o r g r a n u l a t e d (bound w i t h c l a y ) .
192
lstep
1
MlXING 2 step
I
DRYING (770°C)
I
MIXING
L
0
I o wder
F i g . 3. M a n u f a c t u r i n g p r o c e s s o f t e m p l a t e - f r e e type zeolite.
Pentasil
PROPERTIES OF THE PRODUCTS When c om pa ri n g t h e p r o p e r t i e s o f t e m p l a t e - c o n t a i n i n g and template-f ree synthesized products,
according t o t h e i r S i O d
A1203
ratio,
i t
A1203
ratio,
mo rp h o l o g y and a c i d i t y cnn be v a r i e d s y s t e m a t i -
can be f o u n d t h a t t h e i r c r y s t a l l i n i t y , S i O d
c a l l y i n t h e t e m p l a t e - f r e e v a r i s n t as w e l l , o n l y p o s s i b l e w i t h i n a s m a l l range.
although t h i s i s
I n t h i s respect,
a r e s i m i l a r and d i f f e r e n t t e n d e n c i e s .
there
These a r e g i v e n i n
T a b l e 5. For both synthesized variants, products,
the C r y s t a l l i n i t y o f the
t h e volume o f t h e m i c r o p o r e s , t h e c o n t e n t o f
alum inium and t h e B r o e n s t e d a c i d i t y a r e i n t h e same p r o p o r t i o n , depending on t h e c o n t e n t o f t h e a v a i l a b l e A l . t o this, cells,
In o p p o s i t i o n
t h e r e a r e d i f f e r e n t t e n d e n c i e s f o r t h e volume o f u n i t
t h e r a t i o o f s t r o n g l y a c i d i c t o weakly a c i d i c c e n t r e s
and f o r t h e volume o f mesopares.
193
TABLE 5 Comparison o f t h e p r o p e r t i e s o f t e m p l a t e - c o n t a i n i n g and t e m p l a t e f r e e P e n t a s i l p r o d u c t s ( g e n e r a l t e n d e n c i e s depending on t h e Si0,./A1203
ratio
S i m i l a r tendencies
D i f f e r e n t tendencies
1. C r y s t a l l i n i t y
1.
dep e n d e n c y o f t h e v o l u m e o f t h e u n i t c e l l upon t h e SiOgAl2O3 r a t i o
2.
r a t i o o f weak a n d s t r o n g a c i d i c centres
3.
volume o f mesopores
2.
volume o f m i c r o p o r e s
3.
r e l a t i o n between t h e Si02/A1 0 r e t i o s o f the reaztlon mixtures a nd t h e p r o d u c t s
4.
dependence o f t h e m o r p h o l o g y upon t h e Si02/A1203 r a t i o
5.
d i r e c t r e l a t i o n between t h e A1 c o n t e n t and t h e Broensted-acidity
F o r example,
t a b l e 6 shows a c o m p a r i s o n o f c h a r a c t e r i s t i c d a t a
o f s e l e c t e d samples f r o m b o t h s e r i e s ,
-
letter S
-
o r without template
Si02/A1203 r a t i o ,
the acidity,
a d s o r b e d q u a n t i t y o f NH3, of t h e micropores, is,
e i t h e r with template
- l e t t e r A.
which i s given here as t h e
decreases i n b o t h s e r i e s .
d e t e r m i n e d f rom n-Hexane
w i t h an e q u a l volume,
With an i n c r e a s i n g Th e v o l u m e
adsorption isotherms,
independent o f t h e SiOdA1203 m o l a r
r a t i o i n both series. T h e vo lume o f t h e mesopores o f t h e s a m p l e s s y n t h e s i z e d w i t h
TPA-J i s t w o o r t h r e e t i m e s l a r g e r i n c o m p a r i s o n t o t h e template-f ree v a r i a n t . By c o m p a r i n g t h e m o r p h o l o g y o f t h e p r o d u c t s o f b o t h t h e series (Figure 4), SiO/A1203
i t i s p o s s i b l e t o see t h a t w i t h an i n c r e a s i n g
m o l a r r a t i o t h e r e i s a l w a y s a change f r o m l o o s e
a g g r e g a t e s t o l a r g e r m o n o c r y s t a l s w i t h a smooth su-face. A l t h o u g h t h e r e i s no d e t a i l l e d i d e n t i t y b e c a u s e t h e f o r m a t i o n o f a s p e c i a l m o r p h o l o g y depends o n t o o many f a c t o r s , t e n d e n c y i s t h e same.
the
194
TABLE 6 Comparison o f p r o d u c t p r o p e r t i e s a c c o r d i n g t o t h e i r s y n t h e s i s w i t h ( S ) and w i t h o u t ( A ) t e m p l a t e s (H-forms)
Ternalate
s 5 5 5 5
Product
TPA-J TPA-3 TPA-J TPA-J TPA-J
1 2 3 4 5
no no no
A 1 A 2
A 3
NH,-TPD
Volume o f p o r e s
38
0,83
56 124 204 3 64
0,bB
0,46 0,20 0,13
1,53 1,21
22 32 58
SiOdA1203 r a t i o
0,BO
0,13 0,16 0,16 0,16 0,16
0,06 0,06 0,05 0,04 0,03
0,17 0,16 0,15
0,02 0,Ol 0,Ol
SPECIAL VARIANTS OF MODIFICATIONS The p o s s i b l e v e r i a t i o n s d e s c r i b e d so f e r o n l y r e s u l t f r o m t h e d i m e n s i o n s o f t h e shown f i e l d o f s y n t h e s i s .
However,
for
many a p p l i c a t i o n p u r p o s e s t h i s r o n c e o f p o s s i b l e v a r i a t i o n s i s not sufficient, retio.
e s p e c i a l l y w i t h regard t o t h e SiOdA1203 molar
Therefore,
i t W F S t h e a i m o f o u r f u r t h e r work t o a c h i e v e
a b e t t e r adjustment o f
t h e p r o p e r t i e s t o s p e c i a l demands b y
o p t i m i z i n g t h e s y n t h e s i s and m o d i f y i n g t h e p r o d u c t . T h i s wes a c h i e v e d i n d i f f e r e n t
WAYS:
1. By u s i n g s y n t h e s i s v a r i a n t s f o r w i d e n i n g t h e r a n g e o f t h e S i 0 2/ A1 203-m o d u 1e
.
2.
V a r i n g t h e p r o p e r t i e s b y isomorphous s u b s t i t u t i o n o f t h e n e t w o rk-f o r m i n g e l e m e n t s .
3.
I n c r e a s i n g t h e module b y subsequent d e a l u m i n i z a t i o n . By u s i n g s p e c i a l v a r i a t i o n s o f t h e s y n t h e s i s c o n d i t i o n s
(concentration o f a l k a l i metals,
c r y s t a l l i z e t i o n time)
,
i t
i s
p o s s i b l e t o extend t h e renge o f t h e module o f t h e s y n t h e s i z e d p r o d u c t s t o 20
-
60 f r o m a s t a r t i n g m o l a r r a t i o o f 20
K i t h an i n c r e a s i n g m o l a r r a t i o ,
- 70.
the rate of crystallization
-
With a s t c r t i n g m o l a r r a t i o o f 80 100 t h e r e p r o d u c i b i l i t y is p o o r , a n d t h e r e w i l l b e a d e c r e a s e i n t h e c r y s t a l l i n i t y decreases.
of t h e products.
@ O D
N
w m L O
I
y.
N
N
e,m m 4 0 E
W
r
D t I r o a m
a I-
0)
W C
L O
..
196
me,
Q)c
mrl E l rl 0 a,@ P N
ma,
0.4 0 0
Or: Le, v c
€ 0
4%
196
On t h e o t h e r hand we can see,
t h a t t h e SiO$Al2O3
molar r a t i o
o f t h e p r o d u c t is a l w a y s a p p r o x i m a t e l y 10 u n i t s l o w e r compared w i t h those s t a r t i n g compositions f o r template-free A s t o n i s h i n g l y enough,
syntheses.
an e x t e n s i o n o f t h e p o s s i b l e S i 02/A 1203
m o l a r r a t i o t o 80 was p o s s i b l e ,
when p r o d u c t s c o n t a i n i n g f i v e -
membered r i n g s as a s t r u c t u r a l e l e me nt w ere u s e d a s n u c l e i . We us ed o f f r e t i t e o r e r i o n i t e a s n u c l e i .
Obviously,
t h e r e i s an
i n t e r m e d i a t e d e c o m p o s i t i o n i n t o b a s i c el ements, w h i c h i s f a v o u r e s p e c i a l l y t h o s e w i t h MFIi s p o s s i b l e t o produce MFI-products w i t h a S i O d A1203 m o l a r r a t i o o f 70 80 f r o m r e a c t i o n m i x t u r e s w i t h a Si02/A1203 m o l a r r a t i o o f 100 and w i t h o f f r e t i t e o r e r i o n i t e as n u c l e i , w i t h o u t any o t h e r c r y s t a l l i n e byproducts. One more p o s s i b l e m o d i f i c a t i o n d u r i n g d i r e c t s y n t h e s i s i s t h e i som or phous s u b s t i t u t i o n o f Al- o r S i -atoms i n t e t r a h e d r a l p o s i t i o n s b y v a r i o u s hetero-atoms. T h e r e a r e many p o s - s i b i l i t l e s f o r such a s u b s t i t u t i o n d u r i n g s y n t h e s e s w i t h t e m p l a t e s . Even t h e c om plet e s u b s t i t u t i o n o f a l u m i n i u m i n t h e l a t t i c e b y e.g. Fe- o r B-atoms i s p o s s i b l e ( r e f s . 18-21). S o - c a l l e d f e r r i t e s o r b o r a l i t e s a r e formed. T h i s has n o t y e t been d e s c r i b e d f o r t e m p l a t e - f r e e S y n t h e s i z e d v a r i a n t s . We s t u d i e d t h e i somorphous s u b s t i t u t i o n o f B and Fe o n t h e b a s i s o f o u r s t a n d a r d t e m p l a t e f r e e s y n t h e s i z e d runs. able t o the synthesis of pentasils, structure.
It
-
The r e s u l t s of t h e s y n t h e s e s w i t h b o r o n a r e shown i n F i g u r e 5. The r e s u l t a n t r e l a t i v e c r y s t a l l i n i t y (compared t o a s t e n d a r d ) o f synthesized products obtained a f t e r a c r y s t o l l i z a t i o n time ( T K ) o f 24 o r 4 8 h o u rs ,
and a t a t e m p e r n t u r e o f 175
OC
(TK),
depending on t h e S i 0 2 / A 1 2 0 3 m o l a r r a t i o i n t h e r e a c t i o n m i x t u r e , a r e shown.
On t h e b a s i s o f a s t a n d e r d Si02/A1203 i n i t i a l r a t i o
o f 40, alum ini u m was c o n s t a n t l y r e p l a c e d t o a c e r t a i n p e r c e n t a g e by B ( u p p e r l i n e o f F i g u r e 4), w h i a l e t h e r a t i o o f S i O d A 1 2 0 3 + +B203 remained c o n s t a n t a t 40. Up t o a s u b s t i t u t i o n o f a p p r o x i m a t e l y 50% A l , w e l l - c r y s t a l l i z e d p r o d u c t s a r e o b t a i n e d . W i t h
l a r g e r boron content, suppressed.
the c r y s t a l l i z a t i o n of p e n t s a i l i s
T h i s can be e x p r e s s e d b y t h e g r e a t d r e c r e a s e i n
crystallinity.
A p a r t fro m M F I - t y p e z e o l i t e ,
amorphous phases
and l a y e r - s i l i c a t e s a r e s t i l l found, and t h e r e a r e a l r e a d y t r a c e s o f more h i g h l y - c o n d e n s e d S I O -phases, such as c r i s t o b a l i t e . 2
197
%&\ '
55 Cr
- sheet silicates - cristoba/i te am - amorphous I
I
I
700 750 21 product 5102/Ab 0 ' -ratio
50
F i g . 5. S y n t h e s i s o f P e n t a s i l t y p e z e o l i t e s w i t h b o r o n s t a r t i n g Si02/A1203 + B 2 0 3 - r a t i o = 40 TK = 175 OC Vie can draw t h e c o n c l u s i o n , t h a t alum inium i n t e m p l a t e - f r e e
a complete s u b s t i t u t i o n o f
synthesized products i s impossible,
and t h u s t h a t a l u mi n i u m p l a y s an i m p o r t e n t r o l e i n t h e f o r m a t i o n o f t h e MFI-st r u c t u r e .
The e x p e r i m e n t s gave s i m i l a r r e s u l t s w i t h r e g a r d t o t h e s u b s t i t u t i o n o f i r o n . W i t h a Fe:Al
m o l a r r a t i o o f 0,6
there i s
o n l y a s l i g h t d e c re a s e f r o m 100% t o about 95%. B e g i n n i n g w i t h a Fe : A l- m olar
ratio,
crystallinity, r a t i o o f 1:1
o f 0,7, t h e r e i s a s l i g h t decrease i n
w h i c h becomes q u i t e e v i d e n t a t 59% w i t h a m o l a r
( t h i s means a 50% s u b s t i t u t i o n o f t h e A1 b y Fe).
From t h e s e r e s u l t s , s u b s t i t u t i o n w i t h b o r o n i s c o n f i r m e d i n t h i s case a s w e l l . However, t h e s e p a r t i a l s u b s t i t u t i o n s can a l r e a d y e f f e c t c e r t a i n changes i n t h e p r o p e r t i e s o f t h e p e n t s s i l p r o d u c t s . I n p a r t i c u l a r , t h e y have an i n f l u e n c e on t h e morphol o g y , t h e a c i d i t y , t h e l a t t i c e c o n s t a n t s and, t h e r e f o r e , o n t h e i r a p p l i c a b i l i t y a s c a t a l y s t s o r components o f c s t s l y s t s . Further p o s s i b i l i t i e s f o r modification are the so-called p o s t - s y n t h e t i c processes. The p u r p o s e o f such a t r e a t m e n t is t o
TABLE 7
c
co
Q,
S t r u c t u r a l and a c i d i c p r o p e r t i e s o f HS 30 (H-forms)
Zeolite
Type o f t r e a tment
Treatment temperature
Extraction process
a f t e r t h e r m a l and h y d r o t h e r m a l t r e a t m e n t
SiOd
Total
Number o f
Lattice
A1 0 soluhe
K atoms
H-HS H-HS H-HS H-HS H-HS
30 30 30 30 30
H-HS 30 H-HS
30
-
thermal thermal thermal hydrothe rmel hydrot hermel hydrothermal
-
6 n HC1 EOTA, i n
2)
1)
-
NH OH, i n HC1 6 HC1 6 n HC1
29.6 32.4 31.4 32.5 313.5
64.0 58.8 60.6 58.6 49.9
66.3 12.0 12.0 12.0 30.7
54.1 11.5 12.7 11.6 32.3
5.2 3.4 5.4 14.1
773
6 n HC1
55.7
34.9
15.1
14.2
29.1
973
6 n HC1
71.8
27.3
2.4
-
36.7
1073 1073 1073 623
8
tic1
" L a t t i c e A l " 8 T o t a l o f t h e number o f Broensted and L e w i s centres.
*)
"Amorphous Al s o l u b l e " :
D i f f e r e n c e between t h e t o t a l number o f A 1 atoms o f t h e i n i t i a l z e o l i t e and t h e r e s p e c t i v e sample.
199 remove A 1 f r om t h e l a t t i c e i n o r d e r t o a c h i e v e t h e r e q u i r e d a c i d i t y . We t e s t e d Zeosorb HS 30 as a b a s i c m a t e r i a l i n t h e r m a l and h y d r o t h e r m a l t r e a t m e n t processes. These p r o c e s s e s can b e f o l l o w e d b y e x t r a c t i o n s t e p s f o r t h e d i s p o s a l o f t h e removed alum inium ( t h e e x t r a - l a t t i c e a l u m i n i u m ). The e f f e c t s t h a t
can b e a c h i e v e d a r e shown i n f a b l e 7.
Th er m al t r e a t m e n t was c a r r i e d o u t a t t e m p e r a t u r e 8 f r o m 500 t o 1000
OC
i n a f l a t - b e d f o r a p e r i o d o f 0,5 t o 24 h.
some examples a r e g i v e n . p e r i o d o f treatment,
OC
I n Table 8
0 y c h o o s i n g t h e t e m p e r a t u r e and/or
t h e c o n t e n t o f A1 i n t h e l a t t i c e can b e
varied systematically. TABLE 8 P e n t a s i l t y p e z e o l i t e Zeosorb HS 30 f i e l d s o f a p p l i c a t i o n i n t h e GDR
-- cc aa tt aa ll yy ss tt -- cc aa tt aa ll yy ss tt However,
for for for for i t
i n t h e samples,
a h y d r o - c r a c k - f o r m i n g p r o c e s s (Leuna) a dewaxing p r o c e s s (Schwedt, BBhlen) h y d r o r a f f i n a t i o n (Schwedt) h y d r o t r e a t i n g (Schwedt)
i s i m p o s s i b l e t o re d u ce t h e t o t a l c o n t e n t o f A1 even u n d e r d r a s t i c e x t r a c t i o n c o n d i t i o n s .
The
s p e c i e s o f A 1 o b t a i n e d b y t h e r m a l t r e a t m e n t remai n i n t h e sample as e x t r a - l a t t i c e aluminium. H y d r o t h e r m a l t r e a t m e n t makes a s p e c i f i c a d j u s t m e n t o f t h e content o f A1 i n the l a t t i c e possible,
end,
apart from t h 3 t ,
t h e p c r t i o l e x t r a c t i o n o f t h e removed a l u m i n i u m ( u p t o 60%) from t h e l a t t i c e system o f t h e z e o l i t e . T h i s i s i m p o r t a n t f o r t h e i r a p p l i c a b i l i t y as c a t a l y s t o r c a t s l y s t c a r r i e r s .
Comparing t h e
e f f i c i e n c y o f b o t h processes o f after-treatment,
we f i n d t h a t :
- with
-
b o t h p r o c e s s e s p r o d u c t s cen b e manufactured,
that other-
w i s e can o n l y be s y n t h e s i z e d w i t h t h e h e l p o f t e m p l a t e s , d u r i n g hydrothermal treatment the tempersture o f adjustment f o r a c e r t a i n c o n t e n t o f A 1 i n t h e l a t t i c e can be 150 O C t o 200 OC l o w e r, i n comparison w i t h t h a t ,
thermal treatment.
l e s s equipment i s needed f o r t h e
A? P L I CAT I ONS F o r s p p l i c a t i o n as c e t o l y s t c a r r i e r s ,
n e c e s s a r y i n most cases.
F o r r e a c t i o n s u s i n g an a c i d i c c a t e l y s t
t h e H-form i s s u i t a b l e i n most coses.
+ !JH4 - i o n
i o n exchange w i l l be
I t can be a c h i e v e d b y
e x c h a n g e o r by t h e d i r e c t a c t i o n o f a c i d s ,
as Zeosorb
HS 30 i s s t a b l e t o a c i d s . F o r s p e c i a l p r o c e s s e s a m o d i f i c a t i o n w i t h v a r i o u s m e t a l c a t i o n s w i l l be necessary. o f HS 30 w i t h Zn2+-
A modification
o r K + - i o n s p r o v e d t o be s u i t a b l e f o r o b t a i n -
i n g a r o m a t i c s by p e t r o c h e m i c a l p r o c e s s e s .
The c a t a l y t i c e x p e r i -
ments were c a r r i e d o u t a t t h e C e n t r a l I n s t i t u t e of O r g a n i c C h e m i s t r y ( Z e n t r a l i n s t i t u t f u r O r g e n i s c h e Chemie) under t h e supervision o f Prof.
at Leipzig
Anders.
O t h e r a p p l i c c t i o n s o f Z e o s o r b I-IS 30 i n i n d u s t r i z l ~ r c c t ? . " E i
in
t i l e ?'X :,r? choi!ii
i n T a b l e 8.
201
REFERENCES L a n d o l t , US-Potent 3 702 886 (1972) ;\rc!cvcr c n d G.R. 1 x.K. R o s i n s k i and Ch.J. P l a n k , DE-Patent 2 M.V. Rubin, E.J. 2 4 4 2 240 ( 1 9 7 5 ) Rollmann, US-Petent 4 1 0 8 881 (1978) 3 L.D. 4 R.M. B a r r e r and P.J. Denny, 3. Chem. SOC. 971 (1961) 5 G.T. K e r r , US-Potent 3 459 6 7 6 (1969) 6 B.M. Lowe and R. Aroya, EP 77 624 ( 1 9 8 3 ) 7 Y.N. Dwyer e n d E.E. J e n k i n s , US-Patent 3 9 4 1 8 7 1 (1976) 8 H. Bremer, V!. R e s c h e t i l o w s k i , D.Q. Sou, K.P. Wendlandt, P.N. Nau und F. Vogt, A c i d i t a t u n d k a t e l y t i s c h e E i g e n s c h a f t e n des Z e o l i t h s ZSM-5, 2. Chem. 2 1 (1981) 77-78 9 M.F.M. P o s t and J.M. Nanne, DE-Patent 2 913 5 5 2 (1979) 10 M.R. K l o t z , US-Petent 4 377 5 0 2 ( 1 9 8 3 ) 11 T.V. W h i t t a m , EP 59 359 (1982) 1 2 H. Hugiwara, Y. I
This Page Intentionally Left Blank
203
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V.,Amsterdam
SELECTIVE CONVERSION
OF SYNGAS TO HYDROCARBONS BY ZEOLITES
Hiro-o Tominaga, Kaoru Fujimoto and Takashi Tatsumi Department o f S y n t h e t i c Chemistry, F a c u l t y o f Engineering. The U n i v e r s i t y Tokyo, Hongo 7-3-1. Bunkyo-ku, Tokyo 113 (Japan)
of
SUMMARY A s e r i e s o f s t u d i e s were made on the use o f z e o l i t e c a t a l y s t s f o r producing selectively some s p e c i f i e d hydrocarbons from synthesis gas. They are e x e m p l i f i e d by t h e s e l e c t i v e production o f gasoline, r i c h i n i s o p a r a f f i n s , by Ru-Pt b i m e t a l l i c c l u s t e r supported on protonated Y z e o l i t e and t h a t of C3 and C4 p a r a f f i n s , s u b s t i t u t e s f o r LPG, by use o f a h y b r i d c a t a l y s t comprising Ytype z e o l i t e and methanol synthesis c a t a l y s t , etc. For e l u c i d a t i n g t h e s p e c i f i c product s e l e c t i v i t y , r e a c t i o n mechanisms are discussed i n terms o f t h e balance and p r o x i m i t y o f the r e l e v a n t d i f f e r e n t k i n d s o f a c t i v e s i t e s f o r consecutive reactions. INTRODUCTION Fischer-Tropsch process i s i n i n d u s t r i a l o p e r a t i o n a t SASOL producing m o s t l y aliphatic
1i q u i d
quantities
of
hydrocarbons
from
synthesis
I t gives,
gas.
methane and o t h e r gaseous hydrocarbons as
however,
byproducts.
In
the
f i x e d bed reactor, operated a t lower temperature, a l a r g e amount o f medium wax is
a l s o produced.
aldehyde strong
A d d i t i o n a l l y , non-hydrocarbon c o n s t i t u e n t s such as
and a c i d are found i n t h e l i q u i d product.
Obviously t h e r e
alcohol, exists
need t o improve the process f o r b e t t e r y i e l d and q u a l i t y o f t h e
a
liquid
fuels. In
t h i s connection,
carbon number d i s t r i b u t i o n s o f F-T
synthesis
products
over b u l k c a t a l y s t s , such as p r e c i p i t a t e d o r fused i r o n employed a t SASOL, g e n e r a l l y found t o f o l l o w "Schulz-Flory-Anderson t h e l a w works, t h e l a r g e s t y i e l d o f gasoline,
(SFA) Law" ( r e f . 1).
C5 t o Cll,
weight a t a chain growth p r o b a b i l i t y f a c t o r a = 0.7 Non-Schulz-Flory-Anderson
encapsulated was
another
formation
of
of
F-T
catalysts:
4) p o s s i b l y due
t h e z e o l i t e supported c a t a l y s t .
zeolite
over which g a s o l i n e
over t h e conventional b u l k
branched-chain p a r a f f i n s ( r e f .
characteristics
0.8.
improved. The use o f z e o l i t e as support f o r
advantage
Provided
can n o t exceed 48% by
k i n e t i c s , however, has been r e p o r t e d f o r
Ru ( r e f . 2 ) and Fe ( r e f . 3) c a t a l y s t s ,
significantly
offers
-
to
catalysts
consisting
supports i n c l u d i n g z e o l i t e .
of
noble metals
namely
as
the
bifunctional
Isomerization such
yield
synthesis
of
p a r a f f i n s f o r octane number improvement i s commercially c a r r i e d o u t on bifunctional
are
Pt
on
light various acidic
204
on
I n view o f these, e x p l o r a t i o n o f t h e Pt-Ru b i m e t a l l i c c a t a l y s t supported zeolites
was attempted and proved s u c c e s s f u l b y t h e p r e s e n t a u t h o r s t o
h i g h octane g a s o l i n e d i r e c t l y f r o m syngas ( r e f .
obtain
5-9).
Another p o s s i b i l i t y t o produce a r o m a t i c r i c h g a s o l i n e from s y n t h e s i s gas may be o f f e r e d by use o f h y b r i d c a t a l y s t c o m p r i s i n g methanol s y n t h e s i s c a t a l y s t and pentasil
type
combination
zeolite.
of
During
the
course
of
this
study,
a
particular
t h e component c a t a l y s t s was found t o g i v e propane
and
butane
h ig h l y s e l e c t ive 1y. SELECTIVE SYNTHESIS OF GASOLINE R I C H I N ISOPARAFFINS C a t a l y s t p r e p a r a t i o n and r e a c t i o n
A
ion-exchange
s e r i e s o f z e o l i t e s c o n t a i n i n g Ru and/or P t were prepared by
w i t h an aqueous s o l u t i o n o f [Ru(NH3)6]C13 and/or [Pt(NH3)4]Clz a t 6OoC f o r 2 h. Bimetallic
catalysts
ion-exchange, was
no
were prepared b y simultaneous ion-exchange.
t h e c a t a l y s t s were f i l t e r e d ,
longer
detected,
After
washed s e v e r a l t i m e s u n t i l
and then d r i e d i n
air
llO°C.
at
The
the
chlorine exchanged
c a t a l y s t s were dehydrated s l o w l y i n s i t u under a He f l o w by s t e p w i s e h e a t i n g up to
4OO0C,
f i n a l l y reduced w i t h HZ a t 4OO0C f o r 12
and
d e s i g n a t i o n s o f c a t a l y s t i s , f o r example,
h.
The
HY2R2P which r e p r e s e n t
abbreviated a
catalyst
w i t h 2% Ru and 2% P t by w e i g h t supported on p r o t o n a t e d Y t y p e z e o l i t e H
RESULTS AND DISCUSSION S y n e r g i s t i c s o f a l l o y i n g Ru w i t h
pt
Table 1 shows s y n t h e s i s gas conversion data,
R e s u l t s o b t a i n e d w i t h t h e HYER monometallic c a t a y s t and
b i m e t a l l i c catalysts. the
HY2RtHY2P
amounts
of
The HY2P c a t a l y s t was a poor c a t a l y s t f o r CO c o n v e r s i o n (ca. 0.1 %) w i t h
95
HYZR+HYZP,
%
a f t e r 5 h on stream, w i h Ru-Pt
mixed c a t a l y s t ,
consisting
of
equal
a r e a l s o included.
selectivity
higher
physically
than
observed
CH4.
for that
The CO c o n v e r s i o n
over
HYPR.
over
HYER2P
No s y n e r g i s t i c e f f e c t
on
significantly
was
CO
conversion
f o r t h e p h y s i c a l l y mixed c a t a l y s t HYER+HY2P, s u g g e s t i n g t h a t a
was close
p r o x i m i t y and i n t e r a c t i o n o f Ru w i t h P t a r e r e q u i r e d f o r h i g h a c t i v i t y .
There
appears
three
t o be l i t t l e d i f f e r e n c e i n carbon number d i s t r i b u t i o n among t h e
catalysts. hydrocarbons,
Whereas Pt
t h e HYER c a t a l y s t gave p r o d u c t s
rich
in
f u r t h e r increased t h e i / n r a t i o , provided i t
branched-chain was
added
by
simultaneous i o n exchange. Fig.
1 shows
distribution
for
the
effect
o f Ru
v a r i o u s Ru-Pt/HY
metal
loading
catalysts, with a
on
the
P t content
carbon of
number
2
T h e i r c a t a l y t i c performances a r e a l s o l i s t e d i n Table 1. The CO c o n v e r s i o n O b v i o u s l y HY2R2P gave t h e b e s t r e s u l t s , w i t h a 66 % s e l e c t i v i t y f o r Cg-
wt%. and
205
TABLE 1 S y n t h e s i s gas c o n v e r s i o n over HY z e o l i t e - s u p p o r t e d Ru-Pt c a t a l y s t a . ~~
CO c o n v e r s i o n Cata 1ys t HYER HY2R2P ( HYZR+HYZP ) b HYO. 5R2P HY 1R2P HY4R2P HY8R2P
(XI 6.5 13
6.1 1.5 3.6 a5
19
S e l e c t i v i t y (wtW) Carbon number i/n ratio ‘5’ ‘4 ‘5 ‘6-‘10 ‘1 ‘4
6.5 7.0 7.8 41 13 90 16
24 22 31
66 66 56
17 31 1.6 18
33 31 5.2
56
6.8
10 6.5 7.5 9.3 0.2 0.3
aReaction c o n d i t i o n s : 240OC. 1.5 MPa, W/F = 7.0 g-cat. a f t e r 5 h on stream. bW/F = 14.0 g-cat. h/mol.
30 40 24 12
27
11 16
7.6 4.8 6.5
4.4
3.0
1.7
1.0
h/mol,
CO/Hz = 2/3,
HYlR2P 0 HY2R2P A
HY4R2P
Carbon number
F i g . 1. E f f e c t o f metal l o a d i n g on carbon number d i s t r i b u t i o n (240 OC, 1 , 5 MPa, W/F = 7.0 g-cat. h/mol. CO/H2 = 3. a f t e r 5 h on stream.
206
TABLE 2 Dispersion o f metal p a r t i c l e s and t u r n o v e r number
Catalyst
CO/Ma
H/Ma
HY2R HY4R HY2R2P HYEP
2.01 1.18 0.28 0.09
0.61 0.53 0.27 0.23
Particlg size (A )
D from
14 18 28 35
TEMa
N(cgIC (10- / s )
0.61 0.47 0.32 0.27
12.9 15.9 65.8 0.94
aRefer t o t h e d e f i n i t i o n i n Ref. 6. bVolume-area mean diameter determined by TEM. CTurnover number f o r CO conversion a t 24D°C, 1.5 MPa and W/F = h/mol.
12.5 g-cat.
C1 hydrocarbons being obtained. Characterization o f catalysts I n t h e p r e p a r a t i o n o f b i m e t a l l i c c a t a l y s t s , t h e main problem i s whether b o t h elements are a l l o y e d o r not. a
broad
diffraction
40.7',
showed t h e presence o f
characteristic
showed
t h a t the HYZR c a t a l y s t has a
a maximum f r a c t i o n below 10 A.
distribution with
seem t o e x i s t i n supercages. The
narrow
o f H2 and
results
Hence t h e
metal
size
particles
On t h e o t h e r hand, HY2R2P and HY2P e x h i b i t e d wide
respectively.
CO chemisorption a r e given i n Table 2,
d i s p e r s i o n o f metal (D) c a l c u l a t e d on t h e b a s i s
the
Ru-Pt
particle
t h e mean m e t a l l i c p a r t i c l e s i z e s were 28 and 35 A,
distributions:
of
or P t phase i n HY2R2P.
There was no XRD evidence o f a monometallic Ru
alloys.
TEM observations
shows
The XRD p a t t e r n o f HY2R2P
band centered a t 28 =
of
which
TEM,
also
assuming
cubic metal c r y s t a l l i t e s .
There seems t o be s a t i s f a c t o r y agreement between t h e
D values from TEM and H/M,
suggesting t h a t t h e adsorption s t o i c h i o m e t r y can
be
reasonably assumed t o be u n i t y f o r b o t h Ru and P t . turnover number f o r CO conversion, Nco,
The H2
turnover
number
was c a l c u l a t e d on t h e b a s i s
and t h e r e s u l t s a r e given i n Table 2.
chemisorption
of
HY2R2P w i t h t h a t o f HYPR i s
not
A comparison
of
straightforward,
considerable amount of hydrogen i s adsorbed on P t , where t h e s p e c i f i c
observed size
trend
(ref.
a
HY2R2P
However, one must c o n s i d e r t h e g e n e r a l l y
o f increases i n t h e t u r n o v e r number w i t h
10).
as
activity
It i s tempting t o speculate t h a t t h e t u r n o v e r number o f
i s much lower.
was increased b y a l l o y i n g Ru w i t h Pt.
of the
increasing
Attempts t o improve t h e a c t i v i t y o f HY2R by
particle
increasing
the
p a r t i c l e s i z e were unsuccessful. It
should
decrease
in
also
be noted t h a t t h e a d d i t i o n of P t
t h e (CO/M)/(H/M)
= CO/H r a t i o .
resulted
Chen e t al.
in
suggested
a
drastic that
for
207
z e o l ite-supported
Ru
turnover
number
increased
as t h e CO/H
10).
(ref.
catalysts,
the
CO
conversion
ratio
decreased
for
T h i s t r e n d was accounted f o r
by l e s s CO i n h i b i t i o n and hence a h i g h e r
CO
hydrogenation
catalyst mechanism
2150
a l o w CO/H
with
for
might
hold
the
ratio.
i n Table 2 suggest t h a t a
data
HY2 R
activity
for
The
similar
the
Ru-Pt
b i m e t a l l i c system. The
2060
HYZR,
in
s t a t e o f Ru and/or P t
HYEP, and HY2R2P was i n v e s t i g a t e d by
IR
spectroscopy.
CO
The
spectra
of
adsorbed over t h e t h r e e c a t a l y s t s
after
e v a c u a t i o n a t 200°C a r e shown i n Fig. 2. The
HY2R2P
bands a t 2076, 2040, and 2010
2070
cm-l on HYEP
Hence
2000
2400
1900
W a v e n u m b e r (cm-') Fig. 2.
I R s p e c t r a o f CO
HYZRZP, HYZR, and
evacuation a t
ZOO
at
disappeared.
band a t 2060 cm-I
could
be
t o CO adsorbed on
w i t h Pt.
As t h e s t r e t c h i n g frequency of
Ru
adsorbed on Ru was h a r d l y
alloyed dependent
on t h e p a r t i c l e s i z e o f Ru, t h e observed
adsorbed
HYZP
almost
ascribed
CO on
the
cm-I
A broad band
were observed on HYPR.
s h i f t seems t o be due t o t h e
(after
electronic
l i g a n d e f f e c t o f P t , which was r e f l e c t e d
OC).
i n t h e CO c o n v e r s i o n a c t i v i t y . I n f l u e n c e o f alkene a d d i t i o n Alkenes a r e p r o b a b l y t h e p r i m a r y p r o d u c t s o f F-T s y n t h e s i s o v e r t h e z e o l i t e supported Ru c a t a l y s t s ( r e f . 11). alkanes .might zeolite
sites,
expected
that
The p r e f e r e n t i a l f o r m a t i o n o f branched-chain
be due t o t h e a c i d - c a t a l y z e d i s o m e r i z a t i o n o f
the
f o l l o w e d d i r e c t l y by h y d r o g e n a t i o n t h e Ru s i t e s . the r e l a t i v e r a t e s o f the
acid-catalyzed
and
alkenes
on
It
be
can
metal-catalyzed
r e a c t i o n s should determine p r o d u c t s e l e c t i v i t y . To check t h e p o s s i b i l i t y of h y d r o c r a c k i n g r e a c t i o n s c o n c u r r e n t w i t h t h e
F-T
c6-c12 1-alkenes were a l l o w e d t o r e a c t t o g e t h e r w i t h s y n t h e s i s
gas
synthesis, over
OAHYZRZP.
1-hexene,
The carbon number d i s t r i b u t i o n s o b t a i n e d w i t h t h e a d d i t i o n
1-octene,
and 1-decene were examined,
By s u b t r a c t i n g
the
i n h e r e n t i n CO hydrogenation, a c a t i o n i c c r a c k i n g p a t t e r n can be achieved. example,
the
extensive
formation
of
C4 f r o m C8
could
be
of
portion
attributed
For to
208 hydrocracking
over a c i d i c s i t e s o f the z e o l i t e .
C2 i s c o n s i s t e n t w i t h t h e proposed r u l e s o f t h e carbenium
and
and Cn-2
Thus t h e f o r m a t i o n o f Cn-l of
cn
the
enhanced p r o d u c t i o n o f c 7 and c 8 f o r 1-hexene
t h a t a dimerization-cracking
process (eq.
mechanism. scission
addition
1) i s t a k i n g place.
C12 s p e c i e s w i l l be f a i r l y u n s t a b l e and form
adsorbed
ion
f r o m Cn c a n n o t r e s u l t f r o m s i m p l e
T h i s t y p e o f c r a c k i n g decreased i n t h e o r d e r c6 > c8 >>
feed.
Concurrently
C1
V i r t u a l l y no i n c r e a s e i n
the
C12.
suggests
The i n t e r m e d i a t e disproportionation
T h i s i s supported by t h e appearance o f an a p p r e c i a b l e amount o f
products.
C12
s p e c i e s on a d d i t i o n o f 1-hexene.
A
sharp
This as
i n c r e a s e i n t h e r a t e o f h y d r o c r a c k i n g was
observed
for
c8-Clo.
i s i n agreement w i t h t h e expected r e l a t i v e easiness o f c a t i o n i c a
function
considered
of
to
chain length.
H y d r o c r a c k i n g on
be preceded by c o n v e r s i o n t o m u l t i b r a n c h e d
chain
alkene
pore
cracking
zeolites
isomers,
which
is is
The d i f f e r e n c e between t h e
e v i d e n t l y more d i f f i c u l t f o r s h o r t e r c h a i n alkenes. apparent
large
c o n v e r s i o n and c r a c k i n g y i e l d was i n c r e a s e d
with
decreasing portion
of
c r a c k i n g b e h a v i o r o f 1-octene under a He f l o w o v e r m e t a l - f r e e DAHY
was
length,
which
was
due
to
the
increase
in
the
disproportionation. The also
investigated.
There
were
striking
similarities
between
simple
CO
hydrogenation o v e r DAHYZR2P and 1-octene c r a c k i n g o v e r DAHY i n t h e C3+ p r o d u c t s d i s t r i b u t i o n and s e l e c t i v e f o r m a t i o n o f i s o a l k a n e s . The
a d d i t i o n o f p r o p y l e n e and cis-2-butene
r e s u l t e d i n t h e increase i n
the
These m i g h t be formed
via
y i e l d o f hydrocarbons h i g h e r t h a n t h e f e e d alkenes.
t h e i n c o r p o r a t i o n o f alkenes i n t o t h e c h a i n p r o p a g a t i o n s t e p o f F-T However,
these i n c r e a s e s a r e most l i k e l y due t o t h e superimposed
synthesis. cracking
of
o l i g o m e r s formed f r o m t h e f e e d alkenes on a c i d s i t e s o f t h e z e o l i t e s .
I n fact,
a f t e r 5 h on stream,
observed
with
r e l a t i v e l y h i g h y i e l d s o f c6 and $ 3 s p e c i e s were
t h e a d d i t i o n o f p r o p y l e n e and cis-2-butene,
respectively,
indicating
the
occurrence o f d i meriz a t ion. Comparison from
o f t h e r e l a t i v e p o r t i o n s found w i t h i n t h e C4-C9
ranges
s i m p l e CO hydrogenation and t h o s e from a d d i t i o n o f 1-octene
demonstrated
that
t h e d i s t r i b u t i o n o f isomers from t h e
resembles t h e one observed w i t h s i m p l e CO hydrogenation. applied
addition Under t h e
obtained
and
propene
of
alkenes
conditions
i n t h i s s t u d y t h e p r o d u c t s c o n s i s t e d m a i n l y o f branched alkanes.
f r a c t i o n s o f monobranched alkanes were f a r above t h e e q u i l i b r i u m v a l u e s
The (e.g..
209
31% f o r 2-methylpentane and 15% f o r 3-methylpentane). compounds
was
negligible. simple
relatively
small
and
those
with
The amount o f
C
quartenary
dimethyl
atoms
were
The s t r i k i n g s i m i l a r i t i e s i n t h e isomer d i s t r i b u t i o n between
CO hydrogenation and a1 kene a d d i t i o n suggest t h a t t h e common
the
mechanism
i s o p e r a t i v e f o r isoalkane formation. The
cracking
Obviously is
probability
o f n-octane i s compared
to
that
of
1-octene.
n-octane was l e s s s u s c e p t i b l e t o hydrocracking than 1-octene.
what
we
would expect because carbenium i o n s can
be
easily
This
formed
from
alkenes. w h i l e t h e c l a s s i c a l mechanism f o r a c i d c r a c k i n g o f alkanes i n v o l v e s
a
bimolecular hydride t r a n s f e r i n t h e r a t e determining step. E f f e c t o f support carbon number d i s t r i b u t i o n o f the products from t h e CO-HE r e a c t i o n
The
t o t h e l i g h t s i d e i n t h e o r d e r : NaYZRZP ( r e f . 5) < HYZR2P <
shifted (ref.
6).
When 1-octene was added t o s y n t h e s i s gas, t h e product
distribution
w i t h apparent conversion o f 1-
was a f f e c t e d by concurrent c r a c k i n g o f 1-octene; octene
a f t e r 2-3 h on stream i n c r e a s i n g i n t h e order: NaY2R2P (15%)
(26%) < OAHYZRZP (65%).
was
DAHYEREP
HYZREP
<
As a r e s u l t , t h e y i e l d o f C3-C7 hydrocarbons increased
by 1-octene a d d i t i o n i n t h e same order: NaY2R2P < HY2RZP < DAHY2R2P.
TPD measurements o f NH3 revealed t h a t t h e amounts o f weak and s t r o n g acids, d e f i n e d as t h e amount o f ammonia desorbed below and above 350 were as follows: NaYZREP (1.51/0.20
mmol g-cat.-l),
HY2REP (4.17/0.43
cat.-’),
and
tendency
o f 1-octene could be r e l a t e d t o t h e amount o f s t r o n g a c i d
the
DAHYEREP (3.42/0.68
respectively,
OC
catalysts:
NaYZRZP
mmol g-cat.-’).
< HYERZP < DAHYERZP.
cracking requires strong acid s i t e s (ref.
34-
Over
It
is
generally
the
metal
g-
cracking sites
known
on that
Isoalkanes
t h e HY and DAHY supported Ru and Ru-Pt c a t a l y s t s , between
mmol
the
12).
Mechanism f o r s e l e c t i v e synthesis o f C -C observed
Therefore,
p a r t i c l e s i z e and
chain
no
correlation
limitation;
a
was metal
p a r t i c l e s i z e dependent mechanism f o r t h e sharp carbon number d i s t r i b u t i o n been r u l e d o u t ( r e f . 2). various
feed
Cracking
P l a u s i b l e secondary r e a c t i o n s and t h e r e a c t i v i t i e s o f
alkenes, dependent on chain length, a r e summarized
01 igomerization is
and
disproportionation
observed
Hydroisomerization
has
to
increase
i s also favorable
proceed
particularly
fast
for
in
the
in
Fig.
light
3.
a1 kenes.
CS-C10
range.
f o r heavy alkenes, e s p e c i a l l y c7 and
c8
a1 kenes.
A p l a u s i b l e r e a c t i o n pathway f o r the synthesis o f g a s o l i n e range is
as
follows:
CO hydrogenation on t h e Ru-Pt b i m e t a l l i c a l l o y
alkenes as the primary products,
isoalkanes
produces
f o l l o w e d by various secondary r e a c t i o n s .
1The
210
Carbon number of fed alkene
4
3
2
6
5
7
9
8
10
11
Cross chain propagation
S i g n i f i c a n c e o f secondary r e a c t i o n s as a f u n c t i o n o f hydrocarbon c h a i n
Fig. 3. length. light
alkenes a r e oligomerized on t h e a c i d s i t e s t o form branched
hydrocarbons, products.
which
Heavy
converted
to
are
subsequently
alkenes
formed
cracked
by t h e
F-T
to
give
reaction
1 i g h t branched hydrocarbons v i a s k e l e t a l
could
be
similarly Since
distribution state
carbenium i o n s on t h e s u r f a c e approximates t o t h e pseudo-equilibrium
and
hence
the
range o f C8-Cl0 this
than
hydrogenation.
products
obtained
from
different
feeds
is
The sudden drop i n t h e chain growth p r o b a b i l i t y i n t h e
i s c o n s i s t e n t w i t h t h e sharp increase i n t h e h y d r o c r a c k i n g r a t e
region.
reactions
of
pattern
s u b s t a n t i a l l y t h e same. in
chain
isomerization.
o l i g o m e r i z a t i o n and c r a c k i n g a r e m u t u a l l y opposing r e a c t i o n s , t h e of
long
disproportionation
Since alkenes a r e more s u s c e p t i b l e a1 kanes,
they
have
to
undergo
to
the
such
acid-catalyzed
reactions
before
Hence t h e p r o x i m i t y o f t h e metals and a c i d s i t e s as w e l l as
an
adequate balance between t h e metal and t h e a c i d f u n c t i o n s has t o be achieved i n o r d e r t o d i r e c t the r e a c t i o n towards t h e d e s i r e d products. Durabi 1it y o f z e o l i t e supported Ru-Pt b i m e t a l 1 i c C a t a l y s t s As
is
catalyst catalyst
o f t e n the case f o r CO hydrogenation c a t a l y s t s ,
the a c t i v i t y
exhibited
General
a
decay
with
time
decay have been w e l l established,
sintering
of
deactivation
the is
solid. essential
A
better
on
stream.
of
the
causes
of
and i n v o l v e poisoning, coking,
and
understanding
t o the development
of
the
mechanism
for
practical
of
catalysts
with
improved a c t i v i t y , s e l e c t i v i t y and l i f e . Fig. 4 shows the change o f CO conversion w i t h time on stream. change
seems t o depend s t r o n g l y on t h e r e a c t i o n v a r i a b l e s .
MPa. t h e i n i t i a l a c t i v i t y o f HYZRZP was r e l a t i v e l y low. sharply
The
At 24OoC
activity and
1.5
The a c t i v i t y increased
d u r i n g t h e f i r s t 3 h and then d e t e r i o r a t e d r a p i d l y .
A t 24OoC and
5.1
211
Fig. 4. A c t i v i t y change w i t h t i m e on stream. : HYERZP, 240 OC, 5.1 MPa, U : HYZRZP, 240 5.1 MPa, 0 : DAHYZRZP, 240 OC. 1.5 MPa).
( 0 : HYZRZP, 240 OC, 1.5 MPa,A 0.2 MPa. A : HYZRZP, 290 O C ,
OC,
MPa, t h e r e was a r a p i d decay s i m i l a r t o t h a t observed a t 1.5 MPa. Under these c o n d i t i o n s methane was t h e main product and t h e
MPa.
C5t hydrocarbons was low.
for
from t h a t o f HYZRZP.
different to
On t h e o t h e r
r a t h e r s t a b i l i z e d a c t i v i t y was obtained a t 24OoC, 0.2 MPa o r 290°C,
hand,
C5t
for
DAHYERZP
were
The behavior o f
TPD
NH3
of
decrease deposits.
in
acidity The
selectivity significantly
those
for
selectivity
HYZRZP,
the
CO
leveled o f f .
revealed t h a t t h e amount o f weak
a c i d was s l i g h t l y
was
Although the i n i t i a l a c t i v i t y and
somewhat lower than
conversion increased g r a d u a l l y and then strong
DAHYZRZP
5.1
a c i d was l e s s
and
that
l a r g e r on f r e s h DAHYZRZP than on f r e s h HYZRZP.
during the reaction
is
probably
due
of The
to
carbonaceous
l o s s o f surface acid, e s p e c i a l l y weaker acid, on
HYZRZP was
reflected
in
suggested
t h a t i s o m e r i z a t i o n i s catalyzed by weaker a c i d s i t e s .
t h e s i g n i f i c a n t decrease
i n t h e isoalkane s e l e c t i v i t y .
A
It
is
relatively
l a r g e amount o f a c i d s i t e s remained on DAHYZRZP a f t e r use, which c o i n c i d e s w i t h the remarkable branched isomer s e l e c t i v i t y on t h e aged DAHYERZP c a t a l y s t . difference could
in
t h e p r o p e r t y o f t h e coke species between HYZRZP
be r e l a t e d t o a c i d - p r o p e r t i e s such
acid sites.
and
as t h e s t r e n g t h and d e n s i t y
The
OAHYEREP of
the
212 SELECTIVE SYNTHESIS OF PROPANE AND BUTANE Another p o s s i b i l i t y of p r o d u c i n g h i g h octane g a s o l i n e f r o m s y n t h e s i s gas to
e n r i c h , by some means, i t s c o n t e n t o f aromatics.
An i d e a had been
presented by Chang e t a1 t o use h y b r i d c a t a l y s t s c o m p r i s i n g methanol
is
already synthesis
c a t a l y s t such as Z r o x i d e o r Zn-Cr mixed o x i d e and p r o t o n a t e d ZSM-5 ( r e f .
13).
I n o r d e r t o e x p l o r e b e t t e r y i e l d o f t h e aromatics i n g a s o l i n e b o i l i n g range, a wide v a r i e t y o f combination o f methanol s y n t h e s i s c a t a l y s t s and z e o l i t e s have been surveyed e x p e r i m e n t a l l y by t h e p r e s e n t a u t h o r s ( r e f . profile,
14-17).
The
product
a
function
i n terms o f carbon number and s t r u c t u r e , was found t o be
n o t o n l y o f a c t i v i t i e s o f the r e l e v a n t catalysts, different
kinds
reactions. zeolite.
o f active sites participating i n the
In
this
regard,
the s i z e o f pore
and
consecutive crystal,
elementary
especially
use o f t h e two
t o g e t h e r w i t h t h e way o f mixed, o r separate,
catalysts,
the
but also o f proximity o f
of
component
p l a y t h e key r o l e . T h i s i s i n consequence o f t h e f a c t t h a t t h e f i n a l
p r o d u c t p r o f i l e i s s t r o n g l y dependent on t h e r a t e s o f i n t r a - a n d diffusions,
inter-catalyst
i n competition o f t h e r a t e s o f surface reactions, o f t h e
reaction
intermediates. During
the
combination
course o f t h i s study,
of
a h y b r i d c a t a l y s t made
t h e component c a t a l y s t s , e.g.,
c a t a l y s t and DAY z e o l i t e , was found t o g i v e
methanol
LPG, h i g h l y s e l e c t i v e l y ,
of
commercially
namely more t h a n 70% ( r e f .
a
particular
available
propane and
Cu-Zn
butane,
or
18-20).
These w i l l be mentioned below. Catalyst preparation Several
k i n d s o f methanol s y n t h e s i s c a t a l y s t were employed. Pd/ID
catalyst
4 w t % ) was prepared by i m p r e g n a t i n g a commercially a v a i l a b l e s i l i c a g e l ( F u j i Davison ID, s p e c i f i c s u r f a c e area 270 m2 /g, mean p o r e d i a m e t e r 140 A)
(Pd
w i t h p a l l a d i u m c h l o r i d e from i t s a c i d i c aqueous s o l u t i o n ,
f o l l o w e d by d r y up i n
an a i r oven and r e d u c t i o n i n f l o w i n g hydrogen a t 400 OC.
Cu-Zn(H)
40,
23.
Zn
following
A1 the
27 atom %, r e s p e c t i v e l y ) method
described
by
was
prepared
Shimomura.
in
Cu-Zn(C)
catalyst(Cu
this
laboratory
catalyst
was
a
commercially a v a i l a b l e from BASF (S8-45). For kinds
t h e sake o f c o n v e r t i n g methanol i n t o a r o m a t i c of
synthesis catalysts. the
hydrocarbons,
a c i d i c z e o l i t e s were employed upon m i x i n g w i t h
the
o f hydrothermal synthesis. Co.
methanol
ZSM-5 was prepared a c c o r d i n g t o t h e method d e s c r i b e d
patent: t h e i r c r y s t a l l i t e s i z e was c o n t r o l l e d by changing t h e
Norton
different
above
H-M
in
temperature
stands f o r Zeolon 100H m o r d e n i t e o b t a i n e d
De-aluminated Y t y p e z e o l i t e , DAY, was s u p p l i e d by Shokubai
from Kasei
I n d u s t r i e s , Ltd. The
H y b r i d c a t a l y s t s were o b t a i n e d as f o l l o w s : Equal p o r t i o n by
weight
of
213 methanol
synthesis
c a t a l y s t and z e o l i t e were p h y s i c a l l y mixed and
powders below 100 mesh sieve. a
ground
to into
The powders were f a b r i c a t e d under p r e s s u r e
d i s k , which was f i n a l l y crushed t o g r a n u l e s w i t h a 20/40 mesh s i e v e f o r
use
i n the a c t i v i t y test. In
some
employed
cases,
for
granular
mixtures o f t h e
two
component
catalysts
t h e assessment o f t h e i r performance i n comparison w i t h
powdery m i x t u r e s
-
were
of
that
t h e standard h y b r i d c a t a l y s t s .
RESULTS AND DISCUSSION I n Table 3 summarized a r e t h e observed changes i n t h e performance o f c a t a l y s t due t o t h e d i f f e r e n t c o m b i n a t i o n o f t h e two components. hydrocarbons was o b t a i n e d by a c o m b i n a t i o n o f Cu-Zn(C)
of
presence that
Largest y i e l d
DAY, where
and
o f oxygenates, methanol and d i m e t h y l e t h e r , i n t h e p r o d u c t
the
activity
p r o d u c t on Cu-Zn(C),
of
DAY f e l l s h o r t o f
converting
s u c c e s s i v e l y t o hydrocarbons,
hybrid
methanol,
suggested
the
i n particular to
It
i s a l s o noteworthy t h a t DAY g i v e s propane and butane a t a h i g h
of
65 - 75 %. T h i s i s v e r y much o u t s t a n d i n g when compared,
primary
aromatics. selectivity
i n Fig.5.
with
r e s u l t s so f a r p u b l i s h e d and a l s o w i t h t h e maximum v a l u e
experimental
the
the
C3
for
C4 p r e d i c t e d by t h e SFA law ( r e f . 20).
plus
S i g n i f i c a n t amount o f a r o m a t i c hydrocarbons was o b t a i n e d by combining e i t h e r
ZSM-5
H-M
or
as
z e o l i t e with
methanol
synthesis
catalyst.
However,
the
aromatics were m o s t l y polymethyl benzenes. is
It
generally
accepted
t h a t both proper
acidity
and
pore
size
p r e r e q u i s i t e t o t h e c a t a l y s i s f o r conversion o f methanol t o aromatics. of
the
acidic
ammonia,
programmed
gives
(ref.
17).
t h e second peak, s m a l l e r t h a n t h e f i r s t , a t I n spite o f this,
view
desorption
o f a l l o f t h e z e o l i t e s employed i n t h i s s t u d y have a s i m i l a r
which the
c h a r a c t e r , observed by temperature
are
In
around
330-350
DAY o n l y c o u l d h a r d l y g i v e aromatics.
of
profile OC
Therefore,
reason f o r t h e l a c k o f a r o m a t i c s may be a t t r i b u t e d t o t h e p o r e s i z e
of
Y
t y p e z e o l i t e which i s s i g n i f i c a n t l y l a r g e r t h a n t h o s e o f ZSM and mordenite. The larger are
t h e pore s i z e , t h e l a r g e r t h e r a t e o f d i f f u s i o n o f t h e p r o p y l e n e
which
produced from methanol i n t h e pore o f z e o l i t e . T h i s was demonstrated by
simple
adsorption
experiment
of
propylene
on t h e
zeolites
by
use
of
c o n v e n t i o n a l vacuum system as i n d i c a t e d i n T a b l e 4 . Based on t h e s e f i n d i n g s , is
presumed t h a t t h e small o l e f i n s a r e e a s i l y g e t o u t o f t h e l a r g e p o r e
type z e o l i t e b e f o r e t h e y o l i g o m e r i z e s t o form
of
a a it
Y
aromatics.
Residence t i m e o f t h e o l e f i n s i n z e o l i t e p o r e i s a f u n c t i o n n o t o n l y o f p o r e size
but
identify affect
also
of
c r y s t a l l i t e s i z e . Therefore,
an
experiment
how t h e c r y s t a l l i t e s i z e o f t h e z e o l i t e employed i n
the
product
profile.
A s shown i n T a b l e 5 ,
the
was
hybrid
yield
of
made
to
catalyst aromatics
214 TABLE 3 Synthesis gas conversiona over h y b r i d c a t a l y s t s b : E f f e c t o f t h e combination on product y i e l d . Methanol c a t a l y s t
Cu-Zn( H)
Pd/ID ZSM-5(A) H-M
Zeolite
DAY
Product y i e l d ( C%) Hydrocarbons Oxygenates
9.6 8.0 tr. tr. 5.4 3.3 CO2 Coke 0.6 3.1 Hydrocarbon d i s t r i b u t i o n (CW) 25.7 52.5 c1+c2 33.3 24.8 c3 5.3 c4 6.2 c5+ 4.2 6.4 10.6 Aromatics 30.6
ZSM-5(A)
12.6 tr.
Cu-Zn( C )
DAY
DAY
35.7 5.3 30.7
13.9 24.7 49.6 11.9
10.0
17.0 0.6 16.2
1.9
-
15.9 1.5 14.7 0.4
22.0 41.6 26.3 10.1 tr.
21.2 26.9 12.7 35.5 3.8
11.7 17.8 49.4 21.1 0
-
tr.
aReaction c o n d i t i o n s , H2/CO=2/1; 573(K); 2.1(MPa); W/F=3.4(g.cat h/mol). bPowdery m i x t u r e o f methanol c a t a l y s t s and z e o l i t e a t equal weight.
M
6o
C
A
I
I
1
2
I
3
I
I
4
5
Carbon number /
I
6
-
Fig.5 Comparison o f hydrocarbon d i s t r i b u t i o n s by C number i n conversion f o r lower p a r a f f i n production.
synthesis
gas
215
TABLE 4 In-pore d i f f u s i o n r a t e a o f propylene i n z e o l i t e Zeolite
HZSM-5
105D/r02(min-1)b
H-mordeni t e
3.1
3.3
DAY 9.2
aobserved a t room temperature bD,diffusion c o e f f i c i e n t (cm2min-l);
ro, r a d i u s o f z e o l i t e
c r y s t a 1(cm)
TABLE 5 Synthesis gas conversiona over h y b r i d c a t a l y s t Pd/ID+ZSM-5b : Effect o f z e o l i t e c r y s t a l s i z e on product d i s t r i b u t i o n
(A) 2.1
(6) 1.3
(C) 0.8
Hydrocarbon y i e l d (CX) C number d i s t r i b u t i o n (CX)
10.6
10.0
8.8
C1 c2 c3 c4 c5t Aromatics
12.4 13.3 33.3 6.2 4.2 30.6
4.2 39.5 28.0 4.3 1.8 22.3
13.6 27.7 44.2 6.6 2.3 5.5
ZSM-5 c r y s t a l s i z e
(um)
aReaction conditions, H2/CO=2/1; 623(K): 2.1(MPa); W/F=lO(g-cat. bPowdery m i x t u r e o f the two components a t equal weight.
h/mol).
increased i n p r o p o r t i o n t o t h e c r y s t a l l i t e size. another
P r o b a b i l i t y o f hydrogenation o f t h e o l e f i n s escaped from z e o l i t e i s factor
t o determine t h e product p r o f i l e . The r a t e depends on p r o x i m i t y o f
a c i d s i t e s and t h e hydrogenation s i t e s , two
component
powdery
c a t a l y s t s . H y b r i d c a t a l y s t so f a r employed i n t h i s study
mixture.
the
i n o t h e r words, t h e mode o f m i x i n g
the is
I n comparison w i t h t h i s , g r a n u l a r m i x t u r e s a r e proposed
tested.
Two stage processing o f synthesis gas represents t h e extreme case
separate
use,
catalyst,
while
where
the
f i r s t reactor i s
loaded
with
methanol
t h e second one w i t h z e o l i t e t o which t h e e f f l u e n t s
a and of
synthesis from
the
f i r s t r e a c t o r are supplied w i t h o u t separation o f unconverted s y n t h e s i s gas. The performances o f t h e r e s p e c t i v e mode o f using t h e two component c a t a l y s t s are
compared i n Fig.6,
indicating t h a t the v e r s a t i l i t y o f product
profile
is
216
30
DME
Dp:O. 8mm
2-Stage
D :1.7mm
Granular
30
Dp: 0.5mm
&
Mixture
Dp :0.3mm
--n i JL l 30
-riF1 n D :0.8mm P
Powdery Mixture
1 Cata 1y s t
Product
2
3
4
5
Carbon number
6+ arom.
/ -
Product p r o f i l e s depending on t h e use o f t h e component c a t a l y s t s : Fig. 6 Reaction c o n d i t i o n s , 623K; 2.1MPa; W/F = 3.19 o f c a t . h/mol; H2/CO=2/1; twogranular mixture i s of stage,Cu-Zn(H)+T-A1203 ( f i r s t ) p l u s DAY (second): g r a n u l e s o f Cu-Zn(H) and DAY: powderly m i x t u r e i s t h e s t a n d a r d h y b r i d c a t a l y s t ,
COtH2
CH30H
C nH2nt 2
Fig.7
2CH30CH3
+ -
">
Methyl benzenes
Consecutive c o n v e r s i o n o f s y n t h e s i s gas over h y b r i d c a t a l y s t .
217 due t o t h e c o n s e c u t i v e r e a c t i o n scheme d e s c r i b e d i n Fig.7 ( r e f . 20). For
t h e increased p r o d u c t i o n o f a r o m a t i c s from s y n t h e s i s gas,
two stage r e a c t i o n system has been studied. hybrid permits
catalyst, deep
an
c o m p r i s i n g methanol s y n t h e s i s c a t a l y s t
c o n v e r s i o n o f s y n t h e s i s gas i n t o methanol
and and
alumina,
which
dimethyl
ether.
These a r e converted i n t o a r o m a t i c s i n t h e second r e a c t o r o v e r ZSM-5.
CO
conversion
hydrocarbon
of
91% and
advanced
The f i r s t r e a c t o r i s loaded w i t h a
hydrocarbon y i e l d
of
57% were
Thus
attained.
i s analyzed t o a p r o f i l e v e r y s i m i l a r t o t h a t o f t h e
MTG
the The
process
( r e f . 21).
REFERENCES 1. M.E. Dry, " C a t a l y s i s - Science and Technology", (ed by J.R. Anderson and M. Boudart,), Springer-Verlag, B e r l i n , Vol. 1, 1981, p.159. 2. H.H. N i j s and P.A. Jacobs, J. Catal. 65, (1980) 328-334. 3. D. B a l l i v e t - T k a t c h e n k o and I. Tkatchenko, J. Mol. Catal., 13, (1981) 1-10. 4. D.L. King, J. Catal., 51 (1978) 386-397. 5. 0. Okuda. T. Tatsumi, K. Fujimoto, and H. Tominaga, Chem. L e t t . , (1983) 1153-1 156 6. T. Tatsumi. Y.G. Shul, T. Sugiura, and H. Tominaga, Appl. Catal., 21, (1986) 119-131 7. Y.G. Shul, T. Sugiura, T. Tatsumi, and H. Tominaga, Appl. Catal.. 24, (1986) 131-143. 8. T. Tatsumi, Y.G. Shul, Y. A r a i . and H. Tominaga, "Proc. o f 7 t h I n t e r n l . Z e o l i t e Conference." ed. b y Y. Murakami, A. Iijima and J.W. Ward, Kodansha- E l s e v i e r , Tokyo (1986), pp.891-898. 9. Y.G. Shul, Y. A r a i , T. Tatsumi, and H. Tominaga, B u l l . Chem. SOC. Jpn., 60, (1987) 2335-2341 10. Y.W. Chen, H.Y. Wang and J.G. Goodwin, Jr., J. Catal., 83, (1983) 415-427. 11. Y.W. Chen, H.T. Wang and J.G. Goodwin, Jr., J. Catal., 85. (1984) 499-508. 12. A. Corma and B.W. Wojciechowski, Catal. Rev.-Sci. Eng., 27. (1985) 29-150. 13. C.D. Chang, W.H. Lang and A.J. S i l v e s t r i , J. Catal.. 56 (19790 268-273. 14. K. Fujimoto, Y. Kudo and H. Tominaga, Nippon Kagaku K a i s h i , (1982) 206-212. 15. H. Saima. K. F u j i m o t o and H. Tominaga, Chern. Lett., (1984) 1777-1780. 16. K. Fujimoto. Y. Kudo and H. Tominaga, J. Catal., 87 (1984) 136-143. 17. H. Saima, K. F u j i m o t o and H. Tominaga, B u l l . Chem. SOC. Jpn., 58 (1985) 795-802. 18. K. Fujimoto, H. Saima and H. Tominaga, B u l l . Chem. SOC. Jpn.. 58 (1985) 3059-3060. 19. K. Fujimoto, H. Saima and H. Tominaga, J. Catal., 94 (1985) 16-23. 20. K. Fujimoto, H. Saima and H. Tominaga, Ind. Eng. Chem. Res. 27 (1988) 920926. Fujimoto, K. Asami, H. Saima, T. Shikada and H. Tominaga, Ind. Eng. 21. K. Chem. Prod. Res. Dev., 25 (1986) 262-267.
.
.
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
219
129XeNMR OF ADSORBED XENON FOR THE DETERMIN'ATIOP OF 'OID SPACES Q. CHEN, M.A. SPRINGUEL-HUETAND J. FRAISSARD
Laboratoire de Chimie des Surfaces, AssociC au CNRS, URA 1428, UniversitC P. et M. Curie, 4, Place Jussieu, 75252 Paris, Cedex 05 (France) SUMMARY The 129Xe-NMR of adsorbed xenon is a technique generally used at room temperature for determining some zeolite properties, specially void spaces (dimensions of channels or cages, structural defects...). This study indicates more exactly what experimental conditions should be used, depending on the nature of the sample and the property studied; namely: influence of temperature, of Si/Al ratio of zeolite and more generally of chemical composition of molecular sieves. Applications are related to complex structures and intergrowth of structures. INTRODUCI'ION Many properties or defects of zeolites or metal-zeolites are difficult to detect by classical physico-chemical techniques. For example: short-range intergrowth of structure, short-range crystallinity, size of supported metal particles too small to be detected by electron microscopy .... That is why, in 1978, Fraissard et a1 (ref. 1) started to search for a new molecular probe in order to elucidate some zeolite properties which are normally difficult to handle. The central idea of this project was to find a molecule, non-reactive, particularly sensitive to its environment, to collisions with other chemical species and to the nature of adsorption sites, which could be used as a probe for determining certain zeolite properties in a new way. In addition, this probe should be detectable by NMR since this technique is particularly suitable for investigating electron perturbations in rapidly moving molecules. Xenon is an ideal probe because it is an inert gas, monoatomic. with a large spherical electron cloud. From the NMR point of view, the 129Xe isotope has a spin of one-half. Its natural abundance in xenon is 26% and its sensitivity of detection relative to the proton is 10-2. The high polarizability of t h e xenon atom makes it very sensitive to its environment. Small variations in the physical interactions with this latter cause marked
220
perturbations of the electron cloud which are transmitted directly to the xenon nucleus and greatly affect the NMR chemical shift.
CHEMICAL SHIFT OF XENON ADSORBED IN A ZEOLITE Fraissard et al. (ref. 2) have shown that the chemical shift of adsorbed xenon is the sum of several terms corresponding to the various perturbations it suffers:
6 = 6ref-k 6s -k 6xe + & A S
-k
68 -k 8M
(1)
Gref represents the chemical shift of gaseous xenon at zero pressure, taken as the reference; therefore S r e F 0 . 6s corresponds to the chemical shift extrapolated to zero pressure. For - a solid with no electrical charges, it depends on the mean free path, 1 , of
xenon, imposed by the zeolite structure, which depends on the dimensions of the cages or channels and on the ease of xenon diffusion (ref. 3). 6 x e corresponds to the chemical shift due to Xe-Xe interactions. This term becomes predominant at high xenon pressure. The main information have been generally obtained from the analysis of the variation of the chemical shift against the xenon concentration at 26°C. The amount of xenon adsorbed is expressed as the number, T I , of atoms per gram of anhydrous zeolite or the number of atoms, ns, per cage (zeolites Y, A, erionite etc.). When the Xe-Xe collisions are isotropically distributed (large spherical cage) the relationship 6 = f ( n ) is a straight line (see Fig. 1-1). The slope, dS/dn is proportional to the local xenon density and, therefore, 17 0 inversely proportional to the "void Fig. 1. Variation of the chemical shift volume". If the Xe-Xe collisions are 6 against the xenon concentration n anisotropically distributed (narrow (see the text). channels) the slope of this function increases with n (see Fig. 1-2). When there are now in the void space strong adsorption sites (SAS) which interact with xenon much more than the cage or channel walls, each xenon
221
spends a relatively long time on these SAS, particularly at low xenon concentration (ref. 3). The corresponding chemical shift 6 will be greater than in the case of a non-charged structure (see Fig. 1-3). When n increases, 6 must decrease if there is fast exchange of the atoms adsorbed on SAS with those adsorbed on the other sites. When n is high enough the effect of Xe-Xe interactions becomes again the most important and the dependence of 6 on n is then similar to that of curve 1 in Fig. 1-1. In this case an+o which is the chemical shift extrapolated to zero concentration depends on the nature and number of these strong adsorption sites. Often these SAS are more or less charged, and sometimes paramagnetic, cations. The theoretical curve 1-3 is then displaced downfield (see Fig. 1-4). The difference expresses the effect, 8 ~ of . the electrical field and, if it exists, 8~ due to the magnetic field created by these cations (refs. 3,4). FAST SITE EXCHANGE AND MEAN FREE PATH The previous results have been mainly obtained at 26"C, it being assumed that at this temperature the experimental chemical shift is the average value of the shift of xenon in rapid exchange between a position A on the pore surface (defined by 6,) and a position in the volume V of the cavity or channel (defined by <6+) (see Fig. 2):
Fig. 2. Fast exchange model. 6a and aV are the chemical shifts of xenon adsorbed on the wall and in the free space.
where Na and NV are the number of xenon atoms in each state. This equation is valid whatever the xenon concentration, then for n = 0 6=6s. 6a depends on the void space. < 6 " > is a function of 6a and the distance travelled by the xenon atom between two successive collisions. In fact, 6, = 6a when the xenon leaves the surface. 6 , t h e n decreases during the journey between sites 1 and 2, whence the need to determine a mean value,
*>. In order to obtain by this technique precise data on the void space of a zeolite of unknown structure and on the dimensions of structural defects,
222
-
Fraissard et a1 (refs. 5,6) have calculated the mean free path, 1 , of xenon
-
imposed by the structure and determined the dependence of 6s on 1 : a 6 s = 6a a+ 1 where a is a constant. The case of the infinite cylinder and the sphere can be rigorously solved by calculation:
-
-
infinite cylinder : 1 = Dc - Dxe sphere : 1 = 1/2(Ds-Dxe) Dc, Ds and Dxe are the diameters of the cylinder, the sphere and the xenon atom, respectively.
F: Ferrierite Z-5, 48, 12: Z S M J , 48 and 12 A-5, 11, 17: A1m4-5, 1 1 and 17 S-34, 37: SAPO-34 and 37
0
b 1
2
Fig. 3. Variation of
4
6
e A i
-
against the mean free path 1 .
With a more complicated structure we must feed the computer the real structure of the zeolite considered. Then we circulate a xenon atom within this space, assuming that the velocity vector is random and all directions are
equally probable. The mean free path,
-
223
1 , is obtained by averaging the
distances between successive collisions. The result thus obtained for various model zeolites is shown in Fig. 3-1. This curve has enabled us to determine the characteristics of the void volumes of unknown structures or structural defects. But certain points have to be considered more closely. For example, is 26OC a high enough temperature for the rapid exchange model to be used? Up to what point is the effect of the A1 concentration negligible? And, more generally, what is the effect of the chemical composition of the solids, in particular of AlPO4, SAPO, etc? It is obvious that if one increases 6 a or the probability of state A one must in theory increase the experimental 6 value. But, as we shall see later, the magnitude of this variation depends also on , which depends not only on 6 a but also on 1 , compared to 6a. Therefore, it is necessary to
-
study the influence of the different parameters aforementioned. INFLUEiNCT OFTEMPERATURE The chemical shift variations with xenon concentration 6 = f(n) at different temperatures are illustrated in Figs. 4 and 5 for NaY and ZSM-5 samples, respectively, pretreated at 4OOOC and 10-4 torr. Figure 4 can be divided into two regions, I and 11, corresponding to 6 the xenon concentration range of E a at Ocn,cl and 1O"C, 6 varies linearly with n in both regions. The slope of the 6 = f(n) variation decreases with increasing temperature. When T -80; concentration. In the case of ZSM-5 (see Fig. 5), (0) -60; Q -40; (A) -20; (0) 0; (0) 27; 50; @) 80;e) 100. between -100 and 27°C. the
224
Fig. 5. 6=f(n) variations for ZSM-5 zeolite at temperatures PC:(0) -100
chemical shift at very low xenon concentration is almost independent of temperature. At high xenon loading, the chemical shift decreases slightly with temperature (maximum 2 ppm). Above 27"C, the chemical shift decreases with increasing temperature for a given xenon concentration. At low loadings the slope of the 6 = f ( n ) variation seems to increase very slightly with temperature.
The temperature dependence of 6 , ( T ) is illustrated in Fig. 6. The difference in 6 s between -100 and +lOO"C is about 38 ppm for NaY and 6 ppm for ZSM-5.
I
6s E 0. Q
-
90
70
50 1
T,,,,,,,,,,,,,, -100
0
100 TVC
Fig. 6. 6 , variations against temperature T for samples:
(0)
Y and (o)ZSM-5.
Before we discuss these results it is important to know from what minimum temperature the rapid exchange model is valid.
225
Equation (2) can be transformed by the same formulation as depicted in ref. 6 to:
where Dxe is the xenon diameter, P a and p v are the xenon density in the adsorbed state and in the free space, respectively. A and V are the surface area and the volume of the free space, respectively. K (K=pa/pV) is the reduced probability ratio of xenon residences on the surface (A) and in the free space (V). It depends upon the zeolite structure and the temperature. C (C=V/ADxe) is a constant for a given system. To see whether the fast site exchange model is valid, and if it is valid, then in what temperature range, we rearrange equation (3) to:
Furt hermore : [ dLn(-
6a-6.s )/dT) = - dLnK --d T &<$> 0
h
A
m
Y
“0, 2
co:
l
l
,
l
,
,
AE --%*
l
,
l
,
-\ -
-
O\O
I \
-
-0.5 \O
-
c;l “0, e
\ \ Q -1.0 -
\
-
\O
I
l
l
l
l
l
l
l
,
between the xenon in the two states A and V. It is a constant for a given system. Therefore, if the model is valid, then the Ln [ (6a-6s)/(6s-<6v>)]=f( I I T ) plot should be a straight line. To get the experimental Ln [ (6a-6s)/(6s-<6v>)]=f(1I T ) plot we need to know 6aand . As the 6a value measured at -100°C and the value reported by
226
obtain the experimental plot illustrated in Fig. 7. One can see that at high temperature (T>30°C) the plot is a straight line which means the model is valid (ref. 8). Let us consider then - the rapid exchange model and, firstly, the dependence of on 1 . When the xenon leaves the surface 6v=6a. Then, if
-
-
1 is small is hardly less than tia. On the other hand, when 1 is large the
xenon atom has time to lose the memory of the perturbation, caused by adsorption, before the next collision with the surface. can therefore be very small, even zero, given that the electronic relaxation is very rapid. When the temperature T falls, the residence time of adsorbed xenon increases and, consequently, the probability na/na+nv of this state. The variation of 6 then depends on . It will be very small if <6v>-6a (ZSM-5) or, conversely, very large if < & V > < < & a (Nay). This study of the temperature dependence is a way of checking the size of the pores in which the xenon is adsorbed.
EFFECT OF THE Si/Al RATIO OF ZEOLITES That the effect of the internal electrical field on the chemical shift of adsorbed xenon is negligible at 26OC has only been checked for faujasite either with Na+ cation or decationized. In this case the 6=f(n) relationships
Fig. 8. 6=f(n) (full lines) and line width Av=f(n) (dotted lines) variations for ZSM-I1 with Si/Al: (0) 30; (A) 40;(0) 80; (a) 160; (+) OQ.
221
are almost parallel straight lines and, for constant n , 6 decreases by 4 ppm when Si/A1 increases from 1.28 to 54. But what happens with zeolites which have narrow pores? To answer this question we have studied ZSM-5 and ZSM-11 zeolites with various Si/Al ratios. The 6=f(n) relationships for the ZSM-11 samples are of the characteristic shape observed for these zeolites: continuous increase of 6 with n with a marked change of slope for n=8x1020 Xe atoms/g (see Fig. 8). The chemical shift extrapolated to zero concentration, as, increases with [All. But this variation shows a break at about [A1]=2 Al/u.c. (see Fig. 9). The results for ZSM-5 are similar to those for ZSM-11. It is observed first of all that the dependence of as on [All is greater for narrow channels (ZSM-11 and ZSM-5) than for large cavities (Y).This difference is again due to . In the first case c6,>-6,; consequently, any increase in 6, with Xe-wall interaction leads at the same time to that of the experimental 6. In the second case, the influence of the variation of 6 a on 6 is minimized by the small value of . An important and unexpected feature of this study of ZSM-5 and ZSM-11 is the demonstration 6’ PPA that the Xe-wall interactions I I change for a definite value of the 115 115 I aluminium concentration [A1]=2 I-/ Al/u.c., as indicated by the break 110 Hot in the 6,=f[A1] plot, in agreement I I with several analogous II observations reported (refs. /i/ I II O ZSM-5 1 p P0.AO 1 00 55L, h2 Y I A ZSM-11 9,10,11). This observation allows I us to state that A1 is not I I I I 1 randomly distributed in the lattice but that its distribution Allur. depends on its concentration. Fig. 9. Variation of against [All for The break in (also in the signal (A) ZSM-11 and (0) ZSM-5. width and the amount of xenon adsorbed) for [A1]=2 Al/u.c. could correspond to the presence or absence of an A1 atom at the channel intersections. We would mention finally the difference between ZSM-5 and ZSM-11 for [A1122 A1h.c..
t
228
INFLUENCE OF THE CHEMICAL COMPOSITION: ZEOLITES, AlPO4, SAP0 AND MAP0 We have studied the adsorption of xenon on many zeolites and aluminophosphate based molecular sieves whose names and characteristics are listed in Table 1. TABLE 1 Chemical shift tis of 129Xe adsorbed on molecular sieves and characteristics of the void spaces
--------~----Molecular sieve
6 s PPm
Faujasite Y
60
SAPO-37 A, ZK-4
58 87 90
R
73
Erioni te
98
AlPO4- 17 ZSM-5
72 110
ZSM-12
90
Theta- 1
130
VPI-5
49
AlPO4-5 SAPO-5 MAPO-5
56
A1P04-1 SAPO-11
I
120
Characteristics of the void space Sphere, diameter 13A with four prism openings at 109"; 12-ring: 8A The same structure as Faujasite Sphere, diameter 11.4A; six 8-ring openings 4-5A depends on the cation Unidimensional barrel-shaped channels; 12-ring openings of 7.1A; maximum diameter, 9A Unidimensional channels, regular cylinders, 12-rin diameter: 7.4A Cage, 15.1 x 6.3 ; six 8-ring openings 3.6 x 5.2A The same structure as Erionite Tridimensional interconnecting channels; 10-rings: 5.1 x 5 . 5 4 and 5.4 x 5.6A Unidimensional channels.12-ring diameter: 5.5 x 5.9A Unidimensional channels, 12-ring diameter: 4.4 x 5.5A Unidimensional channels, regular cylinders, 18-ring diameter: 12- 13A
i
Unidimensional channels, regular cylinders, 12-ring diameter: 7.3A Unidimensional channels, 10-ring diameter: 3.9 x 6.3A
-
It is sometimes difficult to calculate 1 for certain structures and the values are probably approximate (ZSM-5, ZSM-11 -and EU-1, etc.). Nevertheless, we must remark that the experimental [tis,1 ] points are either
229
on the initial curve defined by Fraissard et a1 (ref. 5,6) (see Fig. 3-I), on a curve parallel to the first one (Fig. 3-11) or, much more rarely, between the two. More exactly, all the points corresponding to solids-with low or zero cation concentration are located below curve I, for 0 . k 1 <3A. For a given structure, when the xenon diffusion is very easy, the experimental points are on the first curve, whatever the composition of the solid (zeolite Y and SAPO-37). The points for AlP04-5, SAPO-5 and MAPO-5 are identical, whereas those for erionite and AlP04-17 (same structure) are different. Comparison of the 6 = f(n) curves of these two solids shows that xenon diffusion appears to be easier in AlP04-17. In fact, the 6 increase, which is linear in the first part of the curve, expresses the Xe-Xe interactions. For erionite, 6 increases very little in this range. The passage from one cage to another is easier in AlP04-17, perhaps because there are no cations, which cause steric hindrance in erionite, or because of a more symmetrical opening which requires less deformation of the xenon atom. Moreover, the xenon adsorption rate is much greater for AlPO4-17. If the same value of the mean free path is assumed for AlP04-17 as for
-i
erionite, the point of the coordinates [ 6 , , 1 ] is very much below the first
-
6,=f( 1 ) curve (Fig. 3-1). The value of the mean free path in the two solids are not, of course, strictly identical; the Si-0 and P - 0 bond lengths are not equal and the T-O-T angles must vary slightly from one compound to another, which may explain the slight difference in the xenon diffusion. Although we cannot explain in detail why there are two curves in Fig. 3, they can be used to study zeolites and more generally all the porous solids regardless of their composition. APPLICATION TO THE STUDY OF COMPLEX STRUCTURES AND INTERGROWTH OF STRUClVREiS VPI-5 The characteristic signal a shows the expected linear dependence 6 = f(PXe) (6,=49*2 ppm). But for all the samples studied there is also a second, narrow, signal h. characteristic of reproducible defects which could correspond either to the beginning of the transformation into AlP04-8 or, more likely, to the existence of 16-membered T-rings (see Figs. 10,ll). Signal c in Fig. 10 corresponds to AlPO4-11 impurity. PolvtvDe Y We have studied two samples, denoted polytype Y-1 and polytype Y-2. The polytype Y-1 sample shows no detectable defects. For polytype Y-2, the
230
200
100
PPm
0
Fig. 10. NMR spectra of xenon adsorbed on VPI-5 treated at 573K.
Fig. 11. Intercalation of a 16-membered T-ring ribbon (*) in the framework proposed for VPI-5. presence of two signals at temperatures of 26°C or higher indicates that there are two differentiated zones. The higher value of 6s for the second signal corresponds to slower xenon diffusion and/or smaller internal volume. At low temperature (T
231
crystallized free zone, a region where the pores are blocked or limited which makes xenon diffusion more difficult. ZSM-8 The spectra mostly consist of two signals corresponding to structures ZSM-5 and ZSM-11. Table 2 summarizes the spectral characteristics and the conclusions.
TABLE 2 Characteristics of the 129Xe-NMR on ZSM-8 and conclusions I
_
-
~
Si/AI
40 80 160 >1950
-
----
-
Spectral characteristics one two two two
symmetrical large signal signals for Pxe>1200 torr signals for PXe>SOO torr signals for PXe>80 torr
Conclusion intergrowth intergrowth intergrowth intergrowth
at at at at
short distance long distance long distance long distance
CONCLUSION These complementary studies indicate more exactly what experimental conditions should be used for the 129Xe-NMR technique, depending on the nature of t h e sample and the property studied. This technique is increasingly able to provide original results difficult to obtain by other methods.
1
2 3 4 5
T. Ito and J. Fraissard, NMR study of the interaction between xenon and zeolites A, X and Y, in: L.V.C. Rees (Ed.), Proc. 5th Int. Conf. on Zeolites, Naples, Italy, June 2-6, 1980, Heyden, London, 1980, pp. 510-515. J. Fraissard and T. Ito, 129Xe n.m.r. study of adsorbed xenon: A new method for studying zeolites and metal-zeolites, Zeolites, 8 (1988) 35036 1 , (and references therein). T. Ito and J. Fraissard, 129x8 nuclear magnetic resonance study of xenon adsorbed on zeolite NaY exchanged with alkali-metal and alkaline-earth cations, J. Chem. SOC.,Faraday Trans. 1, 83 (1987) 451-462. A. Gedeon, J.L. Bonardet, T. Ito and J. Fraissard, Application of 129Xe NMR to the study of Ni2+Y zeolites, J. Phys. Chem.. 93 (1989) 2563-2569. J. Demarquay and J. Fraissard, 129Xe NMR of xenon adsorbed on zeolites. Relationship between the chemical shift and the void space, Chem. Phys. Letters, 136 (1987) 314-318.
232
M.A. Springuel-Huet, J. Demarquay, T. Ito and J. Fraissard, 129Xe NMR of xenon adsorbed on zeolites: determination of the dimensions of the void space from the chemical shift 8(129Xe). Studies in surface science and catalysis, 37 (1988) 183-190. 7 T.T.P. Cheung, C.M. Fu and S. Wharry, 129Xe NMR of xenon adsorbed in Y zeolites at 144K. J. Phys. Chem. 92 (1988) 5170-5180. 8 Q. Chen and J. Fraissard, in preparation. 9 G. Debras, A. Gourgue, J.B. Nagy and G. De Clippeleir, Physico-chemical characterization of pentasil type materials. 111. High power solid state 27AI. 23Na and 29Si n.m.r. of precursors and calcined samples, Zeolites, 6 (1986) 161-168. 1 0 A. Auroux, P.C. Gravelle and J.C. Vedrine, Microcalorimetric study of the acidity of H-ZSM-5 zeolite, in: L.V.C. Rees (Ed.), Proc. 5th Int. Conf. on Zeolites, Naples, Italy, June 2-6, 1980, Heyden, London, 1980, pp. 433439. 1 1 C. Naccache and Y. Ben Taarit, Recent developments in catalysis by zeolites, in: L.V.C. Rees (Ed.), Proc. 5th Int. Conf. on Zeolites, Naples, Italy, June 2-6, 1980, Heyden, London, 1980, pp. 592-606. 6
233
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
DIFFUSION OF HYDROCARBONS IN A AND X ZEOLITES AND SILICALITE DOUGLAS M. RUTHVEN, MLADEN EIC AND ZHIGE XU Department of Chemical Engineering, University of New Brunswick, Fredericton, NB Canada ABSTRACT Diffusion of a range of different hydrocarbons in A and X zeolites and silicalite has been studied by gravimetric, chromatographic and zero length column (ZLC) methods. The ZLC diffusivities for CHq and C2H6 in 5A and 13X at low temperatures are only slightly smaller than the corresponding NMR self diffusivities, extrapolated to zero concentration. Data for the Cg aromatics in silicalite show satisfactory conformity between gravimetric and ZLC measurements but the diffusivity values are higher than most previously reported values, with only about one order of magnitude difference between the isomers. The diffusional activation energies for p-xylene, ethylbenzene and o-xylene are almost identical. First results of a detailed study of diffusion of large aromatic molecules (dimethyl naphthalenes and tri-isopropyl benzene) in NaX zeolite are also reported. INTRODUCTION Because of the practical importance of zeolites as catalysts and selective adsorbents, coupled with their academic interest as examples of geometrically regular micropore systems, the study of zeolitic diffusion has attracted much attention from both experimentalists ad theoreticians.
A wide variety of
experimental techniques have been applied, including both 'microscopic' and 'macroscopic' methods.
The microscopic methods (principally NMR and neutron
scattering) yield direct information concerning the molecular mobility, and hence the self-diffusivity, whereas the macroscopic measurements generally yield the transport diffusivity (the quotient of the flux and the negative concentration gradient) although, by the use of isotopically labelled species the macroscopic methods can be adapted to measure self-diffusion. Since self-diffusion and diffusive transport are physically different phenomena, their coefficients are in general different. However, simple physical reasoning, as well as most detailed quantitative theories, suggest that the self-diffusivity and the transport diffusivity should converge in the low concentration limit.
This has been
confirmed experimentally for several systems although there are also some well documented discrepancies.( 1 * 2 ) A simple classification of the commonly used macroscopic experimental methods, according to the quantity measured and the nature of the measurement, is given in Table 1. The methods of category A3 account for the large majority of the reported data.
They have the advantage of experimental simplicity but
they are subject to the intrusion of extracrystalline resistances to heat and mass transfer,
so
straightforward.
the analysis and interpretation of the rate data is not always
234
Our own recent experimental studies, by chromatographic and gravimetric methods, have been directed to the study of three classes of zeolite system for which there is either little available data (large hydrocarbons in zeolite X ) or for which there are large discrepancies in the reported diffusivity values (light alkanes in A and X zeolites, aromatics in silicalite).
A summary of these
results is presented here. Table 1 Experimental Methods for Measurinn Transport Diffusion in Microporous Solids A.
A1
Transient Measurements
Following the evolution of the concentration profile in response to a change in external concentration. - Optical techniques - X-ray techniques - NMR spin mapping ESR spin mapping
-
A2
Flux measurements - Time lag measurements for diffusion through a membrane - Transient Wicke-Kallenbach
A3
Uptake rate measurements. (Following the variation of the mean sorbate concentration averaged over a particle). - Batch methods (gravimetric, volumetric, piezometric, frequency response) - Flow methods (chromatography, ZLC etc.)
B. Steady State Measurements B1
Steady state flux through a membrane Permeation - Wicke-Kallenbach
B2
Effectiveness factor in catalytic reaction.
-
EXPERIMENTAL METHODS Gravimetric Gravimetric uptake rate measurements were performed in a Cahn vacuum microbalance systems, using small zeolite samples (10-12 mg) with incremental pressure steps (to avoid the problems associated with the analysis of integral uptake rate data). To maximize heat transfer and minimize external mass transfer resistances, the zeolite sample was spread as thinly as possible over the balance pan. Where possible, measurements were repeated with different crystal sizes and different adsorbent sample configurations in order to confirm the dominance of intracrystalline resistance. In the experiments reported here the volume of the system was sufficiently large relative to the quantity of sorbent that the sorbate pressure, following the initial step, remained essentially constant. The simple expressions for the uptake curve for an infinite isothermal system were
235
therefore used to interpret the uptake curves(3): m
t = I - -6 -
m
I n2 n=l
Spherical Particles: (radius R)
2 2 2 -1 e -n II Dt/R n2
m m -t-- 1 - -8 I mm n2 n=O (2n+1)
-exp[-(Zn+l)
Parallel-sided Slab: (half-thickness 1 )
(1)
2 2 2 n Dt/4?. 1
(2)
ZLC In the zero length column (ZLC) meth~d(~-~) a small sample of zeolite crystals (- 2 mg) held between two sinter discs, is equilibrated with sorbate and then purged with helium (or argon) at a high flow rate. The hydrocarbon content of the effluent stream is monitored continuously using a chromatography detector (thermal conductivity or flame ionization).
The conditions of the experiment are
adjusted to ensure that the desorption rate is controlled by diffusion out of the crystals, rather than by the purge rate.
This may be achieved by using a
sufficiently high purge flow. The absence of significant external resistance may be confirmed by replicate measurements with He and Ar as the purge gases and the dominance of intracrystalline diffusion may be confirmed by varying the crystal size. The mathematical model used to interpret the ZLC desorption curves assumes a linear isothermal system with perfect mixing through the thin bed and adsorption equilibrium maintained at the crystal surface. For such a system the desorption curve (for a set of uniform spherical particles) is given by:
- pn2Dt/R2
m
- -- 2 L 0
I n=l [ p
(3) n
t~(L-1)1
where pn are the roots of the equation:
pn cot pn t L - 1 EVR L = 3(1-~)KDz
and
For large values of L, pn
+
-3
= 1
= 0
2 Purge Flowrate . . . Crystal Volume KD
(4) (5)
nv and Eq. 3 reduces, in the long time region to:
The diffusional time constant is readily found from the slope of a plot of log (c/co) vs t. If external film resistance is significant the analysis remains the same except that the expression for L (Eq. 5 ) is replaced by:
236
It is evident that external film resistance may modify the intercept of the semilog plot of c/co vs t but it has no effect on the limiting slope. For a parallel sided slab of half-thickness Q, the expressions corresponding to Eqs. 3-6 are: m
-= 0
where
2L
-a 2Dt/Q2 n
n=l z e [an2+L(L+1)1
are the roots of the equation: 2
a t a n a = L = EVQ ( 1-E)KZD n n For large L, an
(9)
(2ntl)n/2 and Eq. 8 reduces, in the long time region, to: 2 2 -r- 2 -n Dt/4Q +
co
(Ltl) e
(10)
DIFFUSION OF AROMATIC HYDROCARBONS IN SILICALITE Representative uptake curves showing the difference in kinetic behaviour between o-xylene, p-xylene and ethylbenzene in large crystals of silicalite (40x50~105pm) under comparable conditions, are shown in figure 1.
The curves
all conform well to the simple isothermal diffusion model for a parallel sided slab (Eq. Z), rather than for a sphere (Eq. 11, since the intercepts approximate 8/n2
(rather than 6/n2).
This suggests that
diffusion probably
occurs
predominantly through the straight channels of the silicalite pore system (which run between the large parallel faces of the crystals).
This conclusion has been
confirmed by measurements with different crystal sizes which show the time constant varies with the square of the crystal thickness, rather than with the mean equivalent spherical radius. The diffusivities 'for ethylbenzene are almost independent of concentration in the low concentration region but the diffusivities for p-xylene show a stronger trend (increasing with concentration) presumably reflecting the more favourable form of the equilibrium isotherm. Only limited uptake measurements were performed with o-xylene since, as a result of the low diffusivity, these experiments are tedious and time consuming. Diffusion of the same three sorbates (as well as benzene for which the data have been previously reported was also studied by the ZLC method. Representative response curves are shown in figure 2.
The coincidence between the curves
measured, under similar conditions, with He and Ar vs purge gases confirms the absence of any significant extracrystalline resistance to mass transfer. In view of the evidence from the uptake measurements the parallel sided slab model (Eq. 8) was used to derive the time constants which are summarized in Table 2.
1.0
8
E
'- 0.1 E
-
8
1
0.01
0
Figure 1
\ PX,
p=0;0.12
To;r
p=0-0.103 Torr
,
,
,
50
60
~/12.~~10-3
10
20
30 LO Time (rnins)
Gravimetric uptake curves at 125OC ( %
70 =
20 vm).
1.9 0-
,cc.-c.
I
I
I
I
2.3
2.5
2.7
2.9
3.1
1 0 ~ 1 ~
0
I.L
1
2.1
Figure 3
Temperature dependence of limiting diffusivities for benzene (x), p-xylene ( 0 ) . ethylbenzene ( ) and 0-xylene ( A ) . Filled symbols, gravimetric,open symbols ZLC.
ta w -a
238
Table 2 Summary of ZLC Diffusivity Data for Aromatic Hydrocarbons in Silicalite Crystals Sorbate
Temp. (Deg. C)
Benzene
50 75 100 100 150
p - Xylene
Ethylbenzene
o-Xylene
Pu e Flow (cm'gSTP/min)
4 4 4 2 4
L
DX$O9-,) (cm .s
227 105 91 46 87
1.0 1.8 3.1 2.7 11.0 2.8 2.9 4.1 11.9 11.8 43
80 80 100 150 150 (Ar) 200
60 100 60 60 60 60
11 23 37 55.6 40.0 51.3
70 70 100 150 150 (Ar) 200
60 100 60 60 60
19.5 25 28.5 66.7 55.5 91.0
0.8 0.92 2.0 7.3 5.6 29
30 30 15 15
323 95 476 571
0.36 1.24 2.9 7.84
100 150 200 250
60
E W/mol)
27.0
30
30
33
Purge was He except where indicated as Ar. In ref.(7) the ZLC data for benzene were interpreted according to the spherical particle model. These values have been recalculated based on the parallel sided slab model for consistency with the xylene data. Figure 3 shows a comparison between the ZLC data and the gravimetric values (extrapolated to zero concentration). satisfactory.
It is evident that the agreement is
There is little difference in diffusivity between benzene and
p-xylene and the diffusional activation energies are all similar (- 30 kJ/mol). In contrast with the results of earlier studies, the present data show no evidence of any significant deviation from the simple diffusion model and the differences in diffusivity between the C8 isomers are much smaller than has been previously rep~rted.(~*~)The similarity in the activation energies implies that the difference in diffusivity cannot be attributed to differences in repulsive interaction energies resulting from diameters.
the
difference in
critical molecular
239 DIFFUSION OF METHANE AND ETHANE IN ZEOLITE A AND X The ZLC method provides one of the more useful techniques for studying rapid diffusion processes since, by careful attention to detail it is possible to reduce the dead volume of the experimental system to a fraction of a millilitue and hence to reduce the response time essentially to that of the detector.
By
the addition of a refrigeration system the operating temperature range has been extended to -1OO"C, thus making it possible to measure diffusion of the light alkanes (CH4 and C2H6) in 5A and 13X zeolites. The results of such measurements are summarized in figures 4 and 5 and Table 3 where comparative NMR selfdiffusivity data(1°-12), extrapolated to zero concentration, are also shown. For both 5A and 13X the measurements performed with two different crystal sizes show good agreement, thus providing convincing evidence of intracrystalline diffusion control. In contrast to earlier ZLC data for the higher paraffins in NaX(6)
which
show large discrepancies with the NMR self-diffusivities, the agreement of the present data for CHq and C2Hg in both 13X and 5A can be considered satisfactory. A discrepancy of this magnitude is easily accounted for by the natural tendency of an NMR self-diffusion measurement to err on the high side (as a result of relaxation effects) and for a sorption rate measurement to err on the low side because of the emphasis on the tail of the desorption curve. The activation energies also show qualitative agreement and the surprising observation that the activation energy for C2H6 in 13X is higher than in 5A is confirmed by both NMR and ZLC data. However, the c2%-13X
system at the higher temperatures is close
to the limit of the ZLC technique (half time less than 1 sec.) so these data and consequently the activation energy for this system are subject to a large margin of error. Table 3 Comparison of NMR and ZLC Diffusivity Data for CH4 and C2Hg in 5A and 13X Zeolites ZLC Data Do at 200K system CH4-5A C2H6-5A C2H6-13x
( cm2.s-1)
10-6 31c10-~ 10-5
(kJ/mole)
(cm2. s-1)
E
(kJ/mole)
3.7
4~10-~
6.3
5.9
1.4~10-~
6.3
9.6
2.5 ~ 1 0 - ~
(14.6)
ZLC data are at zero concentration; NMR data extrapolated to zero from data at higher concentrations.
240
:.
I .
; * i
1 .I;
5A
( 55 pm)
0 13X
( 1 0 0 pm)
A 13x
( 50 pm)
0
'.4- '..*
1 I ~
1
ClCO
.01
,001
0
10
t
Figure 4
(I)
ZLC desorption curves for C2H6 in 5A and 13X crystals at 183K.
4
Figure 5
30
20
5
6
4 1 0 ~ 1 (~ K1)
5
6
Temperature dependence of limiting diffusivities (ZLC) and NMR self-diffusivities for CH4 and C 2 h j in 5A and 13X crystals.
241
TRANSITION STATE THEORY Because of its simple cubic structure 5A zeolite provides an ideal model system for the application of transition state theory. This was recognized many years ago(13*14)
but the available diffusivity data used
in the earlier
comparisons were derived from measurements with small commercial 5A crystals. More recent evidence suggests that a substantial fraction of the windows in these samples were probably blocked thus invalidating the quantitative comparisons between theory and experiment. It is therefore appropriate to reconsider such a comparison on the basis of the more reliable diffusivity data now available. According to transition state theory, in the low concentration limit where self- and transport diffusivities become identical, the diffusivity is given by:
*
where uz and uz represent the potential energies of a molecule within the cage (equilibrium state) and within the window (transition state). To avoid the difficulties associated with the evaluation of the partition function for a molecule within the cage (f,)
one may make use of the equilibrium relationship:
-AHo/kT fZ (ug-uz)/kT =K = K e kT'Fe B
(12)
where ug is the potential energy for the vapour phase (zero), thus obtaining: .b
or comparing with the Arrhenius form (D = , D e-E/RT):
*
D = -6 2 m
where Vo
hKo
. -fZ f' g
.
'
*
- = u - u - A H o
R
vO
E -
k
k
is simply the difference in potential between the transition and
equilibrium states.
fi,
the partition function per unit volume of the gaseous
species may be expressed as the product of the internal, rotational and translational contributions:
fi
so,
(15) = fCrans * frot * fint if it is assumed that the molecule in the transition state retains the same
internal and rotational freedom as in the equilibrium state and there is no significant additional freedom of motion in the plane of the window:
* fZ =
frot
*
fint
* . -fZ = 7 1 ;
.
f,
g
trans
D, =
ti2
(16)
;ran,
Table 4
N
Absolute Rate Theory: Coqmrism of Predicted a d IIc.8ured Oiffusivities i n
System
Crystals
6
K x10
(molecule/ cage.Torr)
mmmtaic
tqs'
(a.s
)
2 -1
(an .s
Method
1.06
14
4.4
1.2
3.7
24.2
grav.
3.2 rn ( L i d )
1.26
17.7
1.2
0.12
10
33.8
grw.
Xe-5A
20-50 pn
1.29
22.5
1.0
2.8
0.36
(1.0)
WR PFG
cn -4A
36 Im, 7.3 In
1.5
18
12
1
12
24.2
grw.
CH4-5A
20-50
c
1.9
10.8
10
45
0.22
3.8
mR
(15) Derrah
PFG
CF4-5A
3.6 p (Lindc)
0.38
24.7
3.7
2.5
1.5
3.8
grav.
CF -5A
27.5 p, 55 p
1.49
21-7
0.93
19
0.05
27.6
grav.
0 -4A
3.2 p (Lindc)
1.23
13.4
5.2
6.6
0.8
18.8
grsv.
N -4A
36 P 7.3 m
1.o
17.6
7.0
5.0
1.6
24.2
grw.
55 P, 27.5
0.35
21 .o
41
42
1.o
10.0
UC
36 m, 7.3 m
0.019
47.5
211
9.0
23
grw.
1.39
24.7
5.0
4.8
34.3
grav.
Y'ucel
5.9
WR PFG
(11) Caro
6.7
ZLC
xu
grav.
Yucel
2 W -5A 2 u) -4A
2
c n
-u n m.
2 6
7.3 P
23.4 1.oc
17 1.7 10 20-50 ~n 20.0 0.4 2 6 7.3 17 20.8 2.3 atamic C H -5A 27.5. 55 P 0.4 2 6 3.0 1.o 3.1 nC H -5A 27.5 pin, 55 42.6 0.86 4 10 3.0 3.0 42.6 1.o nC tl -5A 34 p, 7.3 P 0.86 4 10 *The nargin of u u e n a i n t y in the experirrntal a c t i v a t i m energies i s probably about ZkJ/nole. This corrcopmS t o a factor of 2-3 i n the valucs dcrived for 0., Ibpblished experimental data obtained at U.N.B. Experimental 0, values are frm taperatwe depenance of 0 or D o woo' ply-
Ref.
)
3.2 p (Linde)
2
linear triatanic
2 -1
Kr-4A
4
diataic
(kJ/nole)
Em.*
6 ODX1O
oDmlo
Ar-4A
4
'Spherical
Theorg
Ano
b
4A and 5A Zeolites
C H -5A
19.2 18.8
(17)
4.
ZLC
Eic
(19)
(6)
where fkrans = (21nnlcT/h~)~/~e is the translational partition function (per unit volume) for the free gas phase. It is in principle possible to estimate the potential energies of molecules in the cage and windows from theoretical potential calculations. Such estimates are, however, very sensitive to the precise values assumed for the molecular radii since, particularly for the molecule in the window, the attractive and repulsive forces are of similar magnitude.
In comparing experimental data with
the theory it is therefore more satisfactory to consider only the pre-exponential factors. Such a comparison is shown, for several simple sorbates, in Table 4 . In view of the obvious approximations in the theory and the uncertainty in the experimental values of D,,
the agreement between theory and experiment is
remarkable. The ratio (D,)theory/(Dm)expt is, for most sorbates, close to unity. Values less than unity can be rationalized by restricted rotation of the transition state and values greater than unity can be explained by
some
contribution from degrees of freedom in the plane of the window.
Evidently, for most sorbates, these two effects must either be small or they must compensate. The close agreement for polyatomic molecules such as n-butane seems at first sight surprising since one might anticipate significant restriction of rotation. However, the butane molecule is probably not in the linear conformation and even though it probably does not have complete rotational freedom in the window, the additional degrees of freedom (torsional oscillations about the centre of mass) may very well compensate fully for restriction of the three dimensional rotation. The most significant discrepancy is for C02 for which the predicted value of D, is substantial greater than the measured value.
Since the free rotation of a
long linear molecule such as CO2 is bound to be severely reduced in the transition state this deviation is clearly understandable. In the original application of transition state theory to diffusion in 5A zeolite, attempts were made to estimate the effect of restricted rotation and the contribution from degrees of freedom in the plane of the zeolite window. The broader perspective which is now possible as a result of the improved data base suggests that such detailed analysis was probably not justified. The simplified treatment leading to Eq. 16 appears to be quite adequate for most small molecules while Eq. 14 may provide a useful basis for more refined calculations. DIFFUSION OF LARGE AROMATIC MOLECULES IN ZEOLITE X Results of an experimental study of diffusion of large molecules in 13X zeolite are summarized in figures 6 and 7.
The data of tri-isopropyl benzene at
200°C show exactly the same pattern of behaviour as has been commonly observed for smaller molecules in 5A and for the c8 aromatics in 13X.
The strong
244
-
1,3,5 Tri Isopropyl Benzene NaX
N
Y
10
0
30
20
60
50
40
Concent rot ion (mg/g) Concentration dependence of diffusivity (D) and corrected diffusivity (Do) for tri-IPB in 13X crystals at 473K.
Figure 6
10-12
1.5
'
1
1.6
'
1
1.7
'
Gravirnetric Data,
I
1.8
I
' ' ' 1.9
2.0
' ' '
2.1
2.2
1 0 ~ 1 (K-') ~ Figure 7
Temperature dependence of corrected (limiting) diffusivities for dimethyl naphthalene isomers in 13X zeolite.
245 concentration dependence appears to be due entirely to non-linearity of the equilibrium isotherm and corrected diffusivities, calculated according to:
D = Do (dllnp/dllnc) are essentially constant.
(17)
The data obtained by gravimetric and chromatographic methods (for the dimethyl naphthalenes) are quite consistent. Despite the differences in critical molecular diameter (Table 5 ) there is evidently little difference in either diffusivity or activation energy between the DMN isomers. These data thus support the conclusion drawn previously from the similarity in the diffusivities (in NaX) between the xylene isomers, that the diffusional activation energy is determined by the kinetic diameter of the freely rotating molecule rather than by the critical diameter (the diameter of the smallest circumscribing cylinder). Table 5 Critical Molecular Diameters of Some Aromatics Molecules Diameter A 6.9 Benzene 6.9 Toluene 6.9 p-Xylene 7.3 o-Xylene 7.3 Naphthalene 7.3 2-6,2-7 DMN 7.9 1-2,l-5,1-6 DMN 8.6 1-3 DMN 9.1 1-4 DMN 8.4 Mesitylene 9.3 Tri-isopropylbenzene Notation c/co ratio of outlet concentration to initial concentration in ZLC experiment diffusivity (intracrystalline) (cm2.s-1)
D
or corrected diffusivity (Eq. 17) (cm2.s-l) (cm2. s-1)
Do Dm
limiting value of D (as c + molecule diffusivity
D
self diffusivity (intracrystalline)
E
activation energy
fz,f2
0 )
(cm2.s-1) (J/mole.deg)
partition functions for adsorbed molecule in equilibrium and transition states
frot,fint rotational and internal contributions to partition function ( cm-3 partition function per unit volume for free gas fi (erg. sec) h Planck's constant (erg/deg.) k Boltzmann constant pre-exponential factor
K, KO Henry's Law constant II
half-thickness of crystal
L
defined by Eqs. 5 , 7 or 9
m
mass of a molecule
(moleucles/cage.dyne/cm2) (cm)
246
mt/m, fractional approach to equilibrium p sorbate partial pressure R
equivalent radius of crystal
T
absolute temperature
uz-ug difference in potential between transition state and free gas
Vo
defined by Eq. 14
z
thickness of adsorbent bed (ZLC)
Sh
Sherwood Number
%,Bn
roots of Eqs. 9 and 4
E
voidage of adsorbent bed (ZLC)
6
lattice parameter heat of adsorption at zero coverage
-AHo
Acknowledgement:
(cm) (erg/molecule)
We are grateful to Dr. David Hayhurst for providing us with
samples of large silicalite crystals. Financial support was provided, in part, by the Imperial Oil Company of Canada. References 1. 2.
J. Karger and D.M. Ruthven, Zeolites 2, 267 (1989). J. Karger and D.M. Ruthven, Diffusion in Zeolites, John Wiley, New York
3.
J. Crank, Mathematics of Diffusion, Oxford University Press, London (1956). M. Eic and D.M. Ruthven, Zeolites 8, 40 (1988). M. Eic, M. Goddard and D.M. Ruthven, Zeolites 8, 327 (1988). M. Eic and D.M. Ruthven, Zeolites g, 472 (1988). M. Eic and D.M. Ruthven, 8th Internat. Zeolite. Conf., Amsterdam, July 1989, Proceedings, p. 897, P.A. Jacobs and R.A. van Santen eds., Elsevier, Amsterdam (1989). P. Wu, A. Debebe and Y.H. Ma, Zeolites 2, 118 (1983). K. Beschmann. G.T. Kokotailo and L. Riekert, Chem. Eng. Process 22, 223
(1991). 4.
5. 6. 7. 8. 9. 10. 11.
12. 13.
14. 15. 16. 17. 18. 19.
(1987). J. Caro, J. Karger, H. Pfeifer and R. Schollner, Z. Phys. Chem. Leipzig 256, 698 (1975). J. Caro, J. Karger, G. Finger, H. Pfeifer and R. Schollner, Z. Phys. Chem. Leipzig 257, 903 (1976). J. Karger, H. Pfeifer, M. Rauscher and A. Walter, J. Chem. SOC. Faraday Trans I, 76, 717 (1980). D.M. Ruthven and R.I. Derrah, J. Chem. SOC. Faraday Trans I. 8,2332 (1972). J. Karger, H. Pfeifer and R. Haberlandt, J. Chem. SOC. Faraday Trans I, 76, 1569 (1980). D.M. Ruthven and R.I. Derrah, J. Chem. SOC. Faraday Trans I, 2. 2031 (1975). J. Karger, H. Pfeifer, F. Stallmach and H. Spindler, Zeolites - in press. H. Yucel and D.M. Ruthven, J. Chem. SOC. Faraday Trans I, 76, 60 (1980). H. Yucel and D.M. Ruthven, J. Chem. SOC. Faraday Trans I, 76, 71 (1980). H. Yucel and D.M. Ruthven, J. Colloid Interface Sci. 2,186 (1980).
241
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
ATLAS OF ZEOLITE STRUCTURE TYPES: PAST- PRESENT- FUTURE
W.M. MElER Institute of Crystallography, ETH Zurich, 8092 Zurich, Switzerland
ABSTRACT The 'Atlas of Zeolite Structure Types' is a periodically updated data compilation of observed zeolite-type frameworks. Apart from being useful as a work of reference it reflects the progress in zeolite structural chemistry. This paper describes some uses of Atlas data including framework densities, pore diameter values, loop configurations and fault planes. In addition, some recurrent misconceptions are discussed.
INTRODUCTION For over two decades the 'Atlas of Zeolite Structure Types' (refs. 1,2) and its forerunner (ref. 3) have reflected the enormous progress made in the elucidation of the framework structures of zeolites and zeolite-like materials. Fig. 1 shows that the current growth rate of over 6 new structure types per year is quite unprecedented. What looks like an exponential growth is due to two developments: (a) the recent advances in the methods of crystal structure analysis, in particular powder diffraction methods (refs. 4,5), and (b) the greatly expanded range of compositions of zeolite-type molecular sieves explored in more recent years (refs. 6-8).
I
"I 53
Fig. 1.
61
Development of the number of known zeolite structure types over the years.
248
The available structural information on zeolite-type materials is very substantial indeed. It raises the need for the Atlas and the other data compilations published on behalf of the 1ZA Structure Commission (refs. 9,lO). The principal aims of the Atlas project have been: (a) to provide a reference work (both for crystallographers and non-crystallographers) based on data which has been subjected to critical evaluation (normally published), (b) to promote some desirable uniformity in recording such data, (c) to fill in gaps in available data, and (d) to facilitate studies and analyses of data. The following account describes some examples of uses of the Atlas which were not dealt with previously (ref. 11).
SILICATE AND PHOSPHATE MICROPOROUS MATERIALS From a chemical point of view zeolite-type silicates and phosphates apparently constitute two distinctive categories of microporous materials. This appears to be particularly evident in synthesis. Table I (which is based on the isotypes listed in the Atlas) shows, however, that there are three, rather than two distinct groups of framework types. Apart from those associated with silicates and phosphates there is a sizeable group of structure types which have been found to occur both in silicates and phosphates. Comparative in-depth investigations of isotypes in this latter group in particular offers new insight into synthesis/structure relationships. An example is the structure refinement of SSZ-24, the silica analogue of AIP04-5 (ref. 12). Table I also shows that while the present structure-orientedactivity is greatest in the field of phosphates, the silicates still make up the majority of zeolite structure types. The structure types which have been approved by the IZA Structure Commission (SC) since the last Atlas appeared in 1988 have been marked by a plus (+) in the Table. In addition to these, seven structure types (listed in Table I) are under consideration for clearance by the SC at present. A current listing of the zeolite structure type codes can be found in the Appendix.
249
TABLE 1
Microporous zeolite-type materials excluding interrupted frameworks
Silicates
Both silicates and phosphates more abundant as a silicate
ABW AFG BIK BOG + BRE CAS + DAC DDR DOH EAB ED1 EPI EUO FER GME GOO
MON + HEU MOR KFI JBW + MTN LAU M U LIO MTW NAT LOS LOV NON LTL OFF LTN PAU MAZ PHI ME1 + SGT MEL STI MEP THO MER TON MFI YUG MFS + (a) (b)
(a) Beta (b) ZSM-20 (c) AIP04-8 (d) MAPO-39
ANA CAN CHA ERI FAU
Phosphates
first established as a phosphate AFI AST BPH +
GIS LEV LTA RHO SOD
AEL AFS A N + AFY APC APD AllATV + Aww + VFI + (c) (d) );: (g1
(e) AIP04-18 (f) AlP04-31 (9) AIP04-41
+
not included in ref. 1
POROSITY AND FRAMEWORK DENSITY
The pore volume of zeolite-type frameworks can be expressed in terms of the framework density (FD), which is defined as the number of tetrahedral atoms per nm3 and is listed in the Atlas for each structure type. The distribution of FD-values as a function of the smallest ring(s) in the network has been shown to be of considerable interest (refs. 13,14). Fig. 2 is an updated version of the earlier diagram prepared in
250
1988. The gap defining the boundary between microporous and dense networks has gained further support by the new data, and the postulated range for zeolite-type materials has been confirmed. A particularly noteworthy addition is the data point relating to ZSM-18 (MEI), the second framework containing 3-rings, which was reported only recently (ref. 15). Earlier, only lovdarite, a beryllosilicate, was known to contain 3-rings. The new structure type ME1 is very significant for it demonstrates that silica can also form tetrahedral nets containing 3-rings (without Be being needed).
T-atoms / nm3
size of smallest rings
Fig. 2.
Distribution of framework densities (FD) as a function of the smallest ring(s) in the network. Updated version of diagram in ref. 13. Data points refer to dense (+), previous (*) and new (0)zeolite-type structures, and low density tetrahedral networks (O).
251
This bodes well for the synthesis of low-density materials based on 3-ring structures in the silica field. The FD is obviously related to the pore volume but does not reflect the size of the pore openings. Customary values are frequently given or assumed for the free diameters of channels with 8-, 10- and 12-ring apertures. It should be clear from Fig. 3 (which is also based on Atlas data) that the crystallographic free diameters for the various types of apertures fall into a wide range. There is even a considerable overlap of the respective ranges for 8-, 10- and 12-ring openings. Hence, sorption properties have to be interpreted with caution when predicting the type of ring openings in structurally unknown materials.
...._ I...... li
18-ring aperture
........ i........ l i
14-ring
#
12-ring
1O-ring 9-ring
[
I 15
J
........... ........... 1 2 :
8-ring
Fig. 3.
Range of crystallographicfree diameters for different ring openings. The arrows indicate the customary values assigned to 8-, 10- and 12-ring apertures. The number of pertinent structure types is given in each bar.
252
The customary values for 8-, 10- and 12-rings (applying to LTA-, MFI- and FAU-type materials) are marked by arrows in Fig. 3. This Figure also includes relevant data for VPI-5 with 18-ring channels (refs. 16-18) and for AIP04-8 with 14-ring channels (ref. 19). These are the most open framework structures to date, and both have unidimensional channel systems.
SOME MISCONCEPTIONS Among the most widespread misconception is the belief that large unit cell constants are indicative of large apertures. Fig. 4 should make it clear that there is no such correlation, no matter whether minimum or maximum unit cell constants are considered.
LTN-
LTN-
-
PAU-
-
AFS-
FAU-
Fig. 4.
Minimum (a) and maximum (b) unit cell constants compared to largest ring openings.
-
253
A more serious illusion is the belief that close resemblance between powder diffraction patterns is sufficient to assign a structure type. This may hold in some special cases. In general, however, the assignment or the establishing of a structure type based on the similarity of XRD powder patterns is full of pitfalls. This has become evident in a number of instances. For example, PHI and GIS were confused for well over a decade, and later EAB and ERI were repeatedly mixed up. In one instance, even the Structure Commission assigned a code to a structure type which turned out to be in error (ref. 20). The code ATF of the erroneous structure type of AIP04-25 had to be discredited, and according to the rules of the SC is not to be used again. The correct structure has been assigned the code ATV. DLS is certainly an indispensable tool for simulating and evaluating possible framework structures but a reasonable agreement between an experimental XRD pattern and one based on DLS atom coordinates does not give enough evidence for the correctness of a framework topology. Even in cases in which this qualitative approach led to basically correct framework topologies subsequent refinements have invariably revealed significant differences. A good example of this kind is ZSM-12 (refs. 21,22). In the originally proposed structure the channel dimensions were 5.5 x 5.9 A but structure refinement later on showed these to be 5.6 x 7.7 A.
SIGNIFICANCE OF LOOP CONFIGURATIONS AND FAULT PLANES The utility and significance of loop configurations (LC) has been demonstrated before (ref. 11). In a study on the evaluation of hypothetical zeolite frameworks, Brunner (ref. 23) determined the relative abundance of various possible LC. One of the more frequently occurring LC consists of a pair of edge-sharing 4-Yktgs. This is shown in Fig. 5 together with two pertinent conformations. While the 'cis' conformation (i) had been observed in 22 known zeolite frameworks, the 'trans' conformation (ii) had not been recorded but was noted to occur in the proposed framework for VPI-5. Brunner's findings cast some doubt on the correctness of the VPI-5 framework structure at an early stage. Meanwhile, it has been found by extensive structure refinements and NMR studies that the VPI-5 framework is not purely tetrahedral (ref. 18). Instead, it is 4-connected (containing some octahedral At) and it even appears that a strictly tetrahedral network of this type is unlikely to be obtainable.
254
Fig. 5.
A common loop configuration and associated conformation (from Brunner, ref. 23).
For a considerable proportion of the frameworks covered in the Atlas, fault planes have been listed. This allows predictions at a glance as to the likelihood of stacking faults, twinning, and disorder. These possible defects are of considerable importance in the characertistics of these materials and their catalytic applications.
OUTLOOK The number of established zeolite-type framework configurations will soon approach one hundred. Compilations, which are extensive enough and which contain carefully examined data, can be expected to be of increasing interest to both theoreticians and practitioners in the field. The need for computer-based data files can be clearly foreseen. The available printed versions of the compilations issued by the SC are already based on computer files to a large extent . In the future it is to be expected that the huge amount of structural information, once prepared for efficient use in the form of well-documented data bases combined with good search routines, can be used as a source for new insights in zeolite chemistty. The first step in this direction has already been made (ref. 24). One important goal is the ability to predict which of the enormous number of hypothetical networks can be synthesized and in what system(s).
ACKNOWLEDGEMENTS The author thanks Dr. Lynne B. McCusker and other members of his group for helpful discussions. Continued support by the Swiss National Science Foundation (grant 20-25256.88 at present) is also gratefully acknowledged.
255
Appendix List of zeolite structure type codes approved by the IZA Structure Commission (on behalf of IUPAC) up to June 1990 Structure types not included in the 1988 Atlas are marked by a plus (+)
ABW AEL AFG AFI AFS AFT AFY ANA APC APD AST ATT ATV AWW BIK BOG BPH BRE CAN CAS CHA -CHI DAC DDR DOH EAB ED1 EPI ERI EUO FAU FER GIS GME GOO HEU JBW
+
+ + +
+ +
+
Li-A(BW) AIP04-11 Afghanite AIP04-5 MAPSO-46 AIP04-52 COAPO-50 Analcime AIPO4-C AIPO4-D AIP04-16 AIPO4-12-TAMU AIP04-25 (revised structure) AIP04-22 Bikitaite Boggsite Beryllophosphate-H Brewsterite Cancrinite Cs-Aluminosilicate (Araki) Chabazite Chiavennite Dachiardite Deca-Dodecasil 3R Dodecasil 1H TMA-E(AB) Edingtonite Epistilbite Erionite EU-1 Faujasite Ferrierite Gismondine Gmelinite Goosecreekite Heulandite Na-J(BW)
KFI LAU LEV LIO LOS LOV LTA LTL LTN MAZ ME1 MEL MEP MER MFI MFS MON MOR MTN M-n
+
+ +
MMI NAT NON OFF - PAR PAU PHI RHO - ROG SGT SOD STI THO TON VFI -WEN YUG
+
ZK-5 Laumo ntite Levyne Liottite Losod Lovdarite Linde Type A Linde Type L Linde Type N Mazzite ZSM-18 ZSM-11 Melanophlogite Merlinoite ZSM-5 ZSM-57 Montesommaite Mordenite ZSM-39 ZSM-23 ZSM-12 Natrolite Nonasil Offretite Partheite Paulingite Phillipsite Rho Roggianite Sigma-2 Sodalite Stilbite Thomsonite Theta-1 VPI-5 Wenkite Yugawaralite
256
REFERENCES 1 2
3 4 5 6 7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
W.M. Meier and D.H. Olson, Atlas of Zeolite Structure Types (IZA Structure Commission, 1978). W.M. Meier and D.H. Olson, Atlas of Zeolite Structure Types, Second extended edition, Buttetworths, 1988. W.M. Meier and D.H. Olson, Adv. Chem. Series, 101 (1970) 155-168. Ch. Baerlocher, Zeolites, 6 (1986) 325-333. L.B. McCusker, Acta Cryst. (in preparation). S.T. Wilson, B.M. Lok, C.A. Messina, T.R. Cannan and E.M. Flanigen, J. Am. Chem. SOC.,104 (1982) 1146-1147. B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan and E.M. Flanigen, J. Am. Chem. SOC.,106 (1984) 6092-6093. E.M. Flanigen, B.M. Lok, R.L. Patton and S.T. Wilson, Proc. 7th Int. Zeolite Conf., Tokyo (Elsevier, 1986) 103-112. W.J. Mortier, Compilation of Extra-Framework Sites in Zeolites. Butterworths (1982), pp. 67. R. von Ballmoos and J.B. Higgins, Collection of Simulated XRD Powder Patterns for Zeolites. Zeolite Special Issue 10 (1990) 313s-514s. W.M. Meier, Studies in Surface Science and Catalysis, 49 (1989) 691-699. R. Bialek, W.M. Meier, M. Davis and M.J. Annen, Zeolites (submitted for publication). G.O. Brunner and W.M. Meier, Nature, 337 (1989) 146-147. M.E. Davis, Nature, 337 (1989) 117. S.L. Lawton and W.J. Rohrbaugh, Science, 247 (1990) 1319-1322. C.E. Crowder, J.M. Garces and M.E. Davis, Advances in X-ray Analysis, 32 (1988) 507-514. J.W. Richardson, Jr., J.V. Smith and J.J. Pluth, J. Phys. Chem., 93 (1989) 8212-8219. L.B. McCusker, Ch. Baerlocher, E. Jahn and M. Buelow, Zeolites (submitted for publication). R.M. Dessau, J.L. Schlenker and J.B. Higgins, Zeolites, 10 (1990) 522-524. J.W. Richardson, Jr., J.V. Smith and J.J. Pluth, J. Phys. Chem., 94 (1990) in press. R.B. LaPierre, A.C. Rohrman, J.L. Schlenker, J.D. Wood, M.K. Rubin and W.J. Rohrbaugh, Zeolites, 5 (1985) 346-348. C.A. Fyfe, H. Gies, G.T. Kokotailo, B. Marlerand D.E. Cox, J. Phys. Chem. 94 (1990) 3718-3721. G.O. Brunner, Zeolites, 10 (1990) 612-614. J.M. Newsam and M.M. Treacy, Zeofile, in preparation.
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
251
ZEOLITES AS MEMBRANES : THE ROLE OF THE GAS-CRYSTAL INTERFACE
Richard M. Barrer Chemistry Department, Imperial College, London, SW7 2AZ, United Kingdom
ABSTRACT Steady flow through a porous cystal membrane is associated with concentration distributions across the gas/crystal interfaces which cannot represent true equilibrium. Steady state distributions have been modelled from mass balance equations for a membrane where a fraction of the flow enters the crystal directly from the gas phase and the remainder enters via an externally adsorbed layer. The same fractions are assumed for the nett flow through the exit surface. Special cases of the general equations investigated numerically or otherwise include : the extent of departures from equilibrium across interfaces; the ideal flow when these departures tend to zero; the ratios of ideal to measured flows; and ratios of true intracrystalline diffusivities to apparent diffusivities obtained from steady flows by assuming equilibrium across interfaces. In fast sorptions and desorptions non-equilibrium distributions across the gas/solid interface may also result in differences between true intracrystalline diffusivities and those obtained by incorrectly assuming equilibrium. INTRODUCTION Zeolite membranes provide in principle a means of continuous scparation based on differences in the shapes and sizes of molecules. If they contain catalytic centres they can also be envisaged as membrane reactors. In steady flow, independently of any model, more molecules of diffusant enter the crystal than leave at the ingoing surface. T h a c imbalances require that concentrations just within the crystal cannot have their equilibrium values (ref. 1) at either surface, so that it is of interest to examine their scale. The imbalances have recently been modelled quantitatively for gases obeying Langmuir's isotherm when all molecules enter the crystal via an externally adsorbed layer (ref. 2). In this paper the model will be extended to include, for very wide windows, possible entry directly from the gas phase simultaneously with entry from the externally adsorbed layer. Also considered is the effect of imbalance betwecn forward and reverse flows across the gas/crystal interface in transient state sorption and desorption kinetics. Aspects of surface resistances have already formed the bases of other papers (refs. 3-5). The treatment for a single crystal applies also to membrana consisting of a monolayer of porous crystals on a macroporous support, in which gaps between porous crystals in the monolayer are filled by a non-permeable medium which adheres permanently to the sides of the crystals, but leaves their top and bottom surfaces free of the adhesive. In another kind of membrane the porous crystals comprise a discontinuous phase embedded in a continuum. Such membranes will not be considered here.
258
(a) CRYSTAL
GAS
el x=
%
0
DISTANCE
-
GAS
%-I
9n x=l
(b) GAS
CRYSTAL
GAS
AE
x=o
DISTANCE
-
x =I
Fig. I . (a) The energydistanre curve for entry to the rrystal from an externally adsorlml layer. ( b ) The energv4istance curve alien entry to tlic crystal takes plare directly froin t l w gas phase. C C denote constant Concentrations in the gas phase at x = 0 and x = Ircspertively. go’ gl Oo , @,denote fractions of external sites of type A occupied at x = 0 and x = 1 respectively. O1. On denote fractions of internal sites occupied a t x = 0 and x = I respcctively.
.
E, E, denote activation energies for entry from external sites and for intracrystalline diffusioii rcsprtively. AEs , AE are energies of dcsorption from sites of type A and from intrarrystallinr sites respcctivrly.
259
MODEL Sorption is assumed to obey Langmuir's isotherm for which diffusivity is independent of concentration (ref. 6). A single crystal membrane, bounded by planes x = 0 and x = 1 separates a constant gas phase concentration C at x = 1, with C > C Inside the crystal activatcd
d
go
gr
diffusion occurs involving energy harriers of height El. Molecules of diffusant may enter or leave the crystals via an externally adsorbed layer, for which the cncrgy of desorption is A E ~ . Fig. l a shows the energy - distancc configuration, the energy levels and the nomenclaturc involvcd. If the windows are widc cnough some moleculcs may pass directly from the gas phase to intracrystalline sites, for which the energy - distance curves and nomenclature are shown in Fig. lb. Of the nett flow J through unit area of surface the fraction F2 J occurs as in Fig. l a and a fraction F1 J as in Fig. lb, wlicre (F1
+ F2) = 1.
A t and around channel mouths at x = 0 there are sites of type A from which the molecules comprising F2 J enter the crystal and possibly sites of type I3 from which they do not. There is however constant exchange of molecules between the two kinds of site and between each kind ol site and the gas phase. In steady flow there is nett transport f2 J from the gas to the B-typc sites and hcncc from B- to A-type sitcs. There is also nctt transport of fl J from the gas phase to sites of type A and therefore (fl
+ f2) J = F2 J from these sitcs to the interior of the crystal.
In addition there is nett transport F1 J from the gas phase directly into the crystal. These mechanisms are shown schematically in Fig. 2. Exit at x =I Sit,esesp
Entrv at x = 0 Gas Phase
F1
7
/Sites A<-Sitf2es J
B
li2J
Sites 1
Fig. 2. Mechanisms of cntry to and exit from the porous crystal mcmbrane.
STEADY FLOW WITIIIN A POROUS CRYSTAL MEMBRANE The flow J will always refer to that through unit cross-scction normal to x. It was then shown that J is given by (ref.2): sk J=(O1-On) =CXP (- E1/RT) (01-On) (11 - 1) (n - 1 )
260
whcre s is the number of energy barriers of hcight El pcr unit arca in each of (n-1) planes normal to x and k = kl exp (-El/RT) is the rate constant for jumping from onc energy well over the barrier El to the next well. (n - 1) per channel denotes the number of barriers bctwcen x = 0 and x = 1 and n is the corresponding number of cncrgy wells (i.e. sorption sites). O1 and On are the fractional occupancies of sites 1 just within thc crystal at x = 0 and of sites n just within the crystal at x = 1. They are not equilibrium but steady state values. To find them the processes of Fig. 2 at and across the gas/solid interface must be considcrcd.
‘I’HE MASS BALANCE EQUATIONS For direct entry from the gas to the crystal at x = 0 F1 J = sk C ( 1 - 01)- skb exp (-AEIRT) O1 a go (2) where AE is the energy of desorption and sk,. and skb exp (-AE/RT) are rate constants pcr unit area for entry to sites 1 and dcsorption from sites 1, respcctivcly. For cntry to the crystal from sites of type A F2J = skcexp (-Es /RT) 0, ( 1 -el)-skdexp[-(AE-AEs+ Es)/RT] Ol(l-Oo) (3) where Es and (AE-AEs
+ Es) are tlic respcctive activation energies for entry from type-A
to
sites 1 and for the conversc of this process (Fig. la). skccxp (-Es/RT) and skdexp [-(AE, -AEs+ Es)/RT] are the corresponding rate coefficients. 0, is the fractional occupation of sitcs of type A. In addition, for the extcrnally adsorbed molecules we have, in terms of Fig. 2, fl J = ya Cgo (1 - Oo) - yb exp (-AEs /RT)0, (for type A sites) (4 1
f2 J = 7’a Cgo (1 - 0:) - 7; exp (-AES /RT) 0; (for type B sites)
(5)
where ya, 7; are respcctive rate constants pcr unit area for adsorption on type A and type B sites, and yb (exp (-AEs /RT) and 7; exp (-AEi /RT) are the corresponding desorption rate constants.
0: is the fractional occupation of sitcs of typc B from which tlic energy of
desorption is AEg. The sum of equations 2 and 3 tlicn gives the flow J through unit area of interface which is also given by equation 1. This sum can Ix rearranged to give O1 as O1 =
kaC o+ kcexp (-Es/RT)Oo- J/s (6) kaCgifkbexp(-AE/Rr kcexp (-E SIRq 0, k [,exp I-(AE-AE s + E s ) ~ K 1 ’ ~ ( * ~ o ) In the same way at x = lone finds on=
+
+
261
kaC I+ kccxp ( - E s / R T ) O I + J/s (7) kaCgl+kbexp(-AE/RT) kcexp (-E ,IRT)0 l+kdexp[-(AE -AEs+Es)litrl'l( 1 4 3 l )
+
where @/isthe fractional occupation of A-typc sites at x = 1. A t cquilibrium I< = ( k a / k b ) e x p ( A E / R T ) = O 1eq / C g o ( l - O l C q ) = 0 ncq /C g l ( 1 - 0 neq )
(8)
o w /C go ( 1 - 0 o w ) = 0kq/C gl (1-0kq ) KSi = (kc /kd) exp [(AE-AEs) / RT] = 0 (1 - 0 ) / 0 (1 -0 )
(9)
=(ra/rb)exp(AEs/RT)=O
leq mq mq leq =Onec,(l-O ) / 0 (1-0 ) h b neq
Froiii tlicse equations one also lias I<SI. = K/I< S
(10) (11)
and (12)
kc /kd = ka ?b /kb^l, With the aid of equation 9 cquation 4 can bc transformcd (ref. 2) to givc at x = 0
(13) cxprcssion at x = lis
M'itlioiit going through all the sleps, equations 6 and 7 with the aid of the eqnations S to 14 can bc transformed to the cxnressions
wlicre r = cxp [(AE- AEs +Es)/ItT] A=
( k b / k d ) ~ p [ ( -AEs)/RT] E~
, and that the tliffcrcnccs nccl ) are not equal except, in the IIcnry's law rangc. Equations 15 and
Equations 15 and 16 show that O1 < OlCq and On > 0 (Ole(, - 01)and (On - 0 ncq 16 also show that the imbalanccs (Olcc, - O1) and
(Oil - 0ncq ) increasc as the flow J through
unit area increases, i.e. as the permeability increasrs. When f l J / y C in equations 13 and 14 is << 1 0,->
and o1-> Olcq. \I'hcn 0 ocrl and 0 thc tcrms involving O0 antl O1in equations 6 and 7 arc respectively replaced hy 0 kq ocq f l J/yaCg i n eqnations 15 and 16 do not appear and tliese equations become a g
262
(20)
If O1 and On in equation 1 are replaced using equations 15 antl 16 tlic result is a cubic i n J. When this substitution is made using cquations 19 and 20, on the other hand, one obtains lJie linear equation
-
(21)
Sk
('lcq - 'neq) = id Jid is the value of J when the second terms on the r.1i.s of equations 19 and 20 arc respectively
<< 0 1ecl and << OIlet1, i.e. when 0 1 ->
Ole(, antl 0 I1 ->
0 neq'
An important case arises whcn at x = 1 a near-vacuum is niaintained for wliicli f l J / y C:
may no longer he small, while at x = 0, 0, -> thcsc circumstances equation 16 bcconics J [I'/Skd + K f , /?,I 0 =0 11 neq + A (I< - K , ) fl .]/ya
+
+
a gl as assumed in deriving equation 19. I n 0 oeq (22)
1
and ......
sk [ I ' / S k d kl'(1 - Ole ) + ( l - ooeti)l '(11 -1) [ A + (I< - sk - (n-I)(Qleq -Oncq) = J id
+ -
I
+
1
1
Equation 23 is a qiiatlratic i n J. Important quantities inclutle Jitl, J i d / J antl f l .J/y C
a 6'
(23) wliicli
will lx considered in turn.
Estimation of Jid Jitl was estimated in reference 2 for assigned values of I
(n -1) = l/d (24) wliere d , the jump distance per unit, tlirfiision st,ep, wa.s t,a.ken as 5 x 10-8cm and 1 as 10-2cm, which is close to t,he values i n recent siriglc crystal memhrane studies (rcfs.7-9).
263
Tlie calculatcd values of Jid are reproduced i n Table 1 for several values of I
go
and E1/RT. C /C = 100 go .@
KCg0
('leq
50 10 5 1 0.1
-Ones)
0.6471 0.9192 0.7959 0.4901 0.0899
Jid when E1/RT is: 15 10 3.31 x lop2 4.19 x 4.01 x 2.51 x 1 0 8 0 . 1 6 lop2 ~
4.91 6.22 5.96 3.71 O.G8
7 98.6 125.0 119.7 74.7 13.7
5 728.6 923.3 844.7 551.8 101.2
Tlie Ratios JidU With k = k l exp(-E1/RT), and with I? from eqiiation 17 and h from cquatioii IS substituted into equation 21, it is seen that tlic resultant expression contains tlic ratios k l / k d atid kb/kd. Thc cncrgy wells involving thcsc coefficient have identical or vcry similar contours
near tlic minima so tliat with kl-vx we expcct kb-Fl vx ; kc1-F2 vx
(25)
because, of the molecules leaving tlic crystal, the fract,ion F1 enters the gas phasc tlircctly and the fraction F2 enters tlic externally atlsorbctl Iaycr. With (n - 1) = l/d (cqn.24) cqn.21 and 23 become rcspcct ivcly
264
TABLE 2 Particular cases of equations %G and 27 Conditions
J .I tl / J (cqn.26)
F1 = 1; Olcq, Oncq<
I 1
+ tl ( 2 - Olcq - Onctl) exp [(AE- E l ) / R l ] + 2d CXII [(AE - El)/RT]
F$, = 1; all 0's << 1
1
+
I:] = 1; 0 << 1 ncq FI = 1; Olcq, 0 < < 1 ncq
l - t 7 ( 2 - - 0 1cq ) exp [(AE - E,)/RT] 1 2tl esp [ ( A E - E l ) / R T ]
F1 = 1
(2s) (29)
r
csp (bAE/RT]
(32)
tl
+
1
+
7-
(skll\f / y a ) (I\
-
CX~,
[(AEs
I\7s) f , J / T ,
-
+
A15 - E,)/RT]
1
I
(35)
265
Several special cases arc suinmarisctl i n l’able 2, i n which 6AE = (AE - AEs
+ Es - E l ) , and
so is the difference hctwccn tlic a c t i v a t h energy for desorption from within the crystal to the cstcriially adsorbed layer, ( A E - AEs
+ Es) , (sce Fig.
la) and the activation energy for
intracrystalline diffusion, El. In all tlic expressions in Table 2 the terms ((I/[) cxp (6AE/RT) or ((I/[) cxp ( A E - E1)/RT appear. Their numcrical cvaluation shows t1ra.t thcsc terms can be
vcry large down to very small (Tahle 3). I f tlir former, J will he much less than Jid; if thr Iatt.cr, J.Id /.I ->
1. The values EAE must Iiavc, at. several temperatures, to givc thc valucs of
(d/l) cxp (EAE/RT) in the second column arc shown in the last five columns of Table 3. I t is notktl t1ia.t 2/(n - 1) = 2d/l is tlic ratio of tlic numl)cr of harriers across thc two interfaces a t x = 0 antl x = 1 to the numbcr of intracryst.allinc barriers.
6AE/RT
bAE/kcal mo1-l at T/I< = 78 90 19.5
(tl/l)csp (EAE/RT)
15
2.4’LG x lo3 I 1.635 x 10
3.009 2.3%
10
1.101 x l o 1
1.550
7
7.421
1.085
‘LO
3.576 2.GSR 1.78s 1.252
7.75 5.S1 3.S75 2.712
1O.S.5 S.14 5.12 3.SO
I n Table 4 (ref. 2) values of J.I d /.I are given, according to equation 31, with antl C
go
/
301.9
273
12.00 9.00 6.00 4.20
tl
= 5 x 10-scni
Cg l = 10. The figures crnpliasisr again that low tcinpcratrircs and vcry sinall
crgstallitcs will favow high valurs of .Iitl /.I and convcrsrly thc higlrcr the tcmpcraturc and the n
Iilrger
tlic crystal tlic iriorc iicarly will .litl /.J approac1i unity. TIIUS,for I = 10-”cin ant1 at ~ O K
.J is rnuclr less than Jitl /.I for all EAE 6AE 2 3.0 kcal mol-’ . A t 301.91i,
> 2 kcal inol-l
oil
tlre other
, and if 1 = 10-2cm the sa1iic is true for all
liantl,
.I.Id / J is vcry close or close to unity for
all l i a ~< 6.0 kcal rno1-l . Large values of ME could inc~ica.tepartial blockage of wintlows giving access to t,lie int.rrior of the crystal due to special causes (refs. 3-5).
266
Estimation of fl J/Y& Cg I n deriving equation 26 it was assumed that Oo = 0 and Q I = Okq, so tlial i n eqwitions oeq 13 and 14 f l J/X. C must be << I . This assumption will now he examined. h g 2 y C is the number of molccnlcs per second per cm that are adsorbed on sites of type A a
g
(at or around channel mouths). l'liis numhcr will bc less than hut of thc same order as the nnrnhcr of collisions per second per cni2 . This collision numhcr is C c/4 whcrc c is the average g
spcctl of a molecule, so that ya is less than but tlic same order as c/4. Likewise f l will be less hut of the same order as unity so that a reasonable approximation to f l J / y C will be 4.J/tC' . " g g The maximum value of this term is 4JiCl/CC . .JiCl can be taken from Table I , after conversion 6 -I cni-2 to niolcculcs s . C is i n molcciilcs cni-"' and the avrrage spcctl in cni s-I. Tallle 5 g gives values of c for a number of gases as well as the collision nninlxrs (molecnlcs s -'cm -2) according to the ideal gas law at one atmosphere pressure.
TABLE 4 Jitl/.J according to cqnation 31(') for different values of I
-'
10-2
90
I
3.0
4.2
6.0
0.1
l.GS3
IS4
1.50 x lo5
3.53 x lo9
1.o 5.0
1.507 1.300
137 SI.3
1.1%x lo5 4 6.59 x 10
1.067
1S.N
1.47 x 104
2.62 x 10' 1.55 x 109 S 3.46 x 10
7.83
1831
(3.07
136 1
6 1.50 x 10 G 1.12 x 10
4.00 1.G7
80.4 181
6.59 x 10' 1.47 x 105
50.0 I
0.1 1.0 5.0 50.0
90
I
301.9 10-2
301.9
2.0
I
I
I
3.53 x 10 10 2.G2 x 1010 1 5 5 x 1010 9 3.46 x 10
0.1 1.o 5.0 50.0
1.011 1.0062 1.0055 1.0011
1.209 1.154 1.092
0.1 I .o 5.0 50.0
1.11 I .ns 1 ,055 1.011
3.09 2.54 1 .92 I .21
1.021
267
Argon may be taken from Table 5 as a sample diffusant. Then in steady flow through single crystal membranes one obtains for 4Jid /C C the values in Table 6. The mcmhrancs g have been assigned a nuinher of values of El /RT and there is a range in KC In each case go' C has the value appropriate for 1 atm pressure at the temperatures T = 90K or 301.91<, and go Cgo /Cg,is 100. Other parameters uscd in estimating Jid were: 1 = 10-2 cm; (1 = 5 x 1012 5 -1 aiid s = 5 x 1013~ m - ~ as ,for Table 1. Clearly at 1 atm one is well justified i n taking 0 =0 , and, with C = lo-' o oeq gl 0 1-- 0 Pq'
cs
hlolccnle
"2
co
0'2 Ar C02
C
60
as i n Table 6 , one is also justified in taking
901;
at: 301.91;
0.971 0.GS7 0.691 0.345 0.2607 0.2609 0.2440 0.21SG 0.2os1
1.77s 1.257 1.266 0.631 0.475 0.47s 0.447 0.400 0.351
c c /4 90k
g
at: 301 .OK
x
1.950 1.400 1.409
1.051 0.7G4 0.769 0.354 0.2902 0.2904 0.2716 0.2433 0.2316
-
0.532 0.532 0.495 0.44G -
I n equation 27 , which is applicable wlicn a near varitiim is maintained on the exit side of tlie nicnil~ranc,the tcim [(I< - KS)( f l J/ya) + 11 arises. The cqiiation is simplifictl if it can he
shown that ( K -I$)
( f l lily,) is << 1. As previottsly, we assitme f1/ya-4/C
is replaced by its maximiim value, .lid .
and the flow, .J,
For the rlioscn valucs of KC the Langmuit g
eqtiilil~riumconstant, I<, was found when the pressure at the ingoing face of tlie crystal was taken as 1 atm. At this pressitre for the ideal gas C = 2.431 x lo1' rnoleculcs c 1 ~ 1 -at~ 301.91< g
arid S.155 x 10 I9' at 9OK. If one then again t a k a the average speed, C, for A r from Table 5 as the example the valrtes of 41
268
TABLE 6 4 J i d / t Cgo for Ar at 90K and 301.9K ,for membranes having the parameters givrn i n the text and with C / C = 100. go gl
50
0.6471
0.8182 5
0.7857
1
0.4901
90 0 . 5 5 5 ~ 1 0 - ~ 301.9 1.O 17x lo-' 1.509x lo-' 90 0.702x 1or9 301.9 90 30 1.9 90
1.2s6x10-9 0.674~10-~ 1.253~10-~ 0.42 1x I o-{)
301.9 0.772x.10-9 0.1
0.0899
90 0.077 1x 1 0-9 301.9 0.14 13x 10-9
Values of .lid were next calculal~ctlfor 1 atm pressure of tliffrisant at x = 0 and iicar-vaciiulr1
x = 1 (Omq ->
0). As
iii
ill,
Talde 1, the valrics of .Jicl at 90K (as molcciilcs s - ~ c I T - ~arc' ) sriclr
that 4 Jid K/c is always orders of inagnitutlc less f,lra.n 1 so that a fortiori the same' must be I r11e
of 4Jid (I<
-
Ks)/c. It therefore scciiis plol)able that the term (I< - I< ) f , l / can ~ ~be oiiiit.t.ctl
from equation 27, 34, and 35 untlcr inost, cspcriinciital contlitioas.
s
1
TABLE 7 Values of q u a n t i t i e s needed t o find 4Jid K/C f o r Ar. Ingoing pressure 1 atm.
J i n molecules s-l
K i n molecules
-'
cm3., c- i.n cm s-'.
Jid a t 90K f o r E1/RT equal t o :
4K/C a t Kcgo
%eq 90K
50 10 5 1 0.1
0.9804 0.9091 0.8333 0.5000 0.0909
301.9K
20
15
10
-
5
3 . 7 6 ~ 1 0 - ' ~6 . 1 3 ~ 1 0 - 2~ .~5 3 ~ 1 03~. ~ 7 5 ~ 1 05~. ~5 6 ~ 1 01~. 1~ 2 ~ 1 08 .~2~6 ~ 1 0 ~ ' 7 . 5 3 ~ 1 0 - ' ~ 1 . 2 3 ~ 1 0 -2~.~3 4 ~ 1 03~. ~ 4 8 ~ 1 05 ~. 2~ 1 ~ 1 01~. 0~ 4 ~ 1 07 .~6~6 ~ 1 0 ~ ' 3 . 7 6 ~ 1 0 -6. ~ ~I ~ X I O - ' ~2 . 1 5 ~ 1 0 ~3' . 1 9 ~ 1 0 ' 4~. 7 3 ~ 1 0 ' 9~ . 5 0 ~ 1 0 ' 7~. 0 2 ~ 1 0 ~ ' 7 . 5 3 ~ 1 0 - 1~ .~2 3 ~ 1 0 - 1~ .~2 9 ~ 1 01~. ~ 9 1 ~ 1 02 ~. 8~ 4 ~ 1 05 ~. 7~ 0 ~ 1 04 ~. 2~ 1 ~ 1 0 ~ ~ 7 . 5 3 ~ 1 0 - ~ ' 1 . 2 3 ~ 1 0 - 2~ .~3 4 ~ 1 03~. ~4 8 ~ 1 05 ~. 1~ 6 ~ 1 01~. 0~ 4 ~ 1 07 ~. 6~ 6 ~ 1 0 ~ ~
N W 0)
270
greatly cxcccds or approaclics app unit,y. If equation 37 is used to int,crpret measured stcatly flows in terms of D a1w where .I.Id / J
‘1’a.blc 4 thcrcfore shows some conditions untlcr which Did/D
cxceeds unity a variety of values of D
coultl result in the same zeolite according to tho aPl) choice of variablcs which dctcrniine the extent of non+quilihrium across ttic gas/crystal will converge toward interface (inter alia I, T , C antl C ). As 1 and/or T increase tlic D go gl aPP Did. Moreover, diffusivitics can vary among different samples of t,he same zrolito if thcsc samples conta.in variable amounts of part.ial int,racrystallinc blockages, as considered i n ref. 2 for lkiimensional channcl systcnis. From Jid = (skld/l) cxp (-El/R.T) (Olcq -Oncq) = Ditl CSat,(0leq - 0ncq )/[, whcrc
csat= s / ( ~is
wlicii
the npta.ke in molecules
o=
1, wc Iiavc for tlic int.racrysta\linc
tliffusivity in an itlcal crystal: I)icl = k l d 2 CXP (-E1/RT)
(39)
This expression is implicit in thc calcula.t.ion of -1.It1 i n ‘I’ahlcs 1 and 7. k l , tl a n d El will vary according to the porous cryst.al antl the diffnsant. With the prcvioris assignnicnt,s k l = vx = 9x
s-’ and d = 5 x 10
-’
cm , D~~~for (liffcrcnt. valncs of E ~ / R Tis as follows:
El/lt?.
20
I)i(l/cnl‘Ls-l
2 . 6 ~ 1 0 - ~ ’ 3 . S ~ 1 0 - ~ . 5 . 7 ~ 1 0 - ~ 1 . 1 4 ~ 1 0 - ~ S . ~ X I O - ~4 . 5 ~ 1 0 ~ ’
15
10
5
7
3.33
El at 301.91< 12.0
9.0
6.0
4.2
3.0
2.0
kcal nio1-l Thus the simple cqnation 39 yiclrls physically rc1asoiial)le valucs of I). id for t,hc ahovc clioice of k l antl (I. ‘1’ It A NS I ENT FLOW I n non-steady flow the niass halancc cquat ions at and across the gas/crgstal intcrfaccs apply exactly as tlicy do i n stcatly flow. IIowevcr, i n cqiiations 1.5 and 16 for the nori-steady state the flows through unit area a t x = 0 antl x = 1 are functions of time, antl tlicreforc so arc O1 and 01,just witliiii the crystal a t tlicsc 1)oiintlarics. T w o important cxaniplcs of non-steady
flows will tx briefly considered: (i)
The single crysthl niemhranc is i n sorpt,ion equilibrium wit,li the gas at, concent,ration and t.hc system C At t = 0 , C at x = 0 only is inst,anta.neously raised to C gf gl go’ t.licrcaftcr rclaxcs t,owartl the stcatly state of flow, wit.11 C and C licltl constknt. For go
d
greater simplicity, in place of cqiiat.ions 1.5 and 16, for the externally adsorhcd layers
271
it is assumed that 0 = 0 0 ocq and O1= OlCqas in equations 19 and 20. Tlicn J 1 ( t ) = [OleC,- O1 (t)] R1 = - Csat D ( t l O / t l ~= )~
(40)
and at x = 1 J n ( t ) = [On ( t ) - 0 neq ] R2 = -Csat D ( t l O / d ~ = ) ~1
(41)
The respective cocfficicnts, with the aid of equations 17, 18 and 25, bccomc Ill =
did ( A
+1
- OOe )
( 1 - 0 1cq)
skl {F1 -
+
+ F,
(1 - OOe ) exp [ ( A E , - Es)/RT]} (42) ( 1 - 0 1c q ) exp (AEIRT)
+
skd ( A 1 - Ole ) skl {F1 F2 ( 1 - O l e ) e x p [ ( A E s - Es)/RT]} (43) I1 (1 - 0 ( 1 - 0n e q ) exp ncq) Eqiialions 42 and 43 show thosc properties of a sorbatesorbent system which, in addition to .J(t), will determine tlie differences between 0 and 0 ( t ) just within tlie crystal. The
R =
cq
coefficients R1 or RI, arc the same for t,ransicnt, and steady flows and thcrcforc thc ratios J1 ( t ) / [Oleq
-
0 , (t)] or J n (1)
/
[On ( t ) - OnCq] arc constant for all t. If transient
solutions of Fick's second equation arc required they must satisfy tlic boundary condition O1 = O1 ( t ) a.nt1 0n = 0n (1) for a.ll t > 0 (44) ratlicr than solutions bascd upon the assumption O1 = Olcq and OI1 = 0 for all t > 0 (15) ncq (ii) The sccontl example is that of sorption and tlcsorption kinctics, in which flow decays from an initial niaximum va.luc to zero at eqiiilil~riiim. At time t = 0 for sorption the concentration at x = 0 a.nd x = 1 is instantaneously raised from C to C and kcpt gl go constant at that value. Tlic system tliercafter relaxes from its prcvious sorption equilibrium to t,hc new equilibrium. For a cryst,al with a I-dimensional channel system parallel with the xdircctioii J
( t ) = 0 = - J ( t ) = / = [@cq - 0 ( t ) ] R = -Csat D((l@/dx)x = 0
For tlcsorption, when at t = 0, C
go
(46)
is instantancously lowered to C the corresponding
.d
expression is -J (1,)
= 0 = J (t,)
= 1= (0 ( t ) - 0
C(l
] 0 = - C sa.t D ( t l B / d ~ ) ~ =
(47)
I n tlirsc expressions a positive sign is given to J (t) whcre flow is in the direction of x incrca.sing. The symmct,ry of tlic situation at each intcrfacc means that R1 = RI1 = R in
equations 42 and 43. Otherwise comments made for transicnt, flow in tlic membrane apply equally to solption~llesorptionkinetics. Especially if J ( t ) is initially large the time dependence of 0 ( t ) a n d its difference from 0 may be such that solut,ions of Fick's ccl second equation ba.scd on 0 = 0 for all t > 0 at x = 0 and x = 1 (equation 45) arc no cq
212
longer suitable for evaluating tliffusivities. In many past attempts to reconcile tliffusivitics obta.inetl from sorption kinetics with the large diffusivity values mcasurahlc by pulsed field gradient N M R , the sorption kinetics method has involved extremely small half-lives for the sorption and initially large flows .](I,). Initial tirnc tlependcncc of 0 (I,) is tlien masimiscd as arc initial differences between 0 and 0 (t). Time dependence of 0 just within the intcrfa.ccs i n such circiimstanccs m y cq therefore have to he atldctl to the othcr known limitations to obtaining meaningfill tliffusivities from very fast sorpt,ion rates. For the pnrpose of finding diffusivitics the sorpt,ion kinetics method should best he confined to slow or very slow uptakes wlicn nearly all the limitations encountered with very fast uptakes are avoided. t,lie relation between 0 (t,) just wit.liin tlic crystal and tlic While, for consta.nt C go ' t.iine is not known, it coiild for fast upthtkes he a st,rong fnnction of time. One empirical function of tinic for 0 ( t ) just, within s = 0 or x = 1 is @ ( t ) = O (l-cxp(-/3t))fort>O
(48)
crl
when the cryst,al is init,ially free of sorhate antl at t = 0 is exposcd t,o a const,ant external concentra.tion of the sorhate. For p = IX and all t > 0 equation 4s givcs 0 ( t ) = 0 eq while, as t -> 0 and for finite p, it givcs 0 ( t ) -> 0. Solutions of the diffusion equation have hccn given by Crank( lo) for tlic bounclary condition of equation 4s for diffusion i n I-, 2- or klirnensions. For a splicrc of ratliris x his plots of relative upt,akes as a fiinct,ion of tlinicnsionless time, Dt/a2 , become increasingly signioid as /3 tlecreases and t,lie half-lives, tlI2, of the sorption process increase strongly wilh decreasing values of Dtl/'_/a 2 show:
3
/3 for constant D/a- , as thc following approximate
pa"/D = o;, 5.0 1.0 0.5 0.1 3 Dtl/,/a- = 19 SI 1!J0 5S0 2S70 In the above circumstances apparcnt diffusivit,ies ohtainc~lfrom half-lives by neglecting tlic role of 0 can, for finite 8, he very significa,nt,ly less than the true value. The larger the 2 value of pa / D the more nearly is Dt 1/2 /a2 given 11y t ~ i csolution of the diffusion equation for 0 = 0 at the surface of entry for all t > 0. cq
CONCLIJDING REMARKS It has becn the purpose of this paper antl its pictlecessor (ref. 2 ) to model the transpoit piopcrtics of single crystal membranes taking into acconnt the processes occurring at and acioss the mcmbiane surfaces as well as tliose witliin the crystal. In the present paper the iiiotlel has hwn extentled to include the possibility of simultaneous entry to the crystal both via an externally adsorbed lajer antl tliicctly fioin the gas phi1sr. Tlic role of tlic mass balance equations across tlic surface has also 1w.w extentled to sorption~esorption kinetics.
273
I n t.hc previous paper (ref. 2) thc important role of partial blockagcs located along l-dimcnsional ( 1 - D ) channcl systems was niodcllcd and examined numerically. The papcrs show how the modcrating effect of the mass balance equations at and across mcmbranc surfaces will increase with decreasing thickness and falling tempcraturc. Some numerical estimates indicate that undcr conditions when the intcrface cffccts havc littlc influcncc upon flow at room tcrnpcrature these effects may dominatc flow at low tcmpcraturcs, such as OOK. \Vliere flow is reduccd by partial blockages distributed along 1 - D clianncls thc reduct.ion is indcpcndent of membrane thickness and can be vcry large (ref. 2).
REFERENCES R.M. Barrer, in "The structure and Propcrtics of porous Matcrials". Eds., D.H. Everett and F.S. Stone, (Butterworth). 195S, p.170 2 R.M Barrcr, J.Chcm. SOC.Faraday Trans., 1990, SG, 1123. 3 M. Kocirik, P. Struvc, K. Fiedlcr and hi. Bulow, -1. Chcm. Soc. Faratlay Trans. I , lOSS, 81, 1
3001. 4 5 G
7 S
9 10
M. Kocirik, A . Zikanova, P. Struvc and hl. Bulow, i n "Zeolitcs: Facts, Fiqurcs, Future". Eds. P . A . Jacobs and R.A. van Santcn (Elscvicr), 19S9, p.925. J . Kargcr, II. Pfcifcr and W. Ileink, i n Proc., Gth Int. Zeolite Conferencc, Eds. D. Olson and A. Bisio (Butterworth) 19S4, p.lS4. I1.M. Barrcr, Langmuir, 19S7,3, 309 D.L. Wernick and E.J. Osterhuher, i n Proc. 6th Int. Zcolite Confcrcncc, Etls. D. Olson and A. Bisco (Butterworth) 1983, p.123. A. Paravar and D.T Hayhurst, in Proc. Gtli Int. Zcolitc Confcrcncc, Eds. D. Olson and A Bisio (Buttcrworth) 1984, p.217. D.T Hayhurst and A.R Paravar, Zcolitcs, ISSS,g, 27. J.Crank, "Thc Mathematics of Diffusion" 2nd Edition, (Clarcntlon Prcss), 1975, pp.53, 75 and 92.
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 1991 Elsevier Science Publishers B.V., Amsterdam
275
THE ROLES OF METAL AND ORGANIC CATIONS IN ZEOLITE SYNTHESIS
D.E.W. VAUGHAN Exxon Research and Engineering Company, Rt. 22 East, Annandale, New Jersey, U.S.A., 08801.
SUMMARY The position is taken that the crystallization products of a gel are the key to understanding the important gel intermediate structures and therefore the mode of zeolite genesis. Reviewing the products of reactions in sodium and potassium aluminosilicate gels leads to the conclusion that extended sheet (Na) and columnar ( K ) sub-structures are the important structure building units. The smaller base metal cation controls the nature of the sub-structure and the larger cation the ways in which such units interconnect to form specific zeolites.
INTRODUCTION Much recent interest in modes of zeolite genesis has focused on the role of small molecular clusters as major participatory species in the direction and development of zeolite structure types, relying mainly on developments i n such techniques as NMR (1,2,3) and Raman spectroscopy (4) combined with increasingly powerful computer hardware and software capabilities which facilitate infinite peak deconvolution and search-match abilities. S u c h methods reveal much new information about the complexities of the silicate and gel systems of interest, but they give little guidance to those researchers interested in creating, and preferably tailoring, new zeolite structures. Indeed, it is probable that species observed in solution during zeolite crystallization are primarily "spectator species" in a state of dynamic near equilibrium in a reactive system in which the "participatory species" are constantly scavenged by the growing zeolite(s) or their structure additive sub-units. An alternative approach, examined i n part by Liebau for silicates i n general ( 5 ) and by Barrer for zeolites i n particular (6), is to look at larger
216
entities, such as sheet or chain structures, and to examine how these interconnect in various ways to form t h e many different structures which characterize the crystallization systems of interest. The products and their inter-relationships are the starting point rather than the gel systems. T o this author it is this latter approach which not only explains much of the available experimental data, but offers some opportunity for designed syntheses within the limitations of the specific system of interest. This paper will evaluate numerous minor cation substitutions on the products of separate series of syntheses in sodium aluminosilicate and potassium aluminosilicate systems. The focus will be mainly on the low silica: alumina ratio gels and products where the reactivity, and therefore the multiplicity and diversity of zeolites, is high. At lower temperatures less stable products survive longer and can therefore be analyzed and recognized. The more rapid the approach to thermodynamic equilibrium (represented by higher density zeolite mineral equivalents) the lower the probability of detecting novel and interesting products. CRYSTALLIZATION PRODUCT OVERVIEW T h e approximate composition range discussion is:
of
primary
interest
in
this
2-4 (Na,K)20: xM20: A1203: 6-15SiO2: 100-200H20 where M is either a Group 1A or simple alkylammonium cation and x is less than 1.O. T h e initial ("metastable") main crystallization products at temperatures between 90 and 180°C using sodium or potassium silicates or colloidal silica are:
Na
K
FAU LTL FA U - b ss in t ergr o w t h s OFF GME ERI MAZ ECR-I MOR beta The phases stable at long reaction times are CIS, PHI and MER, with the first dominating the sodium system and mixtures of the lattcr two the potassium system. Mixtures are common in most of the experiments in this composition range, but pure products can be made in high yield by suitable "fine tuning"
277
of the gel compositions and choice of raw materials. ANA (pollucite) dominates the crystallization of any gel composition containing caesium and, with the exception of CSZ-1 and -3, frequently inhibits nucleation over most of this composition range. In understanding the products of these systems recent high resolution structure imaging results have been invaluable (7,8,9,10). (Throughout the IZA structure codes will be used to indicate structure type (1 l ) , with lower case indicating proposed structures, or "inter g r o w n " p h a s e s where the i n d i c a t e d s t r u c t u r e p r e d o m i n a t e s .) (Oualification: This discussion is based on broad generalizations made on the basis of an overview of many experiments. Specific phases and phase combinations are influenced by numerous experimental factors, particularly raw materials and nucleation effects. Specific references should be consulted for these details.)
GME GIS
MAZ LTL OFF
MER PHI
MAZ
Figure 1. The building units characteristic of the many structures under discussion can be readily derived from the Si12 unit in the center of the figure by rearranging a few S i - 0 linkages. Rapid re-equilibration of the system maintains the supply of those removed by the growing zeolites.
278
In terms of derived secondary building units (sbu) or similarly sized clusters all the structures may be readily derived from the Si12 unit shown in Figure 1 by a small number of S i - 0 bond changes. The rapid equilibration of such a system would ensure a constant supply of nutrient sbu to the growing zeolite crystals, but these would always be at a low concentration and therefore not expected to be observed in the solution or gel. LAYER STRUCTURES DOMINATE THE SODIUM ALUMINOSILICATE SYSTEM Various Si/AI ratio faujasites are the dominant products in this system for almost the whole range of M species evaluated, as detailed in Table 1 . In the cases of K and Rb substitution only morphology effects are evident, with the tendency to form platelet crystals of FAU. Substitution of larger cations such as Li (hydrated) and C s not only accentuates the development of platelet forms (ZSM-2 and 3; CSZ-I and 3), but introduces faults and defects associated with the bss connectivity of the "faujasite" sheets. (The so called Breck structure-6, first proposed as a theoretical structure by Moore and Smith (1 3), formed by connecting "faujasite" sheets together to give the hexagonal structure shown in Figure 2 (16)) These effects become more pronounced with increasing Si/AI ratios, but phase stability pushes the system more towards MOR at gel Si/AI over about 6, with that zeolite dominating as the primary product when the Si/AI exceeds about 8. The addition of alkylammonium cations yields higher ratio zeolite Y (13), but gmelinite like impurities (gme) co-crystallize more readily than in the organic free gels in the absence of agitation. (Growth of these gme products is usually greatly accelerated in stirred crystallizations, probably as a result of "collision breeding" of nuclei of this "phase". As the crystallizing solutions are homogenized before each sample is removed i n a single batch, sequentially sampled to obtain a timed crystallization profile, the sampling mode may i n fact promote the gme component. One notes that t h e basic building u n i t of GME is the same as that for CIS, MER and PHI, as shown i n Figure 1.) The notable exceptions are the development of MAZ in TMA systems and ECR-1 i n DMDE(0H) containing gels at higher temperatures, indicating that the "mazzite" sheet is more stable than the "faujasite sheet. FAU usually precedes MAZ in these crystallizations at M<0.6. With increasing gel Si/AI ratios TMA continues to promote MAZ; TEA systems move towards beta (a racemic mixture of two enantiomorphic sheets (15,16)), and in some special cases the highly intergrown "defective" structure ZSM-20 ( 1 7): DMDE(OM), TPA and TBA gels promote the formation of high silica FAU (ECR-4 "
279
(18) and 32 (19) respectively). MTE(0H) (methyltrihydroxyethyl-ammonium) directs the product to ECR-30 (20) (bss). TABLE 1 Phase development with sodium as the primary cation Secondary Cation(M) Li Na
K Rb
cs TMA DMDE(0H) TEA TPA TBA
Products at 100°C
Products at 15OOC
pZSM-2/3 -> CIS FAU/GME -> CIS pFAll -> MER pFAIJ -> MER pC~Z-1/3-> ANA FAU -> MAZ FAU FAU -> GIS FAU/gme -> GIS FAU/gme -> GIS
GIS -> ANA MOR/GIS -> ANA GIS/PHI -> ANA PHI -> ANA ANA MAZ ECR-1 ->ANA FAU/gme -> GIS -> ANA FAU/gme -> GIS FAU/gme -> ANA
Arrows indicate progressive development with increasing crystallization time. gme is an material comprising mainly gmelinite with some chabazite like intergrowths. p indicates distinctive platelet morphology. TMA= tetramethylammonium; T E A = tetraethylammonium; DMDE(OH)= dihydroxyethyldimethylammonium; TPA= tetrapropylammonium; TBA=tetrabutylammonium. ZSM (Mobil), CsZ (Grace Co.) and ECR (Exxon) are proprietary designations. Products based on FAU sheets With the exception of MAZ and ECR-1, all the initial products in this system comprise various arrangements of FAU sheets. The sheet itself is seemingly specifically sodium "templated". In the absence of a specific secondary influence these always assemble in the FAU arrangement. One notes that at lower Si/Al ratios the organic templates are not trapped within the zeolite, possibly because of the large number of cations required for charge balancing the structure. At higher Si/Al ratios the symmetrical alkylammonium cations are occluded in the FAU but d o not seem to induce detectable bss intergrowths. (ZSM-20 i s somewhat of an exception here in that the large amount of it is randomly intergrown fau-bss, strongly influenced by the specific structure of the gel.
280
layer type 1
FAUJASITE ( F A 4 Pure 1 or pure 2
layer type 2 (mirror image of layer 1)
BRECK'S 'STRUCTURE 6' (BSS) Alternating 1 and 2 Figure 2. The sheet structure of FAU, consisting of double 6-ring connected sodalite cages, may have two kinds of connectivity. The cubic ABCABC kind of arrangement characteristic of FAU (X, Y , ECR-4, ECR-32)and the hexagonal ABAB arrangement characteristic of ECR-30 (bss). If carbon atoms replaced sodalite cages, the former would have the diamond structure and the latter the lonsdaleite structure. It is interesting to note that the later part of the crystallization - represented by the crystal the overgrowths and outgrowths - is almost pure FAU(17).) The large Li and C s cations d o however induce a significant and measurable amount of mis-stacking in the form of twin planes or bss layers. The asymmetric MTE(0H) clearly stabilizes the ABAB stacked (ECR-30) form. Products based on MAZ sheets MAZ is stabilized by TMA, but as the MAZ sheet is also present in ECR-1 (8), TMA is therefore not a necessary prerequisite for the stability of this structure, although it may stabilized it at low temperature by occlusion of TMA in the gmelinite cages. Rather it is a more stable structure than the FAU sheet at higher temperatures (100-160°C). to be replaced at even higher temperatures (and independently at higher Si/AI ratios) by the more stable mordenite sheet, reflected in frequent low levels of MOR with ANA in the
281
base metal gel products. ECR-1 is a "boundary phase" between MAZ and MOR, specifically templated by DMDE(0H) (21), these materials having structures related as shown in Figure 3. The MAZ and MOR sheets may be connected in two different ways, at c=O and 1/2, and it is possible that ECR-1 may be a mixture of the two forms (22). Coherent overgrowths of MAZ on ECR-1 (8) and MOR on MAZ have also been reported(23).
MAZ
ECR- 1
MOR
Figure 3. ECR-1 is a boundary phase between the crystallization fields of MAZ and MOR, comprising connected alternating mazzite and mordenite sheets. I t can be viewed as a recurrent intergrowth of MAZ and MOR. Coherent overgrowths of one structure on another have also been observed in this system. COLUMNAR STRUCTURES DOMINATE THE POTASSIUM ALUMINOSILICATE SYSTEM. LTL dominates this composition range, demonstrating much morphological variability (size and shape) as functions of secondary cation, temperature and mixing. Na, Cs and TMA have the most pronounced influence, primarily by stabilizing OFF and in some cases ERI. As in the sodium aluminosilicate system, the presence of caesium promotes pollucite to the exclusion of all other phases. Details are given in Table 2. At lower Si/Al ratios highly intergrown chabazite like products (the materials designated G by Barrer and coworkers (24)) frequently appear as minor or major products, and at higher ratios MOR may be a minor component. Apart from TMA inducing the formation of OFF and OFF/ERI mixtures (ERI always the minor component), the alkylammonium gels produced mainly
282
LTL, frequently having relatively high Si/AI ratios near the limit of the Linde L composition range of 2.6 to 3.45 (25). The presence of the alkyl ammonium cations also greatly slow the sequential reaction to more dense products, presumably by inhibiting the nucleation of the denser zeolites PHI and MER. OFF and ERI may crystallize either as separate crystals (26) or as a crystalline intergrowth of the Linde T type (27), but in these experiments they were not differentiated. TABLE 2 Phase development with potassium as the primary cation. Secondary Cation(M) Li Na K Rb
cs TMA
DMDE(OH) TEA TPA TBA
Products at 100°C LTL LTL LTL -> MER LTL -> MER ANA OFFER1 LTL LTL LTL LTL
Products at 15OOC LTL -> PHI LTL LTL -> MER/PHI LTL -> MER ANA OFF/ERI LTL/OFF/ERI LTL LTL LTL
Products based LTL type columns LTL and OFF are assembled from the same columns of double 6-ring linked cancrinite cages and the ERI column is similar, differing only in a "flip" in the cancrinite cage linkages, as shown i n Figure 4. Although much has been made of the ABC 6-ring sheet stacking relationship of OFF and ERI (and many other structures too (see ref. 28 for a good review)), LTL does not belong to that structural group and would not therefore be expected as a c o crystallization product if such a sheet dominated the gel structure. The cocrystallization of LTL and OFF/ERI with an apparent epitaxial interface in hammer shaped crystal associations (with the hammer head LTL and shaft OFF) requires a common interfacial structure (29).
283
LlNDE L
OFFRETtlE
ERlONlTE
Figure 4. The columnar structures and relationships in LTL, OFF and ERI. Structures based on 4-ring chains The major secondary and stable products in both systems - CIS, MER and PHI - all have structures based on the same edge shared 4-ring chain (Figure 1) linked together in the ways shown in Figure 5 , and identical to that in GME. Similar chains are the main feature of OFF, M A 2 and LTL, as shown in Figure 1.
PHILLIPSITE
MERLlNOlTE
QISMONDINE
Figure 5. In GIs, MER and PHI the same chain of 4-rings is connected in the ways shown to form sheets which constitute the extended building unit.
284
The most stable product at temperatures over about 160°C f o r all compositions studied is A N A , which comprises a structure of corner shared 4ring chains. DISCUSSION The data discussed i n this paper represent three groups of related zeolites which crystallize and often co-crystallize over a small range of synthesis conditions - FAU/bss; MAZ/MOR/ECR-1; LTL/ERI/OFF. Their associations can be better understood if the critical structure determining units are seen as being much larger than the sbu small cluster groups generally favoured as the important structure building units. In these cases the structure of the extended structure (sheet, column, chain, etc) is seemingly determined by the Group 1A base metal cation and the manner i n which they join together is determined by the charge and configuration of the larger secondary cation.
Figure 6. Small extended structures (discussed above) have been observed in the early stages of agglomeration for a 12-ring zeolite (electron micrograph by Dr.S.B.Rice). The agglomerated 12-rings are similar to the individual microdomains observed in a crystal rnozaic of zeolite LTL by Teresaki et al (29). When there i s no significant secondary influence a prefered structure characteristic of the primary cation will predominate - FAU or LTL in the
285
cases cited. T h e recent remarkable observation of stable m i c r o clusters o f columnar rings (Figure 6) and the micro-mozaic structure of a n L T L crystal (29) not only support this concept, but indicate a possible m e c h a n i s m o f nucleation and growth in zeolite systems b y aggregation and annealing of sev ral weakly bonded such micro-clusters. T h e advantage of the extended structure approach is that it facilitates some degree of synthesis guidance and the possibility of tailored syntheses. It i s particularly useful w h e n a p p r o a c h i n g g r o u p s of theoretical s t r u c t u r e s having c o m m o n extended structures, many of which are now in stages of characterization (30). However, the detection and characterization of extended structures in solutions and gels will require more extensive research in areas of Laser Raman spectroscopy and low angle scattering with neutrons and X rays. In addition to the groups of structures discussed a b o v e m a n y similar p r o b l e m s a n d o p p o r t u n i t i e s occur with ZSM-5/11 (31), f e r r i e r i t e ( 3 2 ) , mordenite (33), ALPO-5/ VPI-5 (34). Theta-l/ZSM-23 (35) and many others.
ck
REFERENCES 1 . J.P.van den Berg, P.C.deJong-Versloot, J.Keijsper and M.F.M.Post, Stud.Surf. Sci.Catal.#37, Eds. P.J.Grobet,W.J.Mortier, E.F.Vansant and G.Schulz-Ekloff, Elsevier (Amsterdam), (1988) 85-95. 2 . G.lIarvey and L.S.Dent-Glasser, Am.Chem.Soc. Syrnp.Ser.#398, Ed.M.L.Occelli and H.E.Robson, (1989) 49-65. 3. A.V.McCormick and A.T.Bell, Catal.Rev.Sci.Engrg.. 31 (1989) 97-127. 4. P.K.Dutta and D.C.Shieh, J.Phys.Chern., 90 (1986) 2331-4. 5 . F.Liebau, "Structural Chemistry of Silicates.", Ch. 7 and 8, Springer-Verlag. (Berlin), (1985) pp.90-160. 6. R.M.Barrer, "Zeolite Structures" in "Zeolites: Science and Technology.". NATO AS1 Series #80, Ed. F.R.Ribeiro et al, (1984) pp.35-82. 7. M.M.J.Treacy, J.M.Newsam, R.A.Beyerlein, M.E.Leonowicz and D.E.W.Vaughan, J.Chem.Soc.Chem.Commun., (1986)1211. 8. M.E.Leonowicz and D.E.W.Vaughan, Nature, 329 (1987) 819. 9. J.M.Newsam, M.M.J.Treacy, D.E.W.Vaughan, K.G.Strohniaier and W.J.Mortier. J.Chem.Soc.Chem.Commun., ( I 989) 403-4. 10. M.M.J.Treacy and J.M.Newsam, Nature, 332 (1988) 249. 1 1 . W.M.Meier and D.H.Olson, "Atlas of Zeolite Structure Types.", 1ZA/ Butterworths (London), ( 1 987). 12. D.E.W.Vaughan, "Microstructure and Properties of Catalysts.", MRS Symp. Proc. 11 I , Eds. M.M.J.Treacy, J.hl.Thomas and J.M.White. (1988) 89-100. 13. P.B.Moore and J.V.Smith, Mineralog. Mag., 35 (1964) 1008-1014. 14. D.E.W.Vaughan and K.G.Strohmaier, in Proc. 7th.Intl. Zeolite Conf.. Stud. Surf. Sci. Catal. #28, Ed. Y.Murakarni, A.lijirna and J.W.W;ird, (1986) 207-214. 15. 1I.L.Wadlinger and G.T.Kerr, US Patent 3,308,065, (l!lh7).
286
16.J.M.Newsam, M.M.J.Treacy, W.T.Koetsier and C.B. decruyter, Proc.Roy.Soc., A320 (1988) 475. 17. D.E.W.Vaughan, M.hl.J.Treacy, J.hl.Newsam, K.G.Strohmnier and W.J.Mortier, Am.Chem.Soc. Symp.Ser.#398, Ed.hl.L.Occelli and ll.E.Rohson, (1989) 544-559. 18. D.E.W.Vaughan, U S Parent 4,714,601 (1988). I!). D.E. W.Vaughan and K.G.Strohmaier. US Patent 4,93 1,267 ( 1 990). 20. D.E.W.Vaughan, US Patent 4,879,103 ( 1 990). 21, D.E.W.Vaughan and K.G.Strohmaier. Am.Chem.Soc. Symp.Ser.#398, Eds. M.I..Occelli and H.E.Rohson, (1989) 506-517. 2 2 , I). E .W .V iiu g h an, hl .E ,Le on0w i c z and K ,G .S troh m a ier , A mer. C hem. Soc. Sy mp. Scr. #31 I , Eds. S.E.Bradley, M.J.Gattuso and R.J.Rertolacini, (1989) 303-318. 23. F.Fajula, F.Figueras, C.Gueguen and R.Dutartre, US Patent 4,946,580 ( I 990). 24. R .M .B arrer , " H y d rot her m a I C h e m is try of Ze o I i t e s " , A c ad e m i c Pre s s (I.ondon), (1982) 208. 25. D.E.W.Vaughan, US Patent 4,554,146 (1985). 26 I.S.Kerr, J.A.Gard, R.M.Barrer and 1. Galabova, Amer. Mineral., 5 5 (1970) 441-154 27. J.A.Gard and J.M.Tait, Amer.Chem.Soc. Adv.Chem.Ser. 101, Eds.E.M.Flanigen and I,.B.Sand , (1971) 230-236. 28. G.R.Millward, S.Ramdas and J.M.Thomas, Proc. Roy. SOC.,A399(1985) 57-71. 29. O.Terasaki, J.M.Thomas and G.R.Millward, Proc. Roy. SOC., A395 (1984) 153-164. 30. J.V.Smith, Chem.Rev.,88 (1988) 149-182. 3 I . G.T.Kokotailo and W.M.Meier, in "Properties and Applications of Zeolites.", Chem. SOC. Special. Puhl.#33, Ed.R.P.Townsend, (1980) 133-139. 32. R.Gramlich-Meier, W.M.Meier and B.K.Smith. Zeit.Kristallogr.. 169 (1984) 2 0 1 - 2 1 0. 33. J.D.Sherman and J.M.Bennett. Proc. 3rd Intl. Zeolite Conf., Amer.Chem.Soc. Adv.Chem.Ser. 121, Eds. W.M.Meicr and J.R.Uytterhoeven. ( I 973) 52-65. 34. J.W.Richardson,Jr.. J.V.Smith and J.J.Pluth, J.Phys.Chem., 9 3 (1989) 82128219. 35. P.A.Wright, J.M.Thomas, G.R.Millward, S.Ramdas and S.A.I.Barri. J.Chem. SOC. Chem.Commun.,(l985) 1 117-9.
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites
281
0 1991 Elsevier Science Publishers B.V., Amsterdam
TEMPERATURE DEPENDENCE OF NUCLEATION OF ZEOLITES I N ALKALINE ALUMINOSILICATE GELS I N HYDROTHERMAL CRYSTALLIZAIPION CONDITIONS ZHDANOV, N.N. FEOKTISTOVA and L.M. VTJURINA I n s t i t u t e of S i l i c a t e Chemistry of t h e Academy of Sciences of t h e USSR, Nab. Makarova 2 , Leningrad 199034, USSR S.P.
ABSTRACT The paper p r e s e n t s a f u r t h e r development of s t u d i e s i n z e o l i t e c r y s t a l l i z a t i o n k i n e t i c s c a r r i e d out e a r l i e r i n the l a b o r a t o r y of t h e authors. It was shown t h a t by performing a j o i n t a n a l y s i s of c r y s t a l l i z a t i o n curves d e s c r i b i n g z e o l i t e c r y s t a l growth i n a l u m i n o s i l i c a t e g e l s i n t h e process of t h e i r hydrothermal crys t a l l i z a t i o n and of d a t a on s i z e d i s t r i b u t i o n z e o l i t e c r y s t a l s i n t h e f i n a l products of g e l c r y s t a l l i z a t i o n i n combination with t h e i r pretreatment at d i f f e r e n t temperatures before subsequent isothermal c r y s t a l l i z a t i o n , one can a t t a c k t h e problem of studying temperature dependences of z e o l i t e n u c l e a t i o n r a t e i n t h e c r y s t a l l i z i n g gels. INTRODIJCT I O N
Since experimental s t u d i e s s t a r t e d publishing t h e k i n e t i c s of z e o l i t e s c r y s t a l l i z a t i o n /1/ and up t o our days, t h e construc t i o n of S-shaped k i n e t i c curves d e s c r i b i n g t h e growth of z e o l i t e mass i n t h e process of a l u m i n o s i l i c a t e g e l s isothermal crys t a l l i z a t i o n h a s remained t h e most common method of d i s p l a y i n g experimental r e s u l t s of such i n v e s t i g a t i o n s . These k i n e t i c dependences a r e i n most c a s e s w e l l described formally by t h e following equation:
-
Zt/Zf = Z = ( 1 e-ktn) (1 1 where Zt mass of c r y s t a l s i n t h e c r y s t a l l i z a t i o n product which were formed during t h e time t from t h e beginning of c r y s t a l l i z a t i o n ; Zf mass of c r y s t a l s i n t h e f i n a l c r y s t a l l i z a t i o n product; k and n - dimensionless c o n s t a n t s depending on g e l comp o s i t i o n s , temperature and some o t h e r c r y s t a l l i z a t i o n c o n d i t i o n s (seed, g e l ageing e t c . ) . However, such curves do not b e a r any information on t h e b a s i c k i n e t i c parameters of c r y s t a l l i z a t i o n : l i n e a r r a t e of c r y s t a l growth and n u c l e a t i o n rate. The p o s s i b i l i t y of experimental s t u d i e s of t h e s e parameters f o r r e a l conditions of z e o l i t e c r y s t a l l i z a t i o n on t h e example of
-
-
a Due t o i l l n e s s o f Prof. Zhdanov the paper could not be presented i n Leipzig.
288
zeal-ite w a s shown i n /2/.The methods d e s c r i b e d i n /2/ ena b l e d i r e c t d e t e r m i n a t i o n of l i n e a r r a t e of z e o l i t e s c r y s t a . 1 growth, s t u d i e s of t e m p e r a t u r e and c o n c e n t r a t i o n dependences of c r y s t a l growth r a t e and s t u d i e s of n u c l e a t i o n k i n e t i c s i n c r y s t a l l i z i n g g e l s /3-9/. P a p e r s / 7 , 8 / c o n s i d e r g e n e r a l p o s s i b i l i t i e s t o s t u d y t h e t e m p e r a t u r e dependence of n u c l e a t i o n i n g e l s on t h e b a s i s of t h i s method. The p a p e r i n q u e s t i o n p r e s e n t s t h e r e s u l t s of a more d e t e i l e d study i n t h i s f i e l d .
Na-A
SXPXRIT':SrJT The a l u m i n o s i l i c a t e g e l s s t u d i e d were of t h e f o l l o w i n g compos i t i o n s : 2,76 Na20 A1203 1,91 Si02 409 H20 and 3 , 7 2 Na20' 4.l2O3 2,8 S O 2 351 H20 t h e f i r s t of them c r y s t a l l i z i n g i n t o Na-A z e o l i t e s , t h e second i n t o Na-X ones. A l u m i n o s i l i c a t e g e l s were p r e p a r e d by mixing a l k a l i n e s o l u t i o n s o f sodium s i l i c a t e 0 and a l u m i n a t e end were c r y s t a l l i z e d at 90 C a f t e r p r e l i m i n e r y h o l d i n g f o r 8 h o u r s at 0, 7 , 27, 40, 60, 70 and 80°C f o r Ba-A and f o r 72 h o u r s a t -12, - 5 , 0, 9 , 20, 40, 60 and 8 O o C f o r Na-X. One sample of each k i n d of g e l was c r y s t a l l i z e d at 90° w i t h o u t thermal p r e t r e a t m e n t . The above h o l d i n g of samples b e f o r e c r y s t a l l i z a t i o n e n a b l e s one t o f o l l o w t h e e f f e c t of t e m p e r a t u r e of p r e l i m i n a r y t h e r m a l t r e a t m e n t on c r y s t a l l i z a t i o n k i n e t i c s and on i t s f i n a l r e s u l t s . The e x p e r i m e n t s were based on t h e measurements by means of an o p t t c a l microscope o f s i z e s of t h e l a r g e s t c r y s t a l s found i n t h e c r y s t a l l i z i n g g e l s at v e r i o u s p e r i o d s of t i n e from t h e b e g i n n i n g of i s o t h e r m a l h e a t l n g o f samples. F o r Na-A t h e edges of c u b i c c r y s t a l s were measured and f o r Ya-X the axes of o c t ahedre. The above measurements e n a b l e t o d e t e r m i n e g r a p h i c a l l y the l i n e a r r e t e o f c r y s t a l g r o w t h , t o follow i t d u r i n g t h e whole c o u r s e of c r y e t e l l i z e t i o n p r o c e s s and t o determine t h e end of t h e p r o c e s s ( t h e time of c r y s t a l l i z a t i o n ) by t h e d i s c o n t i n c a t i o r i of c r . y s t a l growth. For a l l t h e samples t h e h i s t o g r e m s were determined which c h a r a c t e r i z e c r y s t a l s i z e d i s t r i b u t i o n i n t h e f i n a l c r y s t a l l i z a t i o n products. The time wher t h e growth of each mode o f c r y s t a l s s t a r t e d was determined and t h e c u r v e s c h a r a c t e r i z i n g n u c l e a t i o n k i n e t i c s wer e c a l c u l a t e d from t h e s e h i s t o g r a m s and from t h e date. 0 x 1 l i n e e r growth ret e .
.
-
289
RESULTS AND DISCUSSION The c u r v e s i n f i g . 1 and 2 d e s c r i b e t h e v a r i a t i o n i n dimensions of t h e l a r g e s t c r y s t a l s of Na-A ( f i g . 1 ) and Na-X ( f i g . 2 ) z e o l i t e s i n t h e i n v e s t i g a t e d g e l s i n t h e p r o c e s s of t h e i r c r y s t e l l i z a t i o n at 90OC. Curves 1 i n b o t h f i g u r e s a r e f o r g e l samples c r y s t a l l i zed by h e a t i n g o n l y e.t c r y s t a l l i z a t i o n t e m p e r a t u r e . A l l o t h e r c u r v e s a r e f o r g e l samples which have been t h e r m a l p r e t r e a t e d ( b e f o r e c r y s t a l l i z a t i o n a t R s t a n d a r d 90°C t e m p e r a t u r e ) 8 h o u r s f o r g e l P and 2 4 h o u r s f o r g e l X et t e m p e r a t u r e s lower t h e n t h e s t a n d a r d t e m p e r a t u r e of t h e i r subsequent c r y s t a l l i z a t i o n . The p r e t r e a t m e n t t e m p e r a t u r e OP each sample of the g e l s s t u d i e d is g i v e n n e a r t h e r e s p e c t i v e curve. A l l t h e c u r v e s have the same s l o p e and c o n s i s t of t h r e e p a r t s : t h e i n i t i a l l j n e a r p a r t whose s l o p e i n d i c a t e s t h e c r y s t a l growth r a t e ( 0.5A1/bt ) and the length - the time i n t e r v e l with coxstant grovth r a t e , a s h o r t e r p a r t of n o n s t a t i o n m y growth and R h o r i z o n t a l p a r t where 110 growth of c r y s t a l s i z e s was o t s e i ~ e d .
13
10
5
.L
0" 20
40
60
i-;O
2,h
F i e . 1. GrowtIn i n c r y s t a l . s i z e o f t h e l a r g e s t Na-A z e l i t e c r y s t a l s ( 1 max ) i n c r y s t a l l i z i n g g e l s a t gOOc a f t e r thermal pretreatment ( t i s marked n e a r t h e c u r v e s ). Pr
290
I
Fig. 2. Growth of c r y s t e l s i z e of t h e l a r g e s t ITa-X c r y s t a l s i n cr.~ s t a l l i z i n g g e l s at ?O°C a f t e r t h e n v l p r e t r e a t m e u t a t d i f f e r e n t temperatures. The v e r y f a c t of t h e e x i s t e n c e of e n i n i t j a l l i n e a r p a r t i n e v e r y c c r v e ( f i g . 1 end 2 ) e n a b l e s one t o conclude t h a t a t l e a s t i n t h e time i n t e r v a l c o r r e s p o n d i n g t o t h e s e p a r t s of t h e c u r v e s , t h e c r y s t a l s i n d i f f e r e n t samples of t h e g e l s u n d e r o b s e r v a t i o n had a c o n s t a n t growth rate. Eloreover, t h e f e c t t h a t ell t h e s e c u r v e s have t h e same s l o p e means t h a t t h e l i n e a r r e t e of c r y s t a l growth was t h e same f o r ell the samples and independent of t h e t e m p e r a t u r e of p r e l i m i n a r y g e l t r e a t m e n t ( t 1. However, c u r v e s Pr 1 and 2 d i s p l a y a n obvious e f f e c t of s u c h t r e a t m e n t on t h e s i z e s of t h e l a r g e s t c r y s t a l s f o u n d j n t h e f i n e l c r y s t a l l i z a t i o n prod u c t s and on d u r a t i o n of c r y s t a l l i z a t i o n (Z') determined by t h e time when v i s i b l e c r y s t a l growth i s over. T h i s time I s marked i n f i g . 1 end 2 by a c r o s s i n g of dashed c u r v e s . The r e s p e c t i v e d a t a f o r b o t h z e o l i t e s a r e g i v e n i n T a b l e 1. The c h a r a c t e r i s t i c changes due t o t h e t e m p e r a t u r e of p r e l i m i nary t r e a t m e n t of t h e g e l ( t ) were a l s o observed t o have some Pr e f f e c t on t h e p e c u l i a r i t i e s of z e o l i t e c r y s t a l s s i z e d i s t r i b u t i o n i n t h e f i n a l c r y s t a l l i z a t i o n products. One may see s u c h changes f o r Ra-A z e o l i t e i n t h e h i s t o g r a m ( f i g . 3 ) . F i n a l l y , t h e e f f e c t of t h e t e m p e r a t u r e of gels t h e r m a l p r e t r e a t m e n t ( t ) i s c l e a y l y Pr observed 011 l , t e r u n of t h e c a r v e s d e s c r i b i n g t h e growth o f z e o l i t e c r y s t a l s mass i n t h e p r o c e s s of g e l j s o t h e r n o 1 c r y s t a l l i z a t i o n . Fig.4 e l s o g i v e s such k i n e t i c c u r v e s f o r Ve-A z e o l i t e . They are
TABLE
1
The e f f e c t o f g e l s pi*etr.eetrnent t e m p e r a t u y e on c r y s t e l s i z e s , c r g s t f i l l i z a t i o n t i m e and nuc l e a t i o n ~
00
Na- A
27 O
2.82
2.09
3.55 53
4.79 47
20
18
90 1.84
72 1.43
3.62 2.76 59 -120
10 2 Na-X
66 1.06
40 O 16 62 0.85
7"
23
-5 O 35 5
600 64 1.22
19 66 7-49
70 1.87
90° 23 96 2.44
1.67
2.40
2.94
3.6e
4.81
5.97 43
4.16 43
3.41 43
2.71 44
2.08 61
18
700
800 20
90 35 5.2
200
41 5.6
33 4.8
40° 51 6.7
60 O 54 7.1
5'5 7.5
900 60 7.8
5.71
6.01
7.14
00
800
0.035
1,16
1.18
1.17
1.05
3.95
0.069
2.28
2.33
2.31
2.07
7.78
11.3
11.8
14.10
4.39 5.64
4.29 5.63
4.33
4.82
1.29
0.89
0.84
0.71
5.64
5.68
5.11
4.95
4.93
4-85
145 7.16
292
0.08
0.04
40
700
GOo
d 0.08
a
0.04 S
800 n
90"
0.08 0.04
0 61 16 t7 16 1,Y Pig.3. Iiistograms o f t h e f i n a l c r y s t a l l i z a t i o n p r o d u c t s i n g e l Ka-A samples a t 90°C a f t e r a p r e t r e a t m e n t at d i f f e r e n t temperatures.
._ .
..
I
.
-
I"
Fig.4. The curve8 of z e o l i t e Sa-f, c r y s t a l mass growth ( i n L / b f f r a c t i o n ) i n t h e p r o c e s s of i s o t h e r m a l c r y s t a l l i z a t i o n a t 9b.C a f t e r pretreetment at d i f f e r e n t temperatures. The c u r v e s were c a l c u l a t e d from t h e d a t a i n f i g . 1 and 3.
293
c o n s t r u c t e d on t h e b a s i s of d a t a i n f i g . 1 and t h e h i s t o g r a m s i n f i g . 3 by means of r a t h e r s i m p l e c a l c u l a t i o n s . The method of cons t r u c t i n g such c u r v e s i s d e s c r i b e d i n /5,9/ ( s e e also / l o / and /12/). The S-shaped curve8 i n f i g . 4 a r e well d e s c r i b e d by equat i o n (11, t h e v a l u e s of k and n c o n s t a n t s v a r y i n g w i t h e a c h p a r t i c u l a r case. These c o n s t a n t s g r a p h i c a l l y determined from I n ln 11-z 1 n . t dependence e n a b l e one t o f i n d t h e p o s i t i o n of t h e bend p o i n t o f k i n e t i c c u r v e s tinf a c c o r d i n g t o :
-
(2)
T h i s r e l a t i o n f o l l o w s from t h e second d e r i v a t i v e Z w i t h r e s p e c t to t The p o s i t i o n s and s l o p e s o f t h e c u r v e s i n f i g . 4 which are i n t h i s c a s e determined by tinf v a l u e s one may s e e t h a t t h e k i n e t i c s of c r y s t a l mass growth i n the p r o c e s s o f g e l s i s o t h e r m a l c r y s t a l l i z a t i o n i s s t r o n g l y dependent on I n c i d e n t a l l y , as tPr i t i s seen from t h e d a t a i n Table 1 t h e g e l c r y s t a l l i z a t i o n time depending on g e l p r e l i m i n a r y p r e t r e a t m e n t t e m p e r a t u r e changes unevenly and p a s s e s a minimum c o r r e s p o n d i n g t o a temperature tPr c l o s e t o 4 Oo C f o r Ha-A z e o l i t e and t o 20OC f o r Wa-X z e o l i t e . As t h e l i r - e a r rate of c r y s t a l growth is independent of t the Pr observed e f f e c t of t h e t e m p e r a t u r e of g e l s p r e l i m i n a r y t r e a t m e n t on t h e d u r a t i o n of t h e i r c r y s t a l l i z a t i o n , t h e s i z e s of largest c r y s t a l s and t h e c h a r a c t e r of c r y s t a l s i z e d i s t r i b u t i o n i n t h e f i n a l p r o d u c t s may be account f o r o n l y under t h e c o n d i t i o r . o f nonmonotonous dependence of n u c l e a t i o n r a t e i n gels on t h e temperat u r e i n t h e temperatura j i . t e r \ - r l s t u d i e d . This has been p o i n t o u t e a r l i e r / 7 , 8 / b u t no a t t e m p t s have been made t o e s t i m a t e q u a n t i t i v e l y t h i s dependence. Such a n e s t i m a t i o n seems q u i t e r e a l i s t i c t a k i n g i n t o c o n s i d e r a t i o n t h e above - mentioned independence of c r y s t a l growth l i n e a r r a t e on t h e i r s i z e s which was also n o t e d e a r l i e r /4,5/. I n such a c a s e t h e number of n u c l e i formed i n t h e p r o c e s s of g e l c r y s t a l l i z a t i o n may be determined by t h e number of c r y s t a l s i n the f i n a l c r y s t a l l i z a t i o n p r o d u c t s i f t h e volume of a p a r t i c u l a r number o f c r y s t a l s , f o r i n s t a n c e , of 100 c r y s t a l s i n t h e f i n a l product w a s known. The volume of 100 c r y s t a l s ( Vlo0) may be c a l c u l a t e d from t h e histogram according t o t h e equation:
.
.
where 4 - t h e p r o p o r t i o n of c r y s t a l s of each mode, d i a m e t e r s and o t h e r measurable l i n e a r dimensions, K
di-
-
their t h e geo-
294
.
m e t r i c a l c o e f f i c i e n t l i n k i n g d3 w i t h t h e volume Vi The number of c r y s t a l s p e r u n i t product mass is determined from : Nf
100 / V1oop
(4)
where p i s c r y s t a l d e n s i t y , t h e v a l u e s Vlo0 and Nf being given i n Table 1. One may s e e from t h e t a b l e t h a t t h e d u r a t i o n o f c r y s t a l l i z a t i o n ( 2 )of g e l s p r e t r e a t e d at d i f f e r e n t temperat u r e s ( t ) r e g u l a r l y v a r i e s w i t h Nf which may be c l e a r l y seen Pr when comparing t h e r u n of curve8 1 and 2 i n fig.5. 10-8
2 6
4
.
Fig.5 The dependence of c r y s t a l l i z a t i o n time and of t h e o v e r a l l number of c r y s t a l n u c l e i N growing i n g e l s from pretreatment temperature: t h e f i g . on t h & l e f t i s f o r Na-A, on t h e right for Na-X.
-
Sinee, as it was pointed e a r l i e r , t h e number of c r y s t a l s i n t h e f i n a l products of g e l . c r y s t a l l i z a t i o n e q u a l s t o t h e number of t h e i r n u c l e i , t h e data i n fig.5 show t h a t t h e change i n c r y s t a l l i z a t i o n time r e s u l t i n g from g e l pretreatment a t d i f f e r e n t temperat u r e s i s due t o temperature dependence of g e l n u c l e a t i o n r a t e i n t h e r e g i o n studied. The data i n fig.1 and 2 enable one t o conclude t h a t at t h e temperature of m a x i m a l n u c l e a t i o n and a t lower temperatures down t o O°C n u c l e a r s i z e s do not s i g n i f i c a n t l y grow. A t h i g h e r temper a t u r e s ( t ) not only f r e s h n u c l e i a r e being formed b u t a l l t h e Pr n u c l e i g r o w i n s i z e due t o a considerable i n c r e a s e i n c r y s t a l growth r a t e . It is displayed i n t h e observed v e r t i c a l s h i f t s of
295
l i n e a r p a r t s of t h e curves which i n c r e a s e with temperature growth of g e l pretreatment. The r e s u l t s obtained i n t h i s paper make i t c l e a r t h a t t h e w e l l known ageing e f f e c t i n g e l s ( t h e i r being h e l d a t room temperature) is one p a r t i c u l a r case of t h e e f f e c t of n u c l e a t i o n temperature dependence on c r y s t a l l i z a t i o n r a t e . In t h e a u t h o r s ' opinion t h e development of n u c l e a t i o n i n such g e l s is connected w i t h t h e so-called r i p e n i n g process leading t o i r r e v e r s i b l e changes i n c o l l o i d a l g e l s t r u c t u r e /4,5/. The d r i v i n g f o r c e of t h i s process i s t h e tendency of t h e c o l l o i d a l system t o minimal f r e e energy. This process appears t o be r e s p o n s i b l e f o r t h e development of g e l n u c l e a t i o n under o t h e r temperatures studied. The Fjharp i n c r e a s e of c r y s t a l l i z a t i o n r a t e of Na-X a l u m i n o s i l i c a t e g e l s a f t e r t h e i r f r e e z i n g i s obviously not connected w i t h t h e process of ripening. It seemingly r e s u l t s from a phase disbalance when t h e l i q u e d phase of t h e s e g e l s h a s been f r o z e n out. The mechanism of t h i s e f f e c t is not y e t q u i t e c l e a r . CONCLUSION The i n v e s t i g a t i o n s i n question develop an experimental technique f o r studying n u c l e a t i o n of z e o l i t e s i n a l u m i n o s i l i c a t e g e l s and i t s temperature dependence. Study was made of t h e e f f e c t of a l u m i n o s i l i c a t e g e l s thermal pretreatment on t h e c r y s t a l l i z a t i o n r a t e , t h e degree of c r y s t a l d i s p e r s i o n i n Na-A and Na-X z e o l i t e s formed and on n u c l e a t i o n i n gels.
REFERENCES 1 D.W. Breck, E.M. Flanigen, 'Molecular Sieves', SOC. Chem. Ind., London, 1968, pp. 47-61. 2 S.P. Zhdanov, 'Molecular Sieve Z e o l i t e s - l ' , h e r . Chem. SOC., Washington, 1971, pp.20-43. 3 S.P. Zhdanov, h o c . 3rd Intern. Conf. on Molecular Sieves, Louven Univ. Press, 1973, pp.25-29. 4 S.P. Zhdanov and N.N. Samulevich, Adsorbents, t h e i r Preparat ion, P r o p e r t i e s and Application, Nauka, Leningrad, 1978, pp.10-16. 5 8.P. Zhdanov and N.N. Samulevich, Proc. 5 t h Intern. Z e o l i t e Conf. (Ed. L.V.C. Rees) Heyden, London, 1980, pp.75-84. 6 S.P. Zhdanov, S.S. Khvoshchev and N.N. Samulevich, S y n t h e t i c Z e o l i t e s ( i n Russian), Chimij a , USSR, 1981. 7 S.P. Zhdanov, N.N. Feoktistova, E. Jahn, Izv. &ad. Nauk Ser. K h i m . (1986) 1720-1724. 8 N.N. Feoktistova, L.M. Vtjurina, Izv. &ad. Nauk Ser. Khim. (1 988) 727-730. 9 S.P. Zhdanov, S.S. Khvoshchev and N.N. Feoktistova, S y n t h e t i c
296
Z e o l i t e s , vol. 1, Gordon and Breach S c i e n c e P u b l i s h e r s , 1990. 10 R.Y. B a r r e r , Ilydrothermal Chemistry of Z e o l i t e s , Academic Press, 1982. 1 1 C o l i n S. Cundy, I3.N. Lowe, D.TZ. S i n c l a i r , C r y s t a l Growth, v o l . 100 (1990) 189-202. 12 R.U. B a r r e r , Z e o l i t e s , vol.1 (19131) 130-140.
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
297
SYNTHESIS O F PIPERAZINE AND TRIETHYLENEDIAMINE USING ZSM-5-TYPE ZEOLITE CATALYSTS
J. WEITKAMP', S. ERNST1, H.-J. BUYSCH2 and D. LINDNER3 1 University of
Stuttgart, Institute of Chemical Technology I, Pfaffenwaldring 55,
D-7000 Stuttgart 80 (Federal Republic of Germany). 2 Bayer AG,
Werk Uerdingen, P. 0. Box 166, D-4150 Krefeld 11 (Federal Republic of
Germany) 3 University of
Oldenburg, Department of Chemistry, Chemical Technology, Ammerlaender
Heerstrasse 114-118, D-2900 Oldenburg (Federal Republic of Germany)
SUMMARY The synthesis of piperazine and triethylenediamine from ethylenediamine in ZSM-5-t e zeolite catalysts was investigated. With strongly acidic HZSM-5, large amounts of by-pro ucts were observed (viz., pyrazine, 2-picoline, resin-like substances and cracked products). Reducing the strength of the acid sites by ion exchange with alkali metal cations strongly improved the selectivities for the desired products. This effect was further enhanced by the addition of water vapour to the feed. From the zeolites tested in this study, KZSM;5 appears to be the best catalyst: At T = 340 "C, W/F,, = 196 gh/mol and I ~ H ~ On m: A = 12, conversion of ethylenediamine decreased only from30 % to 80 % after more than three days time on stream. During this time, selectivities for piperazine and triethylenediamine remained nearly constant above 95%.
C P
INTRODUCTION The classical fields of industrial catalysis by zeolites comprise large volume refinery processes, viz. catalytic cracking, hydrocracking, isomerization or dewaxing, and the production of petrochemical commodities (e.g., ethylbenzene synthesis via the Mobil/Badgerprocess or xylene isomerization) (refs. 1-3). More recently, research in catalysis focusses on the use of zeolites for the production of more valuable products, viz. organic intermediates and fine chemicals (refs. 4-6). For this purpose, a broad variety of zeolite structures and compositions, and the complete set of techniques for their modification are nowadays available. The challenge, however, is to identify products which can be manufactured with higher selectivities, in an environmentally more acceptable manner or otherwise more economically by using tailored zeolite catalysts. The present paper reports on the synthesis of piperazine (PIP) and triethylenediamine (TEDA or DABCO, 1,4-diazabicyclo[2.2.2]octane), two organic intermediates in the production of pharmaceutical products and plastics, respectively. They are currently manufactured commercially using non-zeolitic catalysts. Piperazine is produced by reacting an ethanolamine/ammonia-mixture on a Raney nickel catalyst at temperatures from 150 to
298
220°C and pressures between 10 and 25 MPa. The synthesis of triethylenediamine from ethylenediamine, diethylenetriamine or diethanolamine is catalyzed by amorphous silicaalumina at atmospheric pressure and temperatures around 360 "C. Both processes suffer from relatively low selectivities for the desired products. Hence, several expensive separation steps are required downstream of the reactor (ref. 7). In the literature, there is only scarce information on the use of zeolite catalysts for the production of piperazine and triethylenediamine, using as feed such compounds as N-(2hydroxyethy1)-piperazine (ref. 8,9), ethanolamine (refs. 10,l l), ethylenediamine (ref. 11) or diethylenetriamine (ref. 11). In the present study, an attempt was undertaken to replace the conventional catalytic systems by more selective zeolite-based catalysts. Among the advantages offered by zeolites are their shape selective properties and the possibility to tailor the nature, density, strength and distribution of their acid sites. EXPERIMENTAL Zeolite ZSM-5 with Si/AI=40 was synthesized from an alkali-free gel according to the procedure outlined by Ghamami and Sand (ref.12). HZSM-5 was obtained by calcination of the as-synthesized material for 16 hours at 540 "C in air followed by ion exchange with a In aqueous solution of NH,CI and a further calcination at 420 "C. Li, Na, K, Rb and Cs exchanged samples of ZSMd were prepared by dispersing the zeolite powder two times for eight hours in a I n aqueous solution of the respective chloride salt. The zeolite powder was pressed without any binder, crushed and sieved. The size fraction between 0.2 and 0.5 mm was used for the catalytic experiments. They were performed in a flow-type apparatus with a fixed bed reactor. Product analysis was achieved by on-line sampling and temperature programmed capillary glc. Ethylenediamine (EDA), either pure or in admixture with water vapour, was used as feed and nitrogen as carrier gas. RESULTS AND DISCUSSION The first experiments were conducted with H Z S M J as catalyst. Under typical reaction conditions (T=340 "c, W=0.28 g, W/FEDA=186gh/mol, PEDA=2.1 kPa, pN2 cj 100 kPa), severe catalyst deactivation takes place. In addition, a variety of side reactions (e. g., cracking, polymerization to resin-like substances) occur beside the formation of the two desired products. Typical data after 2 hours time on stream are: XEDA=24%, SP,,=35 70,h D , , = 6 %. This unsatisfactory picture is completely changed if H Z S M J is replaced by NaZSM-5. Pertinent results are depicted in Figure 1. Catalyst deactivation, although still present, is considerably slowed down when compared to HZSMJ. Side reactions like cracking or polymerization are now strongly suppressed. The sum of the selectivities for the desired products piperazine and triethylenediamine has improved to ca. 70 %. Based on these results, the unsatisfactory performance of HZSM-5 can be tentatively assigned to the presence of
299
100
80 8?
.-
v,
60
L
0
a
0
40
W
X
20 0 0
2
4
6
8
10
12
14
16
18
20
TIME O N STREAM, h Fig. 1. Conversion of ethylenediamine in NaZSM-5. Reaction conditions: T=340 "C, W=O.28 g, W/F,,= 186 gh/mol, pEoA=2.1kPa, PN*U 100 kPa strong Bronsted acid sites in this catalyst: If the interactions between these acid sites and the basic feed or product molecules are too strong, the desorption of the products will be severely hindered and the residence time of the basic reactants at the acid sites will be uncontrollably high which results in the occurrence of consecutive reactions like cracking or polymerization. To overcome the problem, two principle solutions can be envisaged: (i) the addition of an inert component (e.g., water vapour) to the feed, which competes with the basic reactants for the acid sites or, (ii) decreasing the interactions via a reduction of the acid strength. The latter possibility already proved to be successful in the present case. With the aim to further increase the selectivities to the desired products and the stability of the catalyst, additional experiments were conducted to (i) investigate the influence of water vapour and, (ii) clarify the influence of the strength of the acid sites by varying the alkali metal ions exchanged into ZSMJ. Table 1 summarizes the influence of different amounts of water vapour added to ethylenediamine in the feed stream. From these data, the conclusions can be drawn that with increasing amounts of water in the feed (i) the selectivities for the desired products piperazine and triethylenediamine increase, and (ii) the tendency for the formation of cracked products decreases. Therefore, a molar ratio of 12 for H,O/EDA in the feed was used for all further experiments. It has been found earlier (ref. 13) that the strength of the acid sites in Z S M J zeolites can be tailored by ion exchange with different alkali metal ions: When such catalysts are characterized by temperature programmed desorption of ammonia (TPDA), the temperature
300
niiZo: ~ E D A
0: 1
1:l
4: 1
12:1
45
53
58
71
54.3
56.1
57.6
59.0
16.3
20.1
27.1
35.0
8.4
6.8
3.1
1.0
3.4
3.3
2.0
not determined
8.7
6.3
5.6
3.5
100 Li+
._
v,
60
Rb'
Cs'
A;E - 400
80
w
K'
NCl+
0
*\
0
- 320
4
n
L
0
-
< 40
240
n
a
t )i 0 E
W
X
-
20
0 50
I
I
100
I
I
150
I
160
'
80 200
CATIONIC RADIUS, pm Fig. 2. Activities and selectivities of ZSM-5 zeolites exchanged with different alkali metal cations (reaction conditions: T=340 "C, W/FEDA=186 gh/mol, hH,O : ~ E D A= 12, W = 0.28 g)
301
of maximum desorption rate decreases with increasing ionic radius of the exchanged alkali metal cation. Hence, the strength of the acid sites decreases from LiZSM-5 to CsZSM-5. Therefore, Lit-, K+-, Rbt- and Cs+-exchanged Z S M J zeolites were also included in the investigations. The results are depicted in Figure 2. For the sake of clarity, only the conversion are of ethylenediamine (XED*) and the selectivities for the two major products (S,,,, bDA) plotted against the radius of the alkali metal cation. The data were taken after 16 hours time on stream, when with all catalysts a nearly stable stage was reached. The catalytic activity declines in the sequence KZSM-5 > RbZSM-5 > CsZSM-5 > NaZSM-5 > LiZSM-5. Obviously, there is no direct relationship between the height of ethylenediamine conversion and the strength of the acid sites as characterized by TPDA. Probably, the interactions of the basic reactants with LiZSM-5 and NaZSM-5 are too strong to allow for a fast desorption of the desired products. As a consequence, consecutive reactions, i. e. coke formation are favoured. Starting from KZSMJ with its considerably weaker acid sites, there is a decline of conversion which corresponds to the decrease of acid strength with increasing cationic radius. With all alkali metal cation exchanged ZSM-5 zeolites, high selectivities for the desired products (above 90%) are observed. The remaining percentage of selectivities is made up by pyrazine, 2-picoline and cracked products. To achieve even better selectivities and yields, conversion of ethylenediamine was varied by varying the reaction temperature. The selectivities observed for piperazine and triethylenediamine are presented in Figure 3. The general trend holds for all catalysts: 100
I
I
I
50
I
I
I
R b o , Csc
80
I
P
M = Lio, Nav. KA
x
I
x
40
a
a
n
a
W I-
m
60
30 ~
40
I
I
I
I
20 0
20
40 60 8 0 100
’ %
0
20
40 60 80 100
’ %!
Fig. 3. Variation of selectivities for piperazine and triethylenediamine in MZSM-5 (M=Li, Na, K, Rb, Cs) with increasing conversion of ethylenediamine. , X ( was varied by changing the reaction temperature, all other parameters were kept constant as given in Fig. 2).
302
Increasing conversion leads to a decreasing selectivity for piperazine with simultaneously increasing selectivity for triethylenediamine. This can be easily understood since triethylenediamine forms in a consecutive reaction from piperazine and ethylenediamine (under simultaneous formation of two molecules of ammonia per molecule piperazine converted). Since the sum of the selectivities for PIP and TEDA remains almost constant above 90 %, this conversion/selectivity relationship offers a certain flexibility for a possible industrial process. Depending on the market's demand for piperazine and triethylenediamine, the selectivities can be adjusted such as to meet the requirements. For a commercial process, one could envisage to perfom the reaction at elevated pressure. Therefore, the conversion of ethylenediamine/water mixtures was investigated at pressures up to 3.0 MPa in a fixed bed flow-type apparatus made from stainless steel. Pertinent results are summarized in Table 2. The general trends are in agreement with the data obtained at atmoTABLE 2 Influence of total pressure on activity and selectivity of NaZSM-5 (T-340 "C,W =0.28g, h 2 0: = 12).
1.0 1.0 1 .o 3.0
4.2 8.4 8.2 8.1
97 48 188 65
46.7 61.0 87.1 62.2
63.8 56.1 44.4 57.7
25.8 30.6 37.8 31.7
5.6 3.3 3.4 3.4
4.2 4.2 6.3 5.0
spheric pressure, viz. with increasing conversion the selectivity for triethylenediamine increases at the expense of piperazine. However, the sum of the selectivities never reaches the high values attained at atmospheric pressure, i. e., far above 90 %. Surprisingly, diethylenetriamine (DETA) was formed with selectivities between 2 and 6 %. This is in contrast to the experiments at 100 kPa, where these high selectivities for DETA could not be observed. One selected catalyst, KZSM-5, was tested under atmospheric conditions with regard to its time on stream stability in an experiment extended to 76 hours (cf. Figure 4). Over the whole period, only a slow deactivation is observed. After about three days, conversion still amounts to ca. 80 % with excellent selectivities to piperazine (ca. 59%) and triethylenediamine (ca. 38 %). Only small amounts of 2-picoline (S< 1%) and cracked products (S CII 2 %) are detected. The reasons for the slow but continuous deactivation are not yet clear. There are at least two possible explanations, viz. deactivation by coke formation or extraction of aluminum species from lattice positions due to the hydrothermal environment (high temperature and high concentration of water vapour). More work is underway to clarify the mechanism of deactivation and to develop specific techniques for regeneration of the spent catalyst.
303
100
80
I
I
I
I
I
I
I
I
I
-
-
x .-
cn
L
0
<
n w X
0
8
16
24
32
40
48
56
64
72
80
TIME O N STREAM, h Fig. 4. Conversion of ethylenediamine in KZSM-5; reaction conditions as in Fig. 2.
CONCLUSIONS The synthesis of piperazine and triethylenediamine was investigated using ZSMJ-type zeolite catalysts. Conversion of ethylenediamine over strongly acidic HZSM-5 zeolite resulted in fast catalyst deactivation and unacceptable low selectivities for the desired products. As a working hypothesis, this was attributed to the strong interaction of the basic reactants with the Bronsted acid sites which results in prolonged residence times and favours consecutive reactions. Hence two main directions were followed in order to improve the performance of the catalyst: (i) Reducing the strength of the acid sites by ion exchanging HZSM-5 with alkali metal cations and, (ii) adding water vapour to the feed, to make use of competitive adsorption/desorption effects. The combined application of both measures was successful: With KZSM-5 as catalyst and a water/feed of rh,o : tiem = 12, the sum of the selectivities for piperazine and triethylenediamine could be increased to more than 95 % at conversions above 80 % and with excellent time on stream stability. ACKNOWLEDGEMENTS Financial support by Bundesministerium fur Forschung und Technology (BMFT) of the Federal Republic of Germany is gratefully acknowledged. REFERENCES N. Y. Chen and T. F. Degnan, Chem. En Progr., 48 (1988) 32-41. 1 N. Y. Chen, W. E. Garwood and F. G. bwyer, Shape Selective Catalysis in Industrial 2 A plications, Marcel Dekker, New York, 1989,303 p. 3 d Holderich and E. Gallei, Chem.-1ng.-Tech, 56 (1984) 908-915.
304
4
5 6
7 8
9 10 11
12 13
W. Holderich, M. Hesse and F. Naumann, Angew. Chemie, 100 (1988) 232-251. W. Holderich, in P. A. Jacobs and R. A. van Santen (Editors), Zeolites: Facts, Figures, Future, Studies in Surface Science and Catalysis, Vol. 49, Part A, Elsevier, Amsterdam, Oxford, New York, To 0,1989, pp. 69-93. H. v. Bekkurn and H. . Kouwenhoven, Recl. Trav. Chirn. Pays-Bas, 108 (1989) 283294. G. Heilen, H. J. Mercker, D. Frank, R. A. Reck and R. Jackh, in Ullmann's Encyclopedia of Industrial Chemistry, 5th edn., Vol. A2, Amines, Aliphatic, VCH Verlagsgesellschaft, Weinheirn, 1985, pp. 15-17. R. A. Budnik and M. R. Sandner, Europ. Patent Appl. 158 319, Oct. 16, 1985, assigned to Union Carbide Corp. W. Holderich, K. Schneider, H. Lermer and W. Best, Europ. Patent Appl. 263 463, Oct. 5, 1987, assigned to BASF AG. H. Satoh and M. Tsuzuki, WO 87 03 592, June 18,1987, assigned to Idemitsu Kohsan. J. Disteldorf, N. Finke and W. Hiibel, Europ. Patent Appl. 313 753, May 3,1989, assigned to Hiils AG. M. Ghamami and L. B. Sand, Zeolites 3 (1983) 155-162. J. Weitkamp, D. Lindner and S. Ernst, in Projektleitung Material- und Rohstofforschung, Forschungszentrurn Jiilich (Ed.), Proceedings of the Status-Seminar "Katalyseforschung",at Fritz-Haber-Institute of the Max-Planck-Society, March 1, 1990, in press.
%
G. 6tilmann a!. (Editors ), C'ata!ysis and Adsorption hy Zeolitm 63 1991 Elsevier Science Pul)lishers H.V., Amsterdam
305
DIFFUSION EFFECTS ON THE KINETICS OF TOLUENE HETHYLATION AND XYLENE ISOHERIZATION ON HZSH-5 ZEOLITES F . BAUEB, J. DEBHIETZEL and U. JOCKISCH Central Institute for Isotope and Radiation Research, 7050 Leipzig (CDR)
SUHHARY By using carbon-14 labelled compounds diffusion effects in heterogeneous catalysis can be assessed. The formation of polymethylated aromatics in the conversion of toluene uith methanol on HZSH-5 zeolite catalysts is mainly a consecutive reaction, but different diffusivities of the reactants give gaseous hydrocarbons, toluene and poly-methylated aromatics via a direct synthesis from methanol. In xylene isomerization diffusion influences the concentration profiles leading to wrong mechanistic reaction mode 1 s. INTRODUCTION Kinetic studies only allou to drau conclusions about the rate determining step of a reaction sequence; therefore mass transfer to and from the external surface of the catalyst and/or mass transfer into and out of the catalyst pores can lead to results uhich are contrary to chemical theory. Especially in zeolite catalysis, diffusion disguised kinetics are to be expected methylation of aromatics and isomerization of xylenes on HZSH-5 zeolites may serve as examples. In methanol/toluene mixtures only a fraction of the methanol participates in the reaction as alkylating agent, uith the remainder being converted mainly to gaseous hydrocarbons (ref. 1 ) . By using I3C(ref. 2)- or 14C(ref. 3)-labelled compounds it can be shoun that aromatics are formed from methanol too, and the portion of this pathuay have been assessed. Furthermore, diffusion effects in medium pore size zeolites are responsible for the violation of the validity of Broun's LFE relation ship (ref. 4) estimating the alkylation rate of aromatics by the nature and the number of substituents. In xylene isomerization on medium pore size zeolites different diffusion rates causes concentration profiles uhich are best represented by a triangular reaction scheme (ref. 5 ) including the unrealistic 1,3-methyl shift. EXPERIHENTAL For catalytic purpose, a commercial HZSH-5 zeolite catalyst (Leuna-Uerke, GDR) uith a Si02/A1203 ratio of 56 and containing
20 X A1203 binder was crushed and sieved (0.2 - 0.4 BB). The zeolite was highly crystalline to x-ray diffraction pattern and according to SER pictures had a particle size in the 2 - 5 Pm range. The quartz glass microreactor (ref. 6 ) was packed w i t h 0.25 g catalyst grains diluted uith the same amount of quartz. 14CNitrogen was used as carrier gas (volume ratio 1OOO:l). methanol and 14C-benzene ( Isocommerz 1 uero used uithout further purification. Ring-label led 14C-toluene and 14C (7.81-para-xylene were synthesized and checked for radiochemical purity a s uell as for accuracy of label position (ref. 7). The reaction products uere analyzed by radio gas chromatography using a dinonylon Chromosorb U-AH-DHCS column uhich was phthalate/Bentone-34 connected W i t h a katharometer and a proportional counter as detectors.
RESULTS AND DISCUSSION Hethylation of toluene In accordance uith t h e homogeneous Friedel-Crafts methylation of aromatic hydrocarbons the heterogeneous catalyzed reaction on zeolite catalysts can not be stopped after the first step and in a sequence of consecutive reactions polymethylated aromatics are formed. toluene
+neOH,
xylene
+neoH,
trimethy lbenzene
+neoH
This reaction sequence up to hexamethylbenzene is a consequence of the stabilizing action of alkyl groups on the transition state of reaction. Thus, toluene reacts faster than benzene uith alkylating agents forming favorably ortho- and para-xylene (ref. 8 ) . Unlike the homogeneously catalyzed alkylation and the PischerTropsch process the chain growth does not exceed the formation of C10-hydrocarbons in Hobil's methanol-to-gasoline(HtG) process as a consequence of the shape selective properties of HZSH-5 (ref. 9). Although the high content of aromatics makes the HtG products a good motor fuel the successive methylation reaction yields the undesirable 1.2.4.5 tetramethylbenzene (durene) uhich can lead to "carburetor icing" due to its high melting point (352 K). Therefore kinetic studies on the alkylation of toluene uith methanol on HZSH-5 catalysts have of special interest for the reduction of durene content. Fig. 1 and Table 1 demonstrate the application of the kinetic isotope method (refs. 10 and 3) on the reaction of 14C-toluene with methanol. The time dependence of concentrations may be interpreted as consecutive reactions, But the characteristic concentration maximum for xylenes which is to be expected for
307
such a type of reaction is not observed.
" I
-4
i
i
I
I
Fig. 1. Comparison between observed (dots) and calculated (line, model l ) concentration-time dependence of toluene methylation at 693 K uith a ZSH-5 catalyst
It should be noted that the conversion of toluene does not exceed
and methanol is even at short residence times totally consumed. The product composition shoms that in a side reaction methanol reacts preferably to gaseous hydrocarbons as ethene. propene and butene. Therefore, in analogy to the HtG process the formation of aromatics should be considered in mechanistic studies. It is to be supposed that the diffusivity of methanol in the HZSH-5 framework is higher than that of toluene. The same ratio may be valid f o r the adsorption of the two reactants at the acid centers. In both cases adsorbed methanol molecules react preferably uith each other and form polyaethylated aromatics via intermediates which are not in a adsorption/desorption equilibrium uith the gas phase. That means, the methylation o f toluene on HZSH-5 can be described as .triangular* reaction model. 40
X
toluene
+HeoH,
\I/
xylenes
+HeoH
trimethylbenzenes [model
1)
methanol
Using labelled aromatics the portion of methyl transfer reactions (methylation and disproportionation) and the direct build-up of aromatics from methanol resp., is indicated by the molar radio-
activities of the products (Table 1). TABLE 1 Hethylation of toluene and xylene on HZSH-5 catalyst Product composition (mass and activity X ) and relative molar activities of aromatic components (toluene and xylene i n the reaction mixture= 1.00) at 693 K, 0.1 HPa N2. 3 g/(g.h) methanol/ 14C-to luene ___
component
~~
mass activ. percentage
benzene 0.3 toluene 65.9 par a-xy 1ene 14.8 meta-xylene 8.9 ortho-xylene 4.7 Me-Et-benzenes 2.2 tri-lie-benzenes 3.2
methano 1 / 14CHg-p-xy 1 ene
~
0.1 69.1 13.6 8.2 4.3 1.9
2.8
rel.mo1. activity -
0.88 0.76 0.85
0.89 0.82 0.78
mass activ. percentage 0.1 1.6 55.9
23.7 8.7 0.6 9.4
0.7 59.5 25.1 7.6 0.5 6.6
rel.mo1. activity
0.33 0.87 0.87 0.79 0.77 0.65
In the reaction of an equimolar toluene/methanol mixture only between 76 and 69 X of xylenes are formed via methylation. The portion of the direct synthesized aromatics from methanol depends on the reaction conditions, the diffusion path inside the crystallites and the kinetic diameters of reactants. Dessau and La Pierre (ref. 2) found uith 13C-methanol/toluene mixtures a portion of 5 t o 34 X C1-synthesis of xylenes and 29 - 40 X of trimethylbenzenes if the average diameter of the crystallites is 1 Pa. I f small crystallites of 0.02 P B are used the C1-synthesis of xylenes is reduced to 2 - 10 X (19 - 24 X f o r trimethylbenzens). These results correspond w i t h ours ( 1 1 - 24 X for xylenes, 18 - 22 X for trimethylbenzenes) taking into account the different reaction conditions and crystallite dimensions ( 2 5 Pm). This direct synthesis of aromatics from methanol prevents a total suppression of durene formation in the HtG-process. It should be noted that the presented data are only representative for the participation of methylation and C1-synthesis i n the formation of aromatics. The lou portion of C1resp., synthesis of seta-xylene and ortho-xylene should not be interpreted as an consequence of their formation at or near the external surface. According to Dessau and La Pierre (ref. 2) the formations of methyl-ethyl-benzenes and trimethylbenzenes near the external surface (and a high value of the molar radioactivity) are to be expected, but lower values than those for meta- and ortho-xylene resp. uere observed indicating that the formation of trimethylbenzenes from C1-species is higher than i n the case of xylenes. In connection with this fact the sequence of methylation
309
rate and conversion of aromatics (at 693 K and 2 g/(g'h)) be considered: benzene (53 X conversion)
>
toluene (41 X )
>
have to
xylene (22 X )
As mentioned above, this sequence on medium pore size zeolites does not agree uith those in homogeneously catalyzed reactions (ref. 4) and on large pore size zeolites (ref. 1 1 ) because of different diffusivities of the main reaction products. If the methylation experiments are carried out under the same rate limiting step then the theoretically expected ratio of the reaction rates should be observed. This is t h e case uhen the methylation rates of benzene and toluene are determined under the rate mixture of limiting diffusion of xylenes. Therefore a 1:l:l methanol/benzene/toluene uas transfor8ed in three experiments over the catalyst changing in each run the labelled compound (Table 2). This is a typical tracer experiment in multi-component systems. The distribution of radioactivity, using 14C-benzene and 14C-toluene as tracer, s h o w a higher conversion of toluene than benzene corresponding w i t h a higher incorporation of 14C-methanol in the xylene formation than in the toluene formation. These results on HZSH-5 zeolites confirm the rule that the reactivity of aromatics and the stability of the transition state increases uith the number of methyl groups.
TABLE 2 Hethylation of a mixture of benzeneltoluene uith methanol on HZSH-5 catalyst - Product composition (mass and activity X ) at 693 K, 0.1 HPa N2, 40 g/(g-h) mass 14C-methanol Cg-C7-fract ion ben Zen e toluene ethylbenzene para-xylene meta-xylene ortho-xylene Et-He-benzenes Trine-benzenes
0.8 39.1
52.7 0.1 4.6 0.9 1.0 0.2 0.6
11.0 0.7 21.4
0.9 38.9 7.5
7.4 4.1 8. 1
radioactivity 14C-benzene 14C-toluene
0.2 91.8 7.3 0.1 0.4 0.1 0.1
-
-
0.3 0.2 87.4
1.0
0.4 0.8
Isomerization of xylenes The isomeritation of xylenes has been investigated on a great variety of catalysts. The results of earlier papers on the homogeneously catalyzed conversion (aluminum chloride is an example of a widely used catalyst) were consistent with a sequence of intramolecular 1,2-methyl shifts around the aromatic
ring (ref. 12 ortho-xylene
-
14).
=== meta-xylene
para-xylene
(model 2)
A 1.3-methyl migration, i.e. a direct para-ortho reaction, is not alloued. The scheme of consecutive reaction steps has also been confirmed under heterogeneously catalyzed conditions (refs. 13 and 1 5 ) . But there are exceptions to this mechanism; an intermolecular aethyl shift via toluene and/or trimethylbenzenes sirulates a "direct" para-ortho reaction pathuay (refs. 16 and 17).
1,2,4-trimethylbenzene
/+
ortho-xylene
\
it
reta-xylene
f l \
1,2,3-trimethylbenzene
1
@
para-xylene
(model 3)
1,3,5-trimethylbenzene
Especially on zeolites uith strong acid sites toluene and trimethylbenzenes are produced by disproportionation of xylenes. On larger pore zeolites disproportionation comes up to 60 X of isomerization (ref. 18). In analogy to the 1.3-methyl shift by transmethylation an intramolecular para-ortho conversion via C5,Cg-ring isomerization has been observed on bifunctional catalysts under hydrogen pressure (ref. 1 9 ) .
~-
5
I .5
-.+
Fig. 2. Comparison between observed (dots) and calculated (line. model 2) concentration-time deDendence of xvlene isoaerization a t 693 K uith a ZSH-5 catalyst
l i n e ih)
Besides these chemically understandable para-ortho reaction path-
311
ways mass transfer effects can also result in a triangular reaction scheme. Chutoransky et al. (ref. 18) found a dependence of the reaction rate constant kortho-para and kpara-ortho resp., on the crystalline size of the catalyst particles. For small particles no direct ortho-para pathway was observed, but for large crystallites a triangular reaction network had to be considered. ortho-xylene
7-xylene
\.para-xy 1 ene
(model 4 )
The concentration profiles of xylenes within toluene methylation on zeolites can be used for elucidating the mechanism of isomerization of xylenes. The high para-xylene content of the initial product distribution (Pig. 2) is the result of shape selective diffusion. The estimation of the model parameters uas done by nonlinear regression methods using the thermodynamic equilibrium ratio to minimize the number of unknown parameters. Because of the unknoun initial product distribution of xylenes two additional variables had to be calculated. The triangular reaction scheme (Pig. 3) gives a significant better fit than a series mechanism (Pig. 2 ) . Table 3 contains a comparison of the calculated reaction rate constants of model 4 with literature data.
0
I
I
I
I .5
Fig. 3. Comparison between observed (dots) and calculated (line, model 4 and model 5 resp. 1 concentration-time dependence of xylene isomerization at 693 K uith a ZSH-5 catalyst
llnc l b )
In
spite of differences caused by various reaction
temperatures
312
(573 K - 693 K) the different participation of the rate constants kpara-ortho in the brutto consumption rate of para-xylene on zeolite and amorphous catalysts resp.. is striking. Uhereas on alumina the para-ortho conversion is very slou (and therefore could be nearly neglected), on redium pore size zeolites the theoretically "not alloued" 1.3-methyl shift pathuay is nearly 30 % of the overall para-xylene conversion.
TABLE 3 Isomerization of xylenes on various catalysts - Reaction rate constants (and 95 % confidence intervals) of model 4
kpara-met a kmeta-ortho kpara-ortho kp-olzkpara-
HZSH-5 (this uork)
HZSH-5 (ref. 5 )
(ref. 20)
(ref. 17)
1.0 f 15% 0.88 2130% 0 . 4 2 f 40% 0.30
1.0 0.33 0.39
1.0 0.80 0.09 0.08
1.0 0.29 0.04 0.04
0.28
A1203
A1203
A partition of the para-ortho pathuay into transmethylation and diffusion is possible by including toluene and trimethylbenzene concentrations in reaction modeling (model 4 ) . However this procedure increases the number of unknoun reaction model parameters and needs therefore further studies; but a rough estimation of transmethylation can be made by means of the extent of xylene disproportionation. After Bankos et al. (ref. 21) the disproportionation of xylenes on ZSW-5 zeolites does not exceed 3 - 7 X of the isomerization rate (see content of benzene in Table 1). and therefore transmethylation has only a small contribution to the conversion of xylenes on medium size zeolites. Hence, the rate constant of the "direct" para-ortho pathuay is an indication of mass transfer and/or sorption effects of the catalyst under study. Unlike amorphous A1203 catalysts strong diffusion disguised kinetics in xylene isonerization have to put into consideration on HZSH-5 zeolite catalysts. This assumption is supported by the high error of the rate constant kmeta-ortho (Table 3) - the confidence interval includes the value Zero. As Fig. 3 obviously shous the mechanism of xylene isomerization on the investigated ZSH-5 zeolite uhich consists of 2 - 5 pm large crystallite particles can be described by a virtual parallel reaction scheme (model 5 ) uithout any loss in t h e goodness-of-fit. meta-xylene para-xylene
4-
% ortho-xylene
(model 5 )
313
This conclusion does not mean that from the view point of Chemical theory the isomerization of xylenes on zeolite catalysts consists of two simultaneous reactions, but only that the contact time dependence of the gas phase concentrations (which are the result of CheBiCal and sorption/diffusion effects within the catalyst particles) is well represented by a parallel reaction network. REFERENCES 1 N.Y. Chen, J. Catal., 114 (1988) 17 2 R. Dessau and B. La Pierre, J. Catal.. 78 (1982) 136 3 F. Bauer. R. Hanisch, U. Jockisch and Chr. Wienhold, Chem. Techn., in press 4 H.C. Brown, J. Amer. Chem. Soc., 78 (1956) 6255 5 J.D. Collins, R. J. Hedina and B.H. Davis, Can. J. Chem. Eng., 61 (1983) 29 6 J. Dermiettel, F. Bauer. H. Rosseler, U. Jockisch, H. Franke, J. Klerpin and H.J. Barz, Isotopenpraxis, 12 (1976) 57 7 F.D. Kopinke, J. Dermietzel, U. Jockisch and G. Rauber, Isotopenpraxis, 22 (1986) 388 8 R.H. Allen and I.D. Yates, J. Amer. Chem. SOC., 83
(
1961) 2799
C. Chang and A. Silvestri, J. Catal., 47 (1977) 249 10 H. B. Neimann and D. Gal, The kinetic Isotope Hethod and its application, Acaddmia Kiado. Budapest, 1972 11 E. Dumitriu, S. Oprea and V. Hulea, Rev. Roum. Chim., 9
32 12
(
1987) 525
R.H. Allen and L.D. 81
(
Yats. J. Amer. Chem. SOC.,
1959) 5289
K.L. Hanson and A.J. Engel, AIChE J., 13 (1967) 260 H.C. Brown and H. Jungk, J. Amer. Chem. SOC., 77 (1955) 5579 S. Zimnieuict, H. Pilarczyk and K. Kalinski, Chemia Stosouana, 17 (1973) 235 16 H.A. Lanewala and A.P. Bolton, J. org. Cher., 34 (1969) 3107 17 F. Bauer, J. Dermiettel, H. Rosseler and H. Koch. Chem. Techn. , 28 ( 1976) 144 18 P. Chutoransky and F.C. Duyer, Adv. Chem. S e r . . 13 14 15
121
(
1973) 540
19
J. Dermietzel, H. Rosseler. U. Jockisch, Ch. Uienhold, H. Franke, J. Klempin and H.J. Barz. Isotopenpraxis,
20
K.H.
14 (1978) 14 58 21
Robschlager and E.G. Christoffel. Can. J. Chem. Eng., (
1980) 517
I. Bankos. J. Papp and D. Kallo, Acta Chim. Hung., 119
(
1985) 179
This Page Intentionally Left Blank
G . Ohlmann et al. (Editors),CatQlySiSand Adsorption by Zeolites 01991 Elsevier Science PublishersB.V., Amsterdam
315
IRON-CONTAII.IING ZSI-5 TYPE ZEOLITES USED IN THE C0UP L ED MET HA N 0L- H Y DR 0CAR B 0N CR A CK ING ( CMH C )
A. BIARTIN*’, S. NOWAK’, B. LUCKE’, W. WIEKER2 and B. FAHLKE2 ’Central Rudower 2Central Rudower
Institute of Organic Chemistry, Academy o f Sciences, Chaussee 5 , 1199 Derlin, ( G . D . R . ) Institute of Inorganic Chemistry, Academy of Sciences, Chaussee 5 , 1199 Berlin, (G.D.R.)
ABSTRACT Different iron-containing H-ZSK-5 type zeolites have beenused as catalysts for the CKHC process with n-butane as hydrocarbon feed. Furthermore, these iron-containing zeolites have been investigated as catalysts in methanol conversion and in n-butane pyrolysis, the two single reactions of the coupled process. The incorporation of iron into silicon-aluminium catalysts led to an increased lifetime, a higher hydrocarbon conversion as well as stable olefin yields o f such samples compared to silicon-aluminium-containing zeolites during the CbiHC process. Beside a lower deposition o f coke a higher amount o f carbon dioxide was observed during the reaction. Furthermore, it was found that a partial decoking ( u p to 70 wt.-% of deposited coke) occurs during regeneration by steam treatment at reaction temperatures.
I rl TH 0DUCT I0N In the field of zeolite catalysis, the synthesis of light olefins from methanol has been one of the most important subjects for the last ten gears (e.g. refs. 1-4). Catalysts applied in this reaction are narrow-pore zeolite types (refs. 5 , 6 ) . The short lifetime, connected with a strong coke deposibion, is thc I ost serious p d > o b l e : i cuch zeolite types. 23rL-5 type zeolites are described and used for the olefin production from methanol feed as \!ell, but the olefin selectivity is lover than that of the narrow-pore type zeolites. Recently, metal-ion-containing zeolites (e.g. containing gallium, chromium, vanadium and iron) with pentasil pore structure have becn assumed to be catalysts for methanol transformation into olefins with both high selectivity and prolonged catalyst lifetime (e.g. refs. 7-9). Since the synthesis, characterization, and investigation o f their catalytic performance in various reactions (refs. 7-15) these different metal-ion-containing ZSM-5 type zeolites have got a great influence in some catalytic reactions, io18
316
e.g. methanol conversion to gasoline, lower olefins, and aromatics or paraffin conversion to aromatics. In comparison with silicon-
aluminium, the amounts of lower olefins in methanol conversion could be increased mainly by ZSM-5 type zeolites with different amounts of iron incorporated (refs. 8 , l O ) .
Furthermore, in connec-
tion with the lower acidity of such zeolites, a decrease in the deposition of coke has been observed (ref. 1 0 ) . The combination of methanol and hydrocarbon transformation (CiiIIC)
into lover olefins and aromatics was previouslydescribed
in detail (refs. 1 6 - 2 1 ) .
At high temperatures, up to 953 K , the
addition of hydrocarbon to a methanol feed in a defined ratio leads to the conversion o f both methanol and hydrocarbon into lower olefins, gasoline, and aromatics in a nearly thermoneutral
reaction in the presence of H-ZSM-5
zeolites.
The CI4HC process has been investigated with n-butane as hydrocarbon feed a n d various kinds of iron-containing H-ZSM-5 type catalbsts in comparison to the same investigations with only a silicon-aluminium-containing zeolite catalyst as described earlier (refs. 1 6 , 1 7 , 2 1 ) . Furthermore, some experiments o n the methanolto-olefin reaction ( I d T O ) at lower temperatures (ca. 6 7 3 K ) and o n the n-butane pyrolysis ( 7 7 3 - 9 5 3 K ) have been carried out. De-eking cycles proceeded under steam and subsequent air treatment. Iloreover, some results of the physical characterization of these zeolites will be presented in this paper, e.g., investigations of temperature-programmed
(TPD) desorption of ammonia,
electron paramagnetic resonance ( E P R ) and X-ray diffraction (XRD). EXPE R II4EN ‘I‘A I, Apparatus and procedure Isjethano1 and crude n-butane ( 9 2 X n- and 8 7; i-butane) were used as feed €or the C N H C investigations. Some experiments under LIT0 conditions were carried out with pure methanol and n-butane was used as feed in pyrolysis tests. A l l investigations were carried out in a bench-scale reactor (quartz glass, catalyst volume ca. 30 ml) (ref. 1 7 ) . The reaction products were analyzed by on-line gas-chromatography (ref. 2 2 ) .
All conversion data and product yields described in the tables and figures have been calculated on the C H 2 part in the feed. After the reaction the zeolite catalysts could be regenerated by steam at ca.873 - 9 5 3 1.:
317 Catalyst synthesis The synthesis of iron-ZSId-5 and iron-aluminium-ZSl,~-5 products were carried out according to the method of Szostak and Thomas (ref. 23).
Fe(N0
3 3
solution o r an equimolecular mixture of
solution was acidified with H2S04 (96 :.)and Fe(NO3I3 and A l ( N O < ) 3 3 then mixed with 4.51.1 I!a,H 2 2Si04. These quantities of the reactants were chosen in order to get the required silicon-to-iron and silicon-to-iron-to-aluminium ratios in the reaction mixture. To the resulting gel 0.1 mol of tetrapropylammoniumbromide per mol of Si02 was added and, after vigorous stirring, the reaction mixture was heated in stainless-steel autoclaves at 443 1: for 24 hours under autogeneous pressure. During thermal treatment the autoclaves were rotating. The resulting solids were filtered, washed, and dried. After synthesis the organic template was destroyed by calcination at ca. 823 K. The ion exchange was carried out with 0.511 NH4C1 or NH4N03 solution (each sample four times for two hours at 353 K).
The ammonium-containing zeolites were calcined stepwise
from 573 K t o 823 K to get the H-form. The zeolite powder was then treated with pure silica (65 wt.-% zeolite and 3 5 wt.-% Si02) to form catalyst pellets. In general, three catalysts have been used for the CPiHC investigations (sample A : silicon-to-eluminium ( f o r comparison);
=
22
sample B: silicon-to-aluminium = 25; sample C:
silicon-to-iron-to-aluminium
= 50:l:l).
Characterization methods The acidity of some samples was determined by the TPD of ammonia. These experiments were performed by gas-chromatographic measurement of the amount of ammonia desorbed upon heating ammonia-treated samples (treated with ammonia at 393 K and followed by desorption of physically sorbed ammonia at the same temperature €or 30 min) at a rate of' 1 2 deg/min. The EPR spectra of the products were recorded on a Varian E 4 spectrometer operating at X-band microwave frequency. The X-ray powder patterns of the polycrystalline iron- or ironaluminium-silicate samples obtained were performed with a Guinier equipment.
318
RESULTS AND DISCUSSION Characterization of iron-containing ZSL-5
zeolites
(i) Temperature-programmed desorption of ammonia (TPDA).
The
TPDA plot (P'ig. 1 ) clearly demonstrates the weaker Bronsted acidity of iron-containin& zeolites as well as the ireakness of such siteo in iron-aluminium-containing zeolites in comparison to t h e vell-kriovn ZSL-5
(silicon-eluminium) zeolites. Framework alu-
minium (sample A) results in two peaks at ca. 500 and 710 I<. I'ramework iron (sample B) with the weaker acidity shows one broad peak at 585 1:.
Finally, the framework iron and aluminium in one
sample (sample C ) result in the same peaks known from framework aluminium, but the high-temperature peak indicating the strong Bronsted acidity is decreased.
-I
-
7
673
r
s
673
I
1
873
TIKI
Yig. 1 . TPDA of ZSld-5 samples: ( A ) silicon-to-aluminium=22, (€3) silicon-to-iron=25, L C ) siliconto-iron-to-aluminium=50:1:1
Fig. 2 . EPR spectra o f freshly synthesized (a) and used (b) ZSB-5 zeolite (sample B , silicon-to-iron=25
(ii) Electron paramagnetic resonance spectroscpy (EPR). spectra of the freshly synthesized zeolite (sample B )
The EPR
(Pig. 2a)
suggested evidence for at least three different types of iron sites with g = 4.3, 2 . 3 and 2.0 as described in the literature (e.g.
refs. 2 4 - 2 6 ) .
The signal at
g =
4 . 3 could be assigned to
Fe3+ in tetrahedral coordination in the zeolite framework. The
319
signals at g = 2.3 and 2.0 have been assigned, according to data given in the literature (refs. 24,271, to Fe3+ in occluded oxides and hydroxides (g
=
2.3) and to Fe3’- in 011 symmetry in cationic
sites of the hydrated zeolite (g = 2.0).
After strong hydrothermal
treatment of the catalyst samples during CIHC reaction (873 953 K) the resonance at g = 4.3 disappeared completely (Fig. 2b). In agreement with other methods (e.g. IR spectroscopy),
this ob-
servation gave evidence o f a nearly complete outbreak of the framework iron. We found a corresponding phenomenon o f a rapid and complete withdrawal o f framework species during CLHC reaction on pure aluminium-containing H-ZSM-5 zeolites (refs. 1?,28).
(iii) X-ray diffraction (XRD).
All of the X-ray diffraction
patterns of the investigated iron- and iron-aluminium-silicate samples have been nearly identical to that o f aluminium-ZSM-5 without iron incorporation. This indicates that the materials obtained have the same crystalline structure as aluminium-ZSM-5. Methanol-to-hydrocarbon conversion Table 1 describes comparison between the three catalyst samples
A , E, and C during methanol conversion after two hours time onstream at 673 K and a WHSV o f 0.8 h-’ under atmospheric pressure.
TABLE 1 Methanol conversion and product distribution on ZSM-5 samples A ,
B and C (673 K , WHSV
=
0.8 h-’, atmospheric pressure) ~
Catalyst
Si-A1 Si-Pe Si-Fe-A1 t
Sample
A B C
Conversion* Yield* Methanol C,-C 4
~
L‘
-c
paraffins
olefins
92
30
35
97 93
8
58 33
26
liquid hydrocarbons
17 19 20
”t ,-70
In the four columns the conversion of methanol, the yields of lower paraffins (C1-C4),
lower olefins (C2-C4),
and liquid product
amount are shown for every catalyst under the same reaction conditions. The olefin yields are very high on pure iron zeolite (sample B) in comparison to aluminium-containing zeolites. The amount of paraffins increased with increasing aluminium-content
320
in the catalyst, while the amount of olefins decreased. The catalytic behaviour of the zeolite samples A and C is nearly the same. The lower conversion o f methanol on pure aluminium-containing zeolite (sample A ) and its lower yield in liquid hydrocarbons could be caused by a faster deactivation of this sample. The results show that additional aluminium-containing ironZSIil-5
zeolites are not very selective for the generation of lower
olefins, because the primarily formed olefins are further converted to higher hydrocarbons, such as aromatics and higher paraffins. These observations are related to the stronger acid sites of the additionally incorporated framework aluminium. 2514-5 zeolites with lower Bronsted acidity containing solely iron proved to be more effective and selective f o r methanol conversion into ole-
fins. At higher temperatures the methanol conversion was complete and, as expected, the amounts o f carbon oxides, hydrogen, and methane increased drastically. Furthermore, the formation of coke was advanced. This is also known from aluminium-containing zeolites (refs. 16,171. n-Butane pyrolysis The cracking activity has been checked stepwise with the same catalytic systems from 773 up to 953 K under atmospheric pressure and a WHSV = 1.3 h-I. Table 2 depicts the results of pyrolysis at 773, 873 and 953 I[ on iron-ZSM-5 zeolite (silicon-to-iron = 25). TABLE 2 Products obtained by n-butane pyrolysis using ZSM-5 sample B -1 (silicon-to-iron = 25) as catalyst (WHSV = 1.3 h , atmospheric pressure) Temperature (K)
Conversion* Yield* n- But ane '2-'4 paraffins
C2-C4
Methane
olefins
773
6
1.2
2.4
1.4
873
38
4.8
953
68
7.1
18.5 36.9
14.8 19.8
* wt.-% The n-butane conversion has increased with increasing reaction temperature. Main products have been lower paraffins and olefins and, at higher temperatures, also methane.
321
Furthermore, the n-butane conversion on iron-containing ZSE-5 zeolites is weaker (Fig. 3 ) compared with that on aluminium-containing zeolites. Despite of the lower cracking activity of such it has been observed that there is a p r o -
zeolites (iron-ZSM-5),
longation of the reaction time resulting from a lower deposit of coke. This smaller amount of deposited coke could have also been caused by the lower acidity o f the Bronsted sites in iron-containing zeolites.
I:;
& $
,
,
,
1
2
3
0 -
u r
tlhl
Fig. 3. n-Butane conversion on ZSI-1-5 zeolites in pyrolysis reaction depending on time on-stream: key as for Fig. 1. By additional incorporation of aluminium during the synthesis of iron-containing zeolites (samyle C ) the conversion of n-butane
could be increased, but the level of decrease in conversion remained the same a s in investigations with solely iron-containing sample.
Iron containing ZSM-5 type zcolites in the CWHC reaction The preferred mixture of methanol and n-butane for the CMHC reaction was calculated by adding up the standard reaction enthalpies (law o f constant adding u p o f heat by HESS) as described in ( ref s. 16-18). ‘Ihe C L I I C investigations were carried out at temperatures up to
953 K , atmospheric pressure and a WIiSV of ca. 3 h-I. The feed ratio o f methanol-to-n-butane was 3 : l . A3
expected, from the n-butane pyrolysis results, the n-butane
conversion during CWHC was lower than that with solely aluminiumcontaining sample. The conversion level (ca. 40 wt.-%), however, remained nearly constant during time on-stream (Pi&. 4 2 ) .
Uain
products were lower olcfins and, beside a large amount of carbon monoxide, a defined yield of carbon dioxide was obtained. Apart
322
from this, t i l e yliids of the wanted products were low and caused a l o w cracking activity of these zeolite types under C L H C conditions (Fig. 4b).
l'ig. 4. n-Butane conversion (a) and olefin yields (b) on sample E (silicon-to-iron = 2 5 ) at 9 5 9 K in C I i H C reaction The use of zeolites containing both iron and aluminium effected the supposed rising of the hydrocarbon conversion level shown during the investigations of n-butane pyrolysis. Already after 9 0 minutes time on-stream the n-butane conversion on silicon-ironcluniiiiium zeolite (sample C) was higher than o n silicon-aluminium zeolite (sample A )
(ref. 17).
After four hours a significant de-
crease in hydrocarbon conversion on silicon-aluminium zeolite was observed, whereas the decrease on silicon-iron-aluminium zeolite was very l o i r (I'ig.5a).
The decrease in n-butane conversion led
also to a decrease in the yield of olefins formed (sample A),
as
shown in lig. 5b, but the olefin yields obtained on sample C were nearly constant after three hours time on-stream. These data reflect the more stable hydrocarbon conversion o n iron-containing zeolites and the higher level of hydrocarbon conversion and olefin production on thc silicon-iron-aluminium catalyst as well. Generally, the stable conversion rate connected with stable olefin yields caused the prevention of catalyst deactivation in the C6:HC reaction by incorporation of iron into zeolite. The consequence is the prolongation of' catalyst lifetime.
323
3
6
9
tlhl
. N
u
3
6
9
tlhl
Fig. 5. n-Butane conversion (a) and olefin yields (b) on sample C (silicon-to-iron-to-aluminium =50:1:1) at 953 K in CMHC reaction in comparison with sample A (silicon-to-aluminium = 2 2 ) Regeneration o f iron-containing ZSM-5 type zeolites During the CICHC investigations on iron-containing zeolites a distinct decrease in coke formation in comparison with siliconaluminium zeolites was observed, also found by Inui during methanol transformation tests (ref. 10). Furthermore, a higher amount o f carbon dioxide was formed during the reaction and a lower
amount of water was determined than calculated. Iron-oxide phases formed during the reaction by a nearly complete withdrawal of framework iron should stimulate gasification processes simultaneously (known also from water-gas shift reactions) (refs. 29,30). These points led to considerations concerning a regeneration method only by steam. During decoking steps by steam and argon as carrier gas at ca. 953 K the deposited coke could be removed up to 70 wt.-% (ref. 3 1 ) . A complete decoking could be obtained by a followed decoking by air. Probable reasons for a lower deposit of coke during methanol transformation, CMHC and the gasification activity of the used zeolites during decoking are a lower amount of reobtained water and higher yields of carbon dioxide compared to solely aluminium-containing zeolites. CON C LUS I O N S
Iron- and iron-aluminium-containing 281-5 type zeolites were used as catalysts for the CMHC reaction as well as for single reactions of methanol conversion and n-butane pyrolysis. During the CMHC reaction with the above zeolites a lower decrease in cracking
activity could be obtained together with a prolongation of the catalyst lifetime. One reason for this fact could be due to the
324
lower acidity of these catalysts compared to aluminium-ZSM-5, and to the lower deposit of coke by burning the deposits on formed iron oxide phases.
A CKN 0W L E DG E ME N T The authors would like to thank I r s . H.Poethke and I.Irs. U.Hahn f o r their assistance in the experimental work and Dr. R.Liick f o r
the EPR measurements. REFERENCES 1 S.L.Lleise1, J . P.f-;cCullough,C. H..,echthaler and P.B.Weisz, CHEICTECH, 6(1976/2) 86 2 T.Inui and Y.Takegami, Hydrocarbon Process., (1982/11) 1 1 7 3 C.D.Chang, Catal. Rev. - Sci. Eng., 25 (1983) 1 4 Y.C.IIu, Hydrocarbon Process., (1983/5) 88 5 T.Inui, T.Ishihara, N.Eorinaga, G.Takeuchi, H.1-latsuda and Y.Takecami, Ind. Eng. Chem. Prod. Res. Dev., 2 2 (1983) 26 6 ".Fleckenstein, K.Eelendorff and P'.Petting, Chcri.-1nz. -Tech. , 57 (1 985/9) 800 ?/ R.Szostak, V.llair and 'I'.~.Thomas,3 . Chem. SOC., Paraday Trans. 1 , 8 2 (1587) 487 8 T. Inui, O.Yamase, I:.Yukuda, A. Itoh, J. Tarumoto, El.lCorinaga, T.€lagiwara and T.Takegami, Proc. VIIIth Int.Congr.Catalysis, kestberlin, Verlag Chemie, Weinheim, 1984, Vol. 111, p. 569 9 B.Kichterlova, S. Ceran, S.Bednarova, K. Nedomova, L. Dudiltova and P.Jiru in P.J.Grobet, W.J.Eortier, E.F.Vansaut and G.Schulz-Ekloff (Eds.), Studies in Surface Science and Catalysis, Llsevier, Amsterdam, 1988, Vol. 37, p. 199 10 T.Inui, Proc. IInd Symposium on C1-Chemistry Japan-GDR, Uerlin, 1968, p. 23 11 T.Inui, A.b;iyamoto, H.Iiagata, Y.l{akino, IC.E'ukuda, II.Matsuda and E'.Okazunii in Y.Murakami, A.Iijima and J.W.Hard (Eds.), Proc. VIIth 1nt.Zeolite Conference, Tokyo, Kodansha (Tokyo) Elsevier (Amsterdam, Oxford, New York, Tokyo), 1986, p. 862 12 T. Inui, H.I#latsuda, O.Yamase, A.Nagata, K.Fukuda, T.Ultawa and A.Miyamoto, J. Catal., 98 (1986) 491 13 P.Ratnasamy, React. Kin. Catal. Lett., 35 (1987/1-2) 219 14 S.M.Csicsery, Pure 8 Appl. Chem., 58 (1986/6) 841 15 V . P . Holderich, Pure ci Appl. Chem., 5 8 (1986/10) 1383 16 S.Nowak, B.Giinsche1, A.Martin, K.Anders and B.Liicke in M.J.Phillips and !(.Ternan (Eds.), Proc. IXth Int.Congr. Catalysis, Calgary, The Chemical Institute of Canada, 1988, VOl. IV, p. 173s 17 A.Eartin, S.Nowak, B.Liicke and H.GUnsche1, Appl. Catal. 5 0 (19b9) 149 18 S.Nowak, H.Giinsche1, J.Lantzsch and K.Anders, Neftechimia, XXVII (1987/6) 736 19 S.Nowak, Preiberger Yorschungshefte, A 761 (1987) 116 20 S.I'Jowak, Ii.Giinsche1, J.Lantzsch, B.Liicke, K.Anders, J. Jonsch, 11. I:ii r t ig , :I.€10 s e , I:'. R o s c he r , U Had i c k e and I<. Ilehn e r , DD-VP 230 545 (1983) 21 A.I,;artin,B. ~iicke,S.Nowak, H.Poethke, K.Anders, H.Giinsche1, !I. F'ii r t ig , U Had i c ke , W H6 s e and W R o s che r , DD-V!P 260 060 (1987)
.
.
.
.
325
22 A.Martin, S.IJowak, B.Liicke, W.Wieker and B.Fahlke, Appl. Catal., 57 (1990) 203 23 R.Szostak and T.L.Thomas, J. Catal., 1 0 0 (1986) 555 24 P.Ratnasamy, R.B.Borade, S.Sivasanker, V.P.Shiralkar and S.G.Hegde, Acta Phys. Chem., 31 (1985/1-2) 137 25 B.Borade, Zeolites, 7 (1987) 398 26 A.N.Kotasthane, V.P.Shiralkar, S.G.Hedge and S.B.Kulkarni, Zeolites, 6 (1986) 253 27 B.Wichterlova, Zeolites, 1 (1981) 181 28 A.Lartin, U.Wolf, S.Nowak and B.Liicke, Zeolites, in press 29 G.Baron, E'.Bieger and C.Lohmann in J.Falbe (Editor), Chemierohstoffe aus Kohle, Georg-Thieme-Verlag (Stuttgart),
1977, p . 114 30 B.Wichterlova, L.Kubelkova and J.Novakova in D.Kallo and Kh.M.Minachev (Eds.), Catalysis by Zeolites, Akaderniai Kiado, Budapest, 1988, p . 313 31 A.hartin, S. Peter, B.Liicke, S.l,Jowak, W.Wieker, B. Fahlke, U.Hahn, K.Anders and H.Giinsche1, DD-WP Anm. 331 798.6
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
327
DISPERSION DEPENDENT SELECTIVITIES OF SYNGAS CONVERSION ON FAUJASITE ENCAPSULATED Pt, Pd OR Ir
N.I. JAEGER, G. SCHULZ-EKLOFF and A. SVENSSON Institut fiir Angewandte und Physikalische Chemie, Universitzt Bremen, 2800 Bremen 33
ABSTRACT Metal phases of fine as well as coarse dispersions for Pd. Pt or Ir are grown within a NaX matrix. The mean particle sizes obtained under nearly identical reduction-agglomeration conditions follow the order Pd > Pt > Ir. For comparable metal dispersions the activities for syngas conversion or methanol formation as well as the selectivities for methanol formation exhibit the order Pd > Pt > Ir. A fraction of carbonaceous species deposited during catalysis can be removed by hydrogen as methane, whereas oxygen is needed for the conversion of the residual coke fraction to carbon dioxide.
INTRODUCTION It was found, that the methanol formation activities and selectivities of Pd, Pt or Ir on silica gel follow the order Pd > Pt > Ir [ref. 11. It was suggested that this order corresponds to the tendency of the metal to fill its d-shell [ref. 21. Increasing activities and selectivities for methanol formation with increasing metal dispersion is reported for Pd on silica gel [ref. 31 and Pt in faujasite X [ref. 41. A metal dispersion effect was found for the water-gas shift reaction (WGSR) on Pt/NaX, which was analogous to that one obtained for the methanol formation Iref.51. In the following the methanol formation activities and selectivities of faujasite X supported Pt, Pd or Ir catalysts, exhibiting different metal dispersions, are evaluated to clarify the different influences of metal type or metal particle size. Additionally, the amounts of coke deposited by the reaction are determined.
328
EXPERIMENTAL PreDaration of the catalysts Iron-free faujasites NaX (Si/Al ratio 1.2) were prepared by hydrothermal crystallizations. The metals were introduced as the ammine-ion-complexes from aqueous solutions of the chlorides ([Pt(NHa)4]2'; [Pd(NH3)4]2+; [1r(NIi3)~ClJ2*)The loaded zeolites were washed until chloride-free and dried at 353 K. The autoreduction of the specimens was carried out using a temperature program (2-3 K/min) up to temperatures between 573 and 673 K in either streaming argon or oxygen. In the case of samples treated under oxygen the reduction was completed with hydrogen at 373 - 573 K for 1 h. The materials investigated, the reduction treatment and the surface average diameter, d = Inid13/Znid_Lg of the metal crystallites are listed in Table 1. The surface average diameter ds was calculated from the histograms in order to represent the contribution of the surface more adequately. In the formula ni is the number of crystallites within the diameter di and di+dd while di = &+Ad/2. The histograms of the metal dispersions listed in Fig. 1 were evaluated from electron micrographs. The number of metal surface atoms/g metal was calculated from the particle size distribution. A sphere was taken as a justifiable approximation for the observed crystallites and the specific gravity of the bulk metals was used. The evaluated metal surfaces were used for the calculation of the dispersion D and the normalized activities or turnover frequencies, respectively. The global crystallinity of the samples was found to be maintained even in the case of metal particles growing beyond supercage dimensions within the zeolite matrix [ref.5].The metal loading in atom-% was kept constant for all catalysts in order to maintain a constant Bronsted acidity which results from the complete reduction of the metal ions.
Catalytic measurements The syngas conversion to methanol and methane was studied in a continuous flow stainless steel reactor (volume: 1.6 cm3). A fluidized bed was established by means of an external vibrator.
329
TABLE 1 Composition, reduction, surface average diameter D of the catalyst. Material Metal loading
and dispersion
Reduction treatment
L
D
nm
atom-%
wt-%
PtX/O2
g.aAo.2
5.9
heated to 673 K at 2-3 K/min in flowing 0 2 , held 5 h, HZ passed at 513 K for 1 h
1.8
0.7
PtX/Ar
g.ako.2
5.9
heated to 613 K at 2-3 K/min, in flowing Ar, held 5 h
5.2
0.3
PdX/Oa
10f0.2
3.4
heated to 573 K at 2-3 K/min in flowing 0 2 , held 5 h, HZ passed at 313 K €or 1 h
3.1
0.4
PdX/Ar
10A0.2
3.4
heated to 513 K at 2-3 K/min in flowing Ar, held 16 h
11.2
0.15
IrX/02
10k0.2
6.0
heated to 593 K at 2-3 K/min in flowing 0 2 , held 11 h, HZ passed at 563 K €or 1 h
1.0
0.95
IrX/Ar
10A0.2
6.0
heated to 593 K at 2-3 K/min in flowing Ar. held I h
2.1
0.5
The experiments were carried out at 2 MPa total pressure with H2/CO ratios between 1.5
-
4, temperatures between 523
-
623 K and
a space velocity (STP) of 9315 h-1. 500 mg of the pressed and granulated catalysts (grain size 0.355-0.71
nun) were used in the
experiments. Analysis of the reaction products was carried out by on-line gas chromatography. The concentrations of COZ and CH4 were monitored by infrared analyzers.
Characterization of the carbon deposits Consecutively, hydrogen as well as oxygen were passed through a glass reactor (2 ml), containing a used catalyst (350 mg). Normal pressure and a space velocity around 640 h-' were applied. The
330
Fig. 1. Histograms of particle size, metal mass and metal surface. The bars represent integer numbers. Sections of representative transmission electron micrographs are inserted.
heating rates were 5 K/min, and the final temperature was 723 K. The evolutions of methane under hydrogen and of carbon dioxide under oxygen were monitored by infrared analyzers.
331
RESULTS Metal disDersions The metal phases are located inside the faujasite matrix, as could be gleaned from the uniform distribution of the metal particles in the specimens for electron microscopy obtained by ultrasonic fragmentation of zeolite crystals, and as was assured by photoelectron spectroscopy [ref. 61. The growth of particles beyond supercage dimensions is accompanied by local zeolite framework fragmentations and reconstructions, as evidenced by *qSi-NMR spectra [ref. 71 and adsorption isotherm analysis [ref. 51. The recognized mesopore texture, generated in the metal phase containing zeolite matrix, discloses that the metal particles are surrounded by halos of free space and, thus, are fully accessible for chemisorbing and reacting molecules. The found surface average mean particle sizes follow the order Pd > Pt > Ir for both types of applied reduction conditions (Table 1). The reverse order is valid for the heats of vaporization or
heats of fusion [ref. 8 1 , i.e.
the energy of atom abstraction from
metal clusters. Thus, the tendency for sintering by an Ostwald ripening mechanism should follow the order given above. Sintering by particle migration and coalescence can be excluded for faujasite accomodated metal phases. Only slight increases of the mean particle sizes are observed for the used samples as compared to the fresh ones. Svnaas conversion The results of the syngas conversion (cp. Table 2 ) show several significant features. Firstly, the turnover frequencies of syngas conversion, i.e. for the sum of methane and methanol formation, follow.the order Pd > Pt > Ir at comparable dispersions of the metal. Secondly, for the metals exhibiting high activities, i.e. Pd and Pt, the turnover frequencies increase with increasing metal dispersion. Thirdly, the metal type effect as well as the metal dispersion effect are even more pronounced considering the methanol activities separately. Furthermore, the increase of the methanol selectivity with increasing metal dispersion is valid for all metals studied here. Identical features, i.e. the metal type effect as well as the metal dispersion effect, are observed for the activities of carbon dioxide formation resulting from the simultaneous WGSR. Traces of higher hydrocarbons and dimethylether are found in the product spectrum but are not considered further.
332
TABLE 2 Turnover frequencies TOF (mole of product/gram-atom of metal surface atoms-s),selectivities S ( T 0 F (methanol)/TOF (methane)) and relative selectivities S r e I (selectivity of fine dispersion/selectivity of coarse dispersion) for the syngas conversion at H s / C O = 2.33, 573 K, 2 MPa, SV (STP) = 9375 f 75 h-1 after 250 sin time-on-stream.
4L
TOF ( MeOH )
nm
x 104
PtX/On
1.8
PtX/Ar
TOF (CHI x 103
S
6.8
1.8
0.38
5.2
3.0
4.3
0.07
PdX/Os
3.7
21.0
1.9
1.10
PdX/Ar
11.2
6.1
2.7
0.23
Sample
Sre I
5.4
4.8
IrX/02
1.0
0.5
0.4
0.13
IrX/Ar
2.7
0.9
0.9
0.1
1.3
The found features are valid for all reaction parameters where the kinetic regime is preserved, i.e. where differential conversions and sufficient distances from the thermodynamic equilibrium of methanol formation [ref. 91 are maintained. For instance, the metal dispersion effects become more prominent with increasing pressure or space velocity. For the methane formation the kinetic regime is maintained in any case, due to the sufficient distance from equilibrium [ref. 91 under the chosen reaction conditions. This means, that the mutual influence of the extent of methane formation and the extent of methanol formation can be neglected [ref. lo]. Furthermore, methanol and methane are formed on different sites by two independent routes, presumably [ref. 31. The syngas conversion on fresh catalysts is always characterized by induction periods in the beginning of the reaction, as has repeatedly been reported previously [refs. 3, 11, 121.
Deposition of carbonaceous species The fraction of carbon deposits which can be removed by hydrogen show a relatively simple pattern in the temperature
333 programmed reduction spectrum (Fig. 2a). although it comprises the removal of carbidic carbon from the metal surface [ref. 131 as well as hydrogenation of carbonaceous species on the support by spillover hydrogen [ref. 141. The temperature programmed
PIX/O,PlX/Ar
T b - I 0.2
I
,--
PdX/O,-
!
PdX/Ar
0
----
----
U
,\" 75 >
0.1-
0 -*
r-'
I
I
1
- ---
#
/
----+
\
Fig. 2. a: Temperature programmed hydrogenation of coke deposits to methane on different Pt dispersions. b: Temperature programmed evolution of COZ from oxidation of the residual coke not removable by hydrogen on different Pd dispersions.
334
oxidation of the residual coke fraction, however, can exhibit a more complex pattern (Fig. 2b). This type of carbonaceous species is, presumably, located at the support exclusively [ref. 31. Both types of coke do not affect the activity of the syngas conversion with time-on-stream, contrary to observations reported for the methanol formation on faujasite Y supported Pd [ref. 31.
DISCUSSION
Obviously, faujasite X is a favourable support for the study of metal dispersion effects in catalysis, since the occluded metal particles have a low tendency for sintering, but are fully accessible to reaction mixtures. Although the relatively high Bronsted acidity favors the methanation selectivity in the syngas conversion [cp. ref. 31. the found metal type effect as well as the metal dispersion effect in the methanol formation activity and selectivity are parallel with analogous results obtained with the support silica gel [refs. 1.31. Strong metal-support interactions, like formation of new phases or deposition of zeolite fragments on the metal phase, were never observed for the applied metal/faujasite X system under the reduction and catalytic conversion conditions used here. The observed effects can, therefore, be referred to the variations of the metal type or metal dispersion only. It is interesting to note, that the metal dispersion effect can surmount the metal type effect, e.g. the fine Pt dispersion is more active in the methanol formation than the coarse Pd dispersion (cp. Table 2). Up to now, the effects of metal type or metal dispersion in syngas conversion activities or selectivities are poorly understood. Won-dissoziative chemisorption of carbon monoxide. which might be a prerequisite for methanol formation, is found on each of the three metals. The postulates on the participation of Men+ surface sites in the syngas conversion [ref. 151 and the influence of the COz/CO ratio on the density of Men+ surface sites [ref. 161 cannot be falsified, since the methanol activities and selectivities vary analogously to the corresponding WGSR. An influence of the carbonaceous deposits on the catalytic effects, as has been suggested elsewhere [ref. 41, is still uncertain, since the fraction of carbon located on the metal surface only, could not be evaluated separately.
335
ACKNOWLEDGEMENT Financial support by the Deutsche Forschungsgemeinschaft is gratefully acknowledged. We are indebted to Drs. A. Kleine and R. Lamber for taking the electron micrographs and to Mrs. U. Melville for carryingout the temperature programmed hydrogenation and oxidation of the coke on the used catalysts.
REFERENCES 1 2 3 4
5
6 7 8 9 10 11 12 13 14 15 16
M.L. Poutsma, L.F. Elek, P.A. Ibarbia, A.P. Risch and J.A. Rabo, J. Catal. 52 (1978) 157. K. Klier, in "Catalysis of Organic Reactions" (W.P. Moser, Ed.), Marcel Dekker, New York 1981, p.195. F. Fajula, R.G. Anthony and J.H. Lunsford, J. Catal. 73 (1982) 237. N.I. Jaeger, G. Schulz-Ekloff and A. Svensson, in "New Developments in Zeolite Science Technology" (Y. Murakami, A. Iijima and J.W. Ward, Eds.), Elsevier, Amsterdam 1986; Stud. Surf. Sci. Catal., vol. 28, p. 923. N.I. Jaeger, J. Rathousky, G. Schulz-Ekloff, A. Svensson and A. Zukal, in "Zeolites: Facts, Figures, Future" (P.A. Jacobs and R.A. van Santen, Eds.), Elsevier, Amsterdam 1986; Stud. Surf. Sci. Catal., vol. 49 B, p. 1005. G. Schulz-Ekloff, D. Wright and M. Grunze, Zeolites 2 (1982) 70. G. Schulz-Ekloff and N.I. Jaeger, Catalysis Today 3 (1988) 459. R.C. Weast, CRC Handbook of Chemistry and Physics, CRC Press, Brea Raton 1984/85. H. Landolt and R. Bbrnstein, Zahlenwerte und Funktionen II.Bd., 4. Teil (Kalorische Zustandsgrbpen), Springer, Berlin 1961. E.J. Henley and E.M. Rosen, Material and Energy Balance Computations, Wiley, New York 1969, p. 367 ff. M. Ichikawa and K. Shikakura, in "New Horizons in Catalysis" (T. Seiyama and K. Tanabe, Eds.), Elsevier, Amsterdam 1981; Stud. Surf. Sci. Catal., vol. 7 8, p. 925. N.I. Jaeger, A. Jourdan, G. Schulz-Ekloff, A. Svensson and G. Wildeboer, Chemistry Express 1 (1986) 697. A.T. Bell, in "Structure and Reactivity of Surfaces" (C. Morterra, A. Zecchina and G. Costa, Eds.), Elsevier, Amsterdam 1989; Stud. Surf. Sci. Catal., vol. 48, p. 91. W.C. Conner, in "Hydrogen Effects in Catalysis" (Z. Paal and P.G. Menon, Eds.), Dekker, New York 1988, p. 338. E.K. Poels and V. Ponec-, Catalysis - A Specialist Periodical Report, The Royal Society of Chemistry, London 1983, vol. 6, p. 196; and references therein. K. Klier, V. Chatikavanu, R.G. Hermann and G.W. Simmons, J. Catal. 74 (1982) 343.
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
337
CONTRIBUTION OF 13C NMR SPECTROSCOPY TO THE ANALYSIS OF SURFACE COMPOUNDS FORMED IN THE TRANSFORMATION OF ACETONE ON ZEOLITES V. BOSAEEK, L. KUBELKOVA and J. NOVLKOVL J. Heyrovsky Institute of Physical Chemistry and Electrochemistry Czechoslovak Academy of Sciences, 182 23 Prague 8, Czechoslovakia SUMMARY The formation of the protonized form of acetone on HZSM-5 and HY zeolites at room temperature has been demonstrated by high resolution solid state NMR spectroscopy. As the reaction temperature increases, this form undergoes gradual dimerization folbowed by further transformations. At temperatures of about 180 C cyclization and the formation of undesorbable compounds with polyalkylaromatic character take place. The coke formed mostly exhibits aromatic character; the appearance of the signal at 154 ppm (TMS) indicates the presence of oxygen bonded to an aromatic ring. The presence of either an ethereal bond or of bonding of the phenoxy groups formed with the participation of lattice oxygen is suggested. INTRODUCTION Acetone can be converted to various hydrocarbons over the acid forms of zeolites(refs.1-3). These reactions are known to take place via aldolization and dehydration followed by cyclization, aromatization and cracking (refs. 4-7). Analysis of the mechanism of these reactions in the liquid phase revealed that the protonized form of acetone is formed in acid solutions; the enol form is converted to diacetone alcohol (DAA). After dehydration, this compound is then converted to mesityloxide (MO). Recent studies have shown that a similar mechanism probably occurs in the heterogeneously catalyzed transformations of acetone over acid catalysts (refs. 4-6). Various spectroscopic techniques have been employed in the investigation of the mechanisms of catalytic reactions on solid phase surfaces.The development of high resolution NMR techniques for solids has also began to permit the analysis of adsorbed complexes with low mobility on the catalyst surface (refs. 8-11). In this study, acetone adsorbed on decationated zeolite surfaces was studied at various temperatures by 13C HR NMR in order to identify the surface species which could take part in the reaction mechanism. To obtain information about the intermediate
338
products of the reaction, the samples were analysed at various time intervals both in the static and/or the dynamic ( flow ) arrangement. EXPERIMENTAL Materials Decationized samples of Y and ZSM-5 zeolites differing in their Si/Al module were used. The samples were supplied by the Research Institute of Oil and Hydrocarbon Gases (WRUP), Czechoslovakia. The ammonium form NH4 ,Nay-70 was decomposed in vacuum at 35OoC, HZSM-5 zeolites with Si/Al ratio equal to 13.5 and 22.5 were prepared from the Na parent zeolite by decationization with 0.5 M HN03 and calcination at 550OC. The samples for the NMR measurements were activated overnight in a vacuum at 550OC and transferred in sealed ampouls to the adsorption apparatus, where acetone was adsorbed in defined quantities. The ampoules were then resealed under vacuum They were openeded prior to the NMR measurements under an inert atmosphere ( N2) and transferred into the ceramic rotors for NMR measurements. Technique The HR NMR spectra were measured with a BRUKER MSL-200 spectrometer equipped with magic angle spinning (MAS) technique. As the study was carried out to investigate the adsorbed molecules with restricted mobility at room temperature, the spectra were measured by the cross-polarization method (CP) which is efficient in the investigation of organic solid substances by 13C HR NMR (ref.11). This technique has a favourable effect, not only on the signal intensity of 13C nuclei, but also on their relaxation. It was found from the dependence of the signal intensity of the spectrum on the contact time that the optimum contact time for the energy transfer from H ' to I3C (under the Hartman-Hahn conditions) was 2 - 3 ms. The rr/2 pulse for excitation (5-6 u s ) and 0.05 s aquisition time were employed. The number of scans required to obtain spectra with an acceptable signal/noise ratio depended both on the adsorbed amount and on the isotopic composition of the acetone used. For a small adsorbed amount of acetone, 13C-enriched acetone-2-13C was employed, as indicated in the figure captions. 13C CPMAS NMR spectra were measured at 50.32 MHz with 20 or 25 kHz band width. The repetition delay was 4 s , and 10 s in some cases. The ceramic rotor spinrate
.
339
was 4.7 - 5.0 kHz. The data obtained by IR spectroscopy and mass spectrometry were also included in the discussion. The details of these techniques and results are described elsewhere (ref.12). RESULTS and DISCUSSION Adsorption of acetone at room temperature We were especially interested in the interaction of acetone with the highly acidic structural OH groups of HZSM-5 and HY zeolites under conditions preceding the chemical reaction.The 13C CPMAS NMR spectra of adsorbed acetone were measured with samples containing various adsorbed amounts, as the dependence of the signal position on the adsorbed quantity was expected. This effect should appear as a result of an exchange process as described by Bernstein et al.(ref.l3 ) . for acetone on silica gel. Spectrum (a) in Fig.1. for the small adsorbed quantity N=0.2 mmo1.g-’ of adsorbed, isotopically enriched acetone-2-13C (96%) exhibits an intense signal at 224 ppm (TMS), a weak signal at 209.9 ppm in the carbonyl nuclei NMR region and a signal at 28.9ppm in the -CH region. With increasing acetone adsorption (0.2-2.7 mmo1.g -13) primarily the intensity of the signal at 209.9 ppm is increased whereas the signal at 224 ppm becomes only a shoulder and, in addition, a further weak shoulder appears at 204 ppm (see Fig. 2.).
For samples with higher adsorbed quantities of acetone, further signals appear in the spectrum at room temperature, indicating slow chemical transformation of acetone in the adsorbed phase. When acetone has been adsorbed on zeolite for two months, the original signals of physically adsorbed acetone (209.9 and 3 0 ppm) and its protonized form (224 and 29 ppm) were accompanied by new signals indicating dimerization with the formation of DAA 1 characterized by signals at 210 ppm ( X = O ) , 73.5 ppm (-C-OH), I 56.6 ppm (-CO-CH2-CO), 34 and 30 ppm (-CH3), as well as the formation of dimers of the acetylacetone (AAC) keto-enol tautomer type, characterized by a broad signal at 187 ppm (enol and keto *I *I form = C-OH and -CH2- C=O, respectively) , 93.7 ppm (=*CH-), 56.6 ppm (-CH2-) and 2 8 ppm (-CH3). This dimerization evidently takes place because of the formation of the protonized form of acetone, indicated by a signal at 224 ppm (see Fig.l., curve (a)). For low adsorption of 0.2 mmo1.g-’ acetone on zeolite containing 1.0 mmo1.g-I of acid centers, the signal position should be characteristic for the protonized acetone, even if an exchange
340
a
b
s
300
I
1
8
200
100
I
s
Fig.1. I3C CPMAS NMR spectra of adsorbed acetone at 50.32 MHz on HZSM-5 zeolite (a) 0.2 mmol/g of acetone-2-13C(96%)at 25°C (b) spectrum after admission of NH at 130°C (acetone/NH3 ratio = 2:l ) , NS=270 , (c) spectruii as for (b) but for acetone/NH ratio = 1:l * spinning sidebands, spectra normalized to the highest peak with CY=lO units.
341
4
value of 250 ppm for process takes place (ref.13). The protonized acetone in a superacid published by Olah et al. (ref.14) is higher than the value observed by us, however, this discrepancy can be interpreted in terms of the influence of the interaction of the protonized form with the zeolite lattice. IT addition , this difference could be associated with proto mobility resulting from the equilibrium (CH,),+C-OH + 0 Z ' (CH3) 2C=O---HO Z , but the relatively narrow signal at 224 ppm ar its invariability with the adsorbed amount do not support this explanation. Further evidence for the correspondance of the signal at 224 ppm to the protonized acetone form is the discovery that a reaction with protonized molecules occurs after the admission of ammonia in various quantities; the intensity of the signal at 224 ppm gradually decreases (see Fig. l., curves (b), (c)) and a new signal appears at 200 ppm. In addition, the -CH3 signal intensity at 29 ppm also decreases and a new high-field- shifted signal appears at 23 ppm. As we have already noted, both new signals are in very good agreement with the signals of the dimethylimmonium cation described by Olah et al.(ref.l5). This species is also probably created during the interaction of NH3 with protonized acetone (ref.16). The chemical transformation of acetone at temDeratures above 25OC The processes already detected at room temperature are accelerated by heating the zeolite samples with adsorbed acetone. At the same time, new mechanisms come into action, as can be seen from the spectra in Fig.2. Spectrum (b) measured for a sample after treatment at 18OoC for 1 h indicates further chemical transformations. This is mainly a significant decrease in the intesities of the signals corresponding to the various forms of acetone (224 and 210 ppm) .The disappearance of the signals which are typical for DAA (73.5 ppm) and AAC (93.7 ppm) also indicates their transformation into other types of hydrocarbons. The dehydration of DAA to MO ( observed also by IR spectroscopy - see ref.12) seems to be probable, the presence of the latter being indicated by the signals at 197.0 ppm (>C=O), 153.0 ppm ( >* CP =) and probably by a weak signal hidden in the shoulder of the signal at 124 ppm ( = C " < ) . The signal of this a-carbon would be weak, as it is created from a carbon type which has not been enriched with the 13C isotope.
342
a
~
. 300
Fig.2.
.
.I
.
. . . . , .
200
...
1
.
100
.
.
.
1
.
...
ppm
13C CPMAS NMR spectra at 50.32 MHz on (a) 2.5 mmol/g of acetone-2-13C(lO%) days,after 2 months, NS = 560 (b) sample (a) after heating at 180°C (c) sample (b) after heating at 280°C ( normalized spectra as in Fig.1.).
1
.
.
.
.
1
.
,
0
HZSM-5 at 25'C
....after
for lh, f o r 0.5h, NS = 2 5 6 0
2
343
Another interesting signal in the N M R range of carbonyl compounds is the broad signal with a maximum at 177 ppm which could indicate the formation of a carboxylic compound, e.g. acetic acid, or which could correspond to Some form of a cyclic ketone (isophorone?) The signals at 137 and 129 ppm are characteristic for simple aromatic compounds: their presence has also been demonstrated in the desorption products. Spectrum (c) in Fig.2. obtained for the same sample after further heating to 28OoC for 0.5 h corresponds to the formation of alkylaromatics. In this spectrum, the signals at 138.5 and 129.3 ppm predominate and are very close to those exhibited by trimethylbenzene and similar alkylaromatics (ref.17). The high-field-shifted signal of -CH3 is also evidence for this fact. Both the signal at 180 ppm and the methyl signal at 20 ppm most probably correspond to the acetic acid which was found by mass spectrometry in the desorption products. Similarly, C7 - Cll aromatics were also found in the desorption products. Isobutene was not found in the adsorbed state while it appeared in the desorbable gaseous products.
.
Analysis of the coke deposits after reaction of acetone Similar problems to those ocurring in the analysis of coal and other fossil fuels (refs.18-20) are encountered in the analysis of coke deposits on catalysts. The distinction between aromatic and olefinic hydrocarbons represents an especially difficult problem as they exhibit NMR signals in an overlapping region at 110 - 140 ppm. In this respect, the combination of the chemical treatment of coke deposits with 13C NMR analysis seems to be useful (ref.21); this combination permits us to distinguish between olefinic and aromatic bonds on the basis oftheir different reactivities. The spectra depicted in Figs.3 and 4 illustrate the difference between coked HZSM-5 and HY zeolites. The broad and poorly resolved band at 130 ppm is usually attributed either to aromatic compounds or to polyenic chains. In addition, with both types of zeolite signals in the range of paraffinic carbons (-CH3,-C2H5) bonded to the aromatic rings are observed as follows from the high-field-shifted value with a maximum at 19.8 ppm. HZSM-5 zeolite differs from HY in a clearly separated broad signal at 154 ppm, which has also been observed by other authors on coked zeolites (refs.19-21). However, the interpretation of this signal is not yet clear: some authors assigned it to carbocations
344
20 0
100
0 PPm
Fig.3. 13C CPMAS NMR spectra of coke deposits after acetone reaction on Y-type zeolites at 50.32 MHz : (a) 28 w.% of coke on H,NaY 70 zeolite, (b) 16.0 w.% of coke on HY dealuminated zeolite, NS=8000, (c) 13.9 w.% of coke after pyrolysis of the sample (b) at 600°C for 2h, NS=8000. (normalized spectra as in Fig.1.)
K130*8
19,8
Fig.4. 13C CPMAS NMR spectra of coke deposits after acetone reaction on HZSM-5 zeolite at 50.32 MHz: (a) 1.0 w.% of coke after acetone-1,2,3-13C(lO%) reaction, NS=240, (b)sample as ad (a) after partial oxidation at 200°C , ( c ) sample (b) after pyrolysis at 600°C for 2h, NS=7000.
345
(ref.l9), others to olefines (ref.20). In order to solve this problem, we have carried out a number of experiments on the chemical treatment of coke deposits. Partial oxidation (see spectra in Fig.3,4) at temperatures higher than 5OO0C did not lead to any change in the spectrum. Therefore, the creation of carbocations does not seem probable. The easy addition of bromine to olefinic bonds led us to perform bromination experiments with bromine vapours at 25 and 100°C. As the effective radius of bromine permits its adsorption in the microporous structure and no significant effect of bromination in the NMR spectra was observed, we concluded that the mentioned signal cannot be associated with olefins. The pyrolysis of the deposits in vacuum at 6OO0C yielded only evidence that the dealkylation took place (the signal at 19.8 ppm disappeared) but did not lead to any change inthe signalat 154 ppm. We assume (on the basis of its resistance to chemical treatments) that this signal could be associated with carbons belonging to an aromatic ring bonded to oxygen, as is true of phenols, cresols or ethers (ref.l7).We suggest that phenoxy groups are formed on the surface of HZSM-5 zeolites, analogous to the surface methoxy groups (ref.22-26). For the wide pore Y zeolite, the condensation of the aromatic rings is preferred, as is indicated by the relatively strong shoulder at 140 ppm, whereas the signal at 154 ppm is almost indistiguishable from that of HZSM-5.
.
CONCLUSIONS The 1 3 C CPMAS NMR represents a significant contribution to the solution of problems connected with less mobile chemisorbed species which can participate as intermediates in the transformation of acetone. In combination with chemical treatment of catalysts with deposits the NMR spectroscopy can help considerably in elucidating problems in the formation and analysis of coke residues on zeolites. REFERENCES 1 2
3
C.D. Chang and A.J. Silvestri, J.Cata1. C.D. Chang, W.H. Lang and W.K. Bell, Catalysis in Organic Reactions, M. 1981,p.73. Y. Servotte, J. Jacobs and P.A. Jacobs, on Zeolite Catalysis, Siofok 1985, Acta Szegediensis, Szeged, 1985, p. 609.
47 1977) 249. in: W.R. Moser (Ed.), Dekker, New York, in: Proc. Int. Symp. Physica et Chimica
346 4
J. Novakova, L. Kubelkova and Z. Dolejsek, J. Molec. Catal.,
5
L. Kubelkova, J. Cejka, J. Novakova, V. Bosacek, I. Jirka and P. Jiru, in: P.A. Jacobs and R.A. van Santen (Eds.), Zeolites, Facts, Figures, Future, Elsevier, Amsterdam, 1989, p. 1203. V. Bosacek and L. Kubelkova, Zeolites 10 (1990) 64. L.M. Baigrie et al., J. Am. Chem. SOC., 107 (1985) 3640. H. Pfeifer, W. Meiler and D. Deininger, Annual Rep. NMR Spectroscopy, 15 (1983) 291. G. Engelhardt and D. Michel, High Resolution Solid State NMR of Zeolites and Related Systems, J.Wiley & Sons, 1987, Chap.7. D. Freude, in: Advances in Interface Sci., Elsevier, 23 (1985)
39 (1987) 195.
10
21 11
M.W. Anderson and J. Klinowski, J. Am. Chem. SOC., 112 (1990)
12 13
22
L. Kubelkova, J. Cejka and J. Novakova, Zeolites, in press. T. Bernstein, D. Michel and H. Pfeifer, J. Coll. Interfaces Sci., 84 (1981) 310. G.A. Olah and A.M. White, J. Am. Chem. SOC., 90 (1968) 1884. G.A. Olah and D.J. Donovan, J. Org. Chem., 43 (1978) 860. Z. Dolejsek, J. Novakova, V. Bosacek and L. Kubelkova, Zeolites, in press. H.O. Kalinowski, S. Berger and S. Braun, 13C NMR Spektroskopie, G. Thieme Verlag, 1984. C.E. Snape, D.E. Axelson, R.E. Botto, J.J. Delpuech, P. Tekely, B.G. Gerstein, M. Pruski, G.E. Maciel and M.A. Wilson, Fuel, 68 (1989) 547. J.P. Lange, A. Gutsze, J. Allgeier and H.G. Karge, Appl. Catal., 45 (1988) 345. . E.A. Lombardo, J.M. Dereppe, G. Marcelin and W.K. Hall, J. Catal., 114 (1988) 167. L. Carlton, R.G. Coppertwhite, G.J. Hutchings and F.C. Raynhardt, J. Chem. SOC. Chem. Commun., (1986) 1008. V. Bosacek and Z. Tvaruzkova, Coll. Czech. Chem. Commun., 36
23 24
P. Salvador and J.J. Fripiat, J. Phys. Chem., 79 (1975) 1842. P. Salvador and W. Kladnig, J.C.S. Faraday Trans.1, 73 (1977)
10.
14 15 16 17 18 19 20 21
(1971) 551. 1153. 25
G. Senkyr and H. Noller, J.C.S.
Faraday Trans.I.,71
(1975)
997. 26
L. Kubelkova, J. Novakova and (1990) 441.
K.
Nedomova, J.
Catal.,
124
G . Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
341
ISOPROPYLATION OF BENZENE OVER LARGE PORE ZEOLITES
A.R. PRADHAN, B.S. RAO and V.P. SHIRALKAR Catalysis Group, National (India)
Chemical Laboratory, Pune 411 008
ABSTRACT Isopropylation reaction of benzene is carried out over large pore zeolites with characteristic structural differences (namely La-H-Y, H-mordenite and H-ZSM-12). The activity and deactivation pattern are correlated with structural and acidic properties. The deactivation of La-H-Y is due to blocking of active sites while that o f H-mordenite is due to blocking of channels. The stable activity and selective nature o f H-ZSM-12 for cumene can be attributed to siliceous nature, lower acidity and presence of non-interpenetrating channels. INTRODUCTION Isopropylation o f benzene using solid phosphoric acid (SPA) catalyst and Fridel-Crafts catalysts (refs. 1-3) for the production of cumene is an industrially important reaction. The drawbacks suffered by these processes (environmental and corrosion) can be overcome by using solid acid catalysts like zeolites. Major side products formed in this reaction are isomeric diisopropylbenzenes (DIPB) and at higher temperatures n-propylbenzene (nPB). Even though the medium pore zeolite ZSM-5 is reported (ref. 4) a s a potential catalyst for this reaction, better stability and selectivity were observed over large pore zeolites (ref. 5). In view of this, isopropylation of benzene was carried out over large pore zeolites with characteristic structural differences, like (1) La-H-Y with cubic crystal symmetry and three directional channel system having pore opening of 7.4 k , (2) mordenite with orthorhombic structure and unidirect(12 MR) and 2.9 X ional dual pore system with 6.7 X 7.0 5.7 (8 MR) connected via side pockets of 2.9 i, ( 3 ) ZSM12 with monoclinic symmetry and linear non-interpenetrating channels o f 5.7 X 6.1 "A. The stability, selectivity and deactivation pattern were correlated with structural, acidic
348
and sorption properties of these are reported in this communication.
zeolites and
the
results
EXPERIMENTAL Materials Benzene (
X pure) and propylene (having 4
99.98
% propane)
were used for catalytic studies. Catalysts La-H-Y (SK-500) and H-mordenite from M/s
Carbide, USA
Union
and
(Zeolon 100) were procured Norton, USA, respectively.
ZSM-12 was prepared in this laboratory following the procedure reported (ref. 6 ) earlier in the literature. ation of zeolite beta was avoided.
Co-crystallis-
Characterisation the
Crystalline framework
techniques
phase purity and the state of aluminium in of these samples were characterised by the
XRD,
like
of the samples was ammonia. Adsorption
P/Po
=
0.5
using
sensitivity of
4'
IR
and
MASNMR
spectroscopy.
Acidity
measured by the irreversibly adsorbed studies were carried out at 25°C and
a
McBain
balance
with
silica
spring
of
50.0 cm gm-'.
Catalytic reactions The catalyst was pressed and crushed into 10-20 mesh binder free self supported pellets. Prior to catalytic runs, the catalyst was activated hrs.
Catalytic
bed,
down
Benzene while
was
runs were
flow, fed
propylene
in a flow of dry air at 45OOC for 8
silica
by was
a
carried
out
reactor
at
syringe pump
metered
through
in an integral, fixed atmospheric
(Sage a
pressure.
Instruments, USA)
mass-flow
controller
(Matheson, USA). The products were analysed by gas-chromatography (Shimadzu, Model 15A) using Apiezone L column for liquids and Poropak Q column for gaseous samples. RESULTS AND DISCUSSION The structural features, silicon to aluminium ratio, acidity values in terms of irreversibly retained ammonia and equilibrium sorption capacity for benzene are compared in Table 1.
TABLE 1 S t r u c t u r a l and physico-chemical
Catalyst
La-H-Y
Channel structure
H-ZSM-12
i
7.4
Unit c e l l crystal symmetry Si/Al
Cubic
total
5.7 x 7,l 1 2 . 9 X 5.7 ‘A (8 MR)
Monoclinic
Orthorhombic
60.5
0.535
9.8
13.2
20.3
is
acidity
6.4
0.067
1.338
Equil. sorption capacity for benzene (wt Z)
Unidirectional 8-membered i n t e r connecting
5 . 7 X 6.1
2.3
Acidity m mole o f NH3/gm
H-mordenite
Unidirectional l i n e a r non-inte r p en e t r at i n g
Three directional with interconnecting channels
P o r e opening (12-membered ring)
The
p r o p e r t i e s of c a t a l y s t s .
to
found
be
in
accordance
with
the
aluminium c o n t e n t o f t h e z e o l i t e , w h i l e t h e s o r p t i o n of b e n z e n e does
not
the
follow
same
trend
due
to
structural
the
differences. The
catalytic
propylene
over
performance the
all
in
three
alkylation
zeolites
has
of
benzene
been
with
compared
in
T a b l e 2. c a t a l y s t s showed
A l l
But
the
and
higher
and
impurities
of
in
than
200
catalytic
aliphatics,
fractions
after
even
(Fig.
hrs
fast
reaction
Due
were
studies
cumene
to
C8
toluene,
are is
conversions.
aromatics
more very
deactivation
in high
was
La-H-Y i n
the
noticed
w h i l e s t e a d y a c t i v i t y was o b s e r v e d
the
1).
propylene
(H.B.F)
A
catalyst.
and H-mordenite,
H-ZSM-12
X)
( )99
Selectivity
H-ZSM-12
i n La-H-Y
like
boiling
H-mordenite.
case
high
to
its
performed
was
performed
for
steady activity, over
H-ZSM-12
more
further
catalyst.
Influence of temperature In on At
Table
the
3
product
temperatures
not completed.
the
results
on
distribution below
200°C,
the over
the
influence H-ZSM-12
conversion
With t h e i n c r e a s e of
of
temperature
are p r e s e n t e d . of p r o p y l e n e i s
temperature,
a continuous
350 decrease i n the
DIPB formation
is noticed,
while
appreciable
q u a n t i t i e s o f nPB a r e o b s e r v e d a b o v e 230°C.
TABLE 2 I s o p r o p y l a t i o n o f benzene o v e r z e o l i t e c a t a l y s t s R e a c t i o n t e m p e r a t u r e = 230OC; P r e s s u r e = Atmospheric;-TOS = 3 h r s ; Benzene t o p r o p y l e n e molar r a t i o = 6 . 5 ; WHSV = 2 . 5 h r
Catalyst
La-H-Y
H-ZSM-12
Product d i s t r i b u t i o n (wt % ) Aliphatics Benzene T o l u e n e t C8 a r o m a t i c s Cumene nPB C9-C11 a r o m a t i c s DIPB H.B.F
0.74 77.10 0.97 18.52 0.40 0.29 1.75 0.18
0.06 77.80 0.04 20.50 0.02 0.04 1.55 0.01
0.30 78.30 0.27 18.10 0.74 0.12 1.78 0.26
C3 = c o n v e r s i o n Cumene s e l e c t i v i t y S e l e c t i v i t y (cumene t DIPB)
99.4 81.1 88.5
99.9 92.3 99.3
99.8 82.9 91.6
1./F
H-mordenite
t,
w h 19
a
H.ZSM-I2
* H MORDENITE rLa H Y
2
4
6 TIME
ON STREAM (hra)
F i g . 1 . C a t a l y t i c p e r f o r m a n c e o f t h e wide p o r e z e o l i t e s i n t h e i s o p r o p y l a t i o n of b e n z e n e w i t h t i m e o n s t r e a m (TOS). R e a c t i o n temp. = 230OC; WHSV = 2 . 5 h r - 1 ; Benzene t o p r o p y l e n e m o l a r r a t i o = 6 . 5 .
351 The increase in selectivity to cumene with temperature Above is a result of transalkylation of DIPB with benzene. 230°C, the selectivity towards cumene decreases on account of the formation of nPB. Also unwanted products like aliphatics, C9-C11 aromatics and higher boiling fractions increase with the increase in temperature as a result of cracking of higher alkylbenzenes (ref. 7). TABLE 3 Influence of temperature on product distribution Catalyst = H-ZSM-12; Benzene to propylene molar ratio WHSV = 2.5 hr-1
Temperature ("C)
170
190
210
Product distribution (wt X ) Aliphatics 0.03 0.03 Benzene 84.42 82.38 Tol. t C8 arom. 0.26 0.07 Cumene 11.47 13.39 nPB 0.01 C9-C11 arom. 0.06 DIPB 4.01 3.86 H.B.F -
0.05 80.65 0.11 16.68 0.03 0.04 2.41 0.02
0.06 80.00
99.8 86.2 97.1
99.8 91.2 98.0
C3 = conversion 85.3 Select. to cumene 73.6 Select (cumene 99.4 + DIPB)
.
95.3 76.0 97.8
230
0.18 18.24 0.11 0.06 1.35 -
=
6.8;
250
270
0.12 79.71 0.22 18.46 0.31 0.10 1.05 0.03
0.14 78.87 0.23 18.74 0.63 0.17 1.03 0.09
99.6 90.9
99.4
96.2
88.7 93.4
Influence o f mole ratio With increase in benzene to propylene molar ratio, selectivity to cumene increases, even though the total selectivity to cumene and DIPB remained almost constant during present investigation (Fig. 2). This is due to high propylene concentration at lower mole ratios, resulting in the successive alkylations of cumene. Relative amounts of unwanted byproducts are less at higher mole ratios.
352
> 100t >
F
_-
95-
0
w
d
xCUMENE*UPB CUMENE
90.
v)
s
85-
5
IS
10
20
MOLE RATIO
Fig. 2. Influence of reactant mole ratio on the isopropylation of benzene over H-ZSM-12 catalyst. Reaction temp. = 2 3 0 ° C ; WHSV = 2 . 5 hr-l Influence of weight hourly space velocity (WHSV) Low space velocities are found to be favourable selective to
formation
cumene
and
of
(Fig.
cumene
3).
Total
for
selectivity
DIPB again remains constant over the entire
range of WHSV.
100
t
t 1 c
95 90
; 0
. '
CUMENE *DIP6
.
.
:\
CUMENE
85.
fn
s
80
1
2
3
4
5
6
7
SPACE VELOCITY ( W H S V I h i l
F i g . 3.
Influence o f weight hourly space velocity (WHSV) in the isopropylation of benzene over H-ZSM-12 catalyst. Reaction temp. = 2 3 0 ° C ; Benzene to propylene molar ratio = 6 . 5 .
353
Thus
the
optimised condition for propylation of benzene over H-ZSM-12 catalyst are at temperature 230°C, WHSV, 2.5 hr-l, and reactant mole ratio, 6-8. As already mentioned, the ageing studies indicated faster deactivation o f La-H-Y and H-mordenite catalysts. Whereas H-ZSM-12 catalyst did not deactivate even after 200 hrs of time on stream (TOS), therefore faster deacpivation was carried out by accelarated ageing. All the three coked samples were subjected to thermal and sorption studies to find out the Fig. 4 presents the cumene probable cause of deactivation. sorption kinetics on fresh and coked samples.
r
I CATALYST
Lo H Y
o FRSH SAhlPLE DEACTIVATED SAMPLE
1
0
V , 6
n
a w
2
w
5u
I= -=+/ t
l4
10
2
I
f
r
5t
= I
10
30
20
TIME
(
'
120
min 1
Fig. 4 . Kinetics o f cumene sorpkion on fresh ( 0 ) and coked zeolite catalysts at 25 C (P/P, = 0.5).
( 0 )
354
and
The equilibrium sorption capacity for cumene over fresh deactivated samples and the % coke formed for the
cummulative
feed
passed
over
each
catalyst
is
presented
in
Table 4 .
TABLE 4 Physico-chemical studies on fresh and coked catalyst samples
Catalyst
La-H-Y
H-mordenite
H-ZSM-12
Equil. sorption of cumene over fresh sample (wt X)
17.6
4.5
12.55
Equil. sorption of cumene over deactivated sample (wt X)
11.16
1.4
6.36
X sorption capacity retained
63
31
51
7.48
5.80
8
10
250
0.37
0.15
0.003
14.78
Amount of coke formed (wt X) Time required for deactivation,hrs ( 10 of initial activity)
X
Amount of coke formed per 100 gm of catalyst per gm of feed (gm)
and
Although the rate of sorption of cumene is same for fresh coked La-H-Y the equilibrium sorption capacities are
different. to
high
The deactivation of this catalyst may be attributed
acid
irreversibly structure
site
density
adsorbed
wherein
the
(ref.
ammonia)
8)
in
reactants
(as the
revealed three
(especially
by
total
dimentional
propylene)
and
product molecules are strongly adherent to the active sites. Also due to dehydrocyclisation reaction,bulkier coke precursors are (ref.
readily
9).
(14.78 %
formed
in
the
large
intra-zeolite
cavities
This results in higher amount of coke formation calculated b y thermogravimetric methods) in the
catalyst. In spite o f this, the coked sample shows more than half the void volume still available ( 6 3 % o f
case of La-H-Y
355
initial sorption capacity) for sorption of reactant molecules without any further activity. Thus the deactivation to blocking of the active sites.
is due
The phenomenon of deactivation in H-mordenite is relatively slower on account of lower acid site density compared to La-H-Y (ref. 10) (Si/A1 = 6 . 4 and 0.535 m moles of irreversibly adsorbed
ammonia
per
gm
of
catalyst).
ivation in the unidirectional
However,
the
deact-
pore system of mordenite leads
to a drastic decrease in sorption capacity in the coked sample. In
fact,
capacity 30
X)
the in
sorption
the
suggest
kinetics
coked mordenite only
surface
and
equilibrium
(retained
adsorption
sorption
sorption
capacity
indicating
blocking
of the channels. The
least
coking
tendency
is
observed
sample (0.003 gm of coke formed per gm
of
feed
density
passed)
(Si/A1
again
on
60.5 and
=
in
the
H-ZSM-12
100 gm of catalyst
account
0.067 m
adsorbed ammonia per gm o f catalyst).
of
lowest
acid
per site
of irreversibly In addition, the linear moles
non-interpenetrating unidirectional channel system with virtual absence of larger intra-zeolitic cages (like those in zeolite
Y)
and
highly
siliceous nature
of bulkier coke precursors. by
ZSM-12
activity
zeolite for
not
favour the
formation
The sorption capacities exhibited
support
prolonged
do
these
period
of
findings. time
Thus
(more
the
than
stable
200 hrs)
can be explained for H-ZSM-12.
CONCLUSIONS H-ZSM-12
is
a
superior
and
selective
catalyst
in
the
alkylation of benzene with propylene. The catalytic activity and stability are dependent on the acidic and structural properties. Coking
of
La-H-Y
is
due
to
acid
site
blocking,
in
H-mordenite, it is due to channel blocking while deactivation of H-ZSM-12 required accelarated ageing. ACKNOWLEDGEMENT We
sincerely
thank
Dr.
P. Ratnasamy,
for
his
constant
encouragement throughout this investigation. We also thank Dr. V.G. Gunjikar and Mr. S . P . Mirajkar, for helping in
356
the
thermal and s o r p t i o n s t u d i e s .
T h e w o r k was p a r t l y f u n d e d
by t h e UNDP.
REFERENCES
2 3
4 5
6 7 8
9 10
Y.C. Y e n , S t a n f o r d Res. I n s t . E c o n . R e p . , 22 A ( 1 9 7 2 ) and 22 B (1973). I b i d , 49 (1969) E.S. M o r t i k o v , S.R. Mirzabekova, A.G. Pogorelov, N.F. Konov, R . F . M e r h a n o v a , A.Z. D o r o g o c h i n s k i i a n d Kh. M . M i n a c h e v , N a f t e k h i m i y a , 16 ( 1 9 8 8 ) 7 0 1 . W . W . K e a d i n g a n d R.F. H o l l a n d , J . C a t a l . , 109 ( 1 9 8 8 ) 212. B.S. Rao, I. B a l a k r i s h n a n , V.R. Chumbhale, A . R . Pradhan and P. Ratnasamy, P a p e r p r e s e n t e d a t F r i s t Tokyo C o n f e r e n c e o n A d v a n c e d C a t a l y t i c S c i e n c e a n d T e c h n o l o g y (TOCAT l ) , J u l y 1-5, 1990. E . J . R o s i n s k i a n d M . K . R u b i n , U.S. P a t e n t 3 , 8 3 2 , 4 4 9 ( 1 9 7 4 ) . D . A . B e s t a n d B.W. W o j c i c h o w s k i , J . C a t a l . , 4 7 ( 1 9 7 7 ) 11. H.G. Karge and E.P. Boldingh, Catalysis Today, 3 ( 1 9 8 8 ) 53. P.B. V e n u t o a n d P . S . L a n d i s , Adv. C a t a l . , 18 ( 1 9 6 8 ) 2 5 9 . P.E. E b e r l y J r . a n d C.N. K i m b e r l i n J r . , I n d . Eng. Chem. P r o d . Res. D e v . , 9 ( 1 9 7 0 ) 335.
G . Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
357
EXAFS STUDY OF LOCAL STRUCTURE OF Pt-Cr CLUSTERS IN PENTASILS IN RELATION WITH THEIR REACTIVITY IN LOWER ALKANES AROMATIZATION
E.S. S H P I R O ~ , R.W. ~ JOYNER~, 2 G . J . T U L ~ O V A ' . A.V. PFEOBR~IENSKY'. O.P. TKACHENKO , T.V. VASINA , O.V. BRAGIN and Kh.M. MINACHEV
'N.D.Zelinsky Institute of Organic Chemistry, USSR Academy of Sciences, Moscow, YSSR The Leverhulme Centre for Innovative Catalysis, The University of Liverpool, Liverpool, England SUMMARY Pt-Cr/H-ZSM-S samples were characterized by EXAFS and XPS and their catalytic activity in ethane and methane aromatization was examined. EXAFS shows clear evidence of Pt-Cr alloy particles formation in samples reduced at 823 K. Most of these particles is located within zeolite structure and coordinated to zeolitic oxygen. The changes of catalytic activity in C1-C3 alkane aromatization and hydrogenolysis caused by chromium are related to alloying effect.
INTRODUCTION It is well known that VIII Group metal additives play an important role in catalysis of lower alkanes aromatization which proceeds on modified high-silica zeolites (refs. 1-4). Among other metals platinum behavior is established rather in details and such characteristics as dispersion and electronic state was shown to be key factors for high catalytic performance in ethane and propane aromatization (refs. 4-6). Recent EXAFS studies (ref. 7 ) proved earlier suggestion that platinum clusters are located within zeolite structure in the proximate vicinity t o Bronsted acidic centers and they are most probable candidates for active sites in alkane dehydrogenation step (refs. 5-71. Platinum might play even greater role in methane aromatization. This reaction was discovered (ref. 8 ) to proceed over modified pentasils at 1023 K to yield up to 4-14% of aromatics. In connection with aromatization reactions two features of platinum loaded Hpentasils became important: (a) stabilization of very dispersed Pt particles at temperatures as high as 823-1023 K; (b) reducing Pt hydrogenolysis activity and as consequence coke formation and catalyst deactivation. To maintain these characteristics of platinum in reforming catalysis promotion by second metal (Re, Sn, In, Ga, Cr) is widely used, chromium is definitely suppressed platinum activity in hydrogenolysis (refs. 9,101. Our previous studies (ref. 6 ) also suggested the strong influence of chromium on platinum behavior in ethane hydrogenoly-
358 sis but mechanism of this effect remained unclear.
This paper presents a study of the influence of chromium on Pt-HZSM-5 catalyst where we have able to establish a detailed connection between structure and catalytic function via EXAFS and XPS characterization of Pt-Cr/H-ZSM-5 and their reactivity in aromatization of lower alkanes inclusive methane. To elucidate possible alloying effects the activity of Pt-Cr/H-ZSM-5 and Pt/H-ZSM-5 in ethane and propane hydrogenolysis was also compared. EXPERIMENTAL Catalyst containing 0.5 wt % Pt and different contents of chromium (0.13, 0.75 and 1.25 wt % I were prepared by simultaneous impregnation of H-ZSM-5 (Si/A1 a mixture of H PtC16 and (CrO 1 solutions, 0.5% Cr/H-ZSM-5 was ob2 3 x tained by impregnation with (CrO 1 and 0.75% Pt/H-ZSM-5 was prepared by ionic 3 x exchange of [Pt(NH3l41Cl2 with NH -EM-5 (ref. 4). Samples were calcined in an 4 air at 623 K for 2 h and subsequently at 790 K for 3 h. = 16.5) with
EXAFS measurements were performed at the Daresbury synchrotron radiation SQUrce and reduction of the catalyst specimen was carried out in-situ at 1 bar in flowing hydrogen at 623 and 823 K. Data were collected for the Pt L I11 edge and analyzed by standard methods (ref. 11). XPS analysis of the Pt-Cr/H-ZSM-5 has been performed after in-situ reduction in 1 bar of flowing hydrogen at 823-873 K in a specially developed cell attached directly to Kratos XSAM 800 spectrometer and using Mg K radiation. The analysis of Pt-Cr/H-ZSM-5 samples after methane a aromatization reaction was performed with Kratos ES 200B spectrometer according procedure described in (refs. 4 , 6 ) . Pt 4f + A 1 2p and Cr 2p peaks were analyzed by using peak synthesis programme with PDP 11/03L computer. The details of peak synthesis treatment for these particular spectra are described in (ref. 12). Ethane and methane aromatization were tested in microcatalytic pulse
reactor
in He flow at 823 K and 1023 K respectively. Propane and ethane hydrogenolysis were performed in flow reactor at hydrocarbon/H2 ratio equal to 1/10 rential mode (hydrocarbon conversion does not exceed 5-7%).
in diffe-
RESULTS AND DISCUSSION Characterization of the reduced catalysts Studies of Pt/H-ZSM-5 catalysts by XPS, EXAFS and electron microscopy revealed that most of platinum in pre-calcined samples can be stabilized as very dispersed particles within zeolite structure (refs. 4,6,7). The lower limit of ave0
rage particle size, 7-8 A is close to the channel size EXAFS indicating formati-
on of one Pt-Pt shell clusters consisted of 13
5 atoms (ref. 7). Upon increase
of calcination temperature from 723 K to 793 K average particle size increased 0 + 75 up to 13 A and number of atoms in cluster up to 120 - 40. Nevertheless these
359
particles remain to be located inside zeolite matrix (ref. 7). The particular 0 features of 7-13 A particles are 1-4% contraction of nearest Pt-Pt distance with respect to the bulk and bonding between metal and oxygen of zeolitic framework. XPS positive shifts of Pt 4f level observed for such particles have been
inter-
preted in terms of charge transfer from metal to zeolite (refs. 4-6). Fig.1 and Table 1 shown that main structure characteristics of Pt particles remained eventually the same after promotion of 0.5% Pt/H-ZSM-5 with chromium
a
0.03
k)/k
kX(
0.08 0.04 0.00
0.04 0.08 0.12
1 b
I Fig. 1. Pt L I 1 1 EXAFS for the Pt-Cr/HZSM-5 catalyst reduced at 623 K (a) and 823 K (b), solid line experiment; dashed line calculated using the parameters given in Table 1.
360 (0.75%) and reduction at 623 K. The average coordination number
for
the first
0
Pt-Pt shell is equal to 5 and Pt-Pt distance is of 0.08 A shorter than for Ptfoil. Pt-0 distance with C.N. 1.4 is also contributed
to EXAFS spectrum. The
comparison of these data with those obtained for 0.5% Pt/H-ZSM-5 calcined at the same temperature (793 K) (refs. 6,7)indicated some stabilization effect
caused
by chromium. TABLE 1 Best fit Pt L I 1 1 edge EXAFS data for 0.5% Pt-0.75% Cr/H-ZSM-5 ~~
Tred,K
Neighbor
Interatomic 0
distance, A
Coordination
Debye-Waller
number
factor, X2
623
platinum oxygen
2.70:O. 01 1.94-0.03
5.0;1.0 1.4-0.5
0.012:O. 003 0.017-0.005
823
platinum oxygen chromium
2.7110. 01 1.97~0.03 2.65-0.02
4.511.0
0.015:0.003 0.016;O. 005 0.015-0.003
1.8~0.5
1.7-0.5
The increase of reduction temperature up to 823 K did not result in noticeable sintering Pt particles but clear evidence of Pt-Cr distance was
found from
analysis of EXAFS spectra (Fig. lb). Including a Pt-Cr interatomic distance into the analysis decreased the fitting index by 51% and this is significant at excellent statistical level of < 0.5%. As was shown in (ref. 13) a
the
significant
level of 5% is considered sufficient if a new shell of neighbors is to be accepted. Preliminary analysis of EXAFS data for second series of Pt-Cr/H-ZSM-5 samples (0.5%Pt-0.75% Cr and 0.5% Pt-1.25% Cr) which differs from above sample only by Si/A1 ratio demonstrate again Pt-Cr coordination (ref. 1 4 ) . The existence of platinum-chromium bond in the catalysts is thus well established and the observ0
ed distance is in the range expected for metallic bonding (2.65-2.70 A ) . It
was
demonstrated that alloying of chromium with platinum decreases the lattice parameter, the intermetallic of Pt3Cr being often observed (ref. 10). The coordination numbers for 0.5% Pt-0.75% Cr/H-ZSM-5 suggests a Pt/Cr ratio of 3/1 for the first series of samples and up to 1011 for second series (ref. 14). This implies Pt3Cr and more diluted in Cr alloys are formed even in catalysts contained an excess of chromium. The additional information on chemical state of chromium and its distribution over zeolite was deduced from XPS data
(Table 2).
In fresh sample chromium
exists as a mixture of Cr(VII and Cr(II1). the rather strong enrichment of
ex-
ternal surface with Cr was observed. Both calcination and reduction lead to de-
361
TABLE 2 XPS analysis of 0.5% Pt-0.75% Cr/H-ZSM-5 Treatment
Pt ( 0 )
Pt(I1)
fresh
B.E..eV %
air,793 K B.E. ,eV
x
72.3 44.3
Cr(0)
Cr(II1)
Cr(V1)
0.44a)
74.0 100
-
577.5 74.8
579.5 25.2
73.9
-
577.0 78.8
580.8 21.2
0.033
55.7
H2,823 K B.E. ,eV %
72.3 100
-
573.4 12.3
-
577.2 87.7
02,298K
B.E.,eV
x
%
0.032
0.037 577.2 100
72.8 100
H2,873 K B.E. ,eV
Cr/Si
0.029 72.2
-
100
577.4
573.0 16.2
83.8
a) Cr/Sibulk = 0.009 crease of Cr/Si ratio which can reflect migration of part of Cr
ions into the
channels. At 623 K in hydrogen only Cr(II1) was found but increase of
reduction
temperature up to 823 K gives rise new doublet in Cr 2p envelope which is characteristic of Cr(0) (refs. 6.12). The degree of Cr(0) reduction reached to 10-15% at 823-873 K. Catalytic activity of Pt-Cr/H-ZSM-5 (1) Hydrogenolysis reactions. Adding 0.75% Cr to Pt/H-ZSM-5 suppressed
its
activity in both ethane and propane hydrogenolysis (Table 3). The difference between two hydrocarbons is that an decrease of reaction rate with C2 was observed only at relatively low temperatures (623-648K), the activity also strongly depends on chromium content. As temperature increased up to 673 K
the difference
in activity of Pt- and Pt-Cr-samples becomes negligible and at higher temperatures the activity of bimetallic sample even exceeds the activity of purePt/H-ZSM-
5. This is formally explained by higher activation energy observed for Pt-Crsamples. In contrast, with C3 the drop of reaction rate by an order of magnitude was observed on whole temperature range (573-648K) with Pt-Cr sample and difference of activation energies for Pt-, and
the
Pt-Cr-samples is smaller than
with ethane. On other hand, the main decrease of reaction rate was found for 0.5% Pt-0.75% Cr and further increase of Cr effects C hydrogenolysis. 3
content to
1.25% only slightly
362
One of the explanation of the observed difference is that zeolite could contribute to hydrogenolysis (dehydrogenation) or cracking of alkanes at higher temperatures. This explanation is rather speculative because nor H-ZSM-5 neither 0.75% Cr/H-ZSM-5 shown no noticeable activity in C2H6 or C3H8 hydrogenolysis in the temperature range studied. But we could not ruled out possible dualfunctional behavior of Pt-Cr/H-ZSM-S. Despite of complications connected with possible zeolite-catalyzed reactions it is obvious that chromium suppressed intrinstic hydrogenolysis activity of platinum. This is particularly confirmed by the fact that 0.5% Pt-0.75% Cr/H-ZSM-5 reduced at 623 K, when no metallic Cr
is
formed, has the same activity as Pt/H-ZSM-5. TABLE 3 Hydrogenolysis activities of Pt- and Pr-Cr/H-ZSM-S Sample
Ethane (Tred = 823 K) 623K
648K
673K
698K
723K
Reaction rate, U x 103, pmole/g 0.SLPt O.YLPt-0.7SLCr 0.5%Pt-1.25%Cr
1.2 0.36
3.0 1.8
0.056
0.45
7.2 9.8 7.3
20
748K s
280 2100
39
40
-
36
120
Ea' kJ/mole
134 232 260
Propane (Tred = 823 K) 573K
598K
623K
Reaction rate, W x 104, pmole/g 0.5%Pt 0.5%Pt-0.75%Cr 0.5%Pt-1.25%Cr
4.6 0.35 0.22
8.6 0.69 0.82
29 2.9 2.5
698K s
1.2 2.0
76 119 143
( i i ) Ethane and methane aromatization. The 0.5% Pt-0.75% Cr/H-ZSM-5 exhibit
higher aromatization activity with C H than pure Pt/H-ZSM-5. About 25% or BTX 26 was found on former catalyst at 823 K while 0.5% Pt/H-ZSM-5 gives only 15% of aromatics at equal conditions. XPS studies of Pt state after different number of C H pulses shown some increase of B.E. Pt 4f7/2 similar to observed earlier 26 with Pt/H-ZSM-5 (refs. 4.5). This means that platinum in small Pt-Cr particles also possessed electron deficiency which is favorable for aromatization reactions (refs. 4-6). In addition the increase of aromatics yield can be related to Pt-Cr interaction in alloy particles which can modify Pt electronic state. Fig.2 shows the dependence of yield of aromatics produced from methane at
363
1023 K over different compositions of Pt-, Cr-, and Pt-Cr/H-pentasils. H-
pentasil is inactive but Cr, Ga, Zn/H-ZSM-5 give a few percent of aromatics. Including platinum substantially increased aromatics yield from 4 to 9%. But most dramatic increase of aromatization activity was found when Cr was added t o 0 . 5 % Pt/H-ZSM-5. Starting from 0.13% Cr aromatization activity jumped up to 14%. The other important function of chromium is prolongation of catalyst life. While Pt/H-ZSM-5 activity sharply and irreversibly declined after 10-15 pulses of methane (ref. 8 ) (not shown on the figure) activity of Pt-Cr samples retained on the maximum level especially for samples with higher Cr content (1.25% Cr). chromium improved the resistance of the catalyst to deactivation.
So,
1 2
T = 1023 K g = 500
mg cat.
VHe = 1.2 1 h-'
a
rl 6
2
0
,
0
Number of cH4 pulees Fig.2. The dependence of aromatics (B+T) yield on CH4 pulse number: 1 - 0.5% Pt-0.75% Cr; 2 - 0.5% Pt-0.75% Cr (was studied by P S I ; 3 - 0 . 5 % Pt-1.3% Cr; 4 - 0 . 5 % Pt-0.13%Cr; 5 - 0.5% Pt; 6 - 0.75% Cr. 5
10
15
Post-reaction studies The enhancement of methane aromatization activity with Pt-Cr sample seems to be indicative of platinum-chromium interaction rather than simple additive effect. This interaction may appear in the course of the reaction which proceeds at high temperature and i n purely reducing atmosphere. To verify this assumption
XPS spectra were monitored after treatment of 0 . 5 % Pt-0.75% Cr/H-ZSM-5 with He and with 1, 3, 10 and 17 pulses of methane at 1023 K (Table 4). As was expected (refs. 4-7) platinum was part,?lly reduced even during air calcination and
its
reduction was progressing in He at 1023 K. After interaction with methane extent of Pt reduction increased but about 1 5 2 0 % of Pt gives spectrum with higher B.E. than for pure metal. By analogy with Pt/H-ZSM-5 treated with hydrogen or ethane this spectrum can be assign to Pt"
clusters. The lack of significant change of
364
Pt/Si ratio during reaction could suggest no severe sintering of metallic particles. Although most of chromium is present as Cr203 the formation of a few percent of metallic chromium and its alloying with platinum might be possible. These results imply that structure of Pt-Cr catalysts after methane reaction resembles the structure of reduced catalyst or sample exposed to ethane at 823 K. In latter case EXAFS suggest the formation of diluted Cr-Pt alloy
particles (ref. 14). TABLE 4 XPS analysis of 0 . 5 % Pt-0.75% Cr/H-ZSM-5 after methane aromatization Treatment
Pt(I1)
Pt(0)
Cr (YI
Cr(II1)
air, 793 K B.E. .eV
73.9
72.3 44
579.6 21
577.4 79
0.12 56
%
He, 1023 K B.E. ,eV
73.8 % 18 CH4,1023 K,lth pulse
71.9 82
B.E. ,eV
71.5
72. 14
%
-
0.23
576.9 100 0.23
-
577* 5 100
86
CH4,1023 K, 3 pulses B.E.,eV
0.56
73.44 23
%
71.4
577.0
77
100
CH4,1023 K, 10 pulses B.E. ,eV
73.3a) 26
%
0.27 71.5
576.Sb)
74
100
CH4,1023 K, 17 pulses 73.3a)
B.E.,eV
22
%
a) P t ' +
;
Pt/Cr
0.33 71.6
-
577.5
78
100
b) a few Cr(Ol was found from peak synthesis
CONCLUSION Let us discuss in conclusion some points concerning local structure of Pt-Cr zeolites and its possible effect on catalysis. Both EXAFS and XPS
data clearly
demonstrated that Cr(0) is formed during H2 treatment at 823-873 K part of which is alloyed with platinum. Other properties of Pt particles such as dispersion,
location, crystallographic structure are rather similar to those in Pt/H-ZSM-5 (ref. 7). Thus we can neglect size effect and discuss the data in terms of Cr promotion or poisoning. Chromium in H-ZSM-5 is present in several forms
:
(a)
365
Pt-Cr alloy; (b) separate Cr(0); (c) Cr(II1) ions in channels; (d) chromia on external surface. Higher XPS Cr(O)/Pt(O) ratios with respect to found from EXAFS suggest that a part of chromium exist in separate phase, although there is no indications from TEM o r XANES that big Cr(0) particles are formed. In addition about 90% of total chromium content is not reduced and it can be located or in the channels either on external surface. The latter is more probable when we take into account higher Cr/Si ratios on the surface. Size and composition of the alloy particles may be deduced from EXAFS results. The sum of Pt-Pt and Pt-Cr nearest neighbor coordination number is 6.2, very close to that found in 13-atom clusters which is 6 . 5 .
For
other samples
even less chromium is included into Pt-Cr particles (ref. 14) although chromium content in bulk alloys can reach to 71% (ref. 15). We can therefore assume that in diluted alloys chromium is homogeneously distributed over the particle. The limit of chromium concentration in alloy can be explained by the fact that only
a part of chromium i s within zeolite structure. Also, Pt particles which size is limited by channel cross-section can dissolve only a limited number of chromium atoms. Besides, the surface of platinum particles can be decorated by chromia. Thus, the difference in catalytic behavior of Pt/H-ZSM-5 and Pt-Cr/H-ZSM-5 is likely to be due to platinum-chromium interaction in small platinum-chromium particles and not so well-defined interaction between Pt particles and Cr20g. In agreement with data (ref. 10) we ascribe the suppression of hydrogenolysis activity to alloying of Pt-Cr. The ensemble model can be proposed as simplest one to explain, for instance, stronger poisoning of propane hydrogenolysis than ethane one for sample 0.5% Pt-0.75% Cr/H-ZSM-5. But stronger drop of ethane hydrogenolysis activity on 0.5-1.25% Cr requires more complex model. It should taken into account platinum-chromium electronic interaction, strong hydrocarbon adsorption on Cr sites and so on. Pt and Cr lie on opposite sides of their respective volcano curves for alkane hydrogenolysis (ref. 16). Hydrocarbons are insufficiently strongly adsorbed on Pt while adsorption is too strong on chromium for optimal activity. This can inhibit hydrogenolysis which proceeds at relatively low temperatures and on pure metallic surface (low conversions) but at
the aromatiza-
tion conditions (high temperatures, coking) the situation may change in such way that stronger hydrocarbon adsorption on Pt-Cr would play a positive role. This is especially important for methane which activated very hardly. Taken into account mechanism of methane oxidative coupling and thermodynamic hindrances of
direct methane aromatization (refs. 8,171 one can expect that first stage of methane activation must be oxidative one and zeolite or oxide oxygen can participate in initializing reaction. But even in pulse mode the methane aromatization does not look as simple stoichiometric reaction and it definitely involves catalytic steps. For this particular system the presence of Cr203as source of mobi-
366
le oxygen is not enough to provide high activity. In contrast, the system PtCr/Cr 0 /H-ZSM-5 gives rise to maximum effect. The Cr reduction by reaction mlx23 ture was recently found for 0.5% Pt-1.25% Cr/H-ZSM-5. Similar activity of three Pt-Cr samples in methane aromatlzation 1s in accordance with EXAFS data lndlcatlng formation Pt-Cr alloys of close composition in samples with very different total Cr/Pt ratios (ref. 14). REFERENCES O.V. Bragln, T.V. Vasina, Ya.1. Isakov, N.V. Pallshklna, A.V.Preobrazhensky, B.K. Nefedov and Kh.M. Mlnachev, Stud. Surf. Scl. Catal., 18 (1984) 31-36. C.W.R. Engelen, J.P. Wolthulzen, J.H.C. van Hoof, H.W. Zundbergen. Proc. 7th Int. Zeolite Conf., 1986, pp.709-716. T. Inul, J.Maklno, F. Maganos, A. Mlymoto. Ind. Eng. Chem. Res. and Develop., 26 (1987)647-652. O.V. Bragln, E.S. Shplro. A.V. Preobrazhensky. S.A. Isaev. T.V. Vasina, B.B. Dysenbina, G.V. Antoshin and Kh.M. Mlnachev, Appl.Catal., 27 (1986) 219- 231. Kh.M. Mlnachev and E.S. Shpiro, React. Klnet. Catal. Lett., 35 (1987) 195206. E.S. Shplro, G.J. Tuleuova, V.A. Zalkovskii, O.P. Tkachenko. T.V. Vasina,O.V. Bragin and Kh.M. Minachev, in H.G. Karge and J. Weitkamp (Eds.), Zeolites as Catalysts, Sorbents and Detergent Builders, Elsevier, Amsterdam, 1989, pp. 143-152. E.S. Shplro, R.W. Joyner, Kh.M. Minachev and P.D.A. Pudney, J. Catal., to be published O.V. Bragin, T.V. Vaslna, A.V. Preobrazhensky and Kh.M. Minachev, Izv. AN SSSR, Ser. Khlm., 1989, pp. 750-751. K. Anders, R . Feldhaus, H.-G. Vieweg, S. Engels, H. Lausch et al., Chem. Tech.(Lelpzlg) 37 (1985) 65. 10 S . Engels, H. Lausch, B. Peplinski, M. Wilde, W. Morke, P. Kraak, Appl. Catal., 55 (1989)93-107. 11 S.J. Gurman, N. Blnstead, I . Ross, J. Phys. C (Solid State Phys.), 17 (1984) 143. 12 W. Grunert, E.S. Shpiro, R Feldhaus, K. Anders, G.V. Antoshln and Kh.M. Minachev, J. Catal.,100 (1986) 138-148. 13 R.W. Joyner, K.J. Martln and P. Meehan, J. Phys. C (Solid State Phys.), 20 (1987)4005. 14 E.S. Shpiro, R.U. Joyner , unpublished results 15 M. Hansen, Constitution of Binary Alloys, 2nd edn. (Mc Graw-Hl11, New York,1958). 16 J.H. Sinfelt, Catal. Revs. Scl. Eng., 3 (1970) 175-205. 17 Kh.M. Mlnachev, N.Ya. Usachev, V.N. Udut and Yu.S. Khodakov, Russian Chemical Reviews, 57 (1988)385-404.
G . Ohlmann eta!. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
367
SPLITTING OF METHANE INTO THE ELEMENTS OVER NICKEL CONTAINING ZSM-5 CATALYSTS
J. HOFFMANN. R. BAUERMEISTER. 8. HUNGER, K. HANTEL, G. ULLMANN. K.-H. STEINBERG and H. SIEGEL' Department of Chemistry, Karl Marx University, Leipzig. DDR-7010 (GDR) 'Chemieanlagenbau Grimma, Grimma, DDR-7240(GDR) SUMMARY Methane splitting was studied over nickel containing Na-ZSM-5 catalysts and at a Ni/AI,O, catalyst for comparison The reaction took place between 673 and 1073 K The highest activity was found between 923 and 1023 K The ZSM-5 catalyst showed the best long time behaviour Good working conditions were only obtained in fluid bed realized by a vibration reactor. In the temperature region between 673 and 973 K a rate equation containing a first order term with respect to methane and a reversible second order one concerrung hydrogen best fulfilled the experimental results INTRODUCTION The formation of carbon filaments (or whiskers) as a result of the decomposition of hydrocarbons on metal surfaces or, respectively. catalysts containing Fe. Ni or Co particles has been known for a long time 11-31, Since this time the practical interest has been mainly directed on the question how to avoid the whisker growth because it is an important reason for the formation of carbon deposits on reactor walls and for mechanical break-down of catalyst particles. A good review about the production of filamentous carbon is given by BAKER [ 4 :I who emphasizes that the scientific interest should be drrected not only on the prevention of carbon-filament formation but also on the possibilities to produce such materials for useful aims, thus for electrodes with high surface area and electrical conductivity, for new generations of
3-D composites 141,for carbon fibres [ 5 ] and possibly as materials for adsorption processes. As could be shown by electron microscopy the whisker consists of a tubular filament with a co-axial channel [ 6 ] and a metal particle at the growing end of the filament. The whisker structure is graphitic [7,81 with the basal planes parallel to the whisker axis. The core of the whisker may be void or filled with carbon material of lower density.
It could be shown that after an induction period which is necessary for saturation of the nickel particle with carbon the single whisker grows after a rate equation of zero order 191. Moreover the rate of growth depends on the reciprocal root of the dlameter of the particle. The growth can be inhibited when the leadmg face is encapsulated b y a layer of amorphous carbon. Whereas the whisker growth is obviously determined by the diffusion of carbon through the metal particle leading to a concentration gradient in the metal there are different interpreta-
368
tions about the driving force of the carbon segregation. In the case of the decomposition of alkenes and alkines it is assumed that the end of the particle connected with the whisker is cooler than the leading face because of the exothermic splitting of the unsaturated hydrocarbons and therefore the solubility of carbon in the metal is lower. But this explanation cannot be valid in case of the splitting of methane or other alkanes because of their endothermicity. Therefore in recent work other explanations are favourized. KOCK et al. [ 10 1 suggest that an instable bulk carbide is formed as the intermediate phase whereas ALSTRUP [ 1 1 1 assumes the formation of a surface carbide and the splitting of methane only on (100) and (110) nickel facets and segregation on ( 1 I 1) facets.
Whereas in literature extensive studies are published about whisker formation on metal surfaces and on supported metal catalysts almost no results can be found regarding pssibilities to lead the process of alkane splitting continuously and to use it for the production of larger amounts of whisker carbon. With respect to this the observation made in this work that the process can be effectively brought about for a period of some hours using a fluidbed reactor is of significance.
EXPERIMENTAL Preparation of catalysts Pellets of catalyst material consisting of 70 mass per cent of Na-ZSM-5 zeolite and of 30 mass per cent of y-A1203 were impregnated with such amounts of solutions of nickel nitrate that the nickel content of the dry material was 20, 10 and, respectively, 5 mass per cent. After heating in dry air for about 3 hours at 773 K the pellets were broken up and classified to a fraction of 0.2 _..0.4 mm.
Catalvtic measureme& For catalytic measurements a vibration reactor was used. It consists of a quartz tube with an inner diameter of 1 cm which is elastically suspended from springs. By means of a magnetic oscillation apparatus the reactor can vibrate with an amplitude of about 0.5 cm and a frequency between I and 50 Hz. usually at I5 Hz. The flow rate is controlled by a Brooks flow controller and measured at the reactor outlet. The gaseous reaction products are analyzed by GC using a 2 m PORAPAK T column. Catalytic experiments were carried out using 0.1 ... I g of catalyst. The catalyst was heated in a stream of dry He up to 773 K and kept at this temperature for 3 hours. then cooled down to 473 K, newly heated in a stream of hydrogen up to 673 K and reduced at this temperature for 3 ... 4 hours. After having switched the gas stream to pure methane the vibration of the reactor was started and the sample was heated with a rate of 10 K per min up to the reaction temperature.
369
Tpr emeriments In order to find out the optimum reduction conditions samples of the catalysts were reduced in a tpr apparatus consisting of a quartz reactor, a furnace controlled electronically and a heat conductivity cell. The reducing gas mixture was argon containing 5 per cent of hydrogen. The heating rate could be chosen between I and 20 K p e r min and was usually 10 K per min.
Evaluation of the experiments
It was assumed that because of the mechanical vibration of the reactor and of a relatively high flow rate a fluid bed reactor was approached which can be kinetically described by the model of a perfectly mixed reactor. For kinetic evaluation it must be considered that the
-
CH, C 2 H, +
reaction is connected with a change of the mole number. In order to count the degree of conversion X first of all the molar fractions
xcH4and xH2were determined by GC.
Then the conversion degree could be expressed as
From this the momentary consumption of methane was calculated after
"CH4
=
VCH4
(2)
'
(vCH; inlet flow rate of methane). The whole amount of carbon formed until the reaction had been stopped could be calculated by numerical integration: t
a
SVCHddt
(3)
For the kinetics of methane splitting a rate equation of the following form was assumed:
370
In case of the perfectly mixed flow reactor the rate
results from
I
ngH4= molar flow at the reactor inlet nCH4=molar flow at the reactor outlet V
=
volume of the reactor
The partial pressures of methane and hydrogen can be written in the following manner considering the change of the mole number during reaction ( p
When substituting k, by k , / K ngn4(X3
+
2X2
P +
- total pressure):
it results from equation ( 4 ) :
XI
=
klpV - klpV(l
+
4p/Kp).X2
(7)
From equ (7) k , and Kp can easily be calculated by regression and from the rate and equilibrium constants the activation energy EA1 and the heat of reaction ARH can be determined using the equations of Arrhenius and van t Hoff RESULTS AND DISCUSSIGN Studies of the methane splitting reaction were made mainly over Na-ZSM-5 zeolites containing 20 mass p e r cent of nickel (0.2 NiNa-ZSM-5) because these catalysts provide the best results with respect to activity and resistance against deactivation. For comparison same experiments with NiNaX. NiNaY. Ni/A1,0,
and NiNa-ZSM-5 catalysts of lower nickel content
were carried out. As could be observed in all cases carbon formed by decomposition of methane under
fluid-bed conditions balls together to almost spherical aggregates above the catalyst bed but still containing microscopic particles of the catalyst.
In order to study the conditions of reduction of the Ni" containing catalyst precursors more in detail tpr experiments were carried out. Fig. I shows a typical reductogramme of a ZSM-5 catalyst containing 20 per cent of nickel. The heating rate was 10 K per min. The reduction seems to occur as a relatively
371
uniform process as can be seen from the only less structured peak'with the maximum at about 690 K. Taking into account that the peak-maximum temperature decreases when the heating
rate is diminished it can be concluded that a reduction temperature of about '100 K should be
Fig. 1. Tpr spectrum of a 0.2 NiNa-ZSM-5 zeolite (heating rate 10 K per min)
H
heating period
tim on stream [min]
Fig. 2. Catalytic activity in dependence on time of a 0.2 NiNa- ZSM-5 zeolite
372
sufficient. Indeed, a reduction degree of about 80 per cent could be calculated from the peak area related to the whole amount of nickel. The measured peak-maximum temperature is in good agreement with results of other authors who studied the reduction of NiO species [ 121. Considering the activity of the various catalysts it can be stated that over nickel faujasites only a very low amount of carbon was formed until the reaction broke down by rapid deactivation (at 1 g of 0.45 NiNaX only 0.13 g of carbon were formed. the same amount of
0.5 NiNaY led to 0.2 g of carbon). Na-ZSM-5 catalysts containing 5 or. respectively. 10 mass per cent of nickel showed lower activity and more rapid deactivation compared to catalysts with 2 0 per cent of nickel. Comparison of the activity and long-time behaviour of convenient Ni/AI2O3 hydrogenation catalysts containing about 50 per cent of nickel with that of 0.2 NiNa-ZSM-5 catalysts was of special interest: Whereas the catalytic activity of the N i l A1,0,
catalyst is comparable
to that of the 0.2 NiNa-ZSM-5 its long-time behaviour is much worse. Thus in an experiment with 0.2 NiNa-ZSM-5 after 7 hours of time on stream the fourfold amount by weight of carbon related to the weight of catalyst was formed but the degree of conversion diminished only from 0.6 to 0.4 whereas the Ni/AI,O,
catalyst produced only
twice of its own weight until it deactivated dramatically. In fig. 3 the momentary methane consumption at constant inlet flow rate of methane is shown in dependence on temperature for the 0.2 NiNa-ZSM-5 catalyst. Catalytic activity can be observed between 673 and 1073 K with a maximum in the 1000 . . . 1030 K region.
Fig. 3. Catalytic activity of a 0.2 NiNa-ZSM-5 catalyst depending on temperature
373
Similar behaviour was found by other workers [6 1 which used Ni / MgU catalysts. The loss of activity at temperatures higher than loo0 K should not be due to deactivation by carbon
deposition but by adsorption effects in a rate determining surface reaction [ 13 1. The results obtained in this work for the Ni-ZSM-5 and Nil A1,0,
catalysts are contrary to this explanation
because almost no catalytic activity was observed at temperatures beolw 1000 K when the catalyst had worked before at temperatures over 1100 K (in the temperature region where the reaction rate is decreasing) indicating a poisoning of the active nickel centres by deposition
of amorphous carbon. In fig. 2 the conversion is shown in dependence on time on stream. As can be seen the conversion degree decreases in the course of 7 hours only from 0.6 to 0.4. When carrying out the same experiment without vibration of the reactor the activity dramatically decreases after short reaction times because of sticking together of the catalyst pellets. Kinetic studies were carried out in a temperature region between 673 and 923 K. In contrast to other authors 1141 an inhibition by hydrogen was not observed, thus a rate equation ex-
pressing the conditions for the equilibrium reaction CH,
-
C
+
2H2 could be used (equ. (4)).
In case of the 0.2 NiNa-ZSM-5 catalyst the activation energy of the splitting was (1 13
2.2) kJ
per mole, the reaction heat ( 1 24 : I . 2 ) kJ per mole. For the Ni/A1203 catalyst a n activation energy of ( 1 10 : 1.4) k] per mole and a heat of reaction of ( I 25.4
1.2) kJ per mole could
be calculated. These values are in good agreement with results of other authors
r 151.
In figs. 4 and 5 Arrhenius and van? Hoff plots are shown for the 0.2 NiNa-ZSM-5 catalyst. In Tab. 1 the equilibrium constants calculated from the experiments or, respectively. from thermodynamic data are listed together with the respective equilibrium degrees of conversion.
TABLE I Equilibrium constants and equilibrium degrees of conversion in dependence on temperature
T[Kl theor.
Kplbarl 0.2 NiNa-ZSM-5
xNi/AI,O,
theor.
0.2 NiNa-ZSM-5 ~
873 923 973 1023 1073
2.1 3.92 7.32 12.35 20.7
I .34 3.44 7.89 16.70 33.0
0.921 2.35 5.43 11.6 2 3.0
0.59 0.70 0.8 I 0.8 7 0.9 2
0.50 0.6 8 0.8 2 0.90 0.94
Ni/A1203 0.4 3 0.6 1 0.76 0.86 0.9 2
The results show that the equilibrium conversion can be higher than the value calculated from the theoretical thermodynamic data, especially in the upper temperature region. This can be explained by the fact that the gas phase does not contact with the pure graphitic carbon surface but with a carbidic one at the exposed surface of the nickel partic!e (e.g. Ni3C ).
374
B
6
Fig. 4. Arrhenius plot for a 0.2 NiNa-ZSM-5 catalyst
1
.....
. . -..
...
. _..._ _ . _..-.
'\
C b?
r;
-1
+
-2
'4, \ '
-3
-4
\. ,Q\,\
'\
1
1.6
375 CONCLUSIONS The splitting of methane can be well realized on nickel containing Na-ZSM-5 zeolites. When using a vibration reactor which allows fluid bed condtions a quick deactivation of the system by sticking together of the catalyst pellets can be avoided. The good activity and long-time behaviour of the NiNa-ZSM-5 catalysts could be explained: by their high thermal stability by their properties with respect to fluidized-bed conditions (e. g. low density) by a favourable distribution of the nickel-particle size with respect to whisker formation (in our case between 5 and 30 nm as could be proved by electron microscopy) and possibly by the avoidance of formation of carbon forms other than whisker-like Kinetics of the reaction follows a rate equation which contains the methane concentration in a first order term and the hydrogen concentration in a second order one expressing the reversible step. The activity of carbon does not appear in the rate equation and no inhibition by hydrogen was observed. At the temperature which represents the region of highest activity (about 1023 K ) a degree of conversion of nearly 0.9 can be realized (higher than the equilibrium degree calculated from thermodynamic data). REFERENCES
I P. A. Tesner and I.S. Rafalkes. Dokl. Akad. Nauk SSSR, 87(1952)821 2 R.T.K. Baker and P.S. Harris. J. Phys. E., 5(1970)793 3 J.R. Rostrup-Nielsen. J. Catal.. 2 7 ( 19 7 2 134 3 4 R.T.K. Baker. Carbon. 27(1989)315 5 G.G. Tibbetts. Appl. Phys. Lett.. 42(1983)666 6 J.R. Rostrup-Nielsen, Catalytic Steam Reforming, in: J.R. Anderson and M. Boudart (Ed.). Catalysis - Science and Technology. Vol. 5, Springer Berlin, Heidelberg. New York. Tokyo, 1984. pp. 7 4 - 7 5 7 T. Baird. J.R. Fryer and 8. Grant, Carbon, I2(1974)591 8 S.D. Robertson, Carbon, 8(1970)365 9 R.T.K. Baker, P.S. Harris, R.B. Thomas and R.J. Waite. 1. Catal., 96(1985)468 10 A.J.H.M. Kock. P.K. de Bokx. E. Boellaard. W. Klop and J.W. Geus, J. Catal.. 96(1985)468 I 1 I. Alstrup. J. Catal.. 109(1988)241 I 2 N.W. Hurst. S.J. Gentry, A. Jones and B.D. McNicol. Catal. Rev. - Sci. Eng. 24(1982)2 3 3 I 3 J.R. Rostrup-Nielsen and D.L. Trimm. J. Catal.. 48 (1 977) 155 1 4 C.A. Bernardo. I. Alstrup and J.R. Rostrup-Nielsen. J. Catal.. 96 (1 9 8 5) 5 1 7 15 B.K. de Bokx. A.J.H.M. Bock, E. Boellaard. W. Klop and J.W. Geus. I. Catal.. 96(1985)454
This Page Intentionally Left Blank
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
SULFUR TOLERANT SHIFT REACTION
Ni-Mo-Y-ZEOLITE
CATALYSTS
377
FOR
WATER-GAS
M-LANIECKI and W.ZMIERCZAK Faculty of Chemistry, Adam Mickiewicz University, Grunwaldzka 6, 60-780 Poznan (Poland) SUMMARY Molybdenum sulfide and nickel-molybdenum sulfide catalysts were prepared from Mo(CO), encaged in Y and stabilized HY zeolites. These catalysts were characterized by XRD, IR, ESR spectroscopy and NO sorption capacity. Mo(CO& based catalysts are highly reactive in water-gas shift reaction and this is related to the high dispersion of molybdenum sulfide. Ni and Mo containing zeolites indicate the existence of synergetic effect. The synergy is confirmed by the presence of a band at 2082 cm during adsorption of CO. Catalysts prepared from dealuminated supports show low activity, mainly due to the limited access of molybdenum carbonyl into the zeolite cavities. ESR spectroscopy indicates the formation of sulfur chain biradicals after interaction of HIS with molybdenum containing supports. INTRODUCTION It is known that both high-temperature and low-temperature water-gas shift catalysts, based on iron oxides and copper-zinc oxides respectively, are highly sensitive towards sulfur contamination of the feed. In order to overcome this problem a new class of catalysts, similar to the hydrodesulfurization catalysts, based on alumina-supported Co-Mo sulfides has been prepared and recently introduced into the market. Literature concerning the application of zeolites either as catalysts or as the supports in water-gas shift (WGS) reaction is scarce (ref.1-4). The Ru-X (ref.1) and Ru-Y (ref.2) zeolites have been reported to be active in WGS reaction, but their activity and especially stability were relatively low. Iwamoto et al. (ref.3) studying transition metal ion-exchanged X and Y zeolites established that only C U * ~and Ni” containing zeolites showed good performance in WGS and this was related to the low electronegativity of the exchanged transition metal. Lee and co-workers (ref.4) studying the kinetics of WGS reaction over Ni-Mo-Y-zeolite catalysts in the sulfided state established high reactivity in this reaction and excellent resistance against H,S and NH,.
378
Studies performed in the previous years by different groups (ref.5-7) and concerning molybdenum loaded zeolites indicate that saturation with molybdenum hexacarbonyl leads towards the formation of highly and homogenously dispersed molybdenum. This can be subsequently transformed into highly dispersed supported molybdenum sulfide species (ref.8,9). In contrast, molybdenum containing zeolites prepared from ammonium heptamolybdate (AHM) form bulky, weakly dispersed Mo sulfides. Our recent studies (ref.9) showed that Mo(CO), loaded Y-zeolites in the sulfided state indicate high activity in WGS reaction. In the present study the catalytic activity in WGS reaction over sulfided Ni-Mo-Y and Ni-Mo-USY zeolites was measured and the characterization of these catalysts has been given. EXPERIMENTAL Mat er i a1s NaY from KATALISTIKS (Si/A1=2.56) was applied as a starting material for preparation of nickel and amonium exchanged zeolites. The ammonium containing samples were obtained by a single or multiple exchange with 0.25M aqueous solution of NH,C1 at 345 K. Three series of stabilized Y-zeolites, denoted hereafter as USY-1, USY-2 and USY-3, were prepared from NH4NaY forms, differing in the degree of exchange, by hydrothermal treatment at 875 K for 4.5 hours under 100% steam. Nickel was introduced into the zeolites before molybdenum by ion-exchange applying nickel nitrate solution, both for NaY and USY series. Molybdenum was loaded into the zeolites by saturation with molybdenum hexacarbonyl vapours. Zeolitic supports after activation at elevated temperatures in hydrogen were exposed to the Mo(CO), vapours passing the support in a hydrogen stream for 10-15 hours at room temperature. Samples completely saturated with Mo(CO), were subsequently partially decarbonylated at 425 K for one hour in a stream of hydrogen and next slowly exposed to the air at room temperature with the use of leaking valve. Each sample, before any experiment, was presulfided for two hours at 675 K with a 10 vol.& mixture of hydrogen sulfide in hydrogen, flowing with the rate of 2 dm3 h-'. Chemicals a_nd Gases Doubly sublimed Mo(CO), from Aldrich was used for saturations. Hydrogen and helium were freed from impurities by passing through
379
the MnO/SiO, oxygen-water scavenger (ref.10). CO and NO were from Fluka and H,S from Intermar (The Netherlands) and were used without further purification. Characteri zatLog Zeolitic supports as well as Ni and Mo containing samples were characterized by XRD, NO sorption capacity. FTIR, ESR and MAS NMR spectroscopy. The experimental details concerning NO adsorption and ESR measurements are given elsewhere (ref.9). For infrared measurements self-supporting wafers with the -1 were mounted in the infrared cell "thickness" of 3-5 mg cm operating under dynamic conditions. Samples after activation were exposed for 30 minutes to the Mo(CO), vapours according to the procedure already described in the previous section (Materials). IR spectra were recorded applying FTIR Bruker IFS 113v spectrometer. 29 27 Si and A1 MAS NMR measurements of USY and Ni-USY were performed for non-sulfided samples applying Bruker MSL-300 spectrometer. Catalytic Activity Measurements Grained samples (0.5-lmm) of zeolites weighing 0.25 or 0.59 were pleaced in a continous-flow reactor operating at atmospheric pressure. Samples after presulfidation were purged 30 minutes with He at 6 7 5 K and next the temperature was lowered to 6 2 5 K and the WGS reaction started. Gas mixture introduced into the saturator was composed from equimolar amounts of CO and H, and 2 vol.% of H,S. Reaction was usually performed for 20 hours and the hydrogen sulfide concentration was mantained constant during the reaction time. H20 : CO ratio was changed in different experiments and varied between 0 . 7 to 3. RESULTS AND DISCUSSION It was found previously (ref.9) that molybdenum sulfide supported on sodium or hydrogen form of Y-zeolite indicate relatively high activity in WGS reaction but with the tendency to the slow deactivation. It was expected that certain loss of the zeolite crystallinity can be responsible for the decrease in activity. The XRD measurements indicated, however, negligible changes in the support crystallinity during or after the reaction.In the present study three series of stabilized Y-zeolites have been apllied as the supports in order to minimalize the changes in the catalysts crystallinity during the reaction time as well as to study the influence of the support dealumination on the behaviour in WGS reaction.
380
TABLE 1 Characterization of Ni-, Mo- and Ni-Mo-Y zeolite catalysts Cata 1yst
Si/A1 29 Si NMR AAS
Composition (wt.%)
Ni NaY NaN i Y HNaY HNaN i Y USY-1 USY-I-Ni USY-2 USY-2-Ni USY-3 USY-3-Ni
2.56 2.63 2.60 2.78 2.62 2.48 2.76 2.48 2.56 2.55
10.3 10.0 0 10.3 1.1 8.0 0 5.5 1.2 5.0 0 4.8 1.0 4.6 0 4.5 4.0 0.8 0
2.8
3.28 3.78 4.19 4.47 5.02 5.30
Mo
Surface * area 2
- 1
[m 9
900 830 930
840 860
-
830
-
800 780
I
NO adsorption ** capacity [mmol g- ‘ 1 0.184 0.134 0.166 0.121 0.053 0.136 0.056 0.145 0.097 0.154
* “BET analogous surface area” of the support activated at 675 K * * after activation at 675 K , MO(CO)~ deposition, decarbonylation at 425 K and sulfidation at 675 K Table 1 shows certain properties of the catalysts prepared by standart procedure described in the previous section. A l l supports were activated 2 hours at 675 K before Mo(CO), admission. “BET analogous surface area” was practically the same for sulfided or non-sulfided molybdenum-free supports. After deposition of tiiolybdcnum onto non-stabilized supports, followed by sulfidation, t h e surface area was usually reduced by about 30-40%. This is indicative of the location of molybdenum sulfide-like species inside the zeolite porous system (ref.9). In contrast to the non-stabilized zeolites, steam calcined samples indicate a lower surface area, due to the dealumination process. After deposition of molybdenum and subsequent sulfidation, ultrastable samples indicated only minimal decrease of surface area. High concentration of the non-framework aluminum atoms at the zeolite surface (ref.ll), or the formation of boehmite-like structures inside the supercages (ref.12) can cause a partial blockage of the zeolite porous system towards nitrogen molecules and can protect Mo(CO), molecules from entering zeolite porous system. The lowered ability to adsorb Mo(CO), at the inital preparation for USY based catalysts seems to confirm via steam calcination this suggestion. However, dealumination causes changes in the acidity of the support and these changes also influence the ability to adsorb Mo(CO), during catalyst stage of
381
s-'
,
30Q WAVENUMBER
Fig.l.Infrared spectra of USY supports activated at 675 K. HY-obtained from NH,,Y (78% exchanged) calcined 2 hours at 675 K.
34 CM-1
Fig.2.Infrared spectra of nickel exchanged NaY zeolites (NaNiY) a)after activation at 675 K b)after activation at 475 K c)after activation at 475 K , saturation with M O ( C O ) ~ ,and decarbonylation at 425 K.
preparation (ref.7). It is known that basically one aluminum atom in the zeolite framework of HY zeolite can give rise to one acid group. Hence, a decrease in the intensities of OH stretching bands at 3550 and 3650 cm-l with progresive dealumination was expected. Y-zeolites. Fig.1. shows OH bands for the series of dealuminated In our case, ultrastabilization practically eliminated OH groups from the small cages (absence of the band at 3550cm-') and reduced significantly the amount of acidic OH groups in supercages (3650cm-I). The appearance of the OH stretching bands at 3690 and 3610 cm-', non reactive towards NH, and pyridine (ref.l3), as well as silanol groups (at 3745 cm-') is indicative of strong 29 dealumination (compare also SiNMR results). The shoulder bands at
382
Fig.3. Infrared spectra of molybdenum loaded NaNiY zeolite a)after activation at 675 K b)after activation at 675 K, saturation with M~(CO)~.followed by decarbonylation at 425 K c)as b) with subsequent interaction with HZS at room temperature.
b
1
A
3675 and 3590 cm-' appearing for USY-3 can be attributed to the hydroxyls connected with extraframework aluminum species (ref. 14). Nickel-containing ultrastabia lized samples showed additional BOO 170 WAVENUMBER CM-' decrease of OH bands intensity, similarily to the findings presented in ref.14. The characteristic OH bands formed upon nickel exchange with NaY are shown in Fig.2. Here, after activation at 675 K the typical acidic OH groups (3645 and 3554 cm-') are present. Activation at 475 K leads towards the appearance of new weak band at 3610cm-'. This is usually ascribed to the characteristic dissociation of water molecules over the zeolites containing divalent cations (ref.13) with simultaneous formation of Me+(OH) species. Adsorption of Mo(CO), at room temperature followed by thermal decomposition at 425 K yielded the subcarbonyl species with general formula cd Mo(CO), (ref.8,15). Irrespective of the support used, after partial decarbonylation, more or less complex infrared spectra in the region of carbonyl groups vibrations were observed (example presented in Fig.3.). Usually USY and HY supports give rise to the more complex infrared spectra in the reg on of carbonyl vibrations than for NaY or NaNiY samples. The decr ase in intensity of OH bands for USY and NaNiY after partial decarbonylation at 425 K is indicative of the oxidation of molybdenum by zeolite protons (ref.5). Partially decarbonylated samples after intereaction with H,S at room temperature show a significant reduction in the
383
carbonyl bands quantity (2031 sh, 1984, 1946 sh, 1845 cm-') and absorbance value (see Fig.3.). Exposition of partially decarbonylated samples towards H,S at 425 K resulted in a complete disappearance of carbonyl groups and further decrease in the OH bands intensity, suggesting the formation of sulfided molybdenum species. Interaction with H,S at 675 K additionally lowered the intensity of OH groups vibrations. These observations show that after decarbonylation, interaction with H,S at room temperature leads towards Mo(CO),-SH like species (ref.8), whereas at higher temperatures,MoSx structures are formed (1< x <2) (ref.16). Bands at 2530 and 2570 cm-' observed earlier in similar systems for NaY support (ref.8) in our case were always absent. Y-zeolites with Si/A1 ratio higher than 2.5 do not adsorb H2S dissociatively (ref.17) and the absence of the band assigned to H2S physisorption (2570 cm-') indicates a lack of this type of interaction. TABLE 2 Catalytic activity of the studied samples after 2 hours in WGSR ~
~
Support pretreatment temperature [K] 475 675 825 Cata1yY t
H20/C0 NaY NaN i Y
HNaY HNaN i Y USY-1
USY-2 USY-3 USY-1 -Ni USY-2-N i USY-3-N i
Rate constant k [cm3 g-' min-'I k H20/C0 k H20/C0 28
1 2 3 1 2 3
16 21 355 204 213
1
10
0.7
28
1
32
1
245 203 240 74 83 29 14 9 34 52 80 29 53 48
2 3 1 1 0.7 1 2 0.7 0.7
0.7 --
44
0.1 2 0.7 0.7
k
1
105
1
8
1 1
27
6
1
12 -
~
-
-
384
The calculated pseudo-first order reaction rate constants for the WGS reaction and the series of Mo(CO), based catalysts are presented in Table 2. The highest activity was observed for NaNiY support, whereas an order of magnitude lower activity was found for the series of USY supports. Depending on the degree of dehydroxylation, the highest activity was found for the nickel containing supports activated at 475 K. The increase of support pretreatment temperature above 675 K always reduced catalytic activity. It is known that mainly basic sites are responsible for the adsorption and decomposition of Mo(CO), (ref.7.8). The decarbonylation performed in the absence (Nay) or in the presence of very much limited amount of OH groups (high temperature activation of HY or USY) usually leads towards the formation of molybdenum species with a low oxidation number. The subsequent interaction with H,S produces sulfur deficient (ref.16) molybdenum species with a relatively high dispersion but low catalytic activity. TPDE profiles presented by Okamoto et al. (ref.15) show that formation of subcarbonyl species of Mo(CO),ads.-type occurs at temperatures lower than 425 K. It was found that below this temperature of decarbonylation, subsequent sulfidation produces a highly dispersed molybdenum sulfide-like species. The increase of decarbonylation temperature above this level resulted in a significant decrease of molybdenum sulfide dispersion and consequently, in low catalytic activity. Our previous studies showed that NO sorption capacity (represented here as the dispersion) is an important factor influencing catalytic activity (ref.9). A high activity of non-stabilized, nickel containing zeolites is related to a high dispersion of molybdenum sulfide as well as to the existence of synenergetic effect between Ni and Mo sulfided species. The synergy between Ni and Mo is confirmed through IB studies by the appearance of a band at 2082 cm-' after interaction of Ni-Mo species with CO at room temperature. USY supports indicate an activated at 675 K after deposition of Mo(CO), increasing activity with the degree of dealurnination, while the molybdenum content decreases. A comparison between the results of activity and NO sorption capacity suggests that also in the case of USY supports dispersion of the molybdenum sulfide plays a very important role in catalytic activity. These results suggest also that extra-lattice aluminum favours better dispersion of molybdenum sulfide. On the other hand, nickel containing USY-zeolites,
385
decreases catalytic activity with increasing amount of extra-lattice aluminum and NO sorption capacity. The blank experiments with molybdenum-free USY and USY-Ni supports explained this discrepancy, indicating that significant amounts (about 50%) of NO are irreversibly adsorbed at room temperature over sulfided nickel species. Moreover, the presence of extra-lattice aluminum in Ni-USY zeolites favours the NO adsorption. It was already mentioned that all catalysts indicated slow but constant deactivation in time. The formation of carbonaceous deposits can cause the activity decay. It was assumed that application of high steam concentration will suppress carbon deposition. For these reasons different H,O to CO ratios were applied to study this phenomenon. The deactivation in time was always observed apart from steam concentration in the feed . The increase of H,O to CO ratio always reduced catalytic activity, which suggests that adsorption of CO (associative or dissociative), rather than adsorption of H,O, is the rate-limiting step.
Fig.4.ESR spectra of molybdenum loaded USY-2 zeolite after sulfidation at 675 K a)USY-molybdenum free b)USY + Mo(C0)6decomposed at 425 K c)Ni-exchange USY f Mo(COIg decomposed at 425 K d)Ni-exchanged USY - molybdenum free .
The mechanism of WGS reaction proposed by Hou et al. (ref.18) for sulfided Mo on alumina, in the case of zeolitic supports should be ruled out because of the absence of Mo'~ species detectable v i a ESR spectroscopy. Fig.4 shows only the typical ESR spectra obtained after sulfidation of USY-2 based samples but similar results were obtained for other studied samples.The shape of resonance lines and g-values are characteristic of sulfur chain biradicals (see literature in ref.9). In Ni and Ni-Mo containing supports a presence of wide resonance line with g=2.285 was additionally observed.
386 The results presented in th s paper indicate that high activity of Ni-Mo sulfides supported on Y-zeolite certainly is related to the high dispersion of these su fides and synenergetic effect between Ni and Mo species. However, futher studies are required in order to clarify the influence of support acidity, presence of sulfur chain biradicals and other factors on catalytic behaviour. CONCLUSIONS (i) Mo- and Ni-Mo loaded Y-zeolites idicate high activity in water-gas shift reaction in the presence of H,S. ( 1 1 ) Synergetic effect between Ni and Mo sulfided species for water-gas shift reaction has been found. ( i i i ) Acidic OH groups of the support enhance catalytic activity in water-gas shift reaction. (iv) ESR measurements excluded the importance of M o ’ ~ ions in the mechanism of WGS reaction over zeolitic sulfided catalysts. REFERENCES 1 J.J. Verdonck, P.A. Jacobs and J.B. Uytterhoeven, J.Chem.Soc. Chem.Comm., (1979) 181-182. 2 B.L. Gustafson and J.H. Lunsford, J.Catal., 74 (1982) 393-404. 3 M. Iwamoto, T. Hasuwa, H. Furukawa and S. Kaqawa, J.Cata1. 79 (1983) 291-297. 4 A.L. Lee, K.C. Wei, T.Y. Lee and J. Lee, in B.Imelik et al. ‘Eds.), Studies in Surface Science and Catalysis, -101.5, Catalysis by Zeolites, Elsevier, Amsterdam 1980, pp.327-333. 5 S . Abdo and R.F. Howe, J.Phys.Chem., 87 (1983) 1713-1722. 6 T. Komatsu and T. Yashima, J.Mol.Catal., 40 (1987) 83-92. 7 M. Laniecki, in H.G. Karge and J. Weitkamp (Eds.), Studies in Surface Science and Catalysis, vo1.46, Zeolites as Catalysts, Sorbeiits and Detergent Builders, Elsevier, Amsterdam, 1989, 259-269. 8 Y. Okamoto, A. Maezawa, H. Kane and T, Imanaka, J.Mol.Cata1.. 52 (1989) 337-348. 9 M. Laniecki and W. Zmierczak, Zeolites, in press. 10 R. Moseler, B. Horvath, D. Lindenau, E.G. Horvath and H.L. Krauss, Naturforsh., Anorq.Chem.,Org.Chem. 31B (1976) 892-898. 11 J. Arribas, A. Corma, V. Fornes and F. Melo, J.Catal., 108 (1387) 135-142. 12 R.D. Shannon, K.H. Gardner, R.H. Staley, G.Berqeret. P. Gallezot and A . Auroux, J.Phys.Chem., 89 (1985) 4778-4781. 13 P. Jacobs and J.B. Uytterhoeven, J.Catal., 22 (1971) 193-203. 14 A . Ezzamarty, E. Catherine, D. Cornet, J.F. Hemidy, A . Janin, J.C. Lavalley, J. Leglise and P. Meriaudeau, in P.A.Jacobs and R.A.van Santen ( E d s . ) Studies in Surface Scie ce and Catalysis, vo1.49B, Elsevier, Amsterdam 1989, pp. 025-1034. Proc. 9th 15 Y. Okamoto, A. Maezawa, H. Kane and T. Imanaka 1988, International Congress on Catalysis, Calgary V O l . 1 , pp.11-18. 16 M. Laniecki and W. Zmierczak, unpublished results 17 H.G. Karqe and J. Raskd, J.Coll.Interface Sc ., 64 (1978 522-532. 18 P. Hou, D. Mekker and H. Wise, J.Catal., 80 (1983 280-285.
381
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
THE INFLUENCE OF CATIONS ON THE ALKYLATION OF TOLUENE WITH ETHYLENE OVER MODIFIED ZSM-5 ZEOLITES 1
EEJKA, Blanka WICHTERLOVd and Gloria LLabre RAURELL The J. Heyrovsky Institute of Physical Chemistry and Electrochemistry, Czechoslovak Academy of Sciences, 182 23 Prague 8 , (Czechoslovakia)
Ji!?:
'Chemical Research Center Cerro, La Habana (Cuba) ABSTRACT It has been found that H-forms of ZSM-5 zeolites differing in the Si/A1 ratio exhibit straightforward relationship of activity vs. para-selectivity in toluene alkylation to para-ethyltoluene. Addition of Na, K and A1 cations or A1 oxidic species into the zeolite cationic sites suppresses the zeolite void volume. This results in an increase of para-ethyltoluene selectivity in the initial alkylation step occurring in the channel intersections and/or in a preferential transport of para-ethyltoluene from the zeolite channels. Zeolite deactivation by coke is supported by the presence of cations in the zeolite due to narrowing of the channels or eventually due to their function as strong electron acceptor sites. The coke formed in the zeolite does not enhance the zeolite p
388
includes isomerization and transalkylation reactions, different transport rates of individual isomers to the outer surface and possibly also surface isomerization reactions [refs. 6-91. This paper demonstrates the influence of the presence of alkali-metal cations and extralattice aluminium species (incorporated into the zeolite during zeolite synthesis and as post-synthesis modifications) on the zeolite activity, selectivity and deactivation followed during reaction time-on-stream experiments. EXPERIMENTAL
The parent ZSM-5 zeolites with Si/Al ratios ranging from 13.6 to 600 and oi: crystal size about 1 p m were supplied by the Research Institute €or Oil and Hydrocarbon Gases, Czechoslovakia. Acid forms of ZSM-5 zeolites were prepared by the ion exchange of their N a forms with 0.5 M HN03 at ambient temperature. NaH-ZSM-5 and KH-2SI.I-5 'rlcrc prepared by ion exchange of H-ZSM-5 with 0 . 0 5 M NaCl and 0.05 1.1 ICCl, respectively, at 350 K for 3 hours. Aluminium was introduced into the H-ZSM-5 zeolites by ion exchange with 0.5M Al(N03)3 (Alll-ZSM-5) or by impregnation with 1 M A1(N03)J followed by calcination at 770 K in oxygen to provide A1203H-ZSN-5. The Si/Al ratio of the H-ZSM-5 zeolites used, the number of strong acid bridqing hydroxyls and the sorption capacity for argon f o r the 11- and modified ZSM-5 zeolites are given in the Table. Fig. 1 depict.s the infrared (IR) spectra of the hydroxyl band region for tt.e parent and modified zeolites. The IR spectra were recorded on a FT-IR Nicolet MX-1E spectrometer. The zeolite void volume was est-imated from the sorption capacity for argon measured in a static apparatus at 13.33 kPa of argon at a temperature of 8 0 K on samples pretreated at 7 2 0 K in a vacuum of 10y4 kPa for 16 hours. The alkylation of toluene with ethylene was carried out in a vapour phase continuous down-flow microreactor at atmospheric pressure and 623 K. Nitrogen as a carrier gas was saturated with toluene at 3 3 3 K to 18.5 vol. %. The toluene to ethylene molar ratio was 3.8. The reaction products were analyzed in on-line arrangement by a Hewlett-Packard gas chromatograph with MS and FI detectors using a supelcowax 10 capillary column. The toluene conversion was normalized to 100 % of the theoretical conversion corresponding t o a toluene to ethylene ratio 3.8. The details of the catalytic measurements can be found in ref. 4 .
TABLE 1
Characteristics
o f t h e z e o l i t e s used and t h e i r a c t i v i t i e s ,
t o l u e n e a l k y l a t i o n w i t h e t h y l e n e a f t e r 15 m i n . Zeolite
Si/Al
OH g r o u p s
( n'mo l/g
1
of
p-ET
s e l e c t i v i t i e s i n the
time-on-stream.
S o r p t i o n Capa-
WHSV
c i t y (mmol Ar/)
(h-l)
Ccriversion
( %I
Selectivity (wt.
H-ZSM-5( 1 )
22.5
0.72
4.87
20
79.8
31.3
NaH-ZSM-5( 1 )
22.5
0.65
4.77
20
75.7
49.3
KH-ZSM-5( 1)
22.5
0.44
4.65
20
59.5
59.8
AlH-ZSM-5( 1)
22.5
0.67
4.76
20
78.6
56.6
H-ZSM-5( 2 )
45
0.30
5.80
20
40.8
62.8
H-ZSM-5(3)
600
0 . 02a
5.92
20
13.7
88.9
H-ZSM-5( 4 )
13.6
1.00
5.31
60
97.9
34.3
NaH-ZSM-5( 4 )
13.6
0.62
5.12
20
79.8
36.6
H-ZSM-5( 5 )
22.5
0.76
5.15
20
93.9
33.1
AlH-ZSM-5( 5 )
22.5
0.73
5.11
20
92.1
34.6
A1203 H- ZSM-5 ( 5 )
22.5
0.68
4.92
20
85 .O
54.4
a
calculated value
%)
390 RESULTS AND DISCUSSION
The Table and Figure 2 present the dependence of the toluene conversion and the selectivity to p-ET on the number of strong acid bridging OH groups and the relationship of the activity vs. para-selectivity for H-ZSM-5 zeolites differing in the Si/Al ratio (not containinrj a significant amount of extralattice aluminium and the content of Nn cations was lower than 0.05 wt. % ) . It follows that the toluene conversion is proportional and the p-ET selectivity inversely proportional to the number of bridging OH groups. The activity and p-ET selectivity of the H-ZSM-5(1)-(4) were very stable in time-on-stream (T-0-S) (for H-ZSM-5 (1) see Fig. 3 ) . Only a slight decrease in toluene conversion and, on the other hand, a slight increase in the para-selectivity were observed (Fig. 3 ) . The modification of H-ZSM-5(1) with Na and K cations
,
Fig. 1
Infrared spectra of OH groups of H-ZSM-5(5) - a, H-ZSM-5(1) - b, AlH-ZSM-5(1) - c, NaH-ZSM-5(1) - d and KH-ZSM-5(1) e l zeolites.
-
caused a decrease in the zeolite void volume, in the number of strong acid sites and a corresponding decrease in the toluene conversion. The para-selectivity was slightly higher for NaH-ZSM-5 (1) and for KH-ZSM-5 (1) than expected from the activity vs. selectivity relationship (cf. Fig. 2). Simultaneously, zeolite deactivation for Nall- and KH-ZSM-5(1) was a little increased compared with the parent zeolite. The toluene conversion decrease in T-0-S was accompanied by the p-ET selectivity increase which followed the activity vs. para-selectivity relationship. When Na cations were exchanged into H-ZSM-5 ( 4 ) (containing
391
0
0.25
0.5
0.75
25
OH groups (mmoI/g)
Fig. 2
50
75
Conversion
100
(70)
-Dependence of the toluene conversion ( 0 ) and the selectivity ( 0 ) of H-ZSM-5 zeolites on the number of 011 qroups compared with modified zeolites: A - IfaH-ZSM-5(1), 0 KH-ZSM-5(1), 0 AlH-ZSM-5(1), AIII-ZSM-5 ( 5 ) and A1203H-ZSM-5( 5 ) B- Dependence of the toluene conversion vs. the p-ET selectivity for H- and modified ZSM-5 zeolites.
A
p-ET
-
100 I
100
AIB
15
55
95
135
175
15
T - 0 - S (min)
3
.
I
I
Fig.
-
el@-
o,e-
55
95
135
175
,.
Dependence of the toluene conversion and the p-ET selectivity on T-0-S for: A - H-ZSM-5(1) (0 ), N a l l - Z S M - 5 ( 1) ( ,A ) and KH-ZSM-5 (1)( 0 ,a) B A l H - Z S I ~ i - 5 ( 1 ) ( 0 , I ) and NaH-ZSM-5(4) ( O,.).
1.00 mmol OH/q) to such a degree that NaH-ZSM-5(4) possessed nearly the same number of bridging OH groups as NaH-ZSM-5(1), then the initial conversion of these NaH- zeolites was practically
equal, however, the NaH-ZSM-5(4) exhibited faster deactivation due to coke formztion even though that the parent H-ZSM-5(4) zeolite had stable activity in T-0-S (the decrease of conversion wasfrom 97 to 9 2 % at. WHSV 60 h -', 623 K, after 300 minutes). If the toluene conversion and p-ET selectivity is approximated to t = 0 then the resulting p-ET selectivity is again higher than that for H-ZSM-5 zeolite. The above results indicate that the larger radius of the Na and K cations in comparison with the protons contribute to the higher zeolite para-selectivity. However, it is difficult to decide if it is a result of the higher selectivity to p-ET in the initial alkylkation step occurring in the zeolite channel intersections and/or a result of the increased diffusivity of the para-isomer in comparison with the meta- and ortho-isomers. Investigating the relationship of activity vs. selectivity on H-Y and H-ZSM-5 (ref. 4 ) pointed out,that for the H-Y zeolites, o-ET prevailed at low toluene conversion levels (due to the higher reactivity of the ortho-position) and m-ET was found at higher toluene conversion on the expense of ortho isomer. In contrast, for H-ZSEI-5 (cf. Fig. 2 ) the para-isomer dominated at low conversions and the meta-isomer was mainly formed at higher conversions. This indicated that steric hindrances at the H-ZSM-5 channel intersections (in contrast to H-Y) might be important for the initial alkylation step. Therefore, we can suggest that the location of f l a and K cations in the channel intersections is responsible for the higher para-selectivity in the initial alkylation step, however, the contribution of the cation presence to the differences in the diffusivity of ethyltoluene isomers cannot be neglected. As concerns an intensive deactivation of NaH-ZSM-5(4) containing high concentration of Na cations, it cannot be caused by an increasc in ethylene oligomerization which is not catalyzed by Na ions. The reason of zeolite coking could be therefore seen only in steric hindrances for aromatic molecules which cannot be easily removed from the zeolite channel system. It is clear that industrially prepared zeolites (and/or after their thermal or hydrothermal treatment) can contain in some cases extralattice aluminium species in a poorly specified form such as isolated A 1 cations or A1 oxidic species as, alumina or
393
silica-alumina clusters or phases. The H-ZSM-5(5) zeolite investigated here differs in its performance in the alkylation reaction in T-0-S and its tendency to deactivate by coking was more pronounced than that for H-ZSM-5(1). A possible occurrence of some isolated A 1 cations in H-ZSM-5(5) can be indicated from the IR spectra exhibiting a low intensity IR band at 3663 cm-I of OH groups bound to A 1 cations (Fig. 1, ref. 1 0 ) . The addition of
100
100
h
75 $
75
3
v
x U
c
2vl 50 L
ti
6 25
t 1
1
I
I
I
15
55
95
135
175
T - 0 - S (mid Fig. 4
T-0-S dependence of toluene conversion and p-ET selectivity for H-ZSM-5(5) (O), AlH-ZSM-5(5) ( 0 ) and ?.I2 03 ll-ZSM-5(5) (0) zeolites.
some A 1 cations (by ion exchange), to the H-ZSM-5(5) as well as zeolites (see Fig. 1,2 and 4 and the to the H - Z S f 4 - 5 ( 1 ) Table) resulted in a small decrease in the number of bridging OH groups and led to a slight decrease in the toluene conversion and to an increase in the p-ET selectivity. When A13+ ions were exchanged and supported on a zeolite, thus forming besides isolated A 1 3 + some A1 oxidic species, a much substantial decrease in zeolite void volume and an increase of the zeolite para-selectivity was observed (Fig.2,4, andtheTable). It follows that A1 cations and to a greater extent A1 oxidic species in A1Hand A120311-modified zeolites decrease a void space in the zeolite channels and contribute to higher p-ET selectivity in the initial alkylation step and to higher relative diffusivity of the para-isomer. Simultaneously,the presence of A1 species increased the zeolite deactivation by coke in T-0-S in comparison with
394 the particular parent zeolite (Fig. 4 ) . However, the expected increase of p-ET selectivity in T-0-S did not correspond to the activity v s . selectivity relationship (Fig. 2 ) . The increase in the zeolite coking can be caused by greater steric hindrances owing to extralattice aluminium species in the zeolite channel system as for )la, K cation modified zeolites and/or by enhanced ethylene oligomerization by A13+ cations (cf. ref.11) competed with toluene alkylation. It has been stated in the literature [ref. 51 that zeolite coking leads to narrowing of the channel system and increases differences in the diffusivities of ethyltoluene isomers and thus enhances the zeolite para-selectivity. However, the activity and para-selectivity data presented here in dependence on T-0-S for all zeolites evidence that this para-selectivity enhancement is only apparent. For "low" coked zeolites the para-selectivity increase in T-0-S corresponded roughly to a decrease in toluene conversion and, therefore, followed the activity vs. selectivity relation ( e . g . for H-, NaH-, KH-ZSM-5(1) and for short T-0-S values for othrr zeolites. On the other hand for zeolites with higher content of coke the selectivity increase in T-0-S is much lower than that corresponding to a decrease in toluene conversion (e.g. for N ~ l l - ? S M - 5 ( 4 ) , €I-ZSM-5(5) and A1203H-ZSM-5(5)). It has been found [ret. 4 1 that, in the very beginning of the reaction, a simultancouc clccrcase in the activity, sorption capacity and in numhcr of O H groups took place. The p-ET selectivity increase corresponded to the activity decrease. For higher T-0-S values, futher decrease in the activity and sorption capacity occurred while the total number of the strong acid sites remained nearly constant. At this stage in the reaction, the para-selectivity did not increase to an extent that would correspond to the activity decrease. Therefore, at the beginning of the reaction, the channel system is open to reactant molecules: however, after longer T-0-S some portion of the OH groups is poisoned and the zeolite channel system is far more blocked. Accordingly, the wholezeolite channel system cannot be utilized for the reaction and the para-selectivity cannot be improved with increasing coke formation in the zeolite channel system. CONCLUSIONS
It can be summarized that the occurrence of the cations with larger ionic radius than that of protons (Na, K, Al) in the zeolite cationic sites as well as of some oxidic A1 phase
395 increases the para-selectivity of H-ZSM-5 zeolite in alkylation of toluene with ethylene owing to filling partly a zeolite void volume. The cations or oxidic species depending on their number or character stimulate as electron acceptor sites to some extent zeolite deactivation by coke and/or by narrowing the diameter of the channels enabling longer contact time and transformations of bulkier aromatic molecules. Coke itself, in contrast to homogeneously distributed cations in the zeolite cationic sites, does not increase the zeolite para-selectivity. Its apparent increase is only a result of a decrease in toluene conversion caused by coke deposition on strong acid OH groups. Blocking of the channels by coke at higher coke loadings and, therefore, limiting utilization o f . the whole zeolite channel system is reflected in the decrease of the zeolite para-selectivity. REFERENCES 1. 2. 3. 4. 5.
L.R. Young, S . A . Butler and W.W. Kaeding, J. Catal. 7 6 , (1982) t l 8 . W.W. Kaeding, C.-C. Chu, L.B. Young, B. Weinstein and S . A . Butler, J. Catal. 6 7 , (1981) 153. G. PaDarntto. E. Moretti, G . Leofanti and F. Gatti. J. Catal. 105, j l ? C 7 ) 227. J. elks, R . Wichterlovg and S . Bedna/gova/, Appl. Catal., submittcd. W.W. Kaedincj, L.B. Young and C.-C. Chu, J. Catal. 89 (1984) ~~
267. 6.
7. 8.
9. 10.
11.
I. Wang, C.-L. Ay, B.-J. Lee and M.-H. Chen, Appl. Catal. 5 4 (1989) 257. P. Hatnnsaniy and S.K. Pokhriyal, Appl. Catal., 5 5 (1989) 265 N.Y. C h e n a n d W.E. Garwood, Catal. Rev. Sci. Eng. 28 (1986) 185. J. Wei, J. Catal. 76 (1982) 4 3 3 . L.M. Kustov, V . B . Kazansky, L. Kubelkovi, S. Beran and P. Jir6, J. Phys. Chem. 91 (1987) 5247. B. Wichterlovd, J. Novakovfi, L. Kubelkovi and P. Jirg, Proc. 5th 1nt.Zeol.Conf. Napoli (Ed. L . V . C . Rees) p. 373, 1980.
-
This Page Intentionally Left Blank
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
397
MULTINUCLEAR MAS NMR STUDIES ON COKED ZEOLITES H-ZSM-5 H.Ernst, D.Freude, M.Hunger and H.Pfeifer Sektion Physik der Karl-Marx-Universitst Leipzig, Linn6straBe 5, DDR-7010 Leipzig, GDR SUMMARY During the cracking process carbonaceous materials are deposited on the outer or inner surface of the catalyst. These deposits are in many cases the main cause of catalyst deactivation. Magic angle spinning (MAS) NMR investigations and catalytic n-hexane cracking were carried out o H-ZSM-5 zeolites after a mild hydrothermal dealumination. By p 3 C CP MAS NMR it could be shown that the enhanced catalytic activity does not enhance the coke formation and that the chem’cal nature of these deposits is essentially aromatic. From H’ MAS NMR tudies qGrformed on shallow-bed activated sealed samples and 2$7Al and Si MAS NMR on rehydrated samples it follows that for high coke concentrations the catalyst deactivation is caused mainly by blocking of Bronsted acid sites. INTRODUCTION The activity, selectivity and deactivation by coking have an important influence on the industrial applicability of zeolites (refs.1-7). In agreement with other authors (refs.9-11) in a previous study (ref.12) using ‘HI 27Al and 29Si MAS NMR as well as the catalytic cracking reaction of n-hexane we have found, that a mild hydrothermal dealumination of zeolite H-ZSM-5 results in an enhanced catalytic activity. Here we report ‘HI 13C, 27AL and 29Si MAS NMR studies on such hydrothermal treated uncoked and coked samples to elucidate where the carbonaceous products are deposited. Whereas in some previous papers H-ZSM-5 was assumed to produce coke preferentially on the external surface of the crystallites (refs.13,14), recent studies give clear evidence of coke formation in the interior pore system of the crystallites (refs.15-19). METHODS Hvdrothermal treatment The zeolite with a framework SifA1 ratio of 15 synthesized without template was the same as used in ref.12. The hydrothermal treatment of the binder free samples was carried out in a tube of
398
50mm inner diameter with a maximum bed-depth of the zeolite material of 8mm. The temperature was increased at a rate of 10K per min. The steaming was carried out at 540aC for 150min under a water vapour pressure, which was adjusted by the temperature of the water bath, through which the nitrogen was flowing. Catalvtic reaction Catalytic activity was determined for the n-hexane cracking reaction in an integral flow reactor. The reactor contained 0.259 zeolite in the form of binder free pellets with diameters of 0.30.4mm. The catalyst was heated at a rate of 6K per min to 450OC and then maintained at this temperature for lh in a hydrogen flow. Since it is known, that the deactivation of H-ZSM-5 zeolites in the n-hexane cracking (refs.6,20,21) and methanol conversion (ref.22) proceeds only slowly with hydrogen a s a strem gas, we have used nitrogen (T=450°C,pN2=6kPa, pnhexane=92kPa). During the rection time of 15h a gaschromatographic analysis of the reaction products was carried out. If U denotes the relative molar conversion, the reaction rate constant k is determined by the equation k=-ln( 1-U) The conditions of the hydrothermal treatment as well as values for the relative rate constant k/ko after coking are given in Table 1. ko denotes the rate constant extrapolated to zero time.
.
Table 1 Conditions of hydrothermal treatment and values for relative rate constant k/kO after n-hexane cracking for the samples under study. Sample
Conditions of hydrothermal treatment t(h) T(OC) Pwater(kPa)
rel. rate constants k/kO f15%
H-ZSM-5/0
2.5
0.76
2.5
540 540
0
H-ZSM-5/10
10.7
0.58
H-ZSM-5/40
2.5
540
40
0.65
measurements 13C, 27Al and 29Si MAS NMR measurements were generally carried
JWR
399
' MAS NMR measurements the out on rehydrated samples. For the H samples were pretreated under shallow-bed conditions in a glass tube of 5.5mm inner diameter and with a bed-depth of lOmm for zeolite material. The temperature was increased at a rate of 10K per h. The samples were kept at the final activation temperature of 4OOOC under a pressure below lOmPa for 24h, and than cooled and sealed. Measurements were made on a homemade spectrometer HFS 270 and on a Bruker MSL 300 spectrometer. The homemade MAS equipment for the rotation of the fused glass ampoules (ref.23) was carefully cleaned to avoid spurious proton signals. As a reference for the 27Al intensity measurements, a wellcharacterized sample of H-ZSM-5 with a framework Si/Al ratio of 15 was used. The total concentration of OH groups in the activated samples was determined by comparison of the initial value of the free-induction decay of the sample with the corresponding value for a standard (water). As a reference for the 13C CP NMR intensity measurements a coked sample of H-ZSM-5 characterized by chemical analysis with respect to the carbon content and the HjC ratio was used. RESULTS H ' MAS NMR The H' MAS NMR spectra of the uncoked as well as of the coked sample H-ZSM-510 are shown in Fig.1. Four different signals can 1a)'CHL l H MAS NMR
A
20
R
0
-10 Clppm
20
10
0
-10 6/ppm
Fig.1 H ' MAS NMR spectra of the sample H-ZSM-5/0 before (A) and after (B) coking. be seen. (i) The line (a) at 2ppm is due to non-acidic hydroxyl groups at the outer surface of zeolite crystallites, at framework
400
defects and in the amorphous part of the sample (ref.16). (ii) For the coked samples this first line is superimposed by a signal of highly mobile molecules of methane, enclosed in the channels of the zeolite. (iii) The line (b) at 4.3ppm is caused by bridging (acidic) OH groups (ref.24). (iv) The very broad line in the spectrum of coked samples is not affected by MAS. Therefore, the line width is determined by proton-proton dipolar interaction. Table 2 summarizes for the six samples under study values for the total concentration (columns (1) and (2)) as well as for the concentration of OH groups contributing to line (b) (columns (3) and (4)). All concentrations are given in protons per U.C. of the zeolite. Table 2 Concentration of protons and of framework aluminium atoms
H-ZSM-510 7.0 H-ZSM-5/10 5.0 H-ZSM-5/40 4.0
14.0
6.0
4.8
6.0
5.0
4.6
5.8
10.5
3.4
2.5
2.7
1.6
1.6
2.0 1.0
6.1
6.5
3.5 1.6
1.6
-
cHtot.:Total concentration of protons; cH (b):Concentration of protons contributing to the line (b); cAlfr.:Concentration of framework aluminium atoms; cHinv.:Concentration of protons per WMR-invisiblellsite; unc. :Uncoked; c. :Coked; # ) :Apparent value. Concentrations in the columns (1)-(7) are given in protons per U.C. of the zeolite. 2 9 ~ iMAS NMR In Fig.2 29Si MAS NMR spectra are shown of the sample H-ZSM510 and of the hydrothermally treated sample H-ZSM-5/40 before and after coking. A fitting procedure using the programm lfLinesimlt yields three lines. It is well-known that the unresolved signal of Si(OA1) groupings in zeolites ZSM-5 consists of a peak at ca. -1llppm with a shoulder at ca. -115ppm. The line at ca. -106ppm arises from Si(lA1) groupings (ref.25). The framework SifAl ratio determined from the relation (ref.25)
401
4
Si/Al
4
nSi(nA1) n=0 where Si(nA1) denotes the intensity of the NMR signal attributable to Si(nA1) groupings is the same for both uncoked and coked sample and agrees very well in the case of uncoked samples with the result of the 27Al MAS NMR measurements. = 4* C
Si(nAl)/
C
n=O
29si MAS NVR
1 h 1
.
I
-100
.
I
.
I
1
.
I
-100
-120
61m
.
I
-1m
.
I
Wpm
Fig.2 29Si MAS NMR spectra of the sample H-ZSM-510 and the hydrothermally treated sample H-ZSM-5/40 before (Arc) and after (B,D) coking. 27~1 MAS NMR
In Table 2 values are given for the number cAlfr. of framework aluminium atoms per U.C. (columns ( 5 ) and ( 6 ) ) Because acid leaching was not performed, it is seen, that the coked samples contain W M R invisible" aluminium.
.
13C CP MAS NMR The 13C CP MAS NMR spectrum of the coked sample H-ZSM-510 shown in Fig.3A is dominated by three narrow lines due to residual n-hexane. This signal disappears by heating the sample up to 4 O O 0 C and after a sufficiently long accumulation time (40.000 repetitions) the signals of the coke deposits can be
402
observed. Figs.3B,CID show the 13C CP MAS NMR spectra of the coked sample H-ZSM-510 and of the coked samples hydrothermally treated at 10.7 and 40kPa, respectively. The spectrum corresponding to Fig.3B but with total sideband supression (TOSS) is shown in Fig.3E.
13C MAS NMR
LA Fig.3 13C CP MAS NMR spectra of the coked samples : ( A ) H-ZSM-510 before heating at 400'C; (B,C ,D ,) H-ZSM-510, H-ZSM-5/10 and H-ZSM-5/40, respectively after heating the samples up to 4OO0C; (E) the same as Fig.3 (B) but after total suppression of the spinning sidebands (TOSS) which are denoted by asterisks in Figs.3(BIC,D).
c
300
200
100
0
6 PPm
DISCUSSION Figs.1 and 3 as well as Table 2 (columns (1)-(6)) show considerable changes in the 'HI 13C and 27Al spectra caused by the coke formation in the channels of the H-ZSM-5 zeolites. The decrease of the intensity of line (b) at 4.3ppm in the H ' MAS NMR spectra and that of the line at -106ppm in the 29Si MAS NMR spectra with increasing water vapor pressure for the uncoked samples is caused by the dealumination of the zeolite (ref.12). On the other hand, coking leaves the 29Si MAS NMR spectra unchanged. That means, that in the n-hexane cracking process no
403
further dealumination occurs. Therefore, the differences between the 27Al MAS NMR spectra of uncoked and coked samples (cf. columns (5) and (6) in Table 2) must be explained by the existence of "NMR-invisible1I framework aluminium. Such "NMRinvisible1#framework aluminium is formed through the coverage of Bronsted sites by carbonaceous deposits which disturb the tetrahedral symmetry of the aluminium atoms (refs.17,26). The same reason is responsible for the decreased intensity of the line (b) and the appearance of the very broad line in the H' MAS NMR spectra after coking. Both the Bronsted OH groups which are poisoned by these deposits and the protons of the deposited molecules are characterized by a strong homonuclear dipolar interaction. The large difference in the H ' and 27Al MAS NMR spectra between uncoked and coked samples is a direct proof for the existence of carbonaceous deposits poisoning the Bronsted acid sites inside of the crystallites of the zeolite crystallites. Assuming on the other hand that 'H(b) cokedz(kfkO) *CAlfr.uncoked (ref.27) as well as CAlfr. kO) *CAlfr.uncoked because cAlfr.=cH (b) (ref.24) we can compare the experimental values given in Table 2, columns ( 4 ) and (6), with the computed values given in column (7). Their agreement is satisfying. Since the ratio k/ko is also effected by pore blockage (ref.5) the last values have to be generally smaller. Furthermore, for the first two samples follows from the values given in Table 2, columns (1)-(4), that about 6 protons are connected with one Bronsted OH group poisoned by carbonaceous deposits (cf. column ( 8 ) ) . Considering that the aromatic to aliphatic ratio derived from the 13C CP MAS NMR spectra shown in Figs.3B and C is ca. 3 we can conclude that (i) the H/C ratio of the carbonaceous deposits is smaller than one and (ii) that the coke content is 3_+lwt.-% for the first two samples and 0.5+0.3wt.-% for the sample 3, respectively. CONCLUSION It has been shown in the present paper that besides the pulsed field gradient technique (ref.16), the 27Al MAS NMR and the 29Si MAS NMR (ref.17) the H ' MAS NMR spectroscopy with sealed samples
404
is a suitable method to elucidate the location of carbonaceous deposites in zeolitic catalysts. REFERENCES 1 N. Y. Chen and T. F. Degnan, Chem. Eng. Progr., 8 4 (1988) 3241 2 3 4 5 6 7 8
S. M. Ciscsery, Zeolites, 5 (1985) 207
s. w. Addison, S . Carlidge, D. A. Harding and G. Mc Elliney, Appl. Catal., 4 5 (1988) 307 J. Abbot, Appl. Catal., 47 (1989) 33 E. G. Derouane, in: B. Imelik et al. (Ed.), Catalysis by Acids and Bases, Elsevier, Amsterdam, 1985, p. 221 M. Guisnet and P. Magnoux, Appl. Catal., 54 (1989) 1 J. Weitkamp, S. Ernst, H. Dauns and E. Gallei, Chem. Ing. Techn., 58 (1986) 623 C. Mirodatos and D. Barthomeuf, J. Chem. SOC., Chem. Commun., (1981) 39
9 10 11
H. Kawakami, S. Yoshida and T. Yonezawa, J. Chem. SOC., Faraday Trans. 11, 80 (1984) 205 R. M. Lago, W. 0. Haag, R. J. Mikowski, D. H. Olson, S . D. Hellring, K. D. Schmitt and G. T. Kerr, Stud. Surf. Sci. Catal., 26 (1986) 677 K . A. Becker and S . Kowalak, J. Chem. SOC., Faraday Trans. I, 83 (1987) 535
12 13 14 15
16 17 18 19 20
E. Brunner, H. Ernst, D. Freude, M. Hunger, C. B. Krause, D. Prager, W. Reschetilowski, W. Schwieger and K.-H. Bergk, Zeolites, 9 (1989) 282 P. Dejafve, A. Auroux, P. C. Gravelle, J. C. Vedrine, Z. Gabelica and E. G. Derouane, J. Catal., 70 (1981) 123 D. H. Olson and W. 0. Haag, ACS Symp. Ser., 248 (1984) 275 M. Guisnet, P. Magnoux and C. Canaff, in: Y. Murakami, A. Jijima and J. P. Ward (Eds.), New Developments in Zeolite Science and Technology, Proc. 7th Int. Conf. Zeolites, Tokyo, 1986, Studies in Surface Science and Catalysis, Vol. 28, Elsevier, Amsterdam, 1986, p. 701 J. Volter, J. Caro, M. Bulow, B. Fahlke, J. Karger and M. Hunger, Appl. Catal., 42 (1988) 15 R. H. Meinhold and D. M. Bibby, Zeolites, 10 (1990) 146 D. M. Bibby, N. B. Milestone, J. E. Patterson and L. P. Aldrige, J. Catal., 97 (1986) 493 T. Ito, J. L. Bonardet, J. Fraissard, J. B. Nagy, C. AndrQ, Z. Gabelica and E. G. Derouane, Appl. Catal., 43 (1988) L5 F. Fetting, E. Gallei and P. Kredel, Chem. Ing. Techn., 54 (1982) 606
24
U. Hammon, G. T. Kokotailo, L. Rieckert and J. Q. Zahn, Zeolites, 8 (1988) 338 M. C. Barrage, F. Bauer, H. Ernst, J. Fraissard, D. Freude and H. Pfeifer, submitted for publication H. Ernst, D. Freude, M. Hunger, H. Pfeifer and B. Seiffert, Z. phys. Chemie, Leipzig, 268 (1987) 304 D. Freude, M. Hunger and H. Pfeifer, Z. phys. Chem. NF, 152
25
J. Klinowski, Progr. Nucl. Magn. Reson. Spectrosc., 16 (1984)
26
D. Freude, J. Klinowski and H. Hamdan, Chem. Phys. Lett., 149
27
W. D. Haag, R. M. Lago and P. B. Weisz, Nature, 30 (1984) 583
21 22 23
(1987) 171 237 (1988) 355
G. Ohlrnann et 01. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
405
COKE OXIDATION IN HZSM-5 ZEOLITES.INTERMEDIATES,FINAL PRODUCTS AND REFORMATION OF OH GROUPS AND VOID VOLUMES
.
I
J NOVAKOVA and L.KUBELKOVA J.Heyrovsk; Institute of Physical Chemistry and Electrochemistry, Czechoslovak Academy of Sciences,Dolej&kova 3, 182 23 Prague 8, Czechoslovakia
SUMMARY HZSM-5 zeolite was completely deactivated during acetone conversion at 35OoC as a result of the poisoning of bridging OH groups by the coke deposit as well asobecause of pore blocking.The coke residue was oxidized at 500 C to various degrees and the restoration of hydroxyl groups and void volumes was checked together with the changes in the remaining coke deposit.The most pronounced changes in the adsorption properties occurred when first coke layers were oxidized and the zeolite pores were opened.The catalytic properties were renewed with the reformation of bridging hydroxyls. During the oxidation, intermediates of ketonic,carboxylate and carbonate character were created,and decomposed yielding CO and COz. INTRODUCTION The coking of zeolites in catalytic reactionstwhich affects the function of the active centers, depends on many factors, of which the dimensions of the pores and cavities and their ordering are of primary importance (refs.1-3) .The zeolite type has also been found to influence the coke oxidation: in HY zeolitestthe coke was oxidized at lower temperatures than that in HZSM-5 (refs.4,5) and the CO,/CO ratio in the final oxidation products was higher (ref.5). The access to coke deposits f o r the oxidizing agent and the formation as well as the decomposition of oxygenated intermediates probably play a decisive role in the coke oxidation path.Consequently, the process of oxidation of coke in a HZSM-5 zeolite was studied in more detail. METHODS NaZSM-5 zeolite (Si/A1=13.5) was supplied by the Research Institute for Oil and Hydrocarbon Gases, Czechoslovakia, and was transformed into the H-form by acid leaching (sample A).It was
406
deactivated in the conversion of acetone at 35OoC (70 vol% of acetone in a helium ~tream,WHSV=5h-~,25g of acetone passed over lg of zeolite) and then purged in a helium stream for 30 min (sample No 1) .The oxidation of this sample was performed at 5OO0C in a 20% oxygen-helium stream (600 ml h-') and the progress of the oxidation was checked by mass spectrometry ( a Finnigan 400 quadrupole mass filter) by recording the m/z ratio 32 for oxygen,28 for CO and 44 for C02.Water evolved during the oxidation was frozen out and its amount as well as the amount of coke in the zeolites were determined from weight changes. The oxidation was interrupted in chosen stages ( samples Nos. 2-4 denoted in Fig. 1 by arrows). Samples 1-4 were also pyrolyzed in a vacuum or in a The gaseous pyrolytical helium stream at 6OO0C (Nos. 1'-4'). products were also analyzed by mass spectrometry. Prior to all the experiments, the samples were pretreated either in a vacuum Pa) or in a helium stream at 35OoC. The hydroxyl groups and the coke composition were studied on zeolite platelets (7 mg cm-2) using a Nicolet MX-1E Fourier transform infrared spectrometer. The void volumes were determined by measuring sorption capacities for Ar at -195OC and a pressure of 13.3 kPa. Temperature programmed desorption (TPD) of ammonia and benzene was also employed: the vapour was adsorbed on 0.19 of the zeolite at room temperature for 30 min, excess adsorbate was evacuated up to a pressure of Pa and the TPD curves were measured by recording the pressure changes using a calibrated Pirani gauge with a heating rate of 10°C min-l. The IR bands of the hydroxyl groups, vibrating at 3610 cm-l (bridging hydroxyls) , were related to the values for the parent zeolite (A) and listed in TABLE 1 together with the coke amounts (wt%) and oxidation levels ( 0-1 for the completely coked and completely oxidized sample, respectively). The ability to exchange the hydrogen of hydroxyl groups and of coke for the deuterium of perdeuterated benzene (supplied by the Central Institute for Isotopes and Radiation Studies, Acad. Sci.GDR, enrichment 99.9 at % ) was also studied: 0.13 g of samples A,1, 1',2,3 and 4 were allowed to exchange their hydrogen atoms with 0.02 mmol of C6D6 at 30OoC. The molecular peaks of the benzenes were recorded using the mass spectrometer working at an ionization energy of 20 ev. The catalytic activity of the partially oxidized samples was checked using the temperature programmed conversion (TPC) of
407
preadsorbed acetone: 1 mmol g-lof acetone was preadsorbed at room temperature on 0.039 of the zeolite.The weakly bonded acetone was desorbed at the same temperature for 15 min,and then the TPC was recorded by introducing the gaseous desorbed products directly into the vacuum system of the mass spectrometer. The mass spectra were scanned in 20-30°C steps and the products were normalized with respect to the fragmentation and ionization cross sections. RESULTS Oxidation - coke c o m The oxidation progress of the coked HZSM-5 (No l), yielding partially oxidized samples N o s . 2-4, is depicted in Fig.1. The partially oxidized sample No 2 exhibits the lowest oxidation level and almost the same coke amount as the pyrolyzed sample 1' (TABLE
0
100
200 min
3160
2840 cml
Fig.1. (left) Progress of the oxidation of coked HZSM-5 (No 1) at 50OoC. Curves a,b and c correspond to the peaks with m/z 32 (oxygen),28 (carbon monoxide) and 44 (carbon dioxide),respectively. The interruptions of the oxidation are denoted with arrows. Fig.2. (right) IR spectra of coked,pyrolyzed and partially oxidized samples.Numbering as in TABLE 1.
408
TABLE 1 Characteristics of coked, pyrolyzed and partially oxidized HZSM-5 in comparison with the parent uncoked zeolite Sample A 1 1’
Coke,wt%
0
10.7 9.2
2
9.7
3 4
515
1.5
Oxidation level
0 0 0.07 0.49 0.86
Void vol. Ar ,mmol/g 5.2 0.1 0.2
Bridging OH ( % ) 100 16 19
2.8
34
3,8 5.3
65 93
Fig. 3 . IR spectra of coked, pyrolyzed and partially oxidized samp1es after subtraction of the zeolite spectrum. Numbering as in TABLE 1. 1). Both the pyrolysis (during which mainly hydrogen,methane and
ethylene are released and condensation of the aromatic rings increases,ref.(6)) and the beginning of the oxidation lead to the
409
disappearance of the alkyl substituents on the aromatic rings as follows from Fig. 2 (the IR bands of stretching vibrations of CH3 and CH2 groups in the wave number interval of 2740-2944 cm-1) .Bands at 1643 cm-’which correspond to residues of intermediates of acetone conversion (Fig.3, refs.(6-8)) also disappear. Less condensed aromatics ( bands at 1500-1541 cm-l) are also missing in both samples 1’ and 2. However, in contrast to the pyrolyzed sample 1’ sample No 2 still contains some of these monoaromatics (Fig.3), and, in addition, new bands at 1730 and 1690 cm-’, which can be removed by pyrolysis (samples Nos.2 and 2’,Fig.3) with the evolution of CO and C02 and weight loss smaller than l%.Samples Nos. 3 and 4 with a higher oxidation level exhibit the same IR spectra for coke as sample No 2, but in a lower intensity (Figs.2 and 3) .The concentration of free bridging hydroxyls increases with the proceeding oxidation (TABLE 1). Oxidation - sorption properties The changes of the sorption properties of the test samples are listed in TABLE 1 and depicted in Fig.4 (a) and (b).The Ar sorption capacity, corresponding to the zeolite void volume, is very low for the coked zeolite No 1,and increases only slightly after pyrolysis (1’).In contrast, sample No 2 (of the lowest oxidation level) exhibits a sharp increase in the void volume
Fig.4.TPD of ammonia (a) and benzene (b).Numbering as in TABLE 1.
410
value.The same behaviour can be seen in the values of the LTP (low temperature peak) for ammonia desorption (Fig.4a).The LTP of benzene also exhibits a lower value for sample No 2 (Fig.4b). Isotope exchange The kinetics of the isotope exchange of deuterium between perdeuterated benzene and the hydrogens of OH groups and coke residues is shown in Fig.5.It can be seen that a plot of the In of the concentration vs. time obeys a first order law for samples A and 2-4, while a strong deviation is visible for the completely coked zeolite (1) and its pyrolyzed form (lt).Fig. 5 also shows that the exchange rate is slower for the hydrogen atoms of coke than for the hydroxyl groups. The number of H atoms participating in the exchange is given in TABLE 2. The calculation was based on the exchange equilibrium values obtained after 15 h of reaction and the known amount of benzene employed in the exchange: N = M * (co-coo)/coo (1) where N is the number of H atoms participating in the exchange with M deuterium atoms of benzeneland co and coo are the concentrations in times to and too,respectively. TABLE 2 Number of H atoms per zeolite g exchanged with Sample A
H x
2
1.3
0.63
Sample 1 3
H x 3.8 0.78
C6D6
sample 11
4
at 30OoC H x i-21
2.1 0.56
Catalytic activity The restoration of the active centers for acetone conversion (strong Bronsted acid sitestbridging hydroxyls (refs.7,8)) exhibits different behaviour than the restoration of void volumes. The acid centers are renewed in closer proportion to the decrease in the coke amount (increased oxidation level) which can be seen in TABLE 1 and Fig.4a (HTP -high temperature peak- of ammonia desorption). The HTP of benzene seems not to reflect only the number of strong acid centers. The reformation of the catalytic activity of the zeolites studied can be seen in Fig. 6. In this Fig. , the TPC curves are shown for the parent HZSM-5 (A) and samples 1,1t,2,3 and 4. In agreement with the above described adsorption properties, the amount of physically adsorbed acetone (i.e. not converted, released with maximum rate at 75OC ) is low for samples 1 and l t ,
411
but high and almost the same for the remaining samples. However, this is not true for the catalytic transformation of preadsorbed that the restoration of the acetone. It follows from Fig. 6 catalytic activity corresponds to the number of freed OH groups.
t
4
0
200
Fig.5. (left) Isotope exchange of C6D6 with the hydrogen of the - Coo); COfCt, and coo are concentrations of D atoms at times toft and too,respectively. acetone,--isobutene, Fig.6.(right) TPC of preadsorbed acetone.. .C O ,~Iaromatics. OH groups and of coke at 30OoC. F= (co- Ct )/(co
.
DISCUSSION Effect of partial oxidation on the coke composition and zeolite void volumes Both the pyrolysis of the completely coked HZSM-5 and partial oxidation result in some similar changes in the coke:in the removal of the alkyl substituents on the aromatic rings and the remainders of the acetone conversion intermediate as well as in the decrease of the light aromatic coke components ( especially after the pyrolysis). It can be concluded that the oxidation of these three coke components proceeds more readily than the
412 oxidation of the highly condensed coke.This also follows from the temperature programmed oxidation of coke (5).In contrast to the pyrolysis,the initial stages of the oxidation involve removal of the components of the coke that block the pore openings, as the value of the void volume substantially increases.Nevertheless, the loss of weight due to the low-level oxidation,compared with the pyrolysis, is almost the same.This means that the coke components blocking the pore openings constitute only a small fraction of the total coke amount. A s the oxidation leads to the formation of oxygenated intermediates, their presence can misrepresent the data for the weight losses. These oxygenated intermediates are of ketonic as well as carboxylate- and carbonate-like character.The bands observed at 1730 and 1690 cm-l agree well with those reported in (ref.4) for coke partially oxidized and freed from the zeolite matrix by its dissolution in fluoric acid.The authors assigned these bands to ketone-like compounds.Quinones as intermediates in the coke oxidation in zeolites were also assumed on the basis of EPR measurements (ref.g).Various oxygenated intermediates were found in the oxidation of gaseous aromatic compounds (refs.lO,ll).The oxidized coke components observed in our measurements are thermally decomposed with the evolution of CO and C02:they are most probably the source of the CO in the final oxidation products, as was suggested in (ref.5).Their amount does not exceed 1% of the total coke amount. The results of the isotope exchange measurements ,on one hand,confirm the opening of the pores for oxygen at the beginning of the oxidation ( the deviation of the first order law in the kinetics on samples 1 and l'in contrast to samples 2-4 and sample A ) , and, on the other hand, enable us to evaluate the number of hydrogen atoms taking part in the exchange. For the parent zeolite ( A ) , the agreement between the number of OH groups calculated on the basis of the chemical composition and determined from the exchange data is very good. The number of hydrogen atoms taking part in the exchange between coke (samples 1 and 1') and deuterated benzene constitutes only 6 0 % of the total hydrogen amount.This could be most probably explained by the inaccessibility of the remaining coke hydrogens for benzene (cf. also Fig.4b).
413
Effect of partial oxidation on restoration of active centers and catalytic activity There are a number of assumptions on the deactivation of zeolites in catalytic reactions. Some authors emphasize the blocking of pores, e.g (refs. 2-4), others the simultaneous effect of poisoning of active centers ( refs.12-15). The results presented here favour the latter suggestion: the catalytic activity was clearly restored with reformation of the active centers.Only after the opening of the pores was the oxygenated agent able to penetrate inside the cavities and liberate the active centers. CONCLUSIONS i) The oxidation of coke formed from acetone in a HZSM-5 zeolite begins at the alkyl substituents on the aromatic rings and with unblocking of the zeolite pores. ii) The oxidation proceeds via the formation of oxygenated intermediates which decompose yielding CO and C02. iii) The restoration of the catalytic activity is related to the reformation of the active centers ( bridging hydroxyls). The opening of the pores, enabling the access of the oxidizing agent to the coke components,is a necessary condition for the reformation of the active centers. REFERENCES D.E. Walsh and L.D. Rollmann, Radiotracer experiments on carbon formation in zeolites.11, J.Catal.,56 (1979) 195-197 and references therein. P-Magnoux, P.Cartraud, S.Mignard and M.Guisnet, Coking,aging and regeneration of zeolites, J.Catal.,l06 (1987) 242-250. M.Guis.net and P.Magnoux, Coking and deactivation of zeolites,Appl. Cata1.,54 (1989) 1-27,and references therein. P.Magnoux and M.Guisnet, Coking,ageing and regeneration of zeo1ites.VI. Comparison of the rates of coke oxidation on HY,H-Mordenite and HZSM-SIAppl.Catal.,38 (1988) 341-352. 5 J.Novakova and Z.Dolejsek,A comment on the oxidation of coke deposited on zeolites,Zeolites,lO (1990) 189-192. 6 J.Novakova,L.Kubelkova,V.Bosacek and K.Mach, Zeolites,in press. 7 J.Novakova,L.Kubelkova,P.Jiru,S.Beran and K. Nedomova, A study zeolite, of the interaction of some ketones with HZSM-5 Proc.Int.Symposium on Zeolite Catalysis,Siofok,Hungary,May 13-16,1985,Acta Physica et Chemica SzegediensisIl985,pp.56l-578. 8 L.Kubelkova,J.Cejka,J.Novakova,V.Bosacek,I.Jirka and P.Jiru, Acetone conversion and deactivation of zeolites, in: P.A.Jacobs and R.A.van Santen (Eds), Zeolites: Facts, Figures, Future,Elsevier, Amsterdam,1989,pp.l203-1212. 9 H.G.Karge,E.P.Boldingh,J.-P.Lange and A.Gutsze,Studies on coke formation on dealuminated mordenites by in-situ IR and EPR measurements,Proc. Int. Symposium on Zeolite Catalysis,Siofok, Hungary, May 13-16,1985,Acta Physica et Chemica Szegediensis, 1985,pp.639-648. 1
414 10 G.Rotzoll,Molecular beam sampling. Mass spectrometric study Of
high-temperature benzene oxidation, 1nt.J. Chem. Kinet., 17 (1985) 637-653. 11 E.A. Stemmler and M.V.Buchanan,Negative ions generated by reactions with oxygen in the chemical ionization source, Org. Mass Spectrometry,24 (1989) 94-104. 12 E.C. Derouane,Factors affecting the deactivation of zeolites by coking, in: B.Imelik et al.(Eds.),Catalysis by Acids and Bases, Elsevier,Amsterdam,1985,pp.221-240. 13 D.M. Bibby,N.B. Milestone,J.E. Patterson and L.P. Aldridge,Coke formation in zeolite ZSM-5,J.Catal. 97 (1986) 493-502. 14 L.M. Parker and D.M. Bibby,Effects of coke formation on the acidity of ZSM-5,J.Catal. 99 (1986) 486-491. 15 K.S. Tsakalis,T.T. Tsotsis and G.J. Stiegel, Deactivation
phenomena by site poisoning and pore blockage.The effect of catalyst size,pore size and pore distribution, J.Cata1. 88 (1984) 188-202.
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
415
INFLUENCE OF THE CONDITIONS OF DEALUMINATION AND THE EFFECT OF PARTIAL EXIRACTION OF NON-FRAMEWORK ALUMINIUM ON THE CATALYTIC PROPERTIES OF ZSM-5 CATALYSTS
G.oHLMANN1, H . G . J ~ R S C H K E W I T Z ~ , G-LISCHKEI, R.ECKELT1, B.PARLITZ1, I.SCHULZ1, K.WEHNERz and D.TIMM2
T.GROSS',
l C e n t r a 1 I n s t i t u t e o f P h y s i c a l Chemistry o f t h e Academy o f Sciences zLeuna-Werke AG, 4220 Leuna 3
SUMMARY Samples o t P-modified LSM-5 t y p e z e o l i t e p r e v i o u s l y d e a l u m i n a t e d by steaming a t d i f f e r e n t temperatures were used as c a t a l y t i c a l l y a c t i v e components i n t h e c o n v e r s i o n o f methanol t o l o w e r o l e f i n s a t 5bO"C. R e s u l t s show t h a t d e a l u m i n a t i o n a t 400°C f o l l o w e d by t r e a t m e n t w i t h 2 N n i t r i c a c i d a l l o w s h i g h e s t degrees o f p a r t i a l e x t r a c t i o n o f non-framework aluminium and o f f e r s b e s t r e s u l t s w i t h r e s p e c t t o s e l e c t i v i t y o f C z / C 4 - o l e f i n s and c a t a l y s t s t a b i l i t y .
INTRODUCTION LSM-5 t y p e z e o l i t e s a r e a p p r o p r i a t e c a t a l y s t s f o r t h e methanol(1-5). Cz/C4-olefins reach s a t i s t y i n g to-olefin conversion s e l e c t i v i t i e s , P r o v i d e d t h a t t h e ~ l / A l F r - a r n e ~ ~( ~S -i /kA l F ) r a t i o i s sufficiently h i g h . S e l e c t i v i t y o f o l e f i n s can be i n c r e a s e d a d d i t i o n a l l y by t r e a t i n g t h e samples w i t h p h o s p h o r i c a c i d ( 6 - 1 0 ) . Even w i t h those P-modified z e o l i t e s , h i g h v a l u e s o f t h e S i / A l F r a t i o e x e r t a p o s i t i v e e f f e c t on t h e f o r m a t i o n o f t h e d e s i r e d olef ins. Besides by a p p r o p r i a t e a d a p t a t i o n o f t h e c o n d i t i o n s o f s y n t h e s i s i t s e l f , aluminium d e f i c i e n t z e o l i t e s can be o b t a i n e d by h i g h temp e r a t u r e c a l c i n a t i o n o r steaming o f samples which o r i g i n a l l y cont a i n a h i g h e r f r a c t i o n o f aluminium (11). The degree o f dealuniinut i o n s t r o n g l y depends on t h e e x p e r i m e n t a l c o n d i t i o n s under which t h e steaming o t t h e sample Proceeds. BY v a r i a t i o n o f t e m p e r a t u r e
416
and time, a broad range o f t h e S i / A l F r a t 1 0 1s e x p e r i m e n t a l l y acc e s s i b l e , On t h e o t h e r hand a z e o l i t e w i t h a d e s i r e d S i / A l F v a l u e can be generated by v a r i o u s c o m b i n a t i o n s o f t h e two mentioned e x p e r i m e n t a l parameters. D e a l u m i n a t i o n o f aluminium r i c h z e o l i t e s , w h i c h seems e c o n o m i c a l l y attractive, i s n e c e s s a r i l y accompanied by t h e f o r m a t i o n o f a l u m i nium s p e c i e s i n non-framework p o s i t i o n s ( 1 2 ) . I t i s o f i n t e r e s t whether aluminium d e f i c i e n t z e o l i t e s w i t h i d e n t i c c o n t e n t s o f aluminium i o n s i n t h e framework o f s i l i c a , which, however, a r e o b t a i n e d by d e a l u m i n a t i o n o f an aluminium r i c h p a r e n t sample under d i f f e r e n t experimental conditions, d i f f e r with respect t o t h e e x t r a c t a b i l i t y o f t h e non-framework aluminium and t o t h e i r catalytic Properties. I n t h e p r e s e n t Paper t h e i n f l u e n c e o f t h e d u r a t i o n o f dealuminat i o n on t h e c a t a l y t i c b e h a v i o u r i s s t u d i e d w i t h emphasis on t h e e x t r a c t i b i l i t y o f non-framework aluminium ( A l n ~ ) and on s t a b i l i t y and s e l e c t i v i t y i n t h e c o n v e r s i o n o f m e t h a n o l . A d d i t i o n a l l y some o f t h e samples were c h a r a c t e r i z e d by p h o t o e l e c t r o n s p e c t r o s c o p y (XPS), by temperature programmed d e s o r p t i o n o f ammonia (NHs-TPD) and by d e t e r m i n a t i o n o f t h e coke d e p o s i t s on t h e spend c a t a l y s t .
EXPERIMENTAL
PrePara.tiPc! . . o ~ f s n m f l ~ . P e n t a s i l t y p e ZSM-5 z e o l i t e (HS-30, Si/A1=19, Chemie AG B i t t e r f e l d ) i s c o n v e r t e d i n t o t h e H+-form by a f o u r t i m e s r e p e a t e d t r e a t m e n t w i t h a 0 . 2 N s o l u t i o n o f HNOs i n a s i m p l e b a t c h p r o cedure. F o r d e a l u m j n a t i o n , t h r e e s e p a r a t e samples o f t h e Hi-form a r e exposed t o a stream o f a i r and steam (pweter=90 kPa) a t 400, 470 and 700OC. To a c h i e v e a p p r o x i m a t e l y i d e n t i c S i / A l F r a t i o s , t i m e o f steaming was 70. 6 and 0.25 hours, r e s p . Framework a l u m i nium o f each sample was d e t e r m i n e d u s i n g t h e ammonium exchange methode (11). Subsequently P a r t o f each o f t h e samples was t r e a t e d w i t h 2 N HN03 a t 103°C f o r p a r t i a l e x t r a c t i o n o f non-framework aluminium formed d u r i n g d e a l u m i n a t i o n . E x t r a c t e d aluminium was determined a n a l y t i c a l l y i n t h e s o l u t i o n o f t h e e x t r a c t i n g a g e n t . C a t a l y s t p a r t i c l e s were p r e p a r e d by m i x i n g t h e HS-30 samples w i t h A e r o s i l - 2 0 0 and water and subsequent e x t r u d i n g , t h e ready made e x t r u d a t e s (6 = 1 t o 2 mm) c o n t a i n i n g 65% by w e i g h t o f z e o l i t e . Pm o d i f i c a t i o n was c a r r i e d o u t by i m p r e g n a t i o n w i t h aqueous phosphor i c a c i d . The c o n t e n t o f phosphorus f o r c a t a l y s t s based upon
417
extracted HS-30 was held somewhat lower taking into consideration their diminished contents of total aluminium. Finally catalyst samples were conditioned by steaming at 700°C for 1 hour. (;o.tnlY 51s
Catalytic experiments were carried out in a tubular fixed bed reactor ($ = 12 mm) using a mixture of methanol/Nz (molecular ratio= 1 : l ) as reaction feed (conditions o f reaction see table 1). Products were analyzed by an on-line operated gaschromatographic device. Calculation o f the concentration from analytical data is based on individual calibration of the components.
TABLE 1: Parameters of reaction
t einpera t u r e : concentration of methanol: catalyst volume: catalyst mass: feed rate (mixture): c h.at-a GHSV(mixture) : LHSV(methano1):
560°C
50 Vol-% 10 ml 5.0 3.33 ml/s
(STP)
1200 ml/mlcataivst*h 1 ml/mlcata~ystffh .. .- . . .
.
~~~
. .
. . ..
..
. -. . . . . .
Photoelectron spectrosc.oPr XPS investigations were carried out with an AEI ES200B electron spectrometer, using MgKd radiation. Resulting signal intensities required a spectra accumulation based on the PDP8e/DS800 data system. Ine analyzer system ot the spectrometer was calibrated according to Anthony and Seah ( 1 3 ) . Evaluation of XPS intensity ratios to derive atomic ratios on the surface was based on the model of homogeneous distribution (14).
Temperature programmed dessrc~tiso.~Lammonia Atter cleaning zeolite samples by heating to 500°C in a He gas stream for 1 hour, ammonia is adsorbed at 1 2 O O C from a gas stream contuining 3 Vol-% of NH3. After flushing by Pure He at 120°C for two nours, desorption of ammonia UP to 5OOOC i s started (heat rate 1L OC/min). The concentration of NH3 in the exit gas is determined using a thermoconductivity cell. The curve o f the desorptogramm is
418
r ecorded by a c o n v e n t i o n a l d e v i c e . Temperature Programmed o x i d a t i o n Carbonaceous d e p o s i t s formed d u r i n g r e a c t i o n on t h e c a t a l y s t were det ermin e d by te mp e ra tu re Programmed o x i d a t i o n . Coke was burned oft i n a stream o t d r y a i r i n t h e t e m p e r a t u r e range between 200 and 700 OC, t h e heat r a t e b e i n g 5 " C h i n . T h i s p r o c e s s was monit ored c o n t i n e o u s l y by r e c o r d i n g c a r b o n monoxide and d i o x i d e . The o x i d e s were s e p a ra te d by a Cekatron 5 column and d e t e r m i n e d g aschromot ogr a p h i c a l l y a f t e r c o n v e r s i o n t o methane. Carbon ( m c ) was c a l c u l a t e d from t h e t o t a l amounts o f t h e o x i d e s .
RESULTS
Experiment al r e s u l t s o f methanol c o n v e r s i o n w i t h t h e s i x c a t a based on HS-30 samples o f d i f f e r e n t P r e p a r a t i o n a r e r e p r e sented i n t a b l e s 2 and 3 . Values o f p r o d u c t s e l e c t i v i t y a r e r e l a t e d to t h e p o r t i o n s o f carbon o f fe d methanol w h i c h a r e c o n t a i n e d i n t h e g i v e n component o f t h e p r o d u c t : lysts
vr
=
hl
=
IL .-
nuinber o f carbon atoms Per m o l e c u l e o f p r o d u c t molor- amount o f p r o d u c t formed Per t i m e u n i t molar amount o f methanol c o n v e r t e d Per t i m e u n i t
Data r e p o r t e d i n t h e t a b l e s a r e v a l u e s averaged over t h e whole r e a c t i o n o e r i o d . The t i m e on stream when methanol f i r s t appears i n t h e e x i t a n s o f t h e r e a c t o r i s denoted as t i m e o f d e a c t i v a t i o n . D u r i n g record e d times o f experiment no d i m e t h y l e t h e r was d e t e c t a b l e among t h e p r o d u c t s o f r e a c t i o n . A l l c a t a l y s t s c o n t a i n i n g n o n -e x tra c t e d z e o l i t e samples ( T a b l e 2 ) g i v e r a t h e r h i g h s e l e c t i v i t y values f o r t h e d e s i r e d products - the lower o l e f i n s . Propene e x h i b i t s t h e h i g h e s t p o r t i o n among them. By v a r i a t i o n o f conditions o f dealumination o n l y gradual d i f f e r e n c e s a r e observed w i t h i n t h e d i s t r i b u t i o n o f p r o d u c t s . Values o f t i m e on stream a r e r e l a t i v e l y low, w i t h medium c o n d i t i o n s g i v i n g a maximum ,
419
TABLE
2:
E f f e c t o f steam d e a l u m i n a t i o n on p r o d u c t s e l e c t i v i t i e s and c a t a l y s t s t a b i l i t y Non-extracted HS-30 z e o l i t e Content o f phosphorus 1 . 5 w t - %
catalyst D93P1.5 temperature o f steaming " C 400 ti me o f steaming h 70 molar S i / A l F r a t i o 93 - - -- -_-~- product s e l e c t i v i t i e s _ % ethene 17,O 36,2 Propene b u t enes 23,4 ~
Ca/C4
olefins
- _ _ _ _ _ _ _ _ _ _ - - - - - met hone ethane Propane i - b u t one n-butane
76,6
D76P1.5
DS6P1.0
470 6 76
0,25
86
_ _ _ - ~
- - - - 1,s 0,18 1,5 1.3 0,78
700
17,O 36,2 24,2
17,3 37,O 23,2
77.8
77,s
1,9 0,16 1,6 1,6 0,57
2,4 0,22 1,7 1,3 0,51
- - - - - - - - - - - - - - - - 10,8 Ca-a1 i p ha t i c s
8,6
benzene toluene xvlenes
0,17 0,86 2,2
0,19 1,o 2,3
0,20 1,l 2.6
BTX aromat ics
3,2
3,s
3,9
t i m e o f d e a c t i v a t i o n (h )
47
58
25
_ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - - - - - - - - - - - - - -
The ses
__
10,o
amount o f e x t r a c t e d non-framework aluminium ( Ta b l e 3) i n c r e a as d u r a t i o n o f d e a l u m i n a t i o n i s i n c r e a s e d . C a t a l y s t s t a b i l i t y - expressed i n terms o f ti m e on stream - i s i n c r e a s e d by e x t r a c t i o n i n any case. The s t r o n g e s t e f f e c t i s a c h i e v e d i f dealuminat i o n proceeded under l o n g t i m e c o n d i t i o n s : The v a l u e f o r t i m e on stream i s r a i s e d f o r loo%, as compared w i t h t h e n o n - e x t r a c t e d sample.
420
TABLE
3:
E f f e c t o f steam d e a l u m i n a t i o n on p r o d u c t s e l e c t i v i t i e s and c a t a l y s t s t a b i l i t y E x t r a c t e d HS-30 z e o l i t e , c o n t e n t o f phosphorus 1.1 w t - %
catalvs t temperature o f steaming * C t i m e o f steaming h m olar S i / A l F r a t i o degree o f e x t r a c t i o n o f A l n ~
D93EP1. CI 400 70 93 22 ~-
~
Di-eEPi, n 470 6 76 15 -
De6EPi. 700 0,25 86 6
5
_ .
p roduct s e l e c t i v i t i e s X e t hene 17,O 18,9 19,5 Propene 37,6 30,6 27,9 b u t enes 25,l 18,5 16,9 - - - - - - - _ - - - - - - - - - - - - - - - - _ - - - - - - - _ CZ/&
olefins
79,7
68,O
64,4
methane ethane propane i - b ut ane n-butane
2,8 0,19 l,4 1,6 0,48
3,4 0,34 2.4 1,6 0,57
6,3 0,47 2,8 1,9 0,86
Cs-aliphatics
7,l
6,4
7,2
benzene toluene xylenes
0,25 1,1 2,2
0,41 1,7 3,l
0,62 2,6 4,8
BTX aromat ics
3,5
5,2
8,O
96
68
33
_ _ _ _ _ _ - - - - - - - - - - - - - - - - - - - - - - - - - - -
ti me o f d e a c t i v a t i o n ( h ) - .
...
-
The i n f l u e n c e o f P a r t i a l removal o f non-framework a l u m i n i u m on c a t a l y s i s c l e a r l y depends on t h e c o n d i t i o n s o f d e a l u m i n a t i o n o f t h e z e o l i t e . T h i s can be concluded from t h e comparison o f any o f the e x t r a c t e d w i t h t h e c o rre s p o n d i n g n o n - e x t r a c t e d samples. Whereas a f t e r l o n g t i m e d e a l u mi n a t i o n , e x t r a c t i o n r e s u l t s i n a s l i g h t increase o f o l e f i n s e l e c t i v i t y , d i s t i n c t losses o f t h e i r p o r t i o n among t h e p r o d u c t s a r e found w i t h t h e two o t h e r c a t a l y s t s . The decrease o f t o t a l s e l e c t i v i t y i s caused by ProPene and t h e
421
butenes. S e l e c t i v i t y o f ethene i s n o t n e g a t i v e l y i n f l u e n c e d as t i m e o f d e a l u m i n a t i o n i s decreased: I n c o n t r a s t t o Propene and t h e butenes i t s v a l u e s l i g h t l y i n c r e a s e s w i t h d e c r e a s i n g t i m e o f steaming. I t i s w o r t h m e n ti o n i n g t h a t e x t r a c t i o n o f non-framework a l u minium f avou rs t h e f o r m a t i o n o f methane, b u t s l i g h t l y suppresses t h e accumulati o n o f C s - a l i p h a t e s . F or p r a c t i c a l purposes o f t h e methanol c o n v e r s i o n , a c a t a l y s t based on an e x t r a c t e d , P-modified ZSM-5 z e o l i t e p r e v i o u s l y dealuminat ed by steaming a t temperatures i n t h e r e g i o n o f 4000C s h o u l d be P r e f e r e d . S e l e c t i v i t y o f l o w e r o l e f i n s and c a t a l y s t s t a b i l i t y a chieve a t t r a c t i v e v a l u e s . D e t e r m i n a t i o n o f coke on spend c a t a l y s t s ( Ta b l e 4) r e v e a l s t h a t t h e r a t e o f coke d e p o s i t i o n i s h i g h e r w i t h t h e n o n - e x t r a c t e d samp l e s . These c a t a l y s t s , a l t h o u g h performed o v e r a d e f i n i t e l y s h o r t e r t i m e th a n t h e i r e x t r a c t e d c o u n t e r p a r t s , show p r a c t i c a l l y i d e n t i c amounts o f coke. The s u p p re s s i o n o f t h e d e p o s i t i o n o f coke i s o b v i o u s l y due t o t h e removal o f non-framework aluminium, because t h e e f f e c t i s found w i t h b o t h P-modified and P - f r e e catalysts. TABLE
4:
E f f e c t o f p a r t i a l e x t r a c t i o n o f non-framework aluminium
on c a t a l y s t d e a c t i v a t i o n and d e p o s i t i o n o f coke Catalyst
ti m e o f d e c t i v a t i o n I hours )
amount o f coke d e p o s i t e d (mgcoke/gzeoli
te)
31,5 35,8
The
8
119,4
10 12 26
126,s 114,3 110,6
e v a l u a t i o n o f XPS d a t a (T a b l e 5) r e v e a l s t h a t w i t h t h e p a r e n t HS-30 z e o l i t e t h e S i / A l r a t i o on t h e s u r f a c e i s h i g h e r t h a n t h e average S i / A l v a l u e d e r i v e d from NHe+-exchange d a t a . D i s t r i b u t i o n o f aluminium over t h e c r y s t a l i s s l i g h t l y inhomogeneous. W i t h t h e dealuminated samples, d i f f e r e n c e s between t h e s u r f a c e and t h e
422
TABLE 5 :
XPS i n v e s t i g a t i o n , r e l a t i v e X P S i n t e n s i t i e s ( d e v i a t i o n f 10%). Hs-30
sample
ratio of intensities
Al/Si ratio XPS
IAIZP/ISIZP
n on-t reat ed steamed f o r Wh a t 700°C steamed f o r 72h a t 400°C
0,038
0,031 0,056
0,143
0,213
0,021
Si/A1 ratio XPS (surface)
32 18
Si/A1 ratio NH4 +-exchange (bulk)
18
86
4,7
93
average S i / A 1 r a t i o a r e c o n s i d e r a b l e . S u r f a c e v a l u e s a r e c l e a r l y s h i f t e d i n fa v o u r o f t h e a c c u m u l a ti o n o f a l u m i n i u m . B o t h samples have n e a r l y i d e n t i c S i / A l F r a t i o s and t h e r e f o r e do n o t d i f f e r i n t h e i r t o t a l amounts o f non-framework a l u m i n i u m . The s t r o n g d e v i a t i o n s o f t h e r e l a t i v e X P S i n t e n s i t i e s between t h e two d e a l u m i n a t e d samples show t h a t d i s t r i b u t i o n s o f non-framework a l u m i n i u m a r e d i f f e r e n t . I t can be assumed t h a t d e p o s i t i o n o f a l u m i n i u m on nonmicroporous s u r f a c e areas, i . e . i n t h e system o f mesopores o r on t h e e x t e r n a l s u r f a c e area, i s fa v o u re d by l o n g t i m e steaming. The t o t a l number of acid sites i s i n c r e a s e d by ex> traction. This can be seen from the results o f $ temperat ure p r o 3 srammed desorp:. t i o n o f ammonia o f the P-f r e e zeolite (Fig.1). 7" lhe amount o f 2& jw h c o ,oo T / '2 200 100 .:OO 5W ammonia adsorbed -F-i g_ .- -1: Te_mmp_er_at_ur_e _pr_os_amm_eed_~e.s_ollPJi~no_f -a&m-n_ii_a Dealurnination D93 dealuminated HS-30 ( S i / A 1 ~ = 9 3 ) D93E d e a l u mi n a te d e x t r a c t e d Hs-30 a t 4000 C (degree o f e x t r a c t i o n = 22%) - - - - - - - - - - - - - - - - - - - ._ f o r 70 h D e e dealuminated HS-30 ( S i / A 1 ~ = 8 6 ) Dealumination D e 6 E dealuminated e x t r a c t e d HS-30 a t 7000C (degree o f e x t r a c t i o n = 6%) for 0 2 5 h
:,
I (
i
-
I
1
423
on places o f s t r o n g i n t e r a c t i o n (maximum o f t h e c u r v e o f desorpt i o n a t about 400°C) achieves h i g h e r v a l u e s w i t h b o t h e x t r a c t e d samples. I t is obvious t h a t t h e g ro w th o f a c i d i t y is more s t r o n g l y Inarked w i t n t h e i o n s t i m e dealuminated z e o l i t e . DiSCbSSlUN .The e x p u l s i o n o t t e t r a h e d r a l aluminium from s u b s t i t u t i o n a l P o s i t i o n s i n t h e framework o f s i l i c a g i v e s r i s e t o t h e f o r m a t i o n o f non-framework aluminium o f v a r i o u s c h e m i c a l n a t u r e ( 1 2 ) . Aluminiiiin species w i t h t e t r a h e d r a l l y and o c t a h e d r a l l y c o o r d i n a t e d c e n t r a l i o n s ( 1 s ) o f d i f f e r e n t degrees o i a g g r e g a t i o n a r e d e p o s i t e d i n t h e z e o l i t e channels and channel i n t e r s e c t i o n s . M i g r a t i o n t o t n e e x t e r n a l s u r f a c e i s p o s s i b l e . I t can be assumed t h a t P a r t i a l b l o c k i n g o f pores by a e p o s i t s can cause a h i n d r a n c e o f t h e d i f f u s i o n o f t h e p r i m a r y p r o d u c t s o u t o f t h e p o r e system t h u s i n c r e a s i n g the p r o b a b i l i t y o f f u r t h e r undesired reactions o f the o l e f i n s . Besides, i t cannot be excluded t h a t non-framework a l u m i nium i t s e l f o r - as i n o u r case - aluminium phosphates formed by i n t e r a c t i o n w i t h t h e p h o s p h o ri c a c i d a r e c a t a l y t i c a l l y a c t i v e , P o s s i b l y e n a b l i n g h y d r i d e t r a n s f e r w i t h t h e o l e f i n s o r any o t h e r b y - r e a c t i o n s o f t h e feed component. C a t a l y t i c r e s u l t s show t h a t w i t h n o n - e x t r a c t e d samples t h e v a r i a t i o n o f t h e c o n d i t i o n s o f d e a l u m i n a t i o n has no s i g n i f i c a n t e f f e c t on t h e d i s t r i b u t i o n o f p r o d u c t s . D i f f e r e n c e s w i t h r e s p e c t t o Pore t r a n s p o r t phenomena and t o o t h e r i n f l u e n c e s o f t h e non-framework alumina on t h e f o r m a t i o n o f a n a l y t i c a l l y determined p r o d u c t s a r e n e g l i g i b l e . i t i s , however, e v i d e n t t h a t t h e f o r m a t i o n o f h i g h l y condensed compounds (carbonaceous d e p o s i t s . coke) i s a f f e c t e d by t h e n a t u r e o f non-framework aluminium. Dealuminated z e o l i t e s o b t a i n e d by s h o r t ti me steaming undergo t h e most r a p i d d e a c t i v a t i o n , as i s shown by t h e v a l u e s o f t i m e on stream and c o n f i r m e d by t h e d e t e r m i n a t i o n o f d e p o s i t e d coke. The dependence o f t h e degree o f e x t r a c t i o n o f non-framework a l u m i nium on t h e c o n d i t i o n s o f d e a l u m i n a t i o n c l e a r l y shows t h a t l o n g ti me steaming o t t h e z e o l i t e d u r i n g d e a l u m i n a t i o n f a v o u r s deposit i o n o t aluminium i n Places o f good a c c e s s i b i l i t y t o l i q u i d a g ent s. With growing t i m e o f d e a l u m i n a t i o n t h e f r a c t i o n o f nonframework aluminium whicn can m i g r a t e i n t o t h e r e g i o n o f poremouths o r t o t h e e x t e r n a l s u r f a c e a re a i s r a i s e d . T h i s assumption Is supported by t h e r e s u l t s o f p h o t o e l e c t r o n spectroscopy ( X P S ) . l h e h i g h e s t Al/Si r a t i o on t h e e x t e r n a l s u r f a c e i s found w i t h t h e sample steamed Tor 72 h o u r s .
424
Removal o t non-framework aluminium can improve c a t a l y t i c Propert i e s , I n any case c a t a l y s t s t a b i l i t y , expressed i n terms o f t i m e o f deactivation, i s i n c r e a s e d . A p o s i t i v e i n f l u e n c e on p r o d u c t d i s t r i b u t i o n i s found w i t h t h e l o n g t i m e d e a l u m i n a t e d sample o n l y . where t o t a l olef1.n s e l e c t i v i t v i s s l i g h t l y i n c r e a s e d . W i t h t h e r e m a i n i n g two c a t a l y s t s , total selectivity of olefins i s clearlv decreased i n f a v o u r o f t h e p r o d u c t s o f h v d r i d e t r a n s f e r ( l o w e r alkanes and BTX a r o m a t i c s ) . These l o s s e s o f t o t a l s e l e c t i v i t y o f olefins, t h e s h i f t o f d i s t r i b u t i o n w i t h i n the lower o l e f i n s i n favour o f ethene, and t h e decrease o f s e l e c t i v i t y o t CS a l i p h a t i c s i n d i c a t e t h a t w i t h d e c r e a s i n g t i m e o f dealurninat i o n , suhseauent e x t r a c t i o n o f non-framework aluminium, which i s l e s s e f f e c t i v e w i t h t h e soinole dealuminated f o r 1 5 m i n u t e s o n l v , i s riccomponied by t h e appearance o f a d d i t i o n a l s t r o n g o c i d s i t e s . l h e v m i a h t be c r e a t e d by t h e removal o f aluminium i o n s from non--tramework Phosphate d e p o s i t s . I h e i n c r e a s e o f t h e number o t s t r o n g a c i d s i t e s , which a r e m a i n l y r e s p o n s i b l e for c a t a l y t i c a c t i v i t y , a f t e r e x t r u c t i o n w i t h n i t r i c a c i d i s proved by t h e r e s u l t s o f NH3-IP1).
REFERENCES 1 C.D.Chang, C a t a l . R e v . - S c i . E n g . 2 5 , 1. l ( 1 9 8 3 ) 2 C.D.Chang, C a t a l . R e v . - S c i . E n g . 2 6 , 3 / 4 . 223(1984) 3 V.Ducarme. J . C . V k d r i n e . A ~ ~ l . C a t a l . 1 7175-84(1985) . 4 N.Y.Chan, W.E.Gorwood, C a t a l . K e v . - S c i . E n g . 2 8 , 185-264(1986) S P.Jarlagadda, C.F.R.Lund, E . R u c k e n s t e i n , A p p l . C a t . 5 4 , 135(1989) 6 L.B.Young, S . A . B u t t e r . W.W.Kaeding, J.Cata.L.76, 418-432(1982) i J . C.Vedrine, A . Auroux, P . D e J a i t r e , V . I)ucarire, n . H O S e r , 5 . Lliou J . C a t a 1 . 7 3 , 147-160(1982) 8 L.W . Z a t o r s k y , P . r . Wierzchowski. A . A . Cichowtas. 61111. P o l i s h Acad.Sciences, Chemistry, 32, 217-222(1984) 9 .J . A . I - e r c h e r . G . R~JmolmflVr. H .N o l l e r - . Acto Pbvs. Cheni. 5 1 , 71(1(485) 10 S h i g e r u I k a i , Manabu Okomoto. A u p l . C a t . 4 9 . '143(1989) 11 G.ohlaann, H.-G.Jerschkewitz, G.Lischke, b . P o r l i t z , K . t c k e l t , M . H i c h t e r , L . t.Cheinie 28, 5, 161-168 (1988) 1 2 J .Scherzer i n C a t a l y t i c M a t e r i a l s , ACS SYinposiuni S e r i e s 248, 157, Washington D . C . 1984 1 3 M.T.Anthony, M.P.Seah. S u r f . I n t e r f a c e A n a l . 6 . 95(1984) 14 M.P.Seah, i n P r a c t i c a l S u r f a c e A n a l y s i s ( t d s . D . B r i g g s . M . P . Seoh). W i l e y . 1983. ChaD.5 1 5 A.Samoson, E.Lipmaa, G . t n g e l h a r d t , U.Lohse, H . - G . J e r s c h k e w i t z Chem.PhYs. L e t t e r s 134, b, 589(1987)
425
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
SPECTROSCOPIC
AND
ChTALYTIC
INVESTIGATIONS
OF
HYDROTHERMALLY
DEALUMINATED ZSM-5
E. LOFFLER 1 ,
L.M.
u.
K A Z A N S K Y ~ , G.OHLMANN'
V.L.
KUSTOV',
CH. P E U K E R ~ ,
ZHOLOBENKO~,
n
LOME',
V.B.
'Central Institute of Physical Chemistry, Academy of the GDR, Berlin-1199
Sciences
2N.D.2elinsky Institute of Organic Chemistry, Academy of of t h e USSR, Moscow
of
Sciences
ABSTRACT The influence of different hydrothermal treatment conditions of HZSM-5 on the n-hexan cracking activity wa5 studied. The Bronsted and Lewis acid sites were characterized by diffuse reflectance IR spectroscopy using various probe molecules. The spectroscopic and catalytic results were discussed with t h e emphasis to t h e role of Lewis sites in the activity enhancement after steaming and in the mechanism o f t h e paraffin cracking. INTHODUCTION The influence of hydrothermal treatment on zeolites is of particular interest
from
the
the
properties
point
of
view
of of
preparation of cracking catalysts with high activity,
selectivity
and stability. Recently, an effect of a
enhancement
considerable
of t h e cracking activity for HZSM-5 zeolites pretreated under mild steaming conditions was reported (refs. (ref. 1 ) b y a strengthening of acidic bridged OH groups.
However,
It
1-3).
was
properties
hydroxyl
groups
of
explained part
a
exhibiting
of
acidic
properties stronger than OH groups of t h e initial zeolite have not been detected (refs.
1-3). In the
present
work
properties of Bronsted and Lewis acidic sites in deal urn1 nated under mi 1 d o r severe
the
nature
HZSM-5
hydrothermal
and
zeolites
condi ti o n s
studied by the highly sensitive method of diffuse spectroscopy. Different probe molecules ( H2 , CH4,
are
reflectance CH3CN) a r e
IR
used
t o characterize the Lewis acidic sites. EXPERIMENTAL The H Z S I I - 5 zeolite (Si/A1=18) was under mild (T = 810 K , pHz0
treated
with
water
= 1.3*104 Pa,t = 10-150 min)
or
vapour severe
426 5
= 10 Pa, t = 1.5 2 catalytic measurements the
conditions (T = 770 K, pH spectroscopic
or
pretreated at 690
-
770 K for 4 h
in
24
Before
samples
or
vacuum
h).
in
He
were
stream.
Diffuse reflectance IR spectra were measured using Perkin Elmer 580 B and Beckman Acta M V I I spectrophotometers according
procedure described by Kazansky
et
al.
measurements were carried out on an IR
(ref.
to
the
Transmission
4).
spectrometer
SPECORD
M80
(Carl-Zeiss-Jena GmbH). For identification of Lewis acid sites t h e low temperature adsorption of molecular hydrogen (ref. 4) a s
well
a s the adsorption o f methane (re+. 5 ) and acetonitrile (ref. 6 ) at 300 K were used. The catalytic data were
reactor with on-line n-hexane
and
GC-analysis
I-hexene
concentrations varying
were
of t h e used
from
obtained
to
using
products. feeds
as
10
vol
%.
a
flow
Mixtures with The
of
olefin reaction
temperature ranged from 670 t o 770 K. RESULTS
AND DISCUSSION
In accordance with t h e data obtained by Lago et al.
(ref.1) t h e
catalytic activity of the samples under investigation in
n-hexane
cracking is strongly depending on t h e steaming conditions, i.e.
on
t h e time of steam treatment.
kt'ko
I
*i
0
40
80
120
160
1440
tlmln Fig. 1.
The dependence of t h e kt/kO ratio on t h e time of
steaming
pretreatment of HZSM-5: 1-cracking of pure n-hexane, 2-cracking of t h e mixture containing 99% of n-hexane and 1X of 1-hexene.
427
Fig.1
shows t h e ratio kt/ko characterizing
the
enhancement
the cracking activity of t h e zeolites dealuminated for
t
of
minutes
a s cornpared with t h e initial sample, as a function of t h e t i m e
of
steaming. It follows from these data that at the initial stages of t = 40
mild dealumination t h e catalytic activity increase6:at it is almost 7 times higher than that of Further dealumination
leads
to
a
the
starting
progressive
decline
activity. A prolonged hydrothermal treatment in severe
of
a s well a s
the
values
of
the
the
product
apparent
the
conditions
(t = 24 h) results in a decrease of t h e activity a s compared t h e initial sample. On t h e other hand,
min
material.
with
distribution
activation
energy
of
50 kJ/mol are practically the same for t h e initial zeolite and for
the
samples
dealuminated
under
mild
steaming
strongly dealuminated zeolites this energy is 67
conditions
For
kJ/mol.
361 0 3665 I
3610 I
3700
3600
3700
3500
-3/cm-1
3600
3500
3 / c m -1
Fig. 2. Diffuse reflectance IR spectra of OH groups of HZSM-5 zeolites measured before ( 1 ) and after stearning under mild conditions: ( 2 ) t = 20 min, (3) t = 150 rnin and severe conditions: (4) t = 1.5 t i , (5) t = 24 h. In order t o explain the catalytic
data
the
HZSM-5
were
studied
hydrothermally
dealuminated
active by
sites
in
diffuse
428 r e f l e c t a n c e I R spectroscopy.
t h e I R s p e c t r a o f OH
I n Fig.2
i n t h e o r i g i n a l and dealuminated z e o l i t e s a r e
spectrum o f t h e unsteamed sample two narrow bands w i t h 3610 and 3740 cm-l
In
<
30 m i n ) under m i l d
groups
are
conditions
the
maxima
belonging t o t h e s t r e t c h i n g v i b r a t i o n s
a c i d i c bridged h y d r o x y l s and t e r m i n a l S i O H Short steaming ( t
groups
presented.
observed.
results
in
decrease o f t h e band i n t e n s i t y o f a c i d i c h y d r o x y l s as w e l l
as
t h e appearance o f two new bands a t 3780 and 3665 cm-'.
After
min of steaming t h e i n t e n s i t y o f t h e band a t 3780 cm-l
starts
d i m i n i s h and t h a t o f t h e band a t 3665 cm-' Fig.2
shows
dealuminated
dl50
treated
under
i n t e n s i t y CC
t h e band a t 3610 cm-l
t h e band a t
3665
cm-l
150 to
severe
hydrothermally
conditions.
The
s t r o n g l y decreases and t h a t
increases
as
a in
continues t o increase.
t h e I R s p e c t r a of OH groups i n
samples
at the
of
compared
with
the
of
other
samples.
HZSM-5 z e o l i t e s t r e a t e d i n stearning
The above data show t h a t i n
OH
conditions there are d i f f e r e n t kinds o f 3780 cm-l i.e.
groups.
The
appears a f t e r a s h o r t t i m e o f hydrother.!nal
a t t h e i n i t i a l stages of dealurnination.
Its
imiensity
increases and then decreases w i t h steaming timeas e x t r a c t i o n o f non-framework
band
well
,
first
as
These OH groups
A1 ( r e f . 7 ) .
at
pretreatment
after
are
able
only t o a weak i n t e r a c t i o n w i t h benzene molecules and hence
their
a c i d i c p r o p e r t i e s a r e n e a r l y t h e same as those o f s i l a n o l s .
On t h e
bases o f
L:hec-e data t h e band a t
3780
U n l i k e t h e band a t 3780 cm-l a t 3665
may
cn-l
t e r m i n a l d l O F ! groups probably i n O=Al-OH
be
assigned
t h e I n t e n s i t y od t h e
g r a d u a l l y increases
witn
incressiog
narrow time
of
dealumination and i s n o t i n f l u e n c e d bv t h e a c j d leaching These
OH
groups
exhibit
relatively
stronq
comparable tr. those of bridged hydrox!/le o p i n i o n t h i s band corresponds t o
the
in
acidic
band mild
(ref.7). properties
X.
zeolite
Ft-otoi:rj
t e t r a h e d r b where t h e aluminium atom is s t i l l z e o l i t e framework
In
Eompensating connected
our A104
with
The a c i d i t y o f t h e OH groups formed a f t e r steaming ranges that
of
Bronsted G i t o s stronger than
OH
bridged the
Therefore,
bridged
groups. OH
t h e e f f r c t of
No
groups
the
from
superacid of
HZSM-5
enhancement
t h e c a t a l y t i c a c t i v i t y o f HZSM-5 z e o l i t e a f t e r stearning c o u l d be
explainer!
b.;
the
by csne or two remaining chsmical bonds ( r e f . 7 ) .
that of s i l a n o l s t o
z e o l i t e were found.
to
species ( r e f .7).
a
eariation
of
the
Bronstod
p a r t i c u l a r l y by an enhancement o f t h e s t r e n g t h o f
acidity
OH groups.
of not and Thus
we may suppose t h a t t h e observed c a t a l y t i c e f f e c t s a r e caused by a
429 change of Lewis acidity of zeolites after mild steaming. For the investigation of Lewis acid sites in t h e zeolites under
study CH3CN, H2 or CH4 molecules were used a s probes.
2 308 I
\
-
2400
2300
-
Fig. 3. Transmission IR spectra of CH3CN adsorbed on t h e fresh HZSM-5 zeolite (1) and on zeolites steamed under mild conditions: (2) t = 10 min and (3) t = 150 min.
2200
31cm-1
Fig.3 shows t h e transmission IR spectra of
adsorbed
the region of CN stretching vibrations. The band at
CH3CN
in
2280 cm-'
is
due t o CH3CN molecules interacting with OH groups (ref. 8 ) whereas the band at 2330 cm-l is caused
by
CH3CN
with Lewis acid sites (ref. 6).
In
the
molecules untreated
interacting
sample
sites are not detected, but they are created already after
Lewis 10 min
under mild steaming conditions. In IR spectra of H2 adsorbed on t h e steamed
zeolites
bands at 4110 and 4135 cm-l are observed
which
complexes with t h e bridged O H groups and
silanols,
The low intensive line at 4070 cm-'
are
(fig.4~1)
assigned
to
respectively.
according t o our previous data
(ref-4) corresponds t o H2 complexes with
Lewis
sites
containing
non-framework aluminium. In fig.4b
IR spectra of
CHq
adsorbed
on
the
steamed zeolites are shown. The bands at 3010 cm-'
and
are attributed t o physically adsorbed CH4 (ref- 5 ) . line at 2860 cm-'
appears
in
the
spectra
of
An
initial 2895
and cm-'
additional
stearned
zeolites
430
previously assigned t o
methane
polarized
by
connected with t h e non-framework aluminium.
Lewis
acid
sites
T h e intensities
3010
I
3080
&110 I
4035
hOB0
a
4150
4100
4050
4000
-
3100
2900
3000
2800
-,
d
-1
---v/cm
Slcrn-1
Fig. 4. Diffuse reflectance IR spectra
of
Hz
(a)
and
CH4
adsorbed on t h e fresh HZSM-5 zeolite ( 1 ) and on zeolites under mild conditions: (2) t=ZO min and (3) t i 6 0 min.
(b)
steamed
of both of these lines and of t h e band at 4070 cm-l in t h e spectra o+ adsorbed Hz rapidly increase at t h e initial stages of
stearning
and saturate after some time. Thus
,a
hydrothermal treatment o f HZSM-5 zeolite leads t o
formation of Lewis molecules.
We
enhancement n-hexane
acidic
assume
of
the
cracking
sites
that
they
catalytic
and
capable are
polarize
responsible
activity
probably
to
HZSM-5
of
participate
in
the
paraffin for
the
zeolite
the
stage
in of
initiation occurring via alkane dehydrogenation. The formation of Carbenium ions is considered in t h e literature a s t h e rate controlled step in alkane cracking. t o b e much more reactive in cracking than protonated easier
than
alkanes.
Then
Olefins a r e
paraffins oleiin
reactant should lead to an increase of t h e
and
addition
cracking
known
may
be
to
the
activity
of
t h e non-steamed sample and should have n o significant influence on t h e activity of mildly steamed zeolites. In fig.1
the
ratio
kt/ko
is
plotted
versus
the
time
of
431 steaming.
This dependence was obtained f o r c r a c k i n g b o t h
n-hexane
(curve 1) and
1-hexene
( 1 % ) (curve 2 ) .
of
observe a sharp maximum m i x t u r e t h e character
the
of
pure
of
n-hexane
(99x) and
I n t h e case o f pure
n-hexane
one
at
of
mixture
could
k /k =7. For t h e cracking of the t o t h i s dependence becomes smoothed as
compared w i t h pure hexane.
'10 o f C6H12
Fig. 5. T h e dependence of t h e r a c k i n g r a t e constant k from t h e (1) the i n i t i a l HZSM-5 concentration o f 1-hexene i n t h e feed: z e o l i t e , ( 2 ) t h e HZSM-5 z e o l i t e treated under m i l d steaming c o n d i t i o n f o r 30 min.
I n fig.5
t h e r e l a t i o n s h i p between t h e
cracking
rate
and t h e o l e f i n concentration i n t h e r e a c t i o n m i x t u r e i s f o r t h e i n i t i a l HZSM-5 z e o l i t e s and f o r 30 min.
lo-'%
The presence of practically
zeolites.
does
1-hexene not
in
sample
the
formation
due
than
of
both
activity
from
steamed
I n our o p i n i o n t h i s seems t o
to
addition
the of
formation
the
result of
reactive
zeolite from
a
carbenium olefin
i n h i b i t i o n o f cracking a t aconsiderable concentration of because of
0.1
o f t h e f r e s h z e o l i t e shows a maximum a t about
competition betneen a c c e l e r a t i o n of ions
for
lower
1-hexene ranging
1%o f I-hexene added whereas t h e a c t i v i t y o f t h e g r a d u a l l y decreases.
presented
steamed
concentrations
influence
However a t concentrations o f
t o 10% t h e a c t i v i t y
the
constant
and
1-hexene
blocking of a c t i v e sites.
I t was e s t a b l i s h e d t h a t i n t h e case o f pure
n-hexane
cracking
432
t h e H /C ratio in t h e reaction products on stearned zeolites is 2 3 about 1.5 times higher a s compared with t h e initial HZSM-5. In t h e presence of 0.5-2.5%
of 1-hexene t h e H2Z/C3
product
of
different
dehydrocyclization,
becomes
reactions
etc.
H2
discussed difference in
such
Nevertheless, yields
for
as
coke
in
our
the
fresh
of
cracking
because
there
are
formation, opinion
zeolites is connected with t h e production of H2 at initiation
almost
It is known that hydrogen may b e formed a s
equal for both samples. a
ratio
and the
no
steamed stage
coke formation decreases
for
accordance
HZSM-5
with
zeolites
of
changes
distribution of other products and in deactivation rates investigated samples. Moreover, in
the
in
for
the
(ref.2)
the
steamed
in
mild
hydrothermal conditions. CONCLUSION Depending on t h e conditions of hydrothermal treatment different acid sites appear in HZSM-5 non-framework
zeolites
due
aluminium. Most of these
to
fragments with acidic properties
weaker
OH-groups.
non-framework
Only a small part
of
the
species than
formation
are those A1
of
hydroxylated of
bridged
exhibits
the
properties of Lewis acid sites. In our opinion Lewis acidic sites connected with aluminium
which
are
capable
to
polarize
non-framework
C-H-bond
in
alkane
molecules a r e t h e active s i t e s responsible for t h e acceleration of t h e initial
stage
of
cracking
via
alkane
dehydrogenation
demethanation. The optimum in t h e cracking activity
observed
HZSM-5 zeolites dealuminated under mild steaming conditions
or
for could
b e interpreted in terms of t w o competitive tendencies:
i ) acceleration
of
t h e rate-determining
step of
initiation
t o an increasing concentration of non-framework Lewis
acid
due sites
and i i ) inhibition o f cracking via
the
disappearance
acid sites during
steaming pretreatment which a r e
into t h e reaction.
In our opinion these s i t e s
take
of
Bronsted
also
involved
part
in
cracking process a s centres responsible for t h e transformation
the of
formed olefinr via t h e classical carbenium ion mechanism. T h u s t h e former trend results in an increase of activity at t h e very beginning of dealumination whereas t h e latter trend which
is
well-pronounced at t h e middle o r final stages of steaming leads t o a loss of activity.
Therefore, t h e highest catalytic activity
t h e best cracking performance of HZSM-5 zeolites could be
and
reached
433 via optimization of the ratio between Lewis and Bronsted acidity. REFERENCES 1
R.M. Lago, W. 0.Haag, R. J. Mi kovsky, D. H. Olson, S. D. He1 Iring, K.D.Schmitt, G.T.Kerr, Proc. 7th Int. Conf. on Zeolites, 1986, Tokyo p.677. E. Brunner, H.Ernst, D.Freude, T.Froh1 ich, M. Hunger, H. Pf eif er, Stud. Surf. Sci. Catal., 1989 v.49 , part CI ,p.623. V.L. Zholobenko, L. M. Kustov, V. B. Kazansky, E. Loef f ler, U. Lohse, Ch. Peuker, G. Oehlmann, Zeolites, 10 (1990) 304-306. V.B.Kazansky, V.Yu.Borovkov, L.M.Kustov, Proc. 8th Int. Congr. on Catalysis West Berlin , 1984 , v.3 , p.3. V.L.Zholobenko, L.M.Kustov, V.B.Kazansky, Dokl. CIN SSSR, 300 ( 1988), 384-388. E. A.Paukshtis, R. J. Sol tanov, E. N. Yurchenko, React. Ki net. Catal. Lett., 23 (1983) 339-342. E.Loeffler, U.Lohse, Ch.Peuker, G.Oehlmann, L.M.Kustov, V.L.Zholobenko, V.B.Kazansky, Zeolites, 10 (1990) 266 - 271 H-Kriegsmann, A-Reklat, E.Loffler, Th.Steiger, Z. phys. Chemie, Leipzig, 271 (1990) 61-67
,
,
,
This Page Intentionally Left Blank
G . Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
435
COMPUTER MODELLING OF P-XYLENE SORPTION IN ZSM-5/SILICALITE-l K.-P. SCHRBDER Central Institute of Physical Chemistry, Academy of Sciences of the GDR, DDR-1199 Berlin-Adlersliof (GDR) ABSTRACT The method of molecular mechanics is used to model pxylene sorption in ZSM-5/silicalite-l on the basis of empirical atom-atom potentials. The experimental atomic coordinates of both monoclinic and highly gxylene-loaded orthorhombic ZSM-5 were taken into account. Considerable differences with respect to the sorption properties were found. It is concluded that the framework structure is locally changing already at very small amounts of sorbed yxylene. There exist two types of pxylene molecules at maximal loading. The XYLl molecule is highly mobile with respect to diffusion along the straight channel, whereas the mobility of the XYLZ molecule in the sinusoidal channel is significantly more restricted. INTRODUCTION The three-dimensional pore system of the silicon-rich zeolite ZSM-5 consists of straight channels parallel to [ O l O l interconnected by sinusoidal channels in [loo] direction, both built of ten-membered rings. The pore diameters are of a size similar to the kinetic diameters of aromatic molecules such as benzene and xylene. Thus the mobility of aromatic molecules in ZSM-5 is restricted (refs. 1-41 and a complicated dependence of the differential heats of sorption on pore filling is observed (refs. 5-8). One of the most interesting features of ISM-5 is its reversible transformation between monoclinic and orthorhombic crystal structures. As-synthesized ZSM-5 including template is orthorhombic with Pnma space group (ref. 9). H-ZSM-5 and its aluminium-free form silicalite-1 are monoclinic. A transformation to orthorhombic symmetry occurs on increasing the temperature up to 320 - 360 K (refs. 10-13) or on loading with various molecules (refs. 9,14,15) including aromatics. This change of crystal structure has been neglected in previous computer modelling studies (refs. 8,16-23) of sorption and diffusion of aromatic molecules in ZSM-5/silicalite-l, which assumed rigid frameworks. A theoretical study of the packing arrangement of pxylene in a ZSM-5 type framework has been performed by Reischman et al. (ref. 2 0 ) . The atomic coordinates of Olson et al. (ref. 2 4 ) with Pnma space group were used. On the basis of an empirical atom-atom potential due to Williams and Starr (ref. 25) the interaction energies of a pxylene molecule with the framework in the
436
different regions of the pore system were calculated. A significant energetic preference for the channel intersection was found. Eor a full occupation of all intersections by one molecule an adsorption energy of -82 kJ.mo1-I was calculated. The energies due to occupation of the straight and the sinusoidal channels were - 4 8 and -30 kJ.mol-', respectively. Packing arrangements for high pxylene loading (eight molecules per unit cell) were also considered. An occupation of each intersection and each segment of the sinusoidal channel both by one molecule is most favoured. The average interaction energy per molecule is -57 kJ.mol-I. This means a decrease of the adsorption energy with increasing amount of sorbed pxylene which is not observed (refs. 8,26,27). It seenis that the interaction energy is underestimated for the channel segments. This leads also to values for the activation energy of diffusjon of more than 50 kJ.rno1-l (ref. 28) in contrast to the value of about 25 kJ.mol-I suggested by sorption uptake measurement (ref. 3 ) . Therefore, the empirical atom-atom potential of Kiselev et al. (ref. 29) was adopted in a recent study (ref. 22) of benzene and gxvlene sorption in ZSM-5/silicalite-l. The atomic coordinates of Olson et al. (ref. 2 4 ) were used again. The framework structure was assumed as rigid, but the methvl groups of pxylene were considered as free rotors. This decreases the interaction energy by 1 - - 3 kJ.mol-I. Only a single molecule was used as a probe, no sorbate-sorbate interaction was taken into account. The adsorption energies for optimized structures were -82.6 kJ.mol-' in the straight channel, -79.3 kJ.mol-I in the sinusoidal channel, and -81.9 kJ.mol-' in the channel intersection. Very recently detailed crystal structures of monoclinic (MONO) and hiqhly pxylene-loaded orthorhombic HZSM-5 (PARA) have become available from single crystal XRD analyses (ref. 3 0 , 3 1 ) . The present study is the first attcnipt to model the sorption of aromatics in ZSM-5/silicalite-l considering different framework structure. Packing arrangements for low and high coverages of p-sylene are studied. CALCIJLATIONS Stable sorbate structures correspond to minima of the potential energy surface for the interaction between the zeolite framework and sorbed molecules. The interaction energy is represented in this study by a pairwise additive atos-atom potential function. The energy for the sorbent-sorbate interaction is calculated using a Lennard-Jones type potential, AE =
r i
(-A j
at?
r -.6. + Bap r -12 , ,) , I J
I J
and that for the interaction between pxylene molecules using a Buckingham type potential,
437
The index i runs over all atoms in the framework cut-out, indices j and k over all atoms in different pxylene molecules, r . is the interatomic distance. The 11 characterizing the interaction between empirical parameters AaS, BaB and C aB atoms of the type a and p are taken from Kiselev et al. (ref. 29) for the Lennard-Jones type potential and from Williams and Starr (ref. 25) for the Buckingham type potential (see Table 1). The minimization was performed by the PENOPT-2 program (ref. 32) using a quasi-Newton method. Energy gradients were calculated analytically. The following restrictions and approximations were made: (i) 'The geometries of the framework (MONO and PARA) and of the pxylene molecules are fixed and not affected by sorption. Only the methyl groups of pxylene are permitted to rotate freely. (ii) No electrostatic interaction is taken into account. The pairwise summation is performed only over the oxygen atoms of the framework, not over silicon or aluminium sites. No cations or protons are considered, so that the interaction with silicalite-1 rather than ZSM--5 is studied. (iii) The zeolite framework is represented by a cut-out of the size 1.5a x 1 . 5 b x 212, where a, b and c are the unit cell parameters. In the high-coverage calculations the interaction of the two pxylene molecules with 16 pairs in the neighbourhood was considered. As previously demonstrated (refs. 8 , 2 1 , 2 2 ) the occupation of sorption sites 'TABLE 1
Potential parameters sorbent
-
sorbate (ref. 29) B
A
(kJ ..k6 .mol 0 - H
(kJ * A" .tool-
5.3781 .lo2
1.5028 .lo5
2.1623~10~
1.4478.106
1.7041 .lo3
1.1410*106
sorbate - sorbate (ref. 25)
c-c
3.6725 . l o 5
3.60
2414
C - H
6.5485.10'
3.67
573
H - H
1.1677 -10'
3.14
136
438
is sensitive t o temperature changes. A simple model is adopted to take temperature etfects into consideration. The sorbate complexes are treated analogously to gas phase complexes within the rigid rotor, harmonic oscillator, and ideal gas approximation (ref. 21.33). The over-all thermodynamic functions are obtained as weighted sums of the contributions from the individual preferential sorption sites (ref. 22). The knowledge of the frequencies of intermolecular vibrations (librations and hindered translations) and of low-frequency torsional motions of the methyl groups is prerequisite, but all other intramolecular modes are not expected to contribute. The frequencies are computed by a recently developed efficient method (ref. 34). RESULTS AND DISCUSSION Low-coverage sorbate structures The MONO and PARA frameworks were probed by a single p-xylene molecule. This actually corresponds to zero loading. For the MONO structure ten sites for the sorption of psylene with an interaction energy lower than -80 kJ.mol-1 were detected. Five of them with energies between -80.6 and -83.2 kJ.mol-I are located at the channel intersection. The energetically most favoured sorption site (-86.7 kJ.rnol-I) was found in the straight channel in the mid between two intersections. In the sinusoidal channel an adsorption energy of -85.5 kJ.mol-' is reached. Under the common assumption, that the framework was not deformed at low loadings, the first pxylene molecule would be sorbed in the straight channel at temperatures below 100 ti. Above this temperature the position at the channel intersection would be thermodynamically more favoured since the mobility of the pxylene molecule is larger at the intersection than in the narrow pores. The sorption heat at 300 K , calculated taking the average over all ten sorption sites, is -74.5 kJ.mol-I. The shape of the pores changes significantly, when the MONO structure of ZSM-5/silicalite-l is transformed into the PARA structure. The limiting ports of the straight channel 10-rings become from 5.78 A x 5.18 A and 5.83 A x 5.27 A to 6.06 -4 x 5.07 .4 and 6.18 .4 x 4.80 .4 (ref.30). Similar changes are reported for the sinusoidal channel (ref. 30). The effect of these structural changes on the energy of pxylene sorption is remarkable. The interaction energy clearly decreases in all regions of the pore system compared to the MONO structure (see 'Table 2 ) . The energy is particularly lowered for the sorption sites in the sinusoidal channel. At 300 ti the calculated over-all sorption heat of a single psylene molecule in a PARA structure amounts to 90.7 kJamol-'. Thus the difference to the value f o r MONO is about 16 kJ.mol-l. Comparing this with the enthalpy change of 3.7*0.5 kJ.mo1-l connected with the temperature-induced transformation from the monoclinic t o the orthorhombic structure of ZSM-5 as found by differential scanning calorimetry (ref. 1 2 ) it is concluded that the
439
TABLE 3 Interaction energy (kJ.mo1-I) of a single pxylene molecule with the rigid framework at different sorption sites
channel intersection straight channel sinusoidal channel
MONO
PARA
-83.2 -86.7 -85.2
-86.6 -91.8 -98.5
assumption of framework rigidity at low loadings of pxylene is not valid. The pxylene molecule is able t o induce a local change of the framework structure. 29 This result is consistent with those from XRD (refs. 35,361 and Si MAS NMR spectroscopic experiments (ref. 37), which show, that already a small amount of sorbed pxylene changes the structure and symmetry of the ZSM-5/silicalite-l framework. High-coveraqe sorbate structure The pxylene sorption capacity of ZSM-5/silicalite-l was reported to be near 8 mol./u.c. (ref. 27) at room temperature. This corresponds to a loading of two molecules per intersection. Since the straight channel segment and the channel intersection cannot be occupied simultaneously, it is concluded, that one molecule (XYL2) is located in the sinusoidal channel. In order to search for the preferential site of the second molecule (XYL1) the energy profile along the straight channel was calculated f o r a dense packing of 8 mol./u.c. (see Fig. 1). Periodic boundary conditions were used. The XYLl molecule was shifted in [OlO] direction through the straight channel passing the intersection with the sinusoidal one. The y coordinate of the centre of the molecule was fixed after
-90-
a
b
c 0
0
-100-
0 0
0 0 0 O 0
0
0
0 0
I
I
0
0 0
0
0 0
ooooo
0 0 0
I
oooooo
I
I
I
I
I
I
I
I
Fig.1. Energy profile (see text) along the straight channel ([OlO]) at high loading (8 mol./u.c.) beginning at the mid of the channel intersection (y/b = 0.25), minimum a: XYLla, b: XYLlb, c: XYLlc
440
Fig.?. Energy profile (see text) along the sinusoidal channel ([loo]) at high loading ( 8 mol./u.c.) beginning at the intersection with the straight channel (x/a = 0.5). XYLl on site XYLla ( 4 , XYLlb ( 0 ) or XYLlc (+) steps of 0 . 3 A and the remaining 15 coordinates corresponding to the locations and orientations of both molecules and their rotable methyl groups were fully optimized. It is surprising, how flat the energy profile is (see Fig. 1). Thus a hiqh diffusional mobility of the XYLl molecule can be expected. In contrast, the energy profile for shifting the pxylene molecule through the sinusoidal channel (Fig. 2) shows, that the diffusion of the XYL2 molecule is much more hindered. The authors of a deuterium solid-state NMR study (ref. 4 ) reached the same conclusion. While there is one clearly preferred sorption site in the sinusoidal channel there are several minima of about the same interaction energy corresponding t o favourable sorption sites in the straight channel. An analysis of the vibrational frequencies shows, that the sites XYLla and XYLlb are not only energetically but also entropically favoured over the others.It is, however, not possible to decide on the basis of our calculations, which of the two sites is preferred. It seems that both are occupied to some extent, what may be the reason, why the theoretical sorption capacity of 8 mol./u.c. is not reached in practice. In both possible packing arrangements the sorbate-sorbate interaction plays an important role in stabilizing the structure. Neither XYLla nor XYLlb (see Fig. 3 , Table 3 ) represents the global framework-sorbate interaction energy minimum inside the straight channel, The atomic coordinates of the XYLla
44 1
4-
+-h -0-
b
-0-
Fig.). High-coverage packing arrangements XYLa (left) and XYLb (right); top: view down [OlO]; bottom: view down [OOl]; for purposes of clarity hydrogen atoms (and at the bottom also zeolite atoms) were left out
molecule are almost identical to those reported by van Ronigsveld et al. (ref. 31) on the basis of their XRD study. Similar packing arrangements were suggested by Mentzen (ref. 3 8 ) and Reischmnn et al. (ref. 2 0 ) . CONCLUSIONS
'The transformation of the ZSM-5/silicalite-l framework induced by pxylene sorption provides a considerable gain of sorption heat. A single pxylene molecule should be able to effect in its neighbourhood a local structural change from the monoclinic t o an orthorhombic space group. At high coverage the sinusoidal channels are fully occupied. Inside the straight channels the p-xylene molecules are highly mobile with respect to translation in the channel direction. At least two distinct sites may be occupied. One of them coincides with experimental findings.
442
TABLE 3 Fractional coordinates ( ) ( l o 4 ) of pxylene molecules for the two high-coverage packing arrangements XYLa and XYLb XYLla atom
X
Y
c1 C2 c3 c4 c5 C6 c7 C8 H2 H3 H5 H6 H7a H7b H7c H8a H8b H8c
5268 4801 4663 4992 5459 5597 5416 4843 4547 4303 5713 5957 5015 5882 5460 4477 5298 4648
3296 3164 2497 1963 2095 2762 4015 1244 3576 2395 1683 2864 4206 4031 4321 1242 1002 970
XYLlb
Z -208 540 816 345 -403 -679 -505 642 903 1392 -766 -1255 -971 -920 160 1240 899 3
X
Y
2
5155 4716 4520 4763 5203 5399 5366 4552 4528 4181 5391 5738 5405 5848 4999 4284 4990 4229
1739 1547 872 389 581 1256 2467 -339 1920 124 208 1404 2600 2537 2791 -408 -662 -~472
105 862 951 284 -472 -561 8 381 1376 1535 -986 .1145 -777 367 366 1079 373 -239
XYL2a atom
X
Y
c1 c2 c3 c4 c5 C6 c7 C8 H2 H3 H5 H6 H7a H7b H7c H8a H8b H8c
1659 1872 2552 3019 2806 2126 926 3752 1511 2716 3167 1962 662 848 737 3934 3833 4020
2429 2828 2925 2623 2223 2126 2324 2727 3062 3233 1990 1818 7798 2156 1944 3132 2862 2263
XYL2b 2
-1164 -1962 -2130 -1500 -702 -534 -982 -1682 -2447 -2745 216 82 -1101 -219 -1495 -1204 -2459 -1513
~-
X
Y
I754 1967 2647 3114 2902 2222 1021 3848 1606 2811 3262 1058 758 944 830 4060 3918 4095
2428 2820 2913 2614 2221
2128 2327 2714 3051 3216 1990 1825 2800 2168 1941 3016 2974 2225
2
-1161 -1967 -2140 -1507 -701 -529 -975 -1693 -2455 -2761 -213 93 1106 -207 -1471 -1095 -2401 -1718
ACKNOWLEDGEMENT
Part of this work was done during a stay at the Institute of Inorganic Chemistry of the Slovak Academy of Sciences at Bratislava. The author thanks dr. J . Noga for hospitality. Thanks are a l s o directed t o dr. J. Sauer (Berlin) for
numerous helpful discussions.
443
REFERENCES 1
B. Zibrowius, M. Bulow and H. Pfeifer, Chem. Phys. Lett., 1 2 0 ( 1 9 8 5 ) 420--423.
2 B. Librowius, J. Caro and H. Pfeifer, J. Chem. SOC., Faraday Trans. I, 84 (1988) 2347-3356.
M. Bulow, J. Caro, B. Rbhl-Kuhn and B. Zibrowius, in: H.G. Karge and J. Weitkamp (Eds.), Zeolites as Catalysts, Sorbents and Detergent Builders, Elsevier, Amsterdam, 1989, pp. 505-517. 4 R. Eckman and A.J. Vega, J. Phys. Chem., 90 (1986) 4679-4683. 5 R. ~ e n d t ,H. Thamm, K. Fiedler and H. Stach, 2. Phys. Chem. (Leipziq), 366 3
( 1 9 8 5 ) 289-301.
H. Thamm, Zeolites, '1 (1987) 341-346. H. 'Thamm,H.-G. Jerschkewitz and H. Stach, Zeolites, 8 ( 1 9 8 8 ) 151-154. B. Grauert, K. Fiedler, H. Stach and J. Janchen, in: P.A. Jacobs and R.A. van Santen (Eds.1 , Zeolites: Facts, Figures, Future, Elsevier, Amsterdam, 1989, pp. 701-713. 9 E.L. Wu, S . L . Lawton, D.H. Olson, A.C. Rohrmann and G.T. Kokotailo, J. Phys. Chem., 83 ( 1 9 7 9 ) 2777-2781. 1 0 H. van Koningsveld, J.C. Jansen and H. van Bekkum, Zeolites, 7 ( 1 9 8 7 ) 6 7 8
564-568. 11 D.G. Hay and
H. Jaeger, J. Chem. SOC., Chem. Commun., ( 1 9 8 4 ) 1 4 3 3 . 13 A. Endoh, Zeolites, 8 (1988) 250-251. 13 J. Klinowski, T.A. Carpenter and L.F. Gladden, Zeolites, 7 (1989) 73-78. 14 C.A. Fyfe, G.J. Kennedy, C.T. De Schutter and G.T. Kokotailo, J. Chem. Soc., Chem. Commun., (1984) 541-542. 1 5 G.W. West, Aust. J. Chem., 3 7 (1984) 455-457. 1 6 S. Ramdas, J.M. Thomas, A.K. Betteridge, A.K. Cheetham and E.K. Davies, Angew. Chem., 9 6 (1984) 629-637. 17 A.K. Nowak, A.K. Cheetham, S.D. Pickett and S. Ramdas, Mol. Simul., 1 (1987) 67-77.
18 S.D. Pickett, A.K. Nowak, J.M. Thomas and A.K. Cheetham, Zeolites, 9 (1989) 123-128.
B. Grauert and K. Fiedler, Adsorpt. Sci. Technol., 5 (1988) 191-198. 20 P.T. Reischman, K.D. Schmitt and D.H. Olson, J. Phys. Chem., 9 2 ( 1 9 8 8 ) 19
5165-5169.
K.-P. Schroder and J. Sauer, Z. Phys. Chem. (Leipzig), 2 7 1 ( 1 9 9 0 ) 289-296. 32 K.-P. Schrdder, Proceedings of the VIIth International Conference on Theoretical Problems of Adsorption, Moscow, 1991, submitted. 23 F. Vigne-Maeder and H. Jobic, Chem. Phys. Lett., submitted. 24 D.H. Olson, G.T. Kokotailo, S.W. Lawton and W.M. Meier, J. Phys. Chem., 85
21
(1981) 2238-2243.
36
D.E. Williams and T.L. Starr, Comput. Chem., 1 (1977) 173-177. C.G. Pope, J. Phys. Chem., 9 0 (1986) 835-837. R.E. Richards and L.V.C. Rees, Zeolites, 8 (1988) 35-39. K.-P. Schroder, unpublished result. A.V. Kiselev, A.A. Lopatkin and A.A. Shulga, Zeolites, 5 ( 1 9 8 5 ) 261-267. H. van Koningsveld, J.C. Jansen and H. van Bekkum, Zeolites, in press. H. van Koningsveld, F. Tuinstra, H. van Bekkum and J.C. Jansen, Acta Cryst. 8, 45 ( 1 9 8 9 ) 423-431. K.-P. Schroder, FORTRAN program PENOPT-2, Berlin, 1986. J. Sauer and Zahradnik, Int. J. Quantum Chem., 26 (1984) 793-822. K.-P. Schroder, Chem. Phys., 1 2 3 (1988) 91-101. G.T. Kokotailo, L. Riekert and A. Tissler, in: H.G. Karge and J. Weitkamp (Eds.) , Zeolites as Catalysts, Sorbents and Detergent Builders, Elsevier, Amsterdam, 1989, pp. 843-848. B.F. Mentzen and F. Bosselet, C. R. Acad. Sci. Ser. 2, 308 ( 1 9 8 9 )
37
C.A. Fyfe, G.T. Kokotailo, H. Strobl, H. Gies, G.J. Kennedy, C.T. Pasztor
35 26 27 28 29 30 31 32 33 34 35
1533-1538.
444
arid G.E. Barlow, in: H.G. Karge and J. Weitkamp (Eds.), Zeolites a s Catalvsts, Sorberits and Detergent Builders, Elsevier, Amsterdam, 1989, pp. 817-842. 38 B.F. Mentzen, in: H.G. Karge and J. Weitkamp (Eds.), Zeolites as Catalysts, Sorbents and Detergent Builders, Elsevier, Amsterdam, 1989, pp. 477-484.
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
445
MICRODYNAMICS O F GUEST MOLECULES IN ZEOLITES STUDIED BY OUASIELASTIC NEUTRON SCATTERING AND NMR PULSED FIELD GRCIDIENT TECHNIQUE Her&
Jobicl, Marc B&e 2, Jiirgen Caro3 , Martin Bulow 3 , Jorg k'Brger4 and Harry Pfeifer4
'Institut de Recherche5 sur la Catalyse, 2 avenue CI. 69626 Villeurbanne, France.
Einstein,
d e SpectromGtrie Physique, Universite 'Laboratoire Fourier, ,78402 St. Martin d'H&res, France.
Joseph
3Zentralinstit~~t fiir physikal ische Chemie der AdW der DDR, Rudower Chaussee 5, 1199 Berlin, G.D.R. 4Sektion Physik der Universitat Leipzig Linnestr. 5, 7010 Leipzig, G.D.R. SUMMARY Application of the quasi-elastic neutron scattering and NMR pulsed field gradient technique to molecular self diffusion studies of hydrocarbons in zeolites ZSM-5 and NaX leads to coinciding results in both the absolute values, the concentration dependence and activation energy of the intracrystalline self-diffusion coefficients. Further agreement between the neutron and NMR diffusion data is obtained by comparing the mean molecular jump lengths which by both techniques are found to be of the order of 1.2...0.25 nm, slightly decreasing with increasing loading. Due to the different the two experimental methods applied, the mean time scales o f in neutron diffusion paths followed are of different magnitude: scattering experiments molecular translation is measured up to 6 nm; in the NMR pulsed field gradient technique the mean molecular displacements amount to some pm. However, in both methods, longrange self-diffusion is detected. INTRODUCTION The investigation of molecular migration of sorbate and reactant molecules in zeolites of different structures has become an important topic in both fundamental research and industrial application. As a most spectacular result obtained about 10 years ago, the application of the NMR pulsed field gradient technique (cf. /1-.3/) to the study of hydrocarbon diffusion in A type zeolites gave evidence that for a number of systems the self-diffusion coefficients determined by the up-to-date sorption uptake methods have been erroneous by up to five orders of magnitude /4,5/. Since this time, the clarification of the correct order of magnitude of the translational mobility of molecules in zeolites is one of the most controversially discussed topics in zeolite research. The analysis of the possible reasons for this discrepancy led to a critical reconsideration of both the sorption measuring techniques and the models of intracrystalline molecular transport in zeolites /6-8/.
446 However,
l i g h t hydrocarbons on Z S M - 5
recent d i f f u s i o n data o f
also
vary over orders o f magnitude 19-141. Besides
the
NMR,
t h e o n l y o t h e r experimental
technique
that
measures molecular s e l f - d i f f u s i o n a t s o r p t i o n e q u i l i b r i u m i s quasi-
(QENS). Up t o now, l i t t l e d i f f u s i o n data
e l a s t i c neutron s c a t t e r i n g
molecules sorbed i n porous media such as z e o l i t e s /12-19/ have
for
been obtained by which
measure
QENS.
Therefore,
self-diffusion
bate/adsorbent
t h e two experimental
on a microscopic
systems have been
applied:
techniques
scale
in
adsor-
quasi-elastic
neutron
s c a t t e r i n g (QENS) and NMH pulsed f i e l d g r a d i e n t technique (PFGT). EXPERIMENTAL a) Q u a s i - e l a s t i c neutron s c a t t e r i n g The
zeolite
samples
720
were heated i n a g l a s s r e a c t o r a t
Y
-7
f l o w i n g oxygen and outgassed t o a f i n a l pressure o f 10
under at
the
same temperature.
transferred
into
geometry
(diameter
amounted
t o about 5 9,
The
I n s i d e a glovebox,
a slab-shaped 50
mm).
aluminium
The mass o f
the
crystals
container the
Pa
of
were
circular
activated
zeolite
thus ensuring a h i g h neutrcsn transmission.
c e l l was connected t o a gas i n l e t system a l l o w i n g t o load
sample
w i t h the hydrocarbons a t d i f f e r e n t pressures
and
the
tempera-
t u r e s d u r i n g the neutron experiment. The neutron r e s u l t s were obtained a t the I n s t i t u t Laue-Langevin, Grenoble,
using t h e t i m e - o f - f l i g h t
wavelengths
o f 0.8
and 1 9 peV,
respectively.
and 0.9
section
o f hydrogen,
largely
incoherent.
avoided
as
nm
spectrometer
and
IN5 w i t h
incident
25
e l a s t i c r e s o l u t i o n s o f ca.
Because o f
the l a r g e
incoherent
cross
the s c a t t e r i n g from hydrocarbon molecules i s The coherent S c a t t e r i n g from the
zeolite
much as p o s s i b l e by t a k i n g s e l e c t e d groupings
is
of
the
d e t e c t o r s between t h e Bragq peaks o f t h e z e o l i t e . b ) NMH pulsed f i e l d q r a d i e n t technique The
zeolite
<
pressure
hydrocarbons amounts
of
Pa was maintained. immediately
up t o 670 K
until
a
sample
activation.
The
sorbed
volumetrically
and
The g l a s s tubes were then sealed o f f .
self-diffusion
measured
aft.er
V~CLIO
The z e o l i t e s were loaded w i t h
hydrocarbons were c o n t r o l l e d b o t h
gravimetrically. The
samples were heated i n
c o e f f i c i e n t s o f the sorbed
molecules
on the home-made NMR p u l s e spectrometer FEGRIS
were
operating
a t a proton frequency o f 60 MHz a t t h e Department o f Physics o f the Universitat Leipzig.
441 THEORY a) Quasi-elastic neutron scattering The long-range translational motion of a molecule can be derived from the broadening of the elastic peak a s a function of the momentum transfer. For hydrocarbons in zeolites, only incoherent scattering has to be considered because of the large incoherent cross section of hydrogen. The intensity scattered by the sample i s then proportional to the incoherent scattering law Sinc(Q,o), which i s related to the self-motion of protons (-0 and nu denote t h e neutron momentum transfer and the neutron energy transfer, respectively). The broadening values obtained for methane diffusing in zeolites can be fitted to a model of a diffusion process proceeding via molecular jumps with a Gaussian distribution of j u m p lengths / 2 0 / . Thus, the scattering law becomes a Lorentzian, S(Q,o)
=
_1_ n
.
&(Q)
w2
+
CaO(Q)12
'
whose half-width at half maximum (HWHM)
is determined by both the mean jump rate r ( t h e inverse of r gives the mean time T between two succeeding jumps) and the mean-square jump length (1 2 >. At small 0 values, using the relationship between continuous and jump diffusion, D = <12>/67,the HWHM becomes equal to h ( Q=) DO2. Clt high (1 values, the half-width approaches t h e mean j u m p rate
r. The broadening values obtained for ethane and benzene in NaX were fitted with the jump diffusion model of Singwi and Sjolander /21/. When the molecule spends more time in i t s oscillary state than diffusing, the shape of the quasi-elastic peak i s still a Lorentzian, but the HWHM i s given by
This model w a s used for ethane and benzene in NaX because the broadening curves converged more slowly towards their asymptotic values at high Q than for methane in NaZSM-5. b) NMH pulsed field gradient technique In pulsed field gradient NMR, spin echo attenuation W d u e to diffusion in an isotropic system such a s zeolite X with a selfdiffusion coefficient D follows the relation /3/
448 with the
respectively,
the i n t e n s i t y ,
which a r e i n the given case t h e protons o f
the hydrocarbon molecules.
-
deno-
t h e w i d t h and t h e separation o f
Y stands f o r t h e gyromagnetic r a t i o o f
f i e l d g r a d i e n t pulses.
the n u c l e i under study,
n
6: and fl
the parameters o f t h e s p i n echo experiment g,
ting,
The measurements a r e based on t h e n / Z
-
echo pulse sequence.
For a n i s o t r o p i c systems l i k e d i f f u s i o n i n Z S M - J / s i l
calite
eq
.
( 4 ) has t o be replaced /3,22/ by
+ -b
Pr(g,S,A) = exp I-Y with
-b + D
vector,
2
2-*--+ 6 g D g
(5
+
and g denoting t h e d i f f u s i o n tensor and t h e f i e l d respectively,
I t has been found, no
( A - 6/3)3
significant
i n s t e a d o f the s c a l a r q u a n t i t i e s however,
gradient
D and 9.
t h a t f o r methane i n ZSM-5
d i f f e r e n c e between t h e values
p r i n c i p a l elements o f the d i f f u s i o n tensor
of
/22/,
r i m e n t a l data are c o r r e c t l y described by eq.
the
there i s
individual
so t h a t the expe-
( 4 ) w i t h a mean s e l f -
+ D Z Z ) denoting one t h i r d c o e f f i c i e n t D = 1/3 (Dxx + D YY o f t h e t r a c e o f t h e d i f f u s i o n tensor /3, 22/. diffusion
RESULTS a ) Methane d i f f u s i o n i n ZSM-5 The
results
elastic
obtained
i n the
QENS f o r t h e
broadening
peak versus Q2 a t 200 K a r e shown i n F i g .
of
the
1 for different
loadings. Fig. 1 Elastic peak broadening versus GI2 for CH i n ZSM-5 a t 2OOK for the loadings 2, ( I4, )and 8 CH4 per U.C.
(v)
(A)
I t f o l l o w 5 from Fig.1
QENS
t h a t i n the
experiments the broadening
of
the e l a s t i c peak as a f u n c t i o n o f Q deviates that
from
a straight
line
distribution
l e n g t h s has t o be interpretation
o f the
considered.
occur
ments vidual
This
lengths
w i t h i n t h e channel
a s w e l l as across channel
a
jump
i m p l i e s t h a t molecu-
l a r jumps w i t h v a r y i n g jump may
so
a jump d i f f u s i o n model w i t h
Gaussian
2
the
intersections.
segindiThe
s e l f - d i f f u s i o n c o e f f i c i e n t s obtained a r e presented i n Table 1.
O
oI2
0:4
'
6.6
a
449
These data yield an energy of activation for t h e self-diffusion o f the order of 4...5 kJ mol-l. The mean jump lengths are found to be larger at small loading and high temperatures /12,14/. For methane in NaZSM-5, the mean jump lengths calculated for small and high loadings at 200 K amount to 1.04 and 0.87 nm, respectively. For the NMR PFGT experiments, Fig. 2 represents the spin echo attenuation due to molecular self-diffusion of methane in the NaZSM-5 at 250 K at different sorbate concentrations. In complete agreement with the behaviour predicted by eq.(4), the plots of 1nPc y2 92 ( A - 1/3 6 ) provide straight lines, whose slopes VS. should coincide with the intracrystalline diffusivities. Cornparing the slopes of the spin echo attenuations and using eq. (41, the mean diffusion coefficients D = 1/3 ( D x x + Dyy + D Z z ) are found t o decrease slightly with increasing sorbate concentration from 6.3 x 10-5 cmzs-l ( 4 CH4/u.c.) to 4.5 x 10-5cm2~-1 ( 1 2 CH4/u.c.). The diffusivities thus obtained are shown in Table 1. From the Arrhenius plots of the NMR self-diffusion coefficients of methane 1251, the energy o f activation for the intracrystalline self-diffusion of being in good methane in NaZSM-5 amounts to (4.7 f 0.7) kJ mole-' o f methane in NaZSM-5
agreement with the QENS data of Ed. In addition to the values of D and Ed, a further agreement between P E N S and NMR PFGT i s found by comparing the individual molecular j u m p lengths. In neutron scattering experiments /12,14/ a s well a s in a previous NMH study /25/ dealing with the microdynamics of hydrocarbons in ZSM-5 type zeolites, it has been shown that the mean jump lengths of light paraffins decrease from 1...1.2 nm at low sorbate concentration to a nm for high loadings. value of 0.7...0.8 Fig.2 Spin echo attenuations d u e to intracrystalline selfdiffusion of CH4 in NaZSM-5 at 250 K for the loadings ( 0 ) 4 ; (.)8and ( 0 ) 12 molecules per u.c., respectively. For calibration, the echo attenuations for neat water ( ) and liquid ) , whose selfammonia diffusion coefficients taken
100
m
80 60
c C
-=!a 2 .-. *
I
*O
(A
10 1
-2
3 4 5 6 7 8 9 $ g 2 6 * ( A - + & I / re\.units
10
v
from literature amount t o 2.3 x loF5 cm2s-' /23/ and 1.5 x cm2s-' /24/, respectively, are included.
450 TABLE 1
Intracrystalline self-diffusion coefficient of methane in NaZSM-5 measured by QENS and NMR PFGT
loading/ CH4iu. c
.
D
T
= 200 K
~
~
<:
T
D ~ ~
~~
/~ D /
1 0 ~ ~ ~ ~ . - ~
1.5
= 250 ~
I(
~
D~
~ ~ ~
/~
x 1o5cm2s-' 5.0
2 2.8
2.8
4 8 12
2.5 2.9
5.9 3.7
6.3
2.8 2.5
5.2 4.5
b ) Ethane self-diffusion in NaX shows a comparison of experimental and calculated energy Fig.3 Table 2 contains spectra for ethane in NaX at different P values. the molecular mobility data determined by QENS.
Frg.3 Energy spectra for ethane in NaX at different Q values (253 K-, 4.3 CZH6 p e r supercage)
-
.
3
-. 0'
m
- 0 4 -02
0
02 04
06
08
1 oad ing/
D/
CZH& Per
105cm2s-1
supercage 1.3 4.3 5.8
5.4 2.5 1.5
hw(meV)
T/ %
10%
1.36 1.48 2.22
<12>1'2
nm
0.66 0.47 0.44
/
451 In Fig.4, the self-diffusion coefficients obtained by QENS (cf.Tab. 2 ) are compared with the corresponding NMR PFGT data. A complete agreement both in the absolute value and the concentration dependence of the self-diffusion coefficient is observed. Furthermore, a very similar trend in the decrease of the mean molecular jump lengths with increasing sorbate concentration a s shown for the QENS results of ethane/NaX in Tab. 2 has been found in previous NMR self-diffusion and NMH relaxation studies for propane/NaX /25/. At 135 K , the mean jump lengths were found t o decrease from 0.78 nm (1.5 C3HB per supercage) to 0.25 nm (5.5 C3H8 per supercage) 1 2 5 1 .
I ,
‘“:I
n
.“-5!
,
1
,
,
,
,
2
3
4
5
0 , 6
1
Fig .4 Comparison of the self-diffusion coefficients of ethane in NaX at 253 K meathis sured by QENS ( A study) and the NMR PFGT ( ,1261)-
7
Loading /C2H6 per supercoge
c ) Benzene self-diffusion in NaX
Taking account of the influences of external heat and mass transfer resistances in limiting the sorption rates, in many instances reasonable agreement between sorption and NMR data could be obtained /4,11,27-29/. However, there is also a number of welldocumented experimental studies showing differences of up to two orders of magnitude /10,28/. An especially large number o f investigations were concentrated on the study of benzene in zeolite NaX. For this system, sorption uptake measurements by different research groups revealed both agreement /13,30,31,34/ and disagreement /10,32/ with the NMR data. Table 3 gives a summary of the QENS self-diffusion coefficients, mean jump lengths and correlation times at 458 K for different sorbate concentrations. In Fig.5, these diffusivities are compared with the results of previous NMR PFGT measurements and the results of adsorption/desorption experiments.
452
D/
loading/ C6H6
per supercage
x
<12>1/2/
6 2 -1 1 0 cm s 7
0.46
5.0
4
0.35
5.1
2.0
1.9
0.24
5.0
Fig. 5 Comparison of the selfdiffusion coefficients of benzene i n NaX a t 458 K directly determined by b o t h the NMH PFGT ( 0 0 / 3 3 / and /28/) and QENS ( A , / 1 3 / ) with the corresponding d i f fusivities determined from non-equilibrium measurements piezometric/30/; :e?b 'Length Column method /32/;n,frequency response and s i n g l e s t e p frequency as response technique /31f-.) NMH w e l l as deduced from l i n e shape a n a l y s i s ./rS4/. The r e g i o n o f the r e s u l t s o f gravimetric measurements with different specimens /28/ is indicated by a hatched area. Asterisked symbols represent data which have been obtained by e x t r a p o l a t i o n from lower temperatures.
5
n 10-6
n
0
+,
10.~
10-8
0
It
1
2 3 1 Loading / C6H6 per supercoge
appears from Fig.5 in
data,
the
uptake
1d1s
1.3
\
trends
x
0.8
N
10-9
T/
nm
5
t h a t i n both the absolute values and
t h e c o n c e n t r a t i o n dependence, NMR
r e s u l t s and the data
experiments
the
derived
applying t h e piezometric o r
quency response techniques agree.
Nevertheless,
neutron from
the
scattering
sophisticated
single-step
fre-
disagreement w i t h
some s o r p t i o n r e s u l t s has t o be stated. Once sing
again,
the decrease i n t h e mean jump l e n g t h s w i t h increa-
sorbate concentration
result
of
previous
(cf.
Tab.3
i s i n agreement
NMR d i f f u s i o n and NMR r e l a x a t i o n
hydrocarbons i n NaX / 2 5 , 2 6 / ,
the of
where t h e decrease i n t h e t r a n s l a t i o -
nal
m o b i l i t i e s could be shown t o be ma n l y due t o a
the
mean
jump
with studies
l e n g t h s r a t h e r than t o i n c r e a s i n g
times between succeeding jumps.
reduction
mean
of
residence
453 DISCUSSION a r e methods enabling t h e investigation B o t h QENS and NMR PFGT, o f molecular self-diffusion, i.e. t h e n e t effect o f a succession o f consecutive elementary s t e p s o f m a s s transfer a t sorption equilibrium. However, d u e to the different t i m e s c a l e s o f t h e t w o methods, the mean diffusion paths followed a r e o f different magnitude: while in QENS, molecular translation i s measured u p t o 6 nm, in NMR P F G T t h e mean molecular displacements may amount t o several pm. Nevertheless, for t h e three adsorbate-adsorbent s y s t e m s under study, the self-diffusion coefficients measured independently by neutron scattering and NMR a r e in very good agreement. the However, in the c a s e o f the methane diffusion in ZSM-5, self-diffusion coefficients given in T a b l e 1 a r e by about s i x o r d e r s of magnitude larger than t h o s e determined by gas-chromatography /35/, t w o o r d e r s of magnitude larger than t h e diffusion coefficients determined by using a permeation method with a zeolite membrane /36/ and by a factor o f 5 larger than diffusion coeffic i e n t s measured by means of t h e frequency-response s i n g l e s t e p Also for t h e benzene diffusion in NaX, d i f f e r e n c e s analysis Ill/. of t w o o r d e r s of magnitude a r e still open f o r explanation / l O / T h e s e strong differences cannot be explained by theory. Following the formalism o f irreversible thermodynamics, t h e diffusion coefficient determined under non-equilibrium sorption c o n d i t i o n s should be expected t o be either equal ( i n t h e c a s e o f a linear sorption isotherm, Henry's law) o r larger ( f o r convex shaped isotherms s u c h a s those o f t h e Langmuir type) than t h e self-diffusion coefficient measured at sorption equilibrium /37/. While in t h e Q E N S and NMR P F G T experiments intrinsic intracrystalline self-diffusivities are measured, diffusion coefficients derived from adsorption/desorption experiments can be affected, additionally, by intercrystalline diffusion (bed d e p t h ) influences, sorption heat release processes and/or by transport resistances near t h e crystal s u r f a c e ( s u r f a c e barriers).
CDNCLUSIUNS (1) T h e joint application of Q E N S and NMR PFGT t o t h e adsorbateadsorbent s y s t e m s CH4/NaZSM-5. C2H6/Nal( and C6H6/NaX leads t o coinciding results in both t h e absolute values o f t h e intracrystalline self-diffusion coefficients, their concentration dependence, t h e e n e r g i e s o f activation o f self-diffusion, t h e mean molecular j u m p lengths and t h e correlation t i m e s o f molecular motions.
454 (ii)
By both QENS and NMR PFGT, the intracrystalline self-diffusion coefficient of methane in ZSM-5 was found in a temperacm s By both tcire region of 2 0 0 K - 2 5 0 K to be methods, the activation energy was determined to be 4...5 kJmol-l. A coinciding value o f ca. 1 nm was obtained f o r the mean molecular jump lengths.
.
.
(iii) The self-diffusion coefficient of ethane in zeolite NaX w a s found to decrease with increasing sorbate concentration. Dependent on the loading, self-diffusion coefficients between 0.5...5 x 10-5cm2cm-1 at 253K were measured by both QENS and NMR PFGT. Mean jump lengths between 0.7 nm and 0.4 nm were found
.
(iv)
Also the comparison of the self-diffusion data of benzene in and 2 x NaX at 458K gave coinciding values o f 7 x cm2s-' at low and elevated sorbate concentrations, respectively. By both techniques it was found that it is a decreasing mean molecular jump length rather than an increasing mean residence time between succeeding jumps that causes this reduction of the self-diffusion coeffitient. Whereas the correlation time of the molecular motions turned out t o be 5 x 10-"s independent of the sorbate concentration, the mean molecular jump lengths decrease from initially 0.46 nm (1 benzene per supercage) to 0.24 nm ( 2 benzene per cage).
ACKNOWLEDGEMENT Kearley (Grenoble) f o r his help in The authors thank Dr. G . J . performing the neutron experiments at the Institut Laue-Langevin Grenoble and Dr. W . Heink (Leipzig) for the development of the NMR spectrometer FEGHIS.
455 LITERATURE /1/ J.E. Tanner and E.O. Stejskal, J. Chem. Phys., 49 (1968) 1768. /2/ J. Klrger and H. Pfeifer, Zeolites, 7 (1987) 90. / 3 / J. K a r g e r , H. Pfeifer and W. Heink, in Rdvances in Magnetic Resonance, ed. J.S. Waugh, 12(1988) 1. /4/ J. Karger and J. Caro, J. Chem. Soc., Faraday Trans. I, 73 (1977) 1363. /5/ J. Karger, H. Pfeifer and W. Heink, Proc. 6 t h Int. Conf. Zeolites, Reno, 1983, Butherworths, 1984, p,104. /6/ M. Bulow, J. Karger, M. KoEiFik and A. Voloscuk, Z. Chem., 2 1 (1981) 175. D.M. Ruthven, L.-K. Lee and H. Yucel, /7/ AIChE J., 26 (1980) 16. /8/ H.-J.Doelle and L. Riekert, Angew. Chem., 9 1 (1979) 309. /9/ J. Caro, M. Biilow and J. Karger, Chem. Engng. Sci., 40 (1985) 2169. /10/ J. Karger and D.M. Ruthven, Zeolites, 9 (1989) 267. /11/ N. Van-Den-Begin, L.V.C. Rees, J. Caro and M. Eiilow, Zeolites, 9 (1989) 287. /12/ H. Jobic, M. B6e, J. Caro, M. Bulow and J. Karger, J. Chem. SOC, Faraday Trans. I, 85 (1989) 4201. Ebe, J. Karger, H. Pfeifer and J. Caro, /13/ H. Jobic, M. J. Chem. S o c . , Chem. Commun., 1990, 341. /14/ H. Jobic, M. Bee and G.J. Kearley, Zeolites, 9 (1989) 312. 1151 H. Jobic, A. Renouprez, M. Bee and C. Poinsignon, J. Phys. Chem, 90 (1986) 1059. 1161 H. Jobic, M. Bee and A. Renouprez, Surface Sci., 140 (1984) 307. /17/ E. Cohen d e Lara, R . Kahn and F . Mezei, J. Chem Soc., Faraday Trans. I, 79 (1983) 1911. /18/ R. Stockmeyer, Zeolites, 4 (1984) 81. 1191 C.J. Wright and C. Riekel, Molec. Phys. 36 (1978) 695. / 2 0 / P.C. Hall and D.K. Ross, Mol. Phys., 42 (1981) 673. /21/ K.S. Singwi and A. Sjolander, Phys. Rev., 119 (196Q) 863. 1221 E . Zibrowius, J. Caro and J. Karger, Z. phys. Chemie (Leipzig), 269 (1988) 1101. 1231 H. Weingartner, Z. phys. Chemie, N.F., 132 (1982) 129. /24/ Landolt-Bornstein, Zahlenwerte und Funktionen, Vol. 5a, Transportphanomene I, p.583, Springer Verlag Berlin, 1969. /25/ J. Caro, M. Biilow, W. Schirmer, J. Karger, W . Heink, H. Pfeifer and S.P. rdanov, J. Chem. SOC., Faraday Trans. I, 81 (1985) 2541. 1261 J. Karger, H. Pfeifer, M. Rauscher and A. Walter, J. Chem. S O C . , Faraday Trans. I, 76 (1980) 717,". 1271 M. Bulow, J. Karger, M. KoEiFik and A. M. Voloscuk, Z. Chem., 2 1 (1981) 175. /28/ J. Karger and D.M. Ruthven, J. Chem. S o c . , Faraday Trans. I, 77 (1981) 1485. Rees, in "Zeolites:Facts, /29/ N.6. Van-Den-Begin and L.V.C. Figures, Future", eds. P.A. Jacobs and R.A. van Santen, Elsevier, Amsterdam, 1989, p. 915. / S O / M. Bulow, W . Mietk, P. Struve and P. Lorenz, J. Chem. S O C . , Faraday Trans. I, 79 (1983) 2457. /31/ Dongmin Shen and L.V.C. Rees, unpublished results. /32/ M. Eic, N.V. Goddard and D.M. Ruthven, Zeolites, 8 (1988) 327. /33/ A. Germanus, J. Karger, H. Pfeifer, N.N. Samulevic" and S.P. tdanov, Zeolites, 5 (1985) 91. /34/ B. Boddenberg and R. Burmeister, Zeolites, 8 (1988) 488. /35/ R.S. Chiang, A.G. Dixon and Y.H. Ma, Chem. Engng. Sci., 39 (1984) 1461. 1361 D.T. Hayhurst and A.R. Paravar, Zeolites, 8 (19881 27. Principles of Adsorption and Adsorption /37/ D.M. Ruthven, Processes, Wiley-Interscience Publ., New York, 1984.
This Page Intentionally Left Blank
G.Ohlrnann et al. (Editors), Catalysis and Adsorption by Zeolites 1991 Elsevier Science PublishersB.V.. Amsterdam
457
PERMEABILITY STUDIES ON A SILICALITE SINGLE CRYSTAL MEMBRANE MODEL
E.R. Geusl, A.E. Jansen', J.C. Jansen', J. Schoonmanl, and H. van Bekkum' Delft University of Technology, Laboratory of Applied Chemistry, Julianalaan 136, 2628 BL Delft, The Netherlands TNO-MT, Utrechtseweg 48, 3704 HE Zeist, The Netherlands SUMMARY
Permeation experiments on large silicalite single crystals (MFI type), embedded in an epoxy resin were performed, using permanent gases and small alkanes. A scaling-up from initially a one zeolite crystal membrane to a multi zeolite crystal membrane is presented. Differences in permeability were small and attributed to a sorption-diffusion mechanism, in qualitative agreement with theory. As permeation may vary within different regimes, maximum selectivities are expected in the Henry region. Some boundary conditions in the preparation of (ceramic) zeolite membranes are formulated.
INTRODUCTION Zeolite membranes may open new horizons in both separation technology and reaction engineering. At present, most gas separation membrane processes are operated using polymer membranes. Ceramic membranes can be used at higher temperatures, which allows regeneration and catalytic application (ref. 1). The sol-gel technique has been used successfully in the preparation of ceramic membranes (refs. 2-5). Alumina gas separation membranes can be produced on a semi-industrial scale with relative ease (ref. 6). Zeolite membranes in an all-ceramic system are expected to exhibit significantly higher selectivities due to the zeolite's molecular sieve characteristics. In the development and use of zeolite membranes it is essential to obtain information on the mass transport through zeolite crystals. Recently an extensive mathematical model for the permeation through porous crystal membranes has been proposed by Barrer (ref. 7). In the experimental field Hayhurst and Paravar (refs. 8 - 9 ) reported on the determination of the diffusivity of alkanes by means of a zeolite membrane configuration. These experiments, however, were performed on a twinned silicalite crystal at low feed pressures. Recently this work was extended by similar experiments on benzene passage through a silicalite single crystal (ref. 10). Wernick and Osterhuber (refs. 11-12) reported on the permeation through a NaX single crystal. Significantly higher feed pressures were used in these experiments, and
the permeation characteristics in time were
emphasized. According to patent literature ceramic zeolite membranes have been prepared by crystallizing zeolites in situ either in (ref. 13), or on (refs. 14-18) a macroporous support. Alternatively, small zeolite crystallites together with
458 a ceramic binder are said to cover a macroporous support completely (ref, 19), or result in an unsupported film (ref. 2 0 ) . Recently Bein et al. (ref. 21) reported on the preparation of inorganic thin films, containing zeolite
crystals (type Y and chabazite). In the above cited patent literature it is emphasized that the size of the zeolite crystals (i.e. the membrane thickness) should be minimalized to realize sufficiently high permeabilities. On the other hand, large zeolite crystals are expected to be most practical in the preparation of zeolite membranes (ref. 2 2 ) . and in the authors' opinion especially on a laboratory scale. In this work the permeation of permanent gases and linear alkanes through silicalite single crystals in a membrane configuration has been subjected to a first scaling-up. To that end several crystals were embedded into an epoxy resin matrix on a perforated metal layer, preliminary to all-ceramic membranes which are in preparation. This study also intends to improve knowledge on (i) pitfalls in the preparation of zeolite membranes, and (ii) the operation under practical conditions (e.g. high feed pressures). In addition, unexpected behaviour of zeolite membranes (caused by e.g. cracking of zeolite crystals or fouling) could be diagnosed. THEORY Barrer (ref. 7) recently proposed a permeation model for systems obeying Langmuir's isotherm in which activated intracrystalline diffusion takes place. In addition external surface effects on both feed and permeate side were included, resulting in a five step process. In this work we will focus on the overall permeability, which is expected to be governed by a sorptiondiffusion mechanism. A convenient way to define the permeation rate of a membrane makes use of the permeability ( @ ) (ref. 23). which can be derived from Fick's first law of diffusion:
in which Qmor denotes the steady state flow, A,,, the surface area of the membrane, 1, the membrane thickness (in this work the crystal size in the bdirection), and Ap the pressure difference over the membrane. The permeability should be considered a phenomenological parameter of the membrane, resulting from specific conditions (i.e. pressure, temperature, feed composition, degree of fouling). Formula [l] will be used to calculate the permeability from the permeation rate at steady state. According to Barrer's model, maximum permeation will occur when the mass transport is only governed by intracrystalline diffusion. This holds if sorption processes on both sides of the membrane are much faster than
459
intracrystalline diffusion: / ' m Dintra,diff * %
'mo1,max
* I
--
in which Acmax denotes the maximum concentration difference over the membrane. Acmax can be derived from sorption isotherms (assuming equilibrium between gas and adsorbed phase). Dintra,diff can be obtained from NMR data, referring to the intracrystalline self-diffusivity (Dself). The PFG NMR technique measures DSelf directly, albeit only for rapid processes (ref. 2 4 ) . The Darken equation enables one to calculate the intracrystalline Fickian diffusivity (ref. 2 5 ) :
Dintra,diff = Dself
*
For the Langmuir isotherm, equation [ 3 ] may be written as: Dintra,diff = Dself *
- 6')
[41
in which 6' denotes the fraction of sites occupied. NMR results (ref. 2 6 ) have shown that for small alkanes (C1-C3) DSelf decreases for increasing carbon number, due to the diminished molecular free volume. For sufficiently high 6'. this effect is overruled by the increase of 1/(1 - 6') (ref. 2 5 ) . During permeation a concentration profile will exist over the zeolite crystals, which will remain constant at steady state. As 6' varies over the membrane, an average diffusivity can be found by integrating over the total membrane. Finally, equation [ 2 ] is rewritten as: (Dintra,diff)av *
@mol,max
m'I%
*
r51
When 6' is high over the whole membrane, the Fickian Dintra,diff will be significantly higher than the NMR DSelf. Thus, the sorption equilibrium will influence the permeability of a species, and both sorption and diffusion should be membrane.
considered in evaluating the permeability through a
zeolite
EXPERIMENTAL Large. flat silicalite single crystals were synthesized as described by Jansen et al. (ref. 2 7 ) . The Si/Al ratio (> 3 0 0 ) was determined by ICP analysis on large crystals from a single batch. Selected crystals of about equal size (a.b,c = 200*100*350 p n ) were carefully calcined at 500' C (heating rate 1" C/min). In a preliminary experiment one large silicalite crystal was mounted in an epoxy resin film (bisphenol A glycidyl ether/poly amine; Araldit, CibaGeigy). As this procedure was rather time consuming and complex, a more
460
practical procedure was developed. Thin copper platelets (thickness 50 fim) were provided with up to 20 holes (diameter 150 pm). Any remaining o i l was removed by cleaning the plate in boiling acetone, and drying in air. The components of the epoxy resin were homogeneously mixed and carefully spread around each hole. A silicalite single crystal was positioned on each hole. The epoxy resin showed good adhesion towards both the zeolite surface and the copper. The starting viscosity of the epoxy upon mixing is very low. This resulted in an epoxy film over the total crystal surface facing the copper plate (b-direction). After 1-2 hours, the viscosity of the epoxy had increased strongly. The adhesion to the zeolite material is still excellent. Therefore, the zeolite crystals are preferably placed on the platelet at this stage of the hardening of the epoxy resin. The epoxy resin had hardened within 24 hours completely. Table 1 gives characteristics of the various membranes. TABLE 1:
Zeolite-epoxy-on copper composites studied in this work
Code
'N
A,'
M1
o
c1
1
MES-21 1 5ZH-1 5 LPZ- 1 10 LPZ-3a,b,c 8
(m2)
1.2*10-~
1, (m) l.0*10-4
1.68*10-'
9.0*10-5
3.0*10-' 4.48*10-' 1.36*10-7 1.10*10-7
1.0*10-4 l.0*10-4 9.0*10-' 9.0*10-5
Remarks epoxy resin film in a perforated platelet as-synthesized single crystal, containing Pr4N+ template initial preparation method measurements at 30" and 120" C measurements at 30" C sequence of measurements at 300, 130", and 30' C ; LPZ-1 after repair
')number of crystals embedded ')effective surface area, determined by light microscopy Figure 1 shows an embedded zeolite crystal in epoxy on a copper plate. The zeolite membranes were inspected by means of light microscopy. Figure 2 demonstrates the determination of the effective surface area of an embedded zeolite crystal. In addition, the integrity of each embedded zeolite crystal was checked by microscope FTIR-spectroscopy. The experimental set-up is described elsewhere (ref. 28). Permanent gases (H2 (99.5%), N2 (99.999%). and O2 (99.9951)). small alkanes (methane (99.5%), ethane (99.9%), and n-butane (99.0%)), and carbon dioxide (99.7%) were used as permeate feeds. Measurements were performed at permeation periods of 2-10 minutes. Feed pressures were within 1-2 bar. Permeate pressures always started at approximately 2-4 Pa, and never exceeded 100 Pa, except for hydrogen (< 250 Pa). As some inleak occurred (approximately 0.01 Pa/s) permeation rates could not be determined at too low a pressure rise. By embedding several silicalite crystals in one membrane the accuracy of the measurements was increased.
461
Fig. 1. Silicalite crystal, embedded in epoxy resin
Fig. 2 . Determination of free area (lighter area in the centre of the zeolite crystal)
RESULTS Some preliminary experiments were performed as to make sure that only transport through the zeolite crystals was to take place. To double check, experiments were carried out several times, using different membranes. The non-permeability of the epoxy resin was checked by membrane M1. Up to 200" C no permeation could be detected, establishing that the epoxy resin meets the requirements. On an as-synthesized crystal (still containing the template molecules; membrane Cl), embedded in a similar way, no permeation could be detected. The expected permeation rate (related from measurements on LPZ-3b at similar conditions) should have been significantly higher than the accuracy of the apparatus. Thus, it was concluded that zeolite crystals could be embedded properly by means of the present procedure. Membrane MES-21 is the initial one crystal composite. For all gases very high permeabilities were found (cf. Table 2 ) . resulting in Knudsen diffusion selectivity. Inspection by scanning electron microscopy revealed a large crack in the zeolite crystal. The high permeability and Knudsen selectivity are attributed to the presence of this crack. Permeation results on membrane LPZ-1 are listed in Table 2 as well. The permeability for each gas seemed to increase slightly after repeated experiments. For this reason LPZ-1 was removed from the cell for inspection. Two out of ten crystals appeared to be cracked, and subsequently these crystals were completely covered by fresh epoxy resin. Permeation rates through the thus repaired membrane (LPZ-3; Figure 3) were
462
nearly two orders of magnitude lower than for LPZ-1 (cf. Table 2). This change in permeability can not be accounted for by the loss of free surface area, which
is only
20%. It
can be
seen from
Table
2
that no high
selectivities for the various gases were found. The permeability for nbutane at 30" C (LPZ-3a) was higher than for other gases tested. TABLE 2:
Experimental permeabilities for permanent gases and alkanes on zeolite-epoxy composites
Membrane
MES-21
T ('C) AP (bar)
25 1-2
Permeability (nmol .m/m2.s . Pa) LPZ - 1 LPZ-3a LPZ-3b
30 1.6-1.8
LPZ-3c
30 1.6
130 1.6
30 1.6
0.0095 0.0046
0.0234 0.0058 0.0057 0.0052 0.0082 0.0065 0.0056
0.0430 0.0134 0.0120
Gas H2 N2 02 COP CH4 'ZH6 n-C4H10
780 170 200 170 250 120
0.69 0.24 0.33 0.20 0.32 0.37 0.42
0.0037 0.0027 0.0145
0.0182 0.0158
Fig. 3. Membrane LPZ-3; the two cracked crystals are completely covered by epoxy resin As the temperature was raised to 130'
C (LPZ-3b). permeabilities of all
gases slightly increased, except for n-butane (decrease of 70%). Again, selectivities were found to be rather small. The measurements on LPZ-3a and LPZ-3b were highly reproducible. Upon cooling to 30" C (LPZ-3c) again, all permeabilities were found to be much higher compared to LPZ-3a. In contrast to the previous experiments, these experiments were not very reproducible. Upon inspection it was indeed
463 found that some crystals were cracked, so the increased permeability is not an effect of the activation at higher temperature. During the first few experiments on LPZ-3a (at 30' C) the zeolite pores proved to be obstructed by adsorbed vater. FTIR-spectra (Figure 4) of calcined silicalite crystals reveal a considerable amount of adsorbed vater under ambient conditions. As the phenomenon vanished after some experiments, the adsorbed water is apparently induced to desorb by the permeating gas.
Fig. 4. FTIR-spectrum of embedded silicalite crystal The above-mentioned phenomenon was also observed during experiments on membrane 5ZH-1 at room temperature. Only few experiments were performed at 30' C, after which the temperature was raised to 90' C. The permeability decreased at this high temperature, and finally the membrane was completely impermeable. Inspection of the zeolite crystals by microscope FTIRspectroscopy revealed C-H-vibrations (Figure 5). As these vibrations do not refer to the zeolite framevork it was concluded that some obstructing guest molecule was stabilized within the zeolite framework at high temperature. 1.8
I . 35
0.9
0. a 5
0.e
4 4000
I 3528
3000
z m
2000 YAVENUMBERS C H - 1
ism
iaaa
t
Fig. 5. FTIR-spectrum of non-permeable silicalite crystal due to fouling
464
DISCUSSION In Table 3 the calculation of the maximum flow according to equation [5] is shown for the linear alkanes C1-C4 at 30", assuming a linear concentration profile. Unfortunately, the required data on sorption (ref. 29) and diffusion (ref. 26) were rather incomplete. The effect of sorption on DFickaV is only demonstrated qualitatively. TABLE 3:
T
=
Maximum theoretical flows of linear alkanes C1-C4 for membrane LPZ3 , calculated from equation [ 5 ]
30" C; Ap
CH4$ C2H6
3600 3472
=
1.5 bar; pressure on permeate side: 100 Pa
0.63 0.67 0.81 0.83
0.00
1.2*10-8 0.02 7.5*10-' 0.07 6.O*lO-lo 0.53 5 . 0*1O-l1
1.48*10-8 5.Ol*lO-' 4.14*10-' 6.31*10-1°
3.56*10-1° 1.16*10-1° 8 . 71*10-11 4.55*10-12
65.3 21.2 16.0 0.84
')fract .ion of sites occupied on feed (F) and permeate (P) side (equilibrium) $ ) data extrapolated from sorption data (ref. 29). cf. text I)data extrapolated from diffusion data (ref. 26), cf. text The intracrystalline self-diffusivity of n-butane can not be determined by the PFG NMR technique (ref. 30). and was estimated from data on methane, ethane, and propane. The correlation of the intracrystalline diffusivity to the carbon number for n-alkanes, found by Eic and Ruthven with the ZLC method (ref. 30) was used. The sorption data on methane were estimated from sorption isotherms of ethane, propane, and n-butane (ref. 29). Because of lack of sufficient data, permeabilities of permanent gases (H2, N2, and 02) and carbon dioxide were not calculated. The sorption behaviour of nitrogen and to a lesser extent carbon dioxide has been shown (ref. 31) to obey Henry's law up to high temperatures (70" C), and high pressures (up to 1 bar). In contrast to the alkanes, no sorption effect is expected for permanent gases, as (aln p/aln c) remains constant. In contrast to theory (Table 3), small differences in permeability are measured (Table 2 ) . Permeabilities are considerably smaller (methane, ethane) than the maximum flows, or of the same magnitude (n-butane)
.
This implicates
that other processes (e.g. sorption to and from the zeolite pores) than intracrystalline diffusion govern the permeation. Applying l o w feed pressures (up to 1.8 kPa), Hayhurst and Paravar have found high selectivities (ref. 9). In their experiments however, permeation occurred through the c-direction (ref. 32). resulting in a high tortuosity factor. Permeation may therefore be hindered, especially for higher alkanes. Negligible selectivity was also found by Wernick and Osterhuber (ref. 11) on binary and ternary mixtures of various alkanes (NaX crystal). The reported
465 rapid and slow permeation regime transition may be (partly) explained by the sorption effect on intracrystalline diffusion, as high feed pressures (1 bar) were applied (permeate pressure < 133 Pa). In the rapid permeation regime the factor (aln p/ah c) will be high, due to high site occupation. This rapid regime was found to be metastable for higher temperatures (65°-1000 C). Sorption experiments by Stach et al. (ref. 33) have shown that sorption on permeate side (13.3-133 Pa) decreases strongly in this temperature region. From these studies it is concluded that the highest selectivities will be realized in the Henry regime. Table 3 indicates that at equal conditions the difference between DSelfmaX and DFickaV will be larger for higher alkanes. As sorption is small in the Henry region, the permeation is governed by the intrinsic diffusivity only. In order to achieve high permeabilities, the Henry regime is favourably reached by an increase of temperature. CONCLUSION Large silicalite single crystals have been found to be permeable for permanent gases and small alkanes. A scaling-up of the mass transfer was achieved by embedding several zeolite crystals into one membrane. Permeation through zeolite membranes may occur within different regimes. Maximum selectivity is expected to be realized in the Henry region, in which the intracrystalline diffusivity is hardly affected by sorption. The application of large zeolite crystals in preparing zeolite membranes is to be preferred from a preparative point of view. Draw-backs, however, may be the sensitivity to high feed pressures, and relatively low permeabilities. As (re)-activated, all-ceramic membrane zeolite membranes have to be configurations are strongly recommended. ACKNOWLEDGEMENT E.R. Geus is very
grateful to Prof. R.M. Barrer
for a
stimulating
discussion on zeolites as membranes. REFERENCES H.P. Hsieh, Inorganic Membrane Reactors - A Review, AIChE Symposium Series 5 , 8 5 , no. 268, (1989), 53-67; A. Larbot, A. Julbe, C. Guizard, L . Cot, J.Membr.Sci., 93, (1989), 289303 ; A. Larbot, J.P. Fabre, C. Guizard. L. Cot, J.Am.Ceram.Soc., 72, (1989), 257-61; W.A. Zeltner, M.A. Anderson, ‘Chemical Control over Ceramic Membrane Processing: Promises, Problems and Prospects’, in: Proc. 1st Int.Conf.Inorg.Membr., (eds. J. Charpin, L . Cot), Montpellier, France, July 3-6, 1989, 213-223; A. Leenaars, Preparation, Structure and Separation Characteristics of Ceramic Alumina Membranes, thesis, University of Twente, Netherlands, (1984) ; H.M. van Veen, R.A. Terpstra, J.P.B.M. Tol, H.J. Veringa, ’Three-Layer Ceramic Alumina Membrane for High Temperature Gas Separation Applications’, in: Proc. 1st Int.Conf.Inorg.Membr., (eds. J. Charpin, L. Cot), Montpellier, France July 3-6, 1989, 329-335;
466
7 8
9 10
11
12 13 14 15 16 17 18 19 20 21 22 23
24 25 26 27 28 29 30 31 32 33
R.M. Barrer, J.Chem.Soc.Faraday Trans., 86(7), (1990). 1123-1130; A.R. Paravar, D.T. Hayhurst, 'Direct Measurement of Diffusivity for Butane Across a Single Large Silicalite Crystal', 6th Int.Zeol.Conf., (Eds. D. Olson, A. Bisio), Reno, USA, July 10-15, 1983, 217-224; D.T. Hayhurst, A.R. Paravar, Zeolites, !3, (1988), 27-29; V.V. Nesarikar, D.B. Shah, D.T. Hayhurst, poster on BZA meeting, Chislehurst, England, July 15-20, 1990; D.L. Wernick, E.J. Osterhuber, 'Diffusional Transition in Zeolite NaX: 1. Single Crystal Gas Permeation Studies', 6th Int.Zeol.Conf., (Eds. D. Olson, A. Bisio), Reno, USA, July 10-15, 1983, 122-130; D.L. Wernick, E.J. Osterhuber, J.Membr.Sci., 22, (1985). 137-146; H. Suzuki, European Patent Application 180,200, (1985) to H. Suzuki; M. Oyama, Japanese Patent 63,291,809. (1988) to Idemitsu Kosan Co.Ltd.; H. Suzuki, US Patent 4,699,892, (1987) to H. Suzuki; I.M. Lachman, M.D. Patil, US Patent 4,800,187, (1989) to Corning Glass Works ; H. Suzuki. European Patent Application 135,069, (1986) to H. Suzuki; S. Satoshi, N. Tagaya, T. Maeshima, T. Isoda, Canadian Patent 1,235,684, (1988) to Toa Nenryo Kogyo Kabushiki Kaisha; K. Miyazaki. Japanese Patent 60,129,119, (1985) to Matsushita Denki Sangyo K.K. ; G. Bellussi, F. Buonomo. A. Esposito, M. Clerici, U. Romano, European Patent Application 265,018 (1988) to Eniricerche, Snamprogetti, Enichem Synthesis ; T. Bein, K. Brown, C.J. Brinker, 'Molecular Sieve Films from ZeoliteSilica Microcomposites', Stud.Surf.Sci.Catal., 49, (1989). 887-896; U. Muller, A. Reich, K.K. Unger, Stud.Surf.Sci.Cata1, 5 2 , (1989), 241251; K. Kammermeyer, Gas and Vapor Separations by Means of Membranes, Progress in Separation and Purification, Vol. 1, (Ed. E.S. Perry), Wiley, (1968). 335-372; J. KBrger, D.M. Ruthven, Zeolites, 9, (1989), 267-281; J. KBrger, S.P. Shdanov, A. Walter, Z.phys.Chemie, Leipzig 256, (1975). 319-329; J. Caro, M. Btllow, W. Schirmer, J . KBrger, W. Heink, H. Pfeifer, S.P. Zdanov, J.Chem.Soc.Faraday Trans., 6 l , (1985). 2541-2550; J.C. Jansen, E. Biron, H. van Bekkum. Stud.Surf.Sci.Cata1.. 37 (1988). 133-141 ; H.C.W.M. Buys, A. van Elven, A.E. Jansen, A.H.A. Tinnemans, J.Appl.Pol.Sci., in press; R.E. Richards, L.V.C. Rees, Langmuir, 3 , (1987). 335-340; M. Eic, D.M. Ruthven, Stud.Surf.Sci.Cata1.. 49, (1989), 897-905; P. Graham, A.D. Hughes, L.V.C. Rees, Gas Sep.Purif., 3 , (1989), 56-64; D.T. Hayhurst, private communication; H. Stach, S.P. Shdanov, K. Fiedler, W. Schirmer, N.N. Samulevic, Z.phys.Chemie, Leipzig. &SO, (1979). 455-464.
G . Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
THE ADSORPTIVE AND THE CATALYTIC
461
DIFFUSION OF 2,3-DIMETHYLBUTANE I N LARGE
CRYSTALS OF (ALUMINATED) SILICALITE P. VOOGD and H. VAN BEKKUM Department o f A p p l i e d Chemistry and Chemical Technology, D e l f t U n i v e r s i t y o f Technology, J u l i a n a l a a n 136, 2628 BL D e l f t (The Net herlands) SUMMARY The a d s o r p t i v e d i f f u s i v i t y o f 2,3-dimethylbutane (2,3-DMB) i n s i l i c a l i t e (373-473 K) o b t a i n e d by t h e v o l u m e t r i c method has been compared w i t h t h a t o b t a i n e d f ro m c r a c k i n g r e a c t i o n s o v e r l a r g e aluminat ed s i l i c a l i t e c r y s t a l s (673-773 K) i n a f i x e d - b e d c o n t i n u o u s - f l o w r e a c t o r . A n a l y s i s o f 2,3-dimethylbutane c r a c k i n g r a t e s w i t h T h i e l e ’ s model r e s u l t s i n experiment a l r e a c t i o n e f f e c t i v e n e s s f a c t o r s which a r e c l o s e l y f o l l o w e d by t h e o r y . The a c t i v a t i o n energy o f t h e r e a c t i o n l i m i t i n g d i f f u s i v i t y o f 2,3di met h y lbut a ne i n aluminated s i l i c a l i t e i s found t o be much h i g h e r t h a n t h e a c t i v a t i o n energy o f t h e c o r r e s p o n d i n g a d s o r p t i v e d i f f u s i v i t y . T h i s phenomenon i s shown t o be c h a r a c t e r i s t i c f o r t h e c a t a l y s t sample. INTRODUCTION D e s p i t e numerous i n v e s t i g a t i o n s t o d e t e r m i ne i n t r i n s i c and s e l f - d i f f u s i v i t i e s o f adsorbates i n z e o l i t e channels and c a v i t i e s , e s t i m a t i o n s o f t hese parameters from
catalytic
high
interest
experiments to
compare
( a t h i g h t e m p erat ures) a r e scarce (1,Z). diffusion
(adsorptive d i f f u s i v i t i e s )
data
It i s o f
determined under i n e r t c o n d i t i o n s
w i t h t h e s e ‘ c a t a l y t i c ’ d i f f u s i v i t i e s . As i t i s n o t
p o s s i b l e t o do b o t h t y p e s o f experiments a t t h e same t emperat ure, comparison can o n l y be made t h r o u g h t h e a c t i v a t i o n energy o f d i f f u s i o n , assuming t h a t t h e ze ol it i c d i f f u s i v i t y i n c r e a s e s exponenti a1 l y w i t h temperature. Zeolite
ZSM-5
and
can
be
experiments have in
an
attractive
o p p o r t u n i t y t o compare c a t a l y t i c and
synthesised
i n crystals
large
enough f o r a d s o r p t i v e d i f f u s i o n
( 3 ) . U n f o r t u n a t e l y , d i r e c t l y s y n t hesised ZSM-5 c r y s t a l s i n v a r i a b l y
c o n t a i n more aluminium a t t h e i r r i m t h a n a t t h e i r core. such a aluminium g r a d i e n t and i t s c o n comit ant heterogeneous d i s t r i b u t i o n
been
Since
offers
d i f f u s i o n r a t e s o f hydrocarbons, f o r i t i s a h i g h l y a c t i v e c a t a l y s t
adsorptive
found
catalytically
to
active
s i t e s impedes t h e successf ul a p p l i c a t i o n o f T h i e l e ’ s
t h e o r y t o ’ c a t a l y t i c ’ d i f f u s i o n , h e r e we r e p o r t on l a r g e ZSM-5 c r y s t a l s o b t a i n e d by a l u m i n a t i o n o f p u r e l y s i l i c e o u s ZSM-5 frameworks ( s i l i c a l i t e ) . 2,3Dimethylbutane i s used as probe molecule because i t ’ s k i n e t i c d i a m e t e r i s comparable t o t h a t o f t h e ZSM-5 channels and t h u s w i l l r e n d e r e x t e r n a l h e a t and m a s s - t r a n s f e r r e s i s t a n c e n e g l i g i b l e i n a d s o r p t i v e methods.
468
EXPERIMENTAL D i r e c t s v n t h e s i s and c h a r a c t e r i s a t i o n o f s i l i c a l i t e and o f ZSM-5 Two batches o f l a r g e s i l i c a l i t e c r y s t a l s ( S i l l and Si12) were prepared by
an a l k a l i n e - f r e e s y n t h e s i s method o f Ghamami e t a l . ( 5 ) . A t h i r d s i l i c a l i t e
sample
(Si13)
was
s y n t h e s i s e d a c c o r d i n g t o p a t e n t l i t e r a t u r e ( 6 ) . Elemental,
s t r u c t u r a l and a d s o r p t i v e c h a r a c t e r i s a t i o n methods as w e l l as methods t o de t ermine c r y s t a l morphologies and c r y s t a l s i z e d i s t r i b u t i o n s have p r e v i o u s l y been d e s c r i b e d ( 3 ) . I n T a ble 1 a summary o f t h e samples’ c h a r a c t e r i s t i c s i s g i v e n . S i l i c a l i t e was dealuminated w i t h H20/HC1 a t 1073 K a c c o r d i n g t o a st andard method (7), l e a d i n g to
samples S i l l d , S i l L d and S i l 3 d . X - r a y powder d i f f r a c t i o n (XRD) and i n f r a r e d
spectroscopy crystallinity
(IR)
showed
that
the
p a r t i a l dealumination d i d n o t a f f e c t the
o f t h e samples. Dealuminated samples were used i n t h e a d s o r p t i v e
d i f f u s i o n experiments. TABLE 1 C h a r a c t e r i s a t i o n o f (dealuminated) s i l i c a l i t e samples Sample
Sill
Si12
Si13
34.9
89.3
5&
Mean c r y s t a l l e n g t h , c - d i r e c t i o n (pm) Mean c r y s t a1 he ig ht a- o r b - d i r e c t i o n (pm)
#
S i 1 i c o n - t o - a 1 umi n i um r a t i o6
11.4, 8 . 4
39.7
503/600*
400/ 500*
150/400*
r10
Yes 10.75/10.72*
Yes 10.66/10.83*
Occurrence o f c r y s t a l agglomerates n-Hexane ads. (wt%)
10.09/10.19*
&) As determined f r o m SEM photographs. #)
Crystal
bre a d t h s
( a - a x i s ) were equal t o h e i g h t s ( b - a x i s ) , except f o r S i l l
f o r which s i z e s i n a - and b - d i r e c t i o n a r e g i v e n (a, b ) . * ) As d et e rmined w i t h ICP-AES.
*
) As-synthesised/after dealumination.
A ZSM-5 sample was s y n t h e s i s e d (8), c a l c i n e d a t 823 K ( 1 K/min) and r e p e a t e d l y ion-exchanged w i t h NH4C1 (1M) f o l l o w e d by a f i n a l c a l c i n a t i o n . X-ray d i f f r a c t i o n and I R proved t h e sample t o be h i g h l y c r y s t a l l i n e ZSM-5. According t o elemental a n a l y s i s i t s s i l i c o n - t o - a l u m i n i u m r a t i o was 86.
469
A1 umination o f s i l i c a l it e S i l i c a l i t e S i l l and Si12 were aluminated a t 773 K w i t h A1C13 according t o Yamagishi e t a l . (9), followed by a repeated ammonium exchange. The aluminated samples are denoted S i l A l l and SilA12, r e s p e c t i v e l y . The
aluminium d i s t r i b u t i o n
in
SilA12 c r y s t a l s was determined by Electron-
.
Probe Microanalysis (EPMA)
AdsorDtive d i f f u s i o n and adsorDtion constants The d i f f u s i v i t y o f 2,3-dimethylbutane (2,3-DMB) i n t h e s i l i c a l i t e samples was determined i n a constant volume v a r i a b l e pressure system as described e a r l i e r (10)The
experimental
Silld 473
and S i l 2 d amount
K;
temperatures. K,
of
diffusion
-
were: pressure = 50 sorbed
-
1.3*10-3
=
measurements performed w i t h
700 Pa a t 373 K and 1000 7*10-3
mmol/g
-
10000 Pa a t
per uptake run a t both
Adsorption c a p a c i t i e s were 0.60 and 0.24 mmol/g a t 373 K and 473
respectively.
silicalite
parameters
I n addition
Sil3d
2,3-DMB
adsorption isotherms were obtained on
at
373, 473 and 523 K. Adsorption data on S i l 3 d above 673 K were determined by gas chromatography as described e a r l i e r (3). Catalvsis Cracking reactions microreactor, n-butane
(3),
(>99.5
were
performed
in
a
fixed-bed
continuous f l o w glass
w i t h 6% 2,3-DMB (>99 %), 6% 3-methylpentane (>99.5%) o r 2%
%) i n d r y
helium a t 101 kPa. The products o f t h e cracking
reactions were gaschromatographical l y analysed. Apparent
first
order
r a t e constants,
kaPP,
were determined according t o
Sill
and Si12 revealed t h a t these
conventional equations (3). RESULTS AND DISCUSSION Characteri sation Scanning crystals show
a narrow
fraction pm i n
the
e l e c t r o n microscopy are
invariably
(SEM)
twinned.
of
No gel phase i s detected. The batches each
p a r t i c l e s i z e d i s t r i b u t i o n . Sample Si12 does c o n t a i n a c e r t a i n
o f agglomerates. SEM o f Si13 shows s p h e r u l i t i c c r y s t a l l i n e m a t e r i a l 5 size. The ZSM-5 sample consists o f smaller s p h e r u l i t e s (0.1-1 pm). From
silicon-to-aluminium
ratios
i n Table
1 it
i s clear that true Al-free
470
silicalite
is
n o t r e a l i s e d i n the a l k a l i n e - f r e e synthesis g e l , mainly because of the i m p u r i t y of the s i l i c a source (Ludox AS40/Dupont, 100 ppm A l ) . A1 uminat i o n Alumination
of
and Si12 has r e s u l t e d i n a s u b s t a n t i a l increase o f the
Sill
aluminium content (Table 2) whereas t h e adsorptive p r o p e r t i e s f o r n-hexane have remained
unchanged.
Results
of
the
elemental
analysis
reveal
TABLE 2 Characterisation o f aluminated s i l i c a l i t e samples SilAll
SilA12
Si/A1 (before NHi-exchange)*
31
46
Si/A1 ( a f t e r NH;
70
85
9.98
10.09
Sample
-exchange)*
n-Hexane ads. (wt%)
t h a t ammonium removes between 20 and 30 % o f t h e deposited aluminium from the sample. This l o s s o f aluminium can be explained w i t h t h e r e a c t i o n scheme proposed by Chang e t a l . ( l l ) , e.g. an i n s e r t i o n o f aluminium i n t o t h e z e o l i t e framework w i t h a concomitant formation o f counter c a t i o n i c aluminium. Figure 1 the s i l i c o n and aluminium d i s t r i b u t i o n across an a- ( b - ) section
In of
a polished c r y s t a l o f sample SilA12 are shown. The A l - l e v e l does n o t vary
significantly
over
the
major
part
of
the
polished crystal
surface.
Nevertheless, some Al-zoning seems t o e x i s t over a very small distance, v i z . 2 3 pm as i n d i c a t e d by t h e i n t e r v a l o f t h e step advancement. Adsorot ive d i f f u s i on An analysis o f adsorption serious samples.
deviations Thus
in
isotherms o f 2,3-DMB revealed t h a t t h e r e are no
adsorptive p r o p e r t i e s o f t h e (dealuminated) s i l i c a l i t e
a comparison
of
data
on
diffusion
obtained w i t h d i f f e r e n t
adsorbent samples i s j u s t i f i e d as the adsorption c h a r a c t e r i s t i c s are comparable f o r each sample. In
Figures
coverage
2 and 3
uptake curves are shown f o r 2,3-DMB d i f f u s i o n a t low
(maximum 3.5 molec. 2,3-DMB/u.c.)
i n a volumetric system a t 373 K and
471
--: 5 0 0 0 . 4000 L
b
SI
A
.r
%
c
Al
3000.
1.2
1.6
1.4
K"
10'/T,
b-dlractlon (mlcron) Flg. 1:
FIg. 4: Tamprraturr rffoct on k- for 2,3-dlmothylbutano crocklng on sarnplrs SIIAll. A and SllA12, and on H-ZSM-5, 0
SI- an Al-dlstrlbutlon across an alurnlnotrd crystal of SIIAIZ.
.
r;
1.0
I 1.0'
Y
0.8 0.6
0.4
0.2
0.0 1 0
Y 0.8
I
100
0.6
0.4
0.2 *
J
300
200
400
0.0
4 0
1000
3000
2000
t, Flg. 2: Uptako of 2.3-dlmrthylbutana In Slldl, 0 and SlldZ, at 373 K. Sorb. covrragr= 0.5 molrc./u.c.
+
FIg. 3: As Flguro 2. T=473 K Sorb. covorago=0.15 maloc./u.c.
472
473 K , respectively. It is clear that the theoretical curve follows the data reasonably well up to high relative uptake. TABLE 3 Apparent and intrinsic diffusivity of 2,3-DMB in dealuminated silicalite samples Silld and Sil2d at 373 K and 473 K
Sil Id 373 473
4.i*io-16 1 .5*10-14
Sil2d 1.0*10-15 1 .4*10-14
Si 1 Id 2.8*10-16 1 .7*10-14
Si 12d 5.5*10-16 1 .6*10-14
In Table 3 results are given o f the curve fitting procedures. Apparent together with intrinsic diffusivities are given, the latter being obtained by calculating Darken’s correction factor from the respective isotherms (10). Furthermore, for each silicalite sample a specific diffusion length is given, taken t o be half the breadth o f a crystal. From the diffusivities obtained at 373 K and at 473 K an activation energy o f diffusion is calculated o f about 52 kJ/mol e . Both the diffusivities and the activation energy compare well with data obtained (using other techniques) on diffusion o f other hydrocarbons in silicalite/ZSM-5 at other temperatures. The diffusivity o f 2,3-DMB is, after extrapolation t o corresponding temperatures, substantially lower than that of 3-methylpentane in ZSM-5, viz. by a factor o f 100 (10). Contrarily, compared to 2,2-dimethylbutane (Deff= 6.5*10-19 m2/s) the migration rate o f 2,3-DMB is enhanced by a factor o f about 500 (2). ’Catalytic’ diffusion First order rate constants have been determined for the 2,3-DMB cracking reaction over the directly synthesised H-ZSM-5 sample and over the aluminated silicalite samples SilAll and SilA12. The rate constants relate t o steady state situations. Figure 4 shows an Arrhenius plot revealing that the apparent activation energy of reaction is highly dependent on the size o f the zeolite
473
c r y s t a l s . I f t h e c o n c e n t r a t i o n s o f a c t i v e s i t e s i n t h e d i f f e r e n t samples do n o t diverge
t o a l a r g e e x t e n t t h i s phenomenon i s i n d i c a t i v e o f a r e a c t i o n which i s
l i m i t e d by zeolite
the
diffusional
r a t e o f r e a c t a n t and/or p r o d u c t molecules i n t h e
channels.
As 2,3-DMB i s a f a r more b u l k i e r molecule t h a n i t s c r a c k i n g pro duc t s i t i s presumed t h a t t h e r e a c t i o n i s l i m i t e d by r e a c t a n t d i f f u s i o n . The t h e ZSM-5 sample has t h e h i g h e s t a c t i v a t i o n energy o f w e l l over
r e a c t i o n ov er
100 kJ/mole. T h i s i s i n d i c a t i v e o f a c o n v e r sion n o t l i m i t e d by i n t r a c r y s t a l l i n e
2,3-DMB d i f f u s i o n . The
use
requires determined
o f t h e Thiele theory i n determining t h e d i f f u s i v i t y o f t h e reactant knowledge o f t h e i n t r i n s i c a c t i v i t i e s o f t h e z e o l i t e samples. We have intrinsic
activities
by
employing
a reference reaction, viz. n-
butane c r a c k i n g . T h i s r e a c t i o n i s expected t o occur w i t h o u t any i n t e r f e r e n c e o f diffusion
i n t h e ZSM-5 c r y s t a l s because t h e k i n e t i c d i a m e t e r o f n-but ane ( 0 . 4 9
nm) i s l o w e r t h a n t h a t o f 2,3-DMB (0.56 nm). F i r s t o r d e r r a t e c o n s t a n t s f o r t h e n-butane c r a c k i n g r e a c t i o n over t h e H-ZSM5 sample and ov e r t h e a l u m i n a t e d s i l i c a l i t e s a r e again present ed i n t h e f orm o f an A r r h e n i u s be to
p l o t i n F i g u r e 5. From t h e r e s p e c t i v e a c t i v a t i o n e n e r g i e s i t can
concluded t h a t a l l c o n v e r s i o n s o c c u r w i t h an e f f e c t i v e n e s s f a c t o r (0) equal unity
and
that
the
cracking
rate
is
therefore
not
retarded
by
i n t r a c r y s t a l l i n e n-butane d i f f u s i o n . I f t h e c a t a l y s t p a r t i c l e i s c o n s i d e r e d t o be o f p l a t e geometry, permeable f o r reactant
molecules
i n one
direction
only,
the
effectiveness
factor
is
t h e o r e t i c a l l y g i v e n by:
4 i s t h e T h i e l e modulus:
R i s equal t o h a l f t h e b r e a d t h o f a twinned c r y s t a l . The a d s o r p t i o n c o n s t a n t Kc is incorporated i n o r d e r t o r e l a t e t h e d i f f u s i v i t y t o t h e adsorbat e concentration
inside
the
zeolite
c r y s t a l s . kintr
can be c a l c u l a t e d f rom t h e
r e s p e c t i v e apparent f i r s t o r d e r r a t e c o n s t a n t s u s i n g t h e n-but ane r e a c t i o n da t a . I t i s assumed t h a t t h e c r a c k i n g o f 2,3-DMB proceeds u n r e t a r d e d over t h e H-ZSM-5 sample.
474
of
I n Figure 6 T h i e l e p l o t s are given. I t can be seen t h a t a p e r f e c t f i t r e s u l t s evaluated data w i t h theory (eq. 1) which i s a marked improvement compared
with
results
obtained
improved homogeneity
e a r l i e r using d i r e c t l y synthesised ZSM-5 (3). Thus the of
the
distribution
of
active
sites
throughout the
aluminated c r y s t a l s now j u s t i f i e s t h e use o f t h e simple T h i e l e theory. Thiele
i n an e f f e c t i v e d i f f u s i v i t y times a dimensionless Deff*Kc. I n order t o c a l c u l a t e t h e t r u e Deff the adsorption constant, based on the 2,3-DMB concentration i n a z e o l i t e c r y s t a l , has t o be determined. Two types o f adsorption experiments have been performed. A t temperatures below 523 K the volumetric method was used. Above 523 K the adsorption constant was determined gaschromatographically from n e t t o r e t e n t i o n times using conventional r e l a t i o n s h i p s . I n both experiments sample Si13 was used. I t was assumed t h a t adsorption c h a r a c t e r i s t i c s o f s i l i c a l i t e do n o t d i f f e r s i g n i f i c a n t l y from t h a t o f the aluminated samples used i n c a t a l y t i c experiments. The r e s u l t s showed t h a t a good c o r r e l a t i o n e x i s t s between the experimental data from both methods, and t h e Van’t H o f f equation: adsorption
analysis
resulted
constant,
e.g.
Kc=Ko*exp( -AH/RT) with
the
pre-exponential
enthalpy, (-AH),
factor,
(3) KO,
equal
t o 1 .8*10-4 and the adsorption
equal t o 57 kJ/mol.
TABLE 4 Reaction l i m i t i n g d i f f u s i v i t i e s o f 2,3-DMB i n aluminated s i l i c a l i t e .
673 7 23 748 773
2.8*10 7. 9.o*10-l2 1 .3*10-11
4.8 2.4
1.7 1.3
5. 8 * d 3 3.0*10-12 5.3*10-12 1.0*10 -
Table 4 shows the r e s u l t s o f estimations o f 2,3-DMB d i f f u s i v i t i e s a t d i f f e r e n t temperatures. I n Fig. 7 an Arrhenius p l o t i s given i n which both adsorptive and ’ c a t a l y t i c ’ 2,3-DMB d i f f u s i v i t i e s are p l o t t e d i n dealuminated and aluminated s i l i c a 1 i t e , r e s p e c t i v e l y . The c o r r e l a t i o n a f t e r e x t r a p o l a t i o n o f
475
1=723 K
-4
k. s-'
m2/s
D,~'K, = 7.3.
1.00
0.10 P
1=773 K
0.0 1 1.2
1.6
1.4 lO'/T,
DeffK, = 1 . 2 6 lo-'' m 2 / s
K-'
Fig. 5: A s Fig. 4 for n-butane
cracking. 0
Fig. 6: Effectiveness f a c t o r as a function of Thiele modulus. 2.3-dimethylbutane cracklng at 7 2 3 K and a t 773 K.
Def f
..
10-14
10-15
1
I -
1.00
1.45
1.90
2.35
2.80
10'/T,
K-'
1.2
1.4
1.6
10-'/T,
K-'
Fig. 7: Adsorptive ( m ) ond 'catalytic'( 0 ) dlffuslvltles of 2,3-dlmethylbulane In (olumlnoted) slllcallte.
Fig. 8: As Fig. 4 f o r 3-methylpentane
cracklng.
476
a d s o r p t i v e d i f f u s i v i t i e s t o l o w e r r e c i p r o c a l t emperat ures i s w i t h i n an o r d e r o f magnitude. However, t h e a c t i v a t i o n e n e r g i e s o f d i f f u s i o n , ED, d e v i a t e , v i z . 121 kJ/mole f o r t h e ' c a t a l y t i c ' and 52 kJ/mole f o r t h e a d s o r p t i v e d i f f u s i v i t y .
I n o r d e r t o f i n d o u t whether t h i s d e v i a t i o n o r i g i n a t e s f r o m t h e p r o p e r t i e s o f t h e r e a c t a n t molecule, 2,3-DMB, o r t h e c a t a l y s t , we have repeat ed t h e c r a c k i n g experiments u s i n g 3-methylpentane. U s i n g d i r e c t l y s y n t h e s i s e d l a r g e ZSM-5 crystals
we
have determined d i f f u s i v i t i e s a t 723 K and a t 811 K which, a f t e r
extrapolation,
corresponded
satisfactorily with
1 ow-temperature
adsorptive
d i f f u s i v i t i e s (3). An
Arrhenius
cracking,
the
plot
of
t h e r e s u l t s ( F i g . 8) shows t h a t , c o n t r a r y t o 2,3-DMB
3-methylpentane
conversion
is
diffusion
l i m i t e d o v e r sample
same method as i n t h e 2,3-DMB case t h e ' c a t a l y t i c ' d i f f u s i v i t y o f 3-methylpentane i n a l u m i n a t ed s i l i c a l i t e has been determined. A d s o r p t i o n d a t a r e p o r t e d e a r l i e r ( 3 ) have been used t o c a l c u l a t e t h e t r u e Deff S ilA 12
only.
from Deff*Kc.
Using
the
I n Table 5 t h e r e s u l t s o f t h e e s t i m a t i o n s a r e compared w i t h d a t a
TABLE 5 Reac t io n l i m i t i n g d i f f u s i v i t i e s o f 3-methylpentane i n aluminat ed s i l i c a l i t e and i n ZSM-5. Data on ZSM-5 and a d s o r p t i o n c o n s t a n t s a r e t a k e n f r o m r e f . 3.
A1 uminated
723 773
1.0*10- lo 1.4*10-1°
4.2 2.3
2 . 4*10-11
Silica1 i t e
81 1
2.3*10-1°
1. 5
1. 5*10-1°
723
5.6*10- lo
4.2
1. 3*10-1°
811
4.6*10- lo
1.5
3. 1 * 1 0 - l o
ZSM-5
on
3-methylpentane
that
correspondence
diffusion exists
6. 1*10-l1
i n d i r e c t l y s y n t h e s i s e d ZSM-5 ( 3 ) . I t i s c l e a r within
an
order
of
magnitude.
However,
the
r e s p e c t i v e a c t i v a t i o n e n e r g i e s o f d i f f u s i o n a r e h i g h l y d e v i a t e , v i z . 48 kJ/mole versus 100 kJ/mole. I t can
diffusion
be
concluded t h a t t h e l a r g e d i s c r e p a n c i e s i n a c t i v a t i o n e n e r g i e s o f
between d i r e c t l y s y n t h e s i s e d ZSM-5 and aluminat ed s i l i c a l i t e a r e due
471
t o t h e c a t a l y s t . I n f a c t , a discrepancy e x i s t s because t h e a c t i v a t i o n energy o f the
cracking r e a c t i o n over
enough that
to of
result the
aluminated s i l i c a l i t e does n o t decrease sharply
i n an a c t i v a t i o n energy o f d i f f u s i o n which i s comparable t o
adsorptive
diffusivity.
This
is
an important d i f f e r e n c e w i t h
r e s u l t s obtained by Voogd e t a l . ( 3 ) on ZSM-5 samples where a c t i v a t i o n energies of
r e a c t i o n on
the
largest
crystals
amounted
t o about h a l f t h e i n t r i n s i c
a c t i v a t i o n energy. This d i f f e r e n c e i n c a t a l y s t behaviour between ZSM-5 and aluminated s i l i c a l i t e i s as y e t unclear. Some
insight
features
of
into this
problem might be gained by i n v e s t i g a t i n g t h e basic
countersorption
behaviour
i n submicroporous systems. Studies on
countersorption o f l i q u i d hydrocarbons i n f a u j a s i t e type z e o l i t e s and i n mordenite have been reported (12-15). The r a t e o f countersorption was observed to
be considerably
phase.
Although
lower than the sorption r a t e o f these sorbates i n the gas
experimental
difficulties
can
be
rather
tedious,
is
it
recommended t o study co- and countersorption behaviour o f gaseous adsorbates i n p e n t a s i l z e o l i t e systems. Results from experiments l i k e t h i s w i l l undoubtedly shed more l i g h t on p e c u l i a r i t i e s o f ’ c a t a l y t i c ’ d i f f u s i o n . CONCLUSIONS I n t h i s study 2,3-DMB has been chosen as probe hydrocarbon species i n d i f f u s i o n experiments w i t h (aluminated) s i l i c a l i t e c r y s t a l s . The uptake o f t h i s compound
in
adsorption
experiments
i n t r a c r y s t a l line diffusion,
not
at
373
influenced
K and a t 473 K was c o n t r o l l e d by by
external
heat
and/or
mass
t r a n s f e r resistances. Cracking experiments using
intracrystalline
aluminated s i l i c a l i t e .
diffusion o f
deposited
randomly
cracking
activity
synthesised
were performed w i t h t h e probe hydrocarbon a t 673-773
of
large crystals
through of
a
2,3-DMB.
large
aluminated
It was
silicalite silicalite
K
Reactions were l i m i t e d by found
that
crystal. equal
to
aluminium
Also that
was
an n-butane of
directly
the
( o f comparable s i l i c o n - t o - a l u m i n i u m r a t i o ) was found. As a c t i v e s i t e s were d i s t r i b u t e d homogeneously throughout t h e z e o l i t e c r y s t a l
the
rates o f
crystal.
H-ZSM-5
diffusion
Therefore
an
and o f r e a c t i o n were equal a t any l o c a t i o n w i t h i n a important
condition
for
using Thiele’s model i n i t s
simple form was f u l f i l l e d . ’ C a t a l y t i c ’ d i f f u s i v i t i e s were estimated between 673 K and 713
K.
478
I t was found t h a t a l a r g e discrepancy e x i s t s between t h e a c t i v a t i o n energy o f
o f 2,3-DMB determined by adsorption k i n e t i c s and by performing cracking experiments. Experiments w i t h 3-methylpentane l e a d t o t h e conclusion t h a t t h i s discrepancy i s caused by t h e a p p l i e d c a t a l y s t r a t h e r than t h e probe
diffusion
molecule.
ACKNOWLEDGEMENTS The authors w u l d l i k e t o thank M r . E.J.A. van Dam, M r . J.F. van Lent and D r . W.G. S l o o f o f t h e Laboratory o f M e t a l l u r g y f o r Electron-Probe Microanalysis, SEM and X-ray d f f r a c t i o n analyses as w e l l as f o r discussions. M r . J. Padmos i s thanked f o r the AAS analyses. REFERENCES 1. W.O. Haag,
R.M.
Lago and
P.B.
Weisz,
Faraday Discuss. Chem. SOC., 72
(1981) 317. Post, J. van Amstel and H.W. Kouwenhoven, i n "Proceedings, 6 t h I n t e r n a t i o n a l Z e o l i t e Conference" (D.H. Olson and A. B i s i o , Eds.), p. 517. Butterworths, Guildford, 1983. P. Voogd and H. van Bekkum, Appl. Catal. 59 (1990) 311. D.B. Shah, D.T. Hayhurst, G. Evanina and C.J. Guo, AIChE J. 34 (1988) 1713. M. Ghamami and L.B. Sand, Z e o l i t e s , 3 (1983) 155. R.W. Grow and E.M. Flanigen, U.S. Patent 4,061,724, (1977). C.A. Fyfe, J.H. O'Brien and H. S t r o b l , Nature 326 (1987) 281. R.J. Argauer and G.R. Landolt, U.S. Patent 3,702,886, (1972). K. Yamagishi, S. Namba and T. Yashima, J. Catal., 121 (1990) 47. P. Voogd and H. van Bekkum, i n "Proceedings o f an I n t e r n a t i o n a l Symposium on Z e o l i t e s as Catalysts, Sorbents and Detergent B u i l d e r s " , (H.G. Karge and J. Weitkamp, Eds.), p. 519, E l s e v i e r , Amsterdam, 1989. C.D. Chang, C.T.-W. Chu, J.N. Miale, R.F. Bridget- and R.B. C a l v e r t , J. Am. Chem. SOC., 106 (1984) 8143. C.N. S a t t e r f i e l d , J.R. Katzer and W.R. Vieth, Ind. Eng. Chem., Fundam., 10
2. M.F.M: 3. 4. 5. 6. 7. 8. 9. 10.
11. 12.
(1971) 478. 13. R.M. Moore and J.R. Katzer, AIChE J., 18 (1972) 816. 14. C.N. S a t t e r f i e l d and J.R. Katzer, Adv. Chem. Ser., 102 (1971) 193. 15. C.N. S a t t e r f i e l d and C.S. Cheng, AIChE J., 18 (1972) 724.
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
479
T R A N S P O R T PHENOMENA AND R E A C T I O N S I N 13X T Y P E ZEOLITES
M.
BULOWl, W .
HILGERT2 a n d G .
EMIG3
' C e n t r a l I n s t i t u t e o f P h y s i c a l C h e m i s t r y , Academy o f S c i e n c e s o f t h e G.D.R., B e r l i n - A d l e r s h o f , DOR-1199 (German Dem. R e p . ) 2 1 n s t i t u t e o f T e c h n i c a l C h e m i s t r y I,U n i v e r s i t y o f E r l a n g e n - N u r e m b e r g , E g e r l a n d s t r . 3, 0 - 8 5 2 0 E r l a n g e n ( F e d . Rep. o f Germany) 3 1 n s t i t u t e o f Chemical Technology, U n i v e r s i t y o f K a r l s r u h e , K a i s e r s t r a a e 1 2 , 0 - 7 5 0 0 K a r l s r u h e 1 ( F e d . Rep. o f Germany)
SUMMARY I n f o r m a t i o n a b o u t Inass t r a n s f e r p r o c e s s e s d u r i n g h e t e r o g e n e o u s c a t a l y t i c r e a c t i o n s on m i c r o p o r o u s s o l i d s h a s become o f s p e c i a l i m p o r t a n c e . To i n v e s t i g a t e b o t h s o r p t i o n a n d k i n e t i c s o f t r a n s p o r t and r e a c t i o n o f e d u c t s and p r o d u c t s , e s p e c i a l l y o f t h e d i s p r o p o r t i o n a t i o n o f e t h y l b e n z e n e o n 13X z e o l i t e s , a h i g h p r e s s u r e a p p a r a t u s with a B e r t y r e c y c l e r e a c t o r i s described. D i f f u s i o n c o e f f i c i e n t s o f a r o m a t i c h y d r o c a r b o n s i n 13X t y p e z e o l i t e s a r e m e a s u r e d as dependences o n t e m p e r a t u r e and c o n c e n t r a t i o n i n t h e n o n - r e a c t i o n s o r p t i o n r e g i o n b y means o f b o t h t h e h i g h - p r e s s u r e c h r o m a t o g r a p h i c t e c h n i q u e and t h e p i e z o m e t r i c method. I n p a r t i c u l a r , a comparative study o f the transport parameters f o r the p - d i e t h y l b e n z e n e / l 3 X z e o l i t e s y s t e m i s c a r r i e d o u t . The d a t a o b t a i n e d a r e i n g o o d a g r e e m e n t , e s p e c i a l l y i n t h e h i g h t e m p e r a t u r e r e g i o n . They complement each o t h e r w i t h r e s p e c t t o t e m p e r a t u r e and c o n c e n t r a tion. INTRODUCTION I n t h i s work i t i s a t t e m p t e d t o d e t e r m i n e t h e c o e f f i c i e n t s o f
d i f f u s i o n a n d s o r p t i o n o f v a r i o u s a r o m a t i c h y d r o c a r b o n s o n 13X t y p e z e o l i t e s b y means o f t h e h i g h p r e s s u r e c h r o m a t o g r a p h i c ( G C ) and t h e s o r p t i o n u p t a k e (SU)
methods i n o r d e r t o a p p l y t h e r e s u l t s
t o t h e m o d e l l i n g and i n t e r p r e t a t i o n o f c o n t i n u o u s d i s p r o p o r t i o n a t i o n o f e t h y l b e n z e n e on a p p r o p r i a t e m o l e c u l a r s i e v e ( M S ) lysts. (373
I n t h e GC (SU)
... 563
s t u d y t h e t e m p e r a t u r e r a n g e i s 373
cata-
...6 8 3
K
K ) a n d t h e p r e s s u r e r a n g e s f r o m v a l u e s less t h a n 1 T o r r
i n vacuo (SU)
t o a b o u t 140 b a r ( G C ) .
The t r a n s a l k y l a t i o n r e a c t i o n
w i l l b e p e r f o r m e d w i t h i n t h e same t e m p e r a t u r e r a n g e , b u t p r e s s u -
res t o b e u s e d w i l l b e h i g h e r t h a n 400 b a r i n o r d e r t o e x c e e d t h e Besides the s o r p t i o n study, p r e l i m i n a r y explo-
c r i t i c a l point.
r a t i o n s a r e needed t o f i n d a most a c t i v e and s t a b l e c a t a l y s t f o r the disproportionation reaction.
480
EXPERIMENTAL The e x p e r i m e n t a l t e c h n i q u e s t o m e a s u r e m a s s f l u x e s o f s o r b i n g s p e c i e s i n m i c r o p o r o u s c a t a l y s t s , e . g . MS c r y s t a l s , a r e c o n f r o n t e d with t h e f o l l o w i n g problems:
-
n o n - i s o t h e r m i c i t y of t h e s o r p t i o n p r o c e s s ;
- v e r y s m a l l p a r t i c l e (MS c r y s t a l ) s i z e s ; - i n t e r p l a y between i n t r a c r y s t a l l i n e and i n t e r c r y s t a l l i n e t r a n s -
p o r t f o r p e l l e t s o r p a c k i n g s o f MS c r y s t a l s . B e s i d e s t h e c l a s s i c a l SU m e t h o d , t h e G C e x p e r i m e n t s h a v e e v o l v e d a s a major source f o r r e l i a b l e diffusionhon-equi1ibrium)data f o r MS s y s t e m s . I n a d d i t i o n , t h e N M R t e c h n i q u e s h a v e e x t e n s i v e l y b e e n u t i l i z e d t o determine s e l f - d i f f u s i o n c o e f f i c i e n t s (equilibrium d a t a ) . U n f o r t u n a t e l y , one f r e q u e n t l y e n c o u n t e r s l a r g e d i s c r e p a n cies between v a r i o u s t y p e s of e x p e r i m e n t a l r a t e p a r a m e t e r s , e s p e c i a l l y f o r h i g h l y m o b i l e systems, e . g . f o r h y d r o c a r b o n s i n 13X t y p e z e o l i t e s . O f t e n t h i s f l a w i s b e i n g a t t r i b u t e d t o u n s a t i s f a c t o r y d e s i g n of experiments or improper measurement c o n d i t i o n s , c f . t h e reviews ( r e f s . 1-3). Therefore, i n t h i s investigat i o n t h e i n f o r m a t i o n o n m o l e c u l a r t r a n s p o r t i s o b t a i n e d by t w o i n d e p e n d e n t e x p e r i m e n t a l m e t h o d s , i . e . t h e GC a n d t h e SU techniques. Chromatographic method I f o n e uses u n p e l l e t i z e d MS c r y s t a l s a s s o r b e n t , t h e d i s t r i b u t i o n o f c o n c e n t r a t i o n of a s o r b i n g m o l e c u l a r s p e c i e s b e t w e e n t h e g a s e o u s a n d s o r p t i o n ( p a r t i c l e ) p h a s e s c a n b e d e s c r i b e d by t h e following equations: gaseous phase:
with
p a r t i c l e phase:
481
The required measurements of transport rate without any chemical reaction were performed by the experimental setup shown in Fig. 1 schematically. The GC column is filled with zeolite Fig. 1. Flow diagram for GC measurements at high pressures.
bl
chrornslopraphlc column
' U nmpllllsr
- --
I
hsnlsd llnes
recorder
crystals (without binder) diluted by inert material to reduce the retention times in actual GC experiments. This arrangement ensures a more uniform gas flow than in the case of the Zero Length Column method (ref. 4). The carrier gas velocity was always high enough to guarantee a quasi-isothermic sorption-diffusion process and to exclude limiting external transport resistance at the particle-gas interface. In addition, the influence of axial dispersion on chromatograms is eliminated by high gas velocities. The pressure drop along the column axis is accounted for by the transport model of the column (Blake-Kozeny equation). A very similar model for GC experiments with zeolite crystals was proposed by Chiang et al. (ref. 5). It is most essential for further use o f the experimental data that the concentration dependence of the coefficients of diffusion and sorption is determined. For this purpose, the carrier gas nitrogen was loaded with the corresponding aromatic in a saturator and a peak is injected on top of the elevated base-line. All measurements were performed at different temperatures. Therefore, it i s possible to calculate both the activation energy and sorption heat for the systems under study. Fig. 2 shows the dependences of diffusion coefficients on temperature for various
482 a r o m a t i c h y d r o c a r b o n s on NaX t y p e z e o l i t e ( c r y s t a l s i z e 1...2
pm).
The d i f f u s i o n c o e f f i c i e n t s a r e t a k e n f r o m t h e r e g i o n w h e r e t h e y F i g . 2. A r r h e n i u s p l o t s f o r v a r i o u s systems aromatic hydrocarbon/ NaX t y p e z e o l i t e (values 0 obtained by t h e GC t e c h n i q u e ; Do... c o r r e c t e d by t h e Darken eqn. benzene ( E / k J mol-1; 0 5 ) ; A t o lu8ne ( 6 9 ) ; O e t h y l b e n z e n e (72); a m - x y l e n e (81).
1.5
1.6 1.7 1 / T [lo3 K - ' I
1.8
remain unchanged w i t h c h a n g i n g c o n c e n t r a t i o n a f t e r a p p l y i n g t h e D a r k e n e q u a t i o n t o t h e p r i m a r y d a t a . An e x a m p l e o f t h e c o n c e n t r a t i o n dependence o f d i f f u s i o n c o e f f i c i e n t s i s g i v e n i n F i g . 3 f o r t h e p-diethylbenzene/NaX
z e o l i t e system.
U
0)
ul N \
-5
Y
10-8 A
c
.-aJ .-U
Y .c -
A
g U C
.-0 ul 3
Y-
'c
u ,040
1
2
3
I
concentration [ m o l e c u l e s / cage]
F i g . 3 . C o n c e n t r a t i o n dependence o f t h e d i f f u s i o n c o e f f i c i e n t o f p - d i e t h y l b e n z e n e o n NaX z e o l i t e a t 6 9 3 K ( o b t a i n e d by t h e GC t e c h n i q u e ) .
483
A second apparatus was built in order to
- decide if the results obtained by the GC technique are transferable to the disproportionation of ethylbenzene (ref. 6 ) ; - derive a kinetic model f o r prediction o f selectivities and conversions on various MS catalysts at different temperatures, pressures and residence times. The corresponding experimental setup is shown in Fig. 4. I n the recycle reactor, pressures up to 400 bar and temperatures up to 770 K can be maintained, so that measurements at the critical
--
I-7-
,
-
I
t
Fig. 4. Flow diagram of the experimental setup f o r kinetic measurements in the supercritical region, point and in the supercritical region are feasible. The experimental principle prevents pulsations in both the gaseous and the liquid phases. First investigations indicate a lower coking tendency at supercritical conditions (refs. 7 - 9 ) . Fig. 5 shows an Arrhenius plot of diffusion data (corrected by the Darken equation) f o r the p-diethylbenzene/NaX zeolite system investigated by both
484
t h e G C ( c r y s t a l s i z e 1...2 s i z e 8 0 pm) m e t h o d s . e a c h o t h e r on t h e
pm) a n d t h e p i e z o m e t r i c S U ( c r y s t a l
The r e s u l t s f r o m b o t h t e c h n i q u e s c o m p l e m e n t
temoerature scale. F i g . 5 . Compar i s o n o f GC ( c i r c l e s ) and SU ( s q u a r e s ) d i f f u s i o n data f o r p-diethylbenzene on NaX t y p e z e o l i tes.
S o r p t i o n method I n t h e SU e x p e r i m e n t s a c o n s t a n t v o l u m e - v a r i a b l e p r e s s u r e system w i t h v a l v e - e f f e c t was e m p l o y e d .
c o r r e c t i o n s as d e s c r i b e d i n ( r e f s .
10,ll)
The v a l v e - e f f e c t c o r r e c t i o n s r e l a t e t o d e a d - t i m e
moments o f t h e a p p a r a t u s a m o u n t i n g t o s 0.2 s .
To a v o i d t h e i n -
f l u e n c e o f i n t e r c r y s t a l l i n e d i f f u s i o n on t h e u p t a k e r a t e and t o r e a l i z e quasi-isothermic conditions,
SU was m e a s u r e d u s i n g ,
r e s p e c t i v e l y , a monolayer o f s i n g l e c r y s t a l s and f a v o u r a b l e e x t e r n a l temperature conditions ( i . e . ,
To<
T v where t h e s u b s c r i p t s
o and v i n d i c a t e t h e dosed gas and z e o l i t e c r y s t a l phases,
tively)
(ref.
12). To e v a l u a t e d i f f u s i o n c o e f f i c i e n t s ,
respec-
the for-
m u l a s o f t h e s t a t i s t i c a l moments t h e o r y a s g i v e n p r e v i o u s l y ( r e f .
1 0 ) a r e u s e d . As i n t h e c a s e o f t h e GC d i f f u s i o n d a t a ,
the diffu-
s i o n c o e f f i c i e n t s t h u s d e r i v e d a r e t r e a t e d by t h e Darken e q u a t i o n (ref.
3).
I n t h e p i e z o m e t r i c SU e x p e r i m e n t s t h r e e d i f f e r e n t l y s i z e d samples o f l a b o r a t o r y - s y n t h e s i z e d NaX
...
6 0 pm;
8 0 pm; NaKX
...
13X z e o l i t e a r e u s e d :
1 2 0 pm,
80% K+.
The s i z e p a r a m e t e r s
given denote the diameter o f spheres c i r c u m s c r i b i n g t h e c r y s t a l s .
485
The standard deviations of crystal sizes are l e s s than 10% for each sample. Therefore, no size distribution function is needed for calculating the diffusion coefficients. The use of NaKX zeolite in S U experiments is due to its larger crystal size compared to that o f the NaX spezimens. Thus, SU measurements at temperatures approaching those of the GC investigation became possible. By the above experimental conditions it is presumed that the diffusion coefficients characterize the intracrystalline molecular mobility of p-diethylbenzene in 13X zeolite. A final decision can only be done by their comparison with the independently determined GC data and by variation of crystal sizes in the investigation. RESULTS Fig. 6 shows concentration dependences of diffusion coefficients for p-diethylbenzene on both NaX and NaKX in the temperature region 323 . . . 373 K. The following main results are evident: - the diffusion coefficients coincide for both 13X specimens(with
different crystal sizes) except the data for lower loadings at higher temperature, whereat the mobility o f p-diethylbenzene in the NaX crystals seems to be lowered; - the molecular mobility decreases strongly with increasing concentration in the intraccystalline void volume (this result corresponds with the behaviour of other hydrocarbon/NaX type systems (refs. 1 1 , 1 3 , 1 4 ) ) . To provide further information on the behaviour of the p-diethylbenzene/NaX zeolite system at low concentration, a treatment of corresponding SU rate curves by means of the computer programme Z E U S (ref. 15) was carried out. An example is given in Fig. 7. At least the conclusion can be drawn that with sufficiently low concentration the molecular mobility of p-diethylbenzene in NaX zeolite decreases most probably. This finding should be connected with the comparatively strong interaction between aromatic molecules and Na' cations. On the other hand, the decrease of diffusion coefficients in the high concentration region can be explained by increasing mutual interaction of molecules sorbed and by a reduced free volume in the micropores (ref. 1 3 ) . As follows from Fig. 8 wherein a r e given the concentration dependences o f diffusion coefficients of p-diethylbenzene on NaKX zeolite a t various tempe-
486
F i g . 6. Comparison o f c o n c e n t r a t i o n dependences o f d i f f u s i o n c o e f f i c i e n t s o f p-diethylbenz e n e o n NaX a n d NaKX type zeolites a t various temperatures (SU experiments).
1 o.8
ro-'O
0.1
1.0
concentration [rnmol/g]
r a t u r e s up t o 5 6 3 K,
t h i s e f f e c t becomes s u p p r e s s e d a t h i g h t e m p e -
r a t u r e s due t o i n c r e a s i n g k i n e t i c e n e r g y o f m o l e c u l e s s o r b e d i n F i g . 7. S i mulation o f a SU c u r v e o f p-diethylbenzen e o n NaX zeolite at 3 7 3 K and 0.04 -1 mmol g ( + ) by t h e software ZEUS ( s o l i d line).
487
the high concentration range. In addition to above conclusions, the following findings are significant: - at temperatures above
560 K the diffusion coefficient remains nearly constant within a wide concentration range; - the piezometric SU rate data become comparable with the high pressure GC data.
The latter conclusion can be drawn especially from the Arrhenius plot at high concentration ( n 2 0.8 mmol 9-l) shown in Fig. 9 .
lo-
r
Fig. 8. Concentration dependences of diffusion coefficients of p-diethylbenzene on NaKX zeolite at various temperatures.
03 1.o concentration [mmol/g]
The difference in diffusion coefficients from SU and GC experi-
ments amounts to a half order of magnitude, only. Taking into account well-known problems connected with such comparisons (ref. 3 ) and the presence of K + cations in one of the zeolites considered, this result is quite satisfactory. The activation energy derived from the SU data amounts to 40 kJ mol-l, a value indicating the intracrystalline character of the underlying molecular transport. It exceeds the well-established value for benzene in NaX zeolite amounting t o % 2 5 kJ mol-' (ref. 14) due to the
488
Fig. 9 . Arrhenius plot of GC (empty symbols) and SU (full symbols) data for p-diethylbenzene on NaX and NaKX type zeolites, r e spect ively.
10-'O
E,
N
at
n
40 ItJ/rnol
N
0.8 rnrnol/g
F
more complex molecular structure of p-diethylbenzene. If one takes into account that the GC measurements were carried out in a temperature region where the diffusion coefficients r e main practically constant over a wide concentration region and, on the other hand, this result is also confirmed by the SU measurements, a good agreement can be stated for the data from both methods. The SU method complements the GC technique at low temperature contributing information about an increasing influence of concentration on intracrystalline molecular mobility. The low diffusion coefficients of p-diethylbenzene at very low micropore fillings of the NaX zeolite - as shown by the GC technique - is suggested by the SU results as well. The molecular mobility data described will serve as reliable information on intracrystalline transport for further consideration of reaction kinetics. CONCLUSIONS The principle of an experimental high pressure chromatographic system with a Berty recycle reactor to study sorption,diffusion and reaction on columns with crystals of molecular sieves is described.
489
By means of this technique at pre-catalytic conditions, coefficients of sorption and diffusion of benzene, ethylbenzene and diethylbenzene isomers on NaX crystals with different sizes are obtained as dependences on concentration approaching values up to 4 molecules per supercage and on temperature as well. A comparative study of diffusion of p-diethylbenzene on various 13X molecular sieves by means of the piezometric sorption uptake method leads to good agreement between diffusivities obtained by both this technique and the high pressure G C method, especially in the high temperature region. The data derived from the measurements by both methods complement each other with respect to temperature and concentration. The diffusion coefficients of p-diethylbenzene in NaX and NaKX zeolites characterize the molecular mobility in the intracrystalline zeolitic void volume. The comparison of intracrystalline molecular mobility data reported here with those from literature for other aromatics in 13X type zeolites shows the sequence: benzene > p-xylene > p-diethylbenzene. ACKNOWLEOGEMENTS W.H. and G . E . thank the Oeutsche Forschungsgemeinschaft for financial support. M.B. thanks the Akademie der Wissenschaften der D D R for the possibility to work on diffusion phenomena in molecular sieve systems. NOTATIONS diffusion coefficient coefficient of axial dispersion ?3x constant of sorption equilibrium K column length L mass flux density across the NRc zeolite crystal interface (cf.eqn.2) external r a d i u s of zeolite crystals RC C gaseous phase concentration 9 total sorbate concentration in CT gaseous phase = Kc , sorbate concentration in C intrgcrystalline void volume o f zeolite r radial coordinate of zeolite crystal time t D
*
cm m m o l cm
-2 s -1
cm mmol cm-3 mmol cm-3 mmol cm-3 cm S
490 V
l i n e a r gas v e l o c i t y
cm s
X
molar f r a c t i o n
'
i n t e r p a r t i c l e v o i d volume of column p a c k i n g (volume f r a c t i o n )
-
7
= r/Rc,
5
dimensionless a x i a l coordinate
of
dimensionless radius zeolite crystal
-1
-
-
REFERENCES 1 2 3
4 5 6 7
8 9
10
11 12 13 14 15
J . K a r g e r a n d D.M. R u t h v e n , Z e o l i t e s , 9 ( 1 9 8 9 ) 2 6 7 - 2 8 1 . H.W. H a y n e s , J r . , C a t a l . Rev. - S c i . E n g . , 30 ( 1 9 8 8 ) 5 6 3 - 6 2 7 . M. B u l o w , J . K a r g e r , M . K o z i r f k a n d A . M . V o l o g z u k , Z . Chem., L e i p z i g , 2 1 (1981) 175-182. D.M. Ruthven and M. E i c , Z e o l i t e s , 8 (1988) 40-45. A . S . C h i a n g , A . G . D i x o n a n d Y . M . Ma, Chem. Eng. S c i . , 3 9 (1984) 1451-1459. W.J. Thomas a n d U . U l l a h , Chem. Eng. Res. D e v . , 6 6 ( 1 9 8 8 ) 138146. N . A . C o l l i n s , P . G . O e b e n e d e t t i a n d 5. S u n d a r e s a n , A I C h E J . , 3 4 (1988) 1211-1214. G . Manos, Ph. 0 . T h e s i s " U n t e r s u c h u n g e n d e r V e r k o k u n g v o n Z e o l i t h e n und i h r e r R e a k t i v i e r u n g m i t u b e r k r i t i s c h e n F l u i d e n " U n i v e r s i t a t E r l a n g e n - Nurnberg, 1989. H . S c h a f e r , Ph. 0 . T h e s i s " U n t e r s u c h u n g e n z u r K i n e t i k d e r Desa k t i v i e r u n g und R e a k t i v i e r u n g p o r o s e r K o n t a k t e m i t u n t e r - bzw. u b e r k r i t i s c h e n R e a k t i o n s g e m i s c h e n " U n i v e r s i t a t E r l a n g e n - Nurnb e r g , 1990. P . S t r u v e , M . KoEi??
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
491
THE EFFECT OF VARIOUS PHYSICAL AND CHEMICAL PARAMETERS ON THE SYNTHESIS OF ZSM-5 FOR PROPENE OLIGOMERIZATION C.T. O’Connor,
S. Schwarz and M. Kojima
Department o f Chemical Engineering, U n i v e r s i t y o f Cape Town, P r i v a t e Bag, Rondebosch 7700, SOUTH AFRICA
ABSTRACT I n t h e p r e s e n t s t u d y i t has been shown t h a t t h e propene igomerization a c t i v i t y o f ZSM-5 a t 5MPa and 220-290°C i s m a i n l y dependent on t h e aluminium c o n t e n t and c r y s t a l s i z e o f t h e c a t a l y s t . Optimum a c t i v i t i e s and l i f e t i m e s were o b t a i n e d w i t h Si/A1 = 40, c r y s t a l s i z e < lpm and a s y n t h e s i s t i m e o f 3 days. T his y i e l d e d under optimum r e a c t i o n c o n d i t i o n s a c a t a l y s t u t i l i z a t i o n v a l u e o f 1100 g.liq/g.cat. E x t r u d i n g t h e c a t a l y s t u s i n g alumina as a b i n d e r d i d n o t a d v e r s e l y affect the activity, selectivity or lifetime o f the catalyst. Pelletizing reduced t h e a c t i v i t y o f t h e c a t a l y s t . A m e c h a n i c a l l y - s t i r r e d autoclave l e d t o a decrease i n t h e t h e average c r y s t a l s i z e and a narrower s i z e d i s t r i b u t i o n . A number o f t e mp l a t e s were used i n t h e s y n t h e s i s o f ZSM-5 and 1,6-hexanediamine and 1,2-cyclohexanediamine produced c a t a l y s t s w i t h s i m i l a r c a t a l y t i c p r o p e r t i e s t o t h o s e prepared u s i n g tetrapropylammonium bromide. INTRODUCTION ZSM-5 has been shown t o be an e x c e l l e n t c a t a l y s t f o r t h e c o n v e r s i o n o f l i g h t o l e f i n s t o l i q u i d f u e l s ( r e f s . 1 - 4 ) . I n general t h e a c t i v i t y i s r e l a t e d t o t h e A1 c o n t e n t ( r e f s .
5-7).
The c r y s t a l s i z e a l s o increases as Si/A1
ratio
inc re as es ( r e f . 8 ) . The a c t i v i t y o f v a r i o u s c a t a l y t i c r e a c t i o n s i n c r e a s e s as c r y s t a l s i z e decreases ( r e f s . 9-10). The need t o f orm e x t r u d a t e s and p e l l e t s fro m t h e powder can a l s o s t r o n g l y i n f l u e n c e t h e s e l e c t i v i t y and a c t i v i t y o f t h e c a t a l y s t ( r e f s . 11-14). Longer c r y s t a l 1 i z a t i o n t imes enhance t h e c o n c e n t r a t i o n o f A1 on t h e e x t e r n a l s u r f a c e sometimes r e s u l t i n g i n a s h o r t e r c a t a l y s t l i f e ( r e f . 15). The use o f a l t e r n a t i v e templates i n t h e s y n t h e s i s o f ZSM-5 has a l s o been e x t e n s i v e l y s t u d i e d ( r e f s . 16-20). T h i s paper p r e s e n t s r e s u l t s o b t a i n e d i n a s t u d y o f t h e e f f e c t o f S i / A l r a t i o , s y n t h e s i s t i m e , t e m p l a t e s and s t i r r i n g mechanism on t h e c h a r a c t e r i s t i c s o f ZSM-5 and on i t s a c t i v i t y , s e l e c t i v i t y and l i f e t i m e f o r t h e o l i g o m e r i z a t i o n o f propene i n a h i g h p r e s s u r e f i x e d bed r e a c t o r . Powdered c a t a l y s t samples were a l s o e x t ruded and p e l l e t i z e d u s i n g v a r i o u s b i n d e r s and t h e e f f e c t s o f t h i s t r e a t m e n t on l i f e t i m e and s e l e c t i v i t y was s t u d ied.
EXPERIMENTAL Samples o f ZSM-5 were s y n t h e s i z e d a c cording t h e method o f Argauer and L a n d o l t ( r e f . 21) u s i n g b o t h m e c h a n i c a l l y and m a g n e t i c a l l y s t i r r e d a u t o c l a v e s .
492
D e t a i l s o f the method are given elsewhere ( r e f s . 22,23).
The samples used are
r e f e r r e d t o as (A)-Z-X-Y where A = template (tetrapropylammonium bromide unless otherwise stated), Z = ZSM-5, X = S i / A l types
of
alcohols.
templates
were
used,
viz.
r a t i o and Y
The f i r s t category included methyl (TMA),
and b u t y l (TBA) ammonium bromide.
synthesis time.
=
tetra-alkylammonium
ions,
e t h y l (TEA),
amines
Three and
propyl (TPA)
The amines used were 1,2 ethanediamine (1,2
ED), 1,2 propenediamine (1,2 PD), 1 butylamine (1 BA), 1 hexylamine (1 HA), 1,6 hexanediamine (1,6 HD) and 1,2 cyclohexanediamine (1,2 CHD). A number o f alcohols were a l s o used. The c a t a l y s t s were characterized using SEM, EDX, XRD, ammonia TPD and, i n some cases, MASNMR. A l l the samples prepared were t e s t e d f o r t h e i r a c t i v i t y and s e l e c t i v i t y
f o r propene o l i g o m e r i z a t i o n and d e t a i l s o f t h e r e a c t o r system used as w e l l as the methods used t o analyse t h e r e a c t i o n products are described elsewhere ( r e f s . 22,23).
Reactions were c a r r i e d out a t 5MPa,
r e a c t i o n temperatures between 220 and 290°C.
WHSV
=
12 h-' and,
i n general,
I n some cases t h e l i q u i d product
was hydrogenated over a Pd/C c a t a l y s t and t h e 'H-NMR spectrum used t o estimate cetane numbers. The amount o f coke deposited on t h e c a t a l y s t was determined using thermogravimetri c anal ys is RESULTS All
.
samples o f c a t a l y s t prepared had X-ray d i f f r a c t i o n p a t t e r n s and
ammonia TPD spectra c h a r a c t e r i s t i c o f ZSM-5 ( r e f . 23). effect of Si/Al
ratio,
Tables 1 and 2 show the
synthesis time and s t i r r i n g procedure on the number
average c r y s t a l size.
TABLE 1 E f f e c t o f S i / A l r a t i o and synthesis t i m e on average c r y s t a l s i z e (Temp1a t e TPABr) Si/Al
18 44 64 131 197 487 48 43 45 19 26
Synthesis time (days)
1 1 1 1 1 1 3 6 10 3 6
Average c r y s t a l
s i z e (pm) 1.8 0.8 1.1 1.4 1.8 4.7 1.1 1.2 1.2 2.9 3.0
493
TABLE 2 E f f e c t o f s t i r r i n g on average c r y s t a l Size (Synthesis time = 3 days) Type o f Stirring
S i / A l Temp1 a t e
46 41 31 44 32 38 487*
TPA TPA 1,6 HD TPA 1,6 HD T PA TPA
Average c r y s t a l size (p) 2.4 0.9 0.4 1.1 1.3 0.7 4.7
none mechanical mechanical magnetic magnetic mechani cal magnetic
*
Synthesis time = 1 day 1 Catalyst u t i l i z a t i o n value = g l i q / g cat As expected the c r y s t a l s i z e i n general decreased as t h e A1 content increased. When t h e s l u r r y was mechanically s t i r r e d smaller c r y s t a l s w i t h a
r e l a t i v e l y narrower s i z e d i s t r i b u t i o n than f o r the magnetically s t i r r e d samples was obtained.
The c r y s t a l s i z e o f samples synthesized using a v a r i e t y o f
tetraalkylammonium bromides, amines and alcohols as templates was also determined. TBA, TPA, lHA, 1,6-HD and 1,2-CHD a l l produced ZSM-5 w i t h c r y s t a l s -1 pm. The other amines y i e l d e d c r y s t a l sizes o f > 3 pm and t h e alcohols c r y s t a l sizes > 4 pm. Fig. 1 shows the r e l a t i v e e f f e c t o f aluminium content on propene conversion t o l i q u i d products. An unusually h i g h r e a c t i o n temperature o f 300°C was used t o ensure t h a t 2-487-1 would show s u f f i c i e n t a c t i v i t y . When samples were prepared w i t h S i / A l r a t i o o f 48 and 64 r e s p e c t i v e l y and c r y s t a l s i z e o f -1.1 pm i n both cases, the aluminium content was more s i g n i f i c a n t than c r y s t a l s i z e f o r propene oligomerization a c t i v i t y . Fig. 2 shows t h e c a t a l y s t u t i l i z a t i o n values obtained throughout t h i s study f o r d i f f e r e n t c r y s t a l sizes i r r e s p e c t i v e o f the method o f synthesis used o r aluminium content. The c r i t i c a l importance o f the c r y s t a l size i s c l e a r l y evident f r o m t h i s f i g u r e . The best c a t a l y s t u t i l i z a t i o n values (CUV
=
g.liq/g.cat)
were obtained
using samples w i t h c r y s t a l sizes < lpm and Si/A1 = 40.
C r y s t a l s o f 2.4 pm
showed very poor oligomerization a c t i v i t y (CUV = 110 g/g).
The CUV values f o r
the 1,6 HD and TPA based samples ( m x h a n i c a l l y s t i r r e d ; Si/A1 = 40) were 409 and
494
C
100
0
n V
e r
80
0
I 0
n
80
0
I P
40
r
0
P
n e
x
0'
0
Fig. 1 MPa, T
=
2
14
16
18
E f f e c t o f A1 content on propene o l i g o m e r i z a t i o n a c t i v i t y ( P 3OO0C, WHSV = 12 h-')
0 0.5 1
Fig. 2
6 8 10 12 Time on stream (hrs)
4
1.5
2
2.5 3 3.5 4 Crystal size (urn)
4.5
5 5.5 6
E f f e c t o f c r y s t a l s i z e on c a t a l y s t u t i l i z a t i o n value.
=
5
495 385 g/g
and t h e i r magnetically
s t i r r e d equivalents were 286 and 258 g/g
r e s p e c t i v e l y . Their c r y s t a l sizes are shown i n Table 2.
A t a constant Si/A1 r a t i o o f about 40 i t was found t h a t an optimum CUV = 520 g/g was obtained f o r a synthesis t i m e o f 3 days. For a Si/A1 r a t i o o f about 20 t h e optimum CUV (380 g/g) was obtained f o r a synthesis time o f 1 day. The former optimum corresponded t o t h e highest concentration o f strong a c i d s i t e s as determined by TPD, v i z . 0.40 mmol/g, whereas the l a t t e r optimum corresponded t o the smallest c r y s t a l size. I n a l l these runs the r e a c t i o n temperature was increased g r a d u a l l y from
220
- 290°C
over t h e e n t i r e run o f about 40h and the
r e p r o d u c i b i l i t y was found t o be very good. The f o l l o w i n g general propene oligomerization a c t i v i t y sequence was observed when d i f f e r e n t templates were used: TPA, TBA, C, diamines > C, amines > Cp,3,4 amines
and diamines
> alcohols.
The f i r s t mentioned group
all
had
comparable CUVs o f -300 g/g. When samples o f ZSM-5 (TPA-Z-41-3) were extruded w i t h s i l i c a o r alumina i t was found t h a t t h e r e was no change i n the CUV o f t h e c a t a l y s t which was found t o be 400 and 375 g/g r e s p e c t i v e l y
as opposed t o 385 g/g f o r t h e powder.
When
s i l i c a - a l u m i n a was used as a binder the CUV increased from 409 t o 497 g/g.
Fig.
3 shows c l e a r l y t h a t the a c t i v i t y and l i f e t i m e o f t h e c a t a l y s t
s i g n i f i c a n t l y when p e l l e t s were used,
decreased
the CUV dropping from 380 g/g
for
extrudates o f TPA-Z-38-3 t o 60 g/g f o r 10 ton p e l l e t s o f t h e same sample. Fig. 4 shows the l i q u i d product analysis obtained from t h e o l i g o m e r i z a t i o n o f propene using powder and extrudates.
The values obtained f o r t h e powder are
representative o f those f o r almost a l l runs i n t h i s work. High r e a c t i o n temperatures l e d t o a s h i f t t o l i g h t e r products l a r g e l y because o f t h e increased e x t e n t o f cracking a t these temperatures.
This f i g u r e shows a d i s t i n c t t r e n d t o
heavier products i n the case o f the extruded c a t a l y s t independent o f t h e binder used.
Gautier's c o r r e l a t i o n ( r e f . 24) was used t o c a l c u l a t e cetane numbers from
the 'H-NMR spectra o f hydrogenated l i q u i d samples. The values obtained were i n the range 39-52 f o r both powder and extrudates. The best c a t a l y s t t e s t e d was alumina-bound extrudates o f (TPA)-Z-38-3 (mechanically s t i r r e d ) which had a CUV o f 1100 g/g and a product cetane number o f 44.
A g r a v i m e t r i c analysis o f the carbonaceous deposit on t h e c a t a l y s t a t the end o f a run showed t h a t i n the case o f extrudates bound w i t h alumina t h e t o t a l mass o f these deposits, v i z . " g r a p h i t i c " material p l u s h i g h b o i l i n g p o i n t hydrocarbons, was 15% o f the t o t a l mass o f deactivated c a t a l y s t . This was about
496 Conversion of orooene (a)
2 Ton9
5 Tons
0
5
10
20
15
25
30
35
40
45
Time on stream (hr)
Fig. 3
Effect of extrusion and pelletizing on propene oligomerization activity (P = 5 MPa, T = 250"C, WHSV = 12 h- )
40
Mass%
30
20
10
0
Dlmer
Tetramer
Trimer
Pentamer
Hexamer,
Oligomer Group Powder
@ Silica
extrudates
Fig. 4 Liquid Product Analysis
0Alumina Extrudetes
497
half that found in the case of powdered, extrudates.
silica or silica-alumina bound
DISCUSSI ON The number average crystal sizes of 2-18-1, 2-19-3 and 2-26-6 were larger than what was expected from the trend observed by Van der Gaag et a1 .(ref. 25) and Jacobs and Martens (ref. 13). For the 6-day catalyst this may be due to amorphous material being deposited on the zeolite surface. The increase in crystal size (2-19-3 and 2-26-6 larger than 2-18-1) has been observed by other workers (ref. 26). Mechanical stirring is expected to cause more attrition than a magnetic stirrer bar in a synthesis mixture and hence as observed in the present study should cause a decrease in crystal size. The narrow size distributions obtained with mechanical stirring could be a result of more effective agitation throughout the autoclave and a more even distribution of stress on the crystals. This is distinctly advantageous with respect to the propene oligomerization activity. When different templates were used it was found that 1 BA, 1,2 ED, 1,2 PD as well as the alcohol-based samples produced larger crystal sizes than TPA-ZSM-5. This may happen if the formation of the template-silicate complex is difficult resulting in the formation of only a few nuclei which then grow to a larger size. The activity for propene 01 igomerization increased with decreasing Si/A1 ratio. TPD studies indicated a high concentration of Bronsted sites (refs. 2326). Moreover little dehydroxylation appeared to have taken place. The relationship between aluminium content and activity is thus probably indicative of a relationship between activity and Bronsted acidity for propene ol igomerization. Calcination temperatures less than 500°C however do not favour optimum catalytic performance (ref. 22,23). The synthesis used in this study was similar to type ‘A’ of Gabelica et al. (ref. 27), who reported an aluminium rich outer layer for this synthesis procedure. Aluminium zoning of this nature was reported by Von Ballmoos and Meier (ref. 28). Suzuki et al. (ref. 29) also confirmed this finding and observed a decrease in activity for methanol conversion as synthesis time increased once maximum crystal1 inity was attained. They attributed the decrease in activity to increased coking on the catalyst surface. The results of the present study support these observati ons. This work has shown that a decrease in crystal size enhances lifetime and activity for propene ol igomerization as has been reported for methanol
498
conversion (ref. 3 0 ) , xylene isomerisation (ref. 31) and propene oligomer sation using N i - and Zn-impregnated ZSM-5 (ref. 3 2 ) . The improved performance of small crystals may be related to coking This has been found by Behrsing et al. (ref. 3 3 ) who compared ZSM-5 crystals of differing crystal sizes using methanol conversion. They observed that ZSM-5 with a crystal size < 1 pm had the longest lifetime for methanol conversion, even though it contained more coke than the other catalysts. The long lifetime of small ZSM-5 crystals may also be attributed to higher pore volumes, higher external surface areas and more external acid sites when compared to large crystals. Higher pore volumes will increase resistance to coking and more external acid sites ensure that a large fraction of the total acid sites are easily available for reaction. In larger crystals, with a lower ratio of external to internal surface area, many acid sites may not be accessible for reaction due to excessively long diffusional pathways. Moreover heavier hydrocarbons produced inside a large crystal may not be able to diffuse out of the crystal and will therefore block the passage of other molecules further into the crystal. This would decrease the amount of acid sites available for reaction and cause quicker deactivation than in small crystals. The low crystallinities and primarily the large crystal sizes of the alcohol-based ZSM-5 samples were probably responsible for their low activities and short lifetimes. Surprisingly extrusion did not decrease the activity for propene ol igomerization although the high conversions involved may have masked any difference in activity between powdered and extruded catalysts. Attempts to verify this observation at low conversions were difficult due to temperature runaway problems. The observed decrease in lifetime and activity with increasing pelletisation pressure is probably due to increased diffusional resistance in the intercrystalline pores and blockage of the ZSM-5 micropores by binder particles. The activity and lifetime of catalysts synthesized using different amine templates increased in the order: 1 , 2 ED = 1,2 PD < 1 BA < 1 HA (crystal size = 3 . 4 prn, 4 . 7 pm, 2 . 9 pm and 1 . 0 pm respectively). At larger crystal sizes this trend became indistinct. No reference has been made to the importance which different morphologies may have on the catalytic properties and this has been considered el sewhere (ref. 2 3 ) . Temperature control for extruded catalysts was much easier than for powdered catalysts due to the dilution effect of the binder and the void spaces between individual extrudates in the catalyst bed. Although it may have been expected that the lighter products would thus be produced by powdered ZSM-5 due to cracking at hotspot temperatures, it was observed that
499
p e l l e t s y i e l d e d t h e l i g h t e s t l i q u i d products. lo we r c o nv ers io n s .
T h i s was
p o s s i b l y due t o t h e
S i l i c a - a l u m i n a e x t r u d a t e s produced more coke t h a n t h e o t h e r e x t r u d a t e s i n v e s t i g a t e d and t h i s may be due t o t h e a c t i v i t y o f t h e b i n d e r f o r propene 01 i g o m e r i s a t i o n . T h i s i n c r e a s e d i d not, however, adversely a f f e c t t h e 1 i f e t i m e o f these extrudates. 1/2
The l o w coke c o n t e n t o f t h e alumina-bound e x t r u d a t e s (-
o f t h e coke c o n t e n t o f t h e s i l i c a e x t r u d a t e s )
did not result
i n any
d i f f e r e n c e i n l i f e t i m e o r a c t i v i t y when compared t o t h e s i l i c a e x t r u d a t e s . The o b s e r v a t i o n t h a t 1 i g h t l y coked e x t r u d a t e s d i d n o t c o n t a i n g r a p h i t i c coke, w h i l e t h e h e a v i l y coked e x t r u d a t e s d i d , seemed t o i m p l y t h a t d e a c t i v a t i o n would o n l y occ ur once h i g h q u a n t i t i e s o f g r a p h i t i c coke were deposit ed. CONCLUSION
T h i s s t udy has shown t h a t a t Si/A1 r a t i o s o f about 40, t h e c r y s t a l s i z e o f ZSM-5 has a c r i t i c a l e f f e c t on propene o l i g o m e r i z a t i o n a c t i v i t y . S m a l l e r s i z e and narrow er c r y s t a l s i z e d i s t r i b u t i o n s a r e o b t a i n e d when u s i n g m e c h a n i c a l l y s t i r r e d a ut o c lav e s .
When t h e powder i s e x t r uded t h e r e i s no adverse e f f e c t on
o l i g o m e r i z a t i o n a c t i v i t y and when alumina i s used as a b i n d e r b e t t e r q u a l i t y distillate
fuel
i s produced.
P e l l e t i z i n g t h e powder reduces a c t i v i t y
in
p r o p o r t i o n t o t h e p r e s s u r e s used. Other t e m plat es such as 1,6 hexanediamine and 1,2 cyclohexanediamine produce ZSM-5 o f s i m i l a r a c t i v i t y t o TPA-based ZSM-5 b u t o t h e r amines and a l l a l c o h o l s t e s t e d produced poor o l i g o m e r i z a t i o n c a t a l y s t s . ACKNOWLEDGEMENT
The a ut h or s w i s h t o acknowledge f i n a n c i a l support f r o m t h e U n i v e r s i t y o f Cape Town, N a t i o n a l Energy Council and Sasol
.
REFERENCES
1 2 3 4 5 6 7 8 9 10 11 12 13
S.A. Tabak, F.J. Krambeck and W.E. Garwood, J.AIChE, 32(9) (1986) 1529. R.J. Ouann. L.A. Green, S.A. Tabak and F.J. Krambeck, I n d . Eng. Chem. Res., '27 (1'988) 567. . M.L. O c c e l l i , J.T. Hsu and L.G. Galya, J. Mol. Cat al. , 32 (1985) 378. K.G. W i l s h i e r , P. Smart, R. Western, T. Mole and T. Behrsing, Appl. C at a l. . 31 119871 345. S. B e s s e l l and 0: Seddon, J. Catal., 105 (1987) 272. S.P. Zhdanov, N.N. Fe o k t i s t o v a , N . I . Kozlova, N.R. Bursian, S.B. Kogan, V.K. Daragan and N.V. Aleksandrova, Acta Phys. Chem., 31 (1985) 131. V.S. Nayak and V.R. Choudary, Appl. C a t al. , 4 (1982) 339. F.J. van d e r Gaag, J.C. Jansen and H. van Bekkum, Appl. C a t a l . , 17 (1985) 264. M. Sugimoto, H. Katsuno, K. Takatsu and N. Kawata, Z e o l i t e s , 7 (1987) 503. P. Ratnasamy, G.P. Babu, A.J. Chandwadkar and S.B. K u l a r n i , Z e o l i t e s , 6 (1986) 98. M.L. O c c e l l i , J.T. Hsu and L.G. Galya, J . Molec. Cat al. , 32 (1985) 379. U. Hammon, M. K o t t e r and L. R i e k e r t , Appl. Cat al. , 37 (1988) 155. P.A. Jacobs and J.A. Martens, S t u d i e s i n Surf ace Science and C a t a l y s i s , 33 (1987) 72.
500
15
W. Holderich, H. Eichorn, R. Lehnert, L. Marosi, W. Mross, R. Reinke, W. Ruppel and H. Schlimper, Proc. 6 t h I n t . Z e o l i t e Conf., Reno, USA, 1983, Butterworths (1984) 545-555. K. Suzuki, Y. Kiyozumi, K. Matsuzaki and S. Shin, Appl. Catal., 42 (1988)
16 17 18 19 20
N . Y . Chen, S.J. Lucki and W.E. Garwood, US Patent 3700585 (1972). S.A. Tabak, US Patent 4254295 (1981). P. Chu, US Patent 3709979 (1973) A. Araya and B.M. Lowe, Zeolites, 6 (1986) 111-118. F.J. van der Gaag, J.C. Jansen and H. van Bekkum, Appl. Catal., 17 (1985)
14
35.
261-271.
25 26 27 28
R.J. Argauer and G.R. Landolt, US Patent 3702886 (1972). S.Schwarz, M.Kojima and C.T.O’Connor, Appl. Catal.,56 (1989) 263-280, S. Schwarz, PhD Thesis, U n i v e r s i t y o f Cape Town, 1990. S . Gautier, Diplome-Ingenieur Thesis, Conservatoire National des A r t s e t Metiers, Paris (1988). N-Y. Topsoe, K. Pedersen and E.G. Derouane, J. Catal., 70, (1981), 46. Z . Gabelica, E.G. Derouane and N . Blom, Catal. Minerals, 12, (1984), 219. R. von Ballmoos and W.M. Meier, Nature (London), 289, (1981). 782. K. Suzuki, Y. Kiyozumi, K. Matsuzaki and S. Shin, Appl. Catal., 42,
29
M. Sugimoto, H. Katsuno, K. Takatsu and N. Kawata, Z e o l i t e s , 7, (1987),
21 22 23 24
(1988), 35. 503. 30
P. Ratnasamy, G.P.
Babu, A.J. Chandwadkar and S.B. K u l k a r n i , Z e o l i t e s , 6,
(19861. 98. 31 32
S.J. M i l l e r , US Patent 4423269 (1983). T. Behrsing, H. Jaeger and J.V. Sanders, Appl. Catal.,
54, (1989), 289.
G. Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 01991 Elsevier Science PublishersB.V., Amsterdam
501
ON CONTROLLED GROWTH OF SAPO-5 MOLECULAR SIEVE CRYSTALS OF DIFFERENT SIZES AND SHAPES
G. FINGER, .J. KORNATOWSKI~,J. RICHTER-MENDAU, K. JANCKE, M. RIJLOW, and M. ROZWADOWSKIl Central Institute of Physical Chemistry, Academy of Sciences of the G.D.R., Berlin (German Democratic Republic) ‘Institute of Chemistry, Nicolaus Copernicus University, Toruh (Poland)
SUMMARY Systematic studies of the SAPO-5 system have been performed. Roth the crystal size and shape as well as the phase purity can be controlled by the reacting gel composition and also by the duration and/or temperature of crystallization. Evidence is given for the interdependence between the shape of crystals and their sorption properties. INTRODUCTION Porous silicoaluminophosphates (ref. 1) are promising materials for numerous scientific and potential practical applications. However, thoroughly refined applications require materials of strictly defined physical featiires, e.g.
crystal size and shape. Unfortunately, the control of these parameters in
the case of zeolitic materials did not represent a field of major interest until now, except for the tailoring of type A zeolites for washing powder production (refs. 2 , 3 cited in ref. 4) and for the growing of MFI type crystals (refs. 5-10). Obviously, the growth of tailored crystals requires the knowledge of both the kinetics of the process and the conditions enabling the formation of larger crystals. The latter problem has been solved for both the AlPO4-5 (refs. 11-16) and the SAPO-5 molecular sieves ( r e f s . 17-19). Recently, the kinetics of SAPO-5 crystallization has been investigated in detail (ref. 2 0 ) but not with respect t o the size of the crystals. We aim at a better understanding of the SAPO-5 crystallization process and at getting control of size and shape of the crystals. The paper presents a number of results and conclusions regarding these goals. EXPERIMENTAL SAPO-5 molecular sieve crystals were synthesized hydrothennally in teflonlined steel autoclaves in air-heated w e n s at 4 5 3 to 473 K from gels of the formal molar composition
a A1203 with a
=
1, b
*
b P2O5 =
-
c TEA
1 t o 2.4, c
=
d Si02
e H20
1.5 t o 12, d
=
*
x H2SO4
0.05 t o 1 , e
(1) =
300 t o 2000, x ? 0.
F i r s t , g e l s were formed by mising a s t a b i l i z e d aluminiumoxidehydrate s o l ( p u r c h a s e d from CTA SAURESCHUTZ, B e r l i n ) w i t h s i l i c a s o l (LUDOX AS 40 from DUPONT/Germany). Next, a s o l u t j o n of 85% p h o s p h o r i c a c i d and t r i e t h y l a m i n e / T E A / (Merck) i n t w i c e d i s t i l l e d w a t e r was added under v i g o r o u s s t i r r i n g . Varioiis amounts o f H3PO4, TEA, and w a t e r were admixed a c c o r d i n g t o t h e g e l rompr~siticln s t u d i e d . A t t h e same t i m e , s u l f u r i c a c i d was added, i f n e c e s s a r y , t o a d j u s t t h e pH v a l u e t o 3 , 3 + 0 , 2 . A s e r i e s of r u n s was made a t d
=
0 t o grow AlP04-5 c r y s -
t a l s a s r e f e r e n c e samples. These g e l s were p u t i n t o t h e a u t o c l a v e s and exposed t o t h e c r y s t a l l i z a t i o n t e m p e r a t u r e . A f t e r a p p r o p r i a t e time p e r i o d s , t h e v e s s e l s were c o o l e d down, t h e samples were washed, s e p a r a t e d from b y p r o d u c t s , i f n e c e s s a r y , by d e c a n t a t i o n ( o b v i o u s l y , a number of smaller SAPO-5 c r y s t a l s were l o s t a t t h e same t j m e ) , and d r i e d a t 3 7 3 K. Furthermore, s e l e c t e d samples were c a l c i n e d under s t a t i c a i r atmosphere a t about 1050 K f o r 24 h r s . The p r o d u c t s were c h a r a c t e r i z e d by XRD ( G u i n i e r t e c h n i q u e ) , MAS NMR, I R s p e c t r o s c o p y , a d s o r p t i o n , and SEM. The c o n t e n t of c r y s t a l s i n t h e samples was e s t i m a t e d and t h e c r y s t a l dimensions were measured by l i g h t microscopy accor-di n s t o t h e methods d e s c r i b e d i n d e t a i l e l s e w h e r e ( r e f . 2 1 ) .
RESUILTS The procedure e l a b o r a t e d ( r e f s . 17-19) i s a r e l i a b l e r c c j p e € o r t h e synt h e s i s of l a r g e hexagonal SAPO-5 p r i s m s , though sometimes i n a broad p a r t i c l e s i z e d i s t - r i b u t i o n . I n comparison w i t h t h e similar p r o c e d u r e t o grow A1P04-5 m o l e c u l a r s i e v e s ( r e f s . 13-16), t h e SAPO-5 s y s t e m is more complex. However, t h e c r y s t a l s o b t a i n e d i n a l l e x p e r i m e n t s e x h i b i t X-ray p a t t e r n s c o r r e s p o n d j n g t o t h e A F I topology and MAS NMR measurements prove s i l i c o n i n c o r p o r a t i o n i n t o t h e framework. The c r y s t a l l i z a t i o n was s t u d i e d w i t h i n t h e range of t e m p e r a t u r e from 423 t o 4 7 3 K and p e r i o d s up t o 10 days. Whereas a t t h e lower l i m i t of t.emperature f o r e i g n phases a r e formed, t h e c r y s t a l l i z a t i o n of SAPO-5 is a c c e l e r a t e d w i t h r i s i n g t e m p e r a t u r e . T h i s l e a d s t o y i e l d s of SAPO-5 up t o 100 p e r c e n t depending on t h e g e l composit.ion. A t 463 K , t h e l a r g e s t c r y s t a l s were obt.ained. Theref o r e , we r e p o r t o n l y about t h e e x p e r i m e n t s a t t h i s t e m p e r a t u r e , The g e l c o m p o s i t i o n r e p r e s e n t s t h e most i m p o r t a n t f a c t o r f o r t h e c o n t r o l of b o t h c r y s t a l s i z e and s h a p e . To t h e d e g r e e t o which t h e r o l e of s j l i c o n can be n e g l e c t e d , i . e .
a t s u f f i c i e n t l y low s i l i c a c o n c e n t r a t i o n i n t h e gel, t h e
e f f e c t s of c o n t e n t s of H 2 0 , TEA, and P2O5 are s i m i l a r t o t h o s e which govern t h e n u c l e a t i o n and growth
of AlPO-5 c r y s t a l s ( r e f . 1 6 ) . The i n f l u e n c e s of b o t h t h e
c o n t e n t of H 2 0 ( F i g . 1 ) and t h e c o n t e n t of TEA o r TEA and P2O5 t o g e t h e r ( F i g . 2 )
503
have been chosen as representative illustrations of the system, The maximum length of the hexagonal prism grown from gels of various compositions are depicted in a graph (Fig. 3 ) . Dilution of the reacting gel results always in the growth of larger crystals (Fig. 1). However, this is accompanied by a rising content of a byproduct and a considerable prolongation of the crystallization time (more than one week for the most diluted gels). Occasionally, a new nucleation of hexagonal prisms occurs in the bulk volume of the gel and on the surface of crystals already formed. A water content 750 < e
c 1000
seems t o be an advant.ageous compromise.
Then, the largest crystals can grow up to a length of 580 pm (Fig. 4D1 and the yield of the SAPO-5 phase (regardless of the size distribution) can amount up to 80 per cent.
Fig. 1. SAPO-5 crystals (after decantation) synthesized from gels with rising water content (a:b:c:d = 1:1:1.55:0.2): ( A ) e = 300, (B) e = 450, ( C ) e= 600, (D) e = 750.
504
Fig. 2. SAPO-5 crystals (after decantation) synthesized from gels with constant water content (e = 300 H20, d = 0 . 2 Si02): ( A ) b = 1 , c = 1.55; ( B ) b = I , c = 3.1,; ( C ) b = 1, c = 6 . 2 ; (D) b = 2, c = 3.1.
6 00I
0
1 1 2
E 400-
-
2 E -.
-
1.55
0.2
3.1 6.2
0.2
3.1
0.2
/
+,
0.2 0
200I
I
I
400
I
I
600
I
I
800 H20ImoLe
Fig. 3. Maximum length of AFI hexagonal prisms depending on the gel composition.
505
F i g . 4 . Various c r y s t a l morphology i n SAPO-5 s y n t h e s i s Gel composition
a
b
c
d
e
vicinal faces
1 1
1 2
3.1 3.1
0.2
( B ) growth aggregate
0.2
450 300
1 1 1
1 1 1
1.55 1.55 1.55
0.3 1.0 0.4
750 300 300
(A) (C)
byproduc t
(D) SAPO-5 c r y s t a l s (E) d i s t o r t e d c r y s t a l s ( F ) secondary n u c l e a t i o n
For AlPO-5, recrystallization of AFI type crystals into tridymite and/or crystobalite type phases have been reported (ref. 2 2 ) . We could not observe this even in runs of one week duration (optimum period for the largest crystals growth). Small contents of a tridymite phase can be observed with SAPO-5 grown from gels with an overabundance of phosphorus, i.e. b:a
=
2.4 (ref. 2 3 ) . Simi-
lar results have been already reported (ref. 20). At a fixed water content in the gel, the crystal size can be influenced by the content of template or template and phosphoric acid toget.her (Fig, 2). Their excess can result in the formation of smaller crystals with a higher yield and in a decreasing crystallization period. Distortions of the cr-ystals or rose-like looking vicinal faces start to occur (Fig. 4A). This behaviour underlines the role of the template in the process of the molecular sieve formation and its significance for the nucleation stage in particular. An excess of both P2O5 and TEA leads to the strongest effect on the morphology of the product and, at low water content., even polycrystalline growth aggregates are formed (Fig. 48). On the other hand, the samples do not contain any byproduct and the crystallization is completed within 24 hrs. Depending on the gel composition, small spherulitic particles can be formed together with the SAPO-5 crystals (Fig. 4C). According to an
MAS
NMR
investigation (ref. 24), this byproduct represents an aluminophosphate which we failed to identify until now. Its occurrence is correlated with HzO/TEA 2 100 (molar ratio), i.e. just with the conditions to synthesize the large SAPO-5 crystals. However, using the differences in s i z e , both phases can be separated, e.g. by decantation. Occasionally, a small number of rhombohedric crystals have been observed in the samples, too (Fig. 4C). Concerning the silicon incorporation into the AFI framework, it is generally accepted that only small amounts can substitute phosphorus atoms esclusively (ref. 2 5 ) . As it was stated for the system investigated, even the smallest content of silica in the gel (i.e. d = 0.05) seems to hitlder the crystals to grow so large as in the AlP04-5 system. With increasing silicon content up to about d = 0.25 the system remains almost insensitive with regard to the crystal size and shape. Surprisingly, the largest crystals ( u p to about 580 pm long) can be synthesized at d = 0.3 (Fig. 4D), whereas at higher d values, only crystals as illustrated in Figs. 4E and 4F are formed. A similar appearance (Fig. 4F) was observed in the FAPO-5 system and attributed to secondary 11uc1eation (ref. 26). Unquestionably, this behaviour of the system should be attrihuted to the presence of heteroatoms. This is underlined by the absence of similar findings in the systems forming large AlP04-5 crystals (ref. 1 6 ) . Up to the same composition limit ( d
= 0.31, silicon replaces almost esclu-
sively phosphorus atoms in the framework of the crystals in this study. This
507 i s proved by 29S1 MAS NMR ( r e f . 24) and I R measurements a r e c o n s i s t e n t w i t h t h i s s t a t e m e n t a s w e l l ( r e f . 2 4 ) . I n r e f . 20, a similar l e v e l of s i l i c a c o n t e n t in t h e g e l ( d = 0 . 4 ) h a s been s u g g e s t e d
as a c r i t i c a l v a l u e € o r t h e s i l i c o n i n -
c o r p o r a t i o n i n t o t h e A F I framework and f o r t h e optimum c r y s t a l l i z a t i o n of g e l s c o n t a i n i n g n-tripropylamine.
1
I
I
-2
I
1
.4
I
.6
I
I
.8 PIP,
F i g . 5 . Adsorption i s o t h e r m s of n-hexane on SAPO-5 m o l e c u l a r s i e v e c r y s t a l s of d i f f e r e n t morphology ( T = 298 K): 1-hexagonal p r i s m s ; 2-growth a g g r e g a t e s ; 3 - d i s t o r t e d c r y s t a l s .
The a d s o r p t i o n behaviour of t h e v a r i o u s SAPO-5 m o l e c u l a r s i e v e s a m p l e s d i f f e r s depending on t h e c r y s t a l morphology ( F i g . 5 ) . A t h i g h r e l a t i v e a d s o r b a t e p r e s s u r e ( p / p s ) , o n l y t h e well-formed hexagonal prisms e x h i b i t t h e i d e n t i c a l a d s o r p t i o n and d e s o r p t i o n i s o t h e r m s , whereas b o t h t h e d i s t o r t e d c r y s t a l s and t h e growth a g g r e g a t e s show a n a d s o r p t i o n h y s t e r e s i s o c c u r r i n g most p r o b a b l y i n t h e secondary p o r e systems. I € t h e l a t t e r p o r e system is assumed as s e r v i n g a s a channel f o r t r a n s p o r t , t h e a d s o r p t i o n k i n e t i c s might b e a c c e l e r a t e d , b u t i n t h e s e p a r a t i o n p r o c e s s e s , t h e a d s o r p t i o n h y s t e r e s i s would be a draw back due t o r e d u c t i o n of t h e a d s o r b e n t - a d s o r b a t e i n t e r a c t i o n . The d i f f e r e n c e s i n t h e amounts adsorbed i n comparison t o t h e f i n d i n g s i n r e f . 27 can be a t t r i b u t e d t o t h e n o t optimized c a l c i n a t i o n c o n d i t i o n s i n t h e present case.
CONCLUSIONS The s t u d i e s performed have shown t h e c o n d i t i o n s € o r t a i l o r i n g t h e c r y s t a l s of t h e SAPO-5 molecular s i e v e w i t h c r y s t a l s i z e s r a n g i n g from 25 t o 580 pm i n
508 l e n g t h . P a r t i c u l a r a t t e n t i o n h a s been p a i d t o t h e c o n d i t i o n s i n f l u e n c i n g c r y s t a l morphology. The f o r m a t i o n and growth of t h e c r y s t a l s a r e i n f l u e n c e d by t h e c o n t e n t of a l l components i n t h e g e l . D i l u t i o n of t h e s y s t e m enhances t h e growth a l o n g t h e c-axis
l e a d i n g t o t h e f o r m a t i o n of l a r g e hexagonal p r i s m s . On t h e o t h e r h a n d ,
an e x c e s s of amine o r amine and p h o s p h o r i c a c i d t o g e t h e r r e s t r i c t s t h e c r y s t a l growth v i a an a c c e l e r a t i o n of t h e n u c l e a t i o n r a t e . Then, t h e d i s t o i . t e d c r y s t a l s
o r even growth a g g r e g a t e s a r e formed. The s i l i c o n c o n t e n t r e p r e s e n t s a n a d d i t i o n a l v a r i a b l e i n f l u e n c i n g t h e c r y s t a l shape. The e u h e d r a l hexagonal p r i s m s c a n c r y s t r a l l i z e o n l y from g e l s c o n t a i n i n g up t n about 0 . 3 s i l i c o n ( m o l a r r a t i o ) which is incorporat.ed i n t o t h e s t r u c t u r e a l m o s t o n l y on phosphorus-T-sites. The f o r m a t i o n of t h e p u r e SAPO-5 phase is c o r r e l a t e d w i t h t h e c o n d i t i o n H?O/TEA < I 0 0 ( m o l a r r a t i o ) . T h e r e f o r e , l a r g e hexagonal SAPO-5 prisms a r e always accompanied by s m a l l p a r t i c l e s of an u n i d e n t i f i e d aluminophosphate which can be s e p a r a t e d , i n p r i n c i p l e , because of t h e l a r g e d i f f e r e n c e s i n t h e s i z e of t.he particles. The shape of t h e c r y s t a l s i n f l u e n c e s c o n s i d e r a b l y t h e i r s o r p t i o n properties.
ACKNOWLEDGEMENTS The authoi
i
a r e i n d e b t e d t o Dr. Janchen f o r a d s o r p t i o n measurements and t o
D I . M . Hndnn f o r h e l p f u l d i s c u s s i o n s . Thr work was p a r t i a l l y s u p p o r t e d by Leuna WilrkF' AG. by t h e fund of t h e P r e s i d e n t of t h e Academy of S c i e n c e s of tht, GDR,
and by t h e P o l i s h M i n i s t r y of N a t i o n a l Education w i t h i n t h e P r o j e c t CPBP 01.06.
REFERENCES
1
2 3
4 5
6 I
8 Y
10 11 12 13
B.M. I n k . C . A . Messina, R . L . P a t t o n , R.T. Gajek, T.R. Cannan, and E.M. Flanigeri, I . Am. Chem. SOC., 106 ( 1 9 8 4 ) . 6092. DE 2 517 218 ( 1 9 8 6 ) . I)F 2 660 726 ( 1 9 8 6 ) . W. Leonhardt and 9.-M. Sax, Stud. S u r f . S c i . Catal., 46 ( 1 9 8 9 ) , 691. R. Mostowicz and J . M . Berak, Proc. I n t . Symp.: " Z e o l 7 t e s - S y n t h e s i s , S t r u c t u r e , Technology, and A p p l i c a t i o n " , P o r t o r o s e , 1984, B. DrZaJ, S . Hocevar, and S . Pejovnik ( E d s . ) , E l s e v i e r , Amsterdam, 1985, pp. 65-72. E.G. Derouane and 2. G a b e l i c a , J . S o l i d S t a t e Chem., 64 (19861, 296. S . F r n s t , P.A. J a c o b s , J . A . Martens, and J. Weitkamp, Z e o l i t e s , L ( 1 9 8 7 ) . 458. J . Kornatowski, Z e o l i t e s , 7 ( 1 9 8 7 ) . 7 7 . D.T. Hayhurst, A. N a s t r o , E. A i e l l o , F. C r e a , and G . Giordano, Z e o l i t e s , (19881, 416. F . Crea, A . N a s t r o , .J.B. Nagy, and R. A i e l l o , Z e o l i t e s , S ( 1 9 8 8 ) , 262. U . M u l l e r , A. Brenner, A . Reich, and K . K . Unger, ACS Symp. S e r . , 398 ( 1 9 8 5 ) , 347. S . Qiu, W . Pang, H . K e s s l e r , and J.-L. Guth, Z e o l i t e s , 2 ( 1 9 8 9 ) . 440. G. F i n g e r , J. Kornatowski, M. Bulow, and M. Rozwadowski, DD WP 330 56512 (1985).
509 Kornatowski, M. Rozwadowski, G. F i n g e r , and M.
Bulow, P 280 471 ( 1 9 8 9 ) . (1990). i n press. and M. Bulow, s u b m i t t e d f o r
14 15
J.
16
C. F i n g e r , J. Kornatowski, J. Richter-Mendau, publication. G. F i n g e r , J . Kornatowski, E. Jahn, M. Bulow, M. Rozwadowski, and B. Zibrowius, DD WP 330 567/7 ( 1 9 8 9 ) . J. Kornatowski, M . Rozwadowski, G. F i n g e r , E. J a h n , M. Bulow, and H. Zibrowius, P 280 470 ( 1 9 8 9 ) . G . F i n g e r and J. Kornatowski, Z e o l i t e s , lo (19901, 615. H . Weyda and H. L e c h e r t , Z e o l i t e s , lo ( 1 9 9 0 ) , 251. H . Beyer and H. Riesenberg, Messen und Zahlen m i t dem Mikroskop, i n : Handbuch d e i Mikroskopie, 3. Auflage, VEB V e r l a g Die Techriik, B e r l i n ( 1 9 8 8 ) . pp. 401-408. S.T. Wilson, B.M. Lok, C . A . Messina, and E.M. F l a n i g e n , Proc. 6 t h I n t . Z e o l l t e Conf., Reno, 1983, D. Olson, A. B i s i o ( E d s . ) , B u t t e r w o r t h , G u i l d f o r d , 1984, p p . 97-109. G . F i n g e r , E. J a h n , D. Zetgan, B. Zibrowius, K . S z u l z e w s k i , J. R i c h t e r Mendau, and M. Bulow, B u l l . S O C . Chim. Belg., % ( 1 9 8 9 ) . 291. B. Zibrowius, E . L o f f l e r , G. F i n g e r , E. Sonntag, M. Hunger, and J . Kornatowski, t h i s volume. N.B. Milestone and N.J. Tapp, Stud. S u r f . S c i . Catal., 36 ( 1 9 8 8 ) , 553. w. P a w , S . Q i u , Q. Kan, Z . W u , S . Peng, G . Fan, and D. T i a n , S t u d . S u r f . S c i . C a t a l . , 49 ( 1 9 8 9 1 , 281. V . R . Choudhary, D.B. Akolekar, A . P . Singh, and S . D . S a n s a r e , J . C a t a l . , 1 1 1 (1988). 23.
17
18 19 20 21
22
23 24 25 26 27
J. Kornatowski, G . F i n g e r , B u l l . SOC. Chim. Belg.
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 1991 Elsevier Science Publishers B.V., Amsterdam
511
APPROXIMATE ASSIGNMENT OF VIBRATIONAL FREQUENCIES OF THE NaX FRAl.iEI.VORK E. GEIOEL',
H.
9uHLIG2,
CH.
PEUKER'
and W.
PILZ'
' C e n t r a l I n s t i t u t e o f P h y s i c a l C h e m i s t r y , Academy o f S c i e n c e s , Rudower Chaussee 5, B e r l i n - 1 1 9 9 (GDR) 2Department o f C h e m i s t r y , K a r l - M a r x - U n i v e r s i t y T a l s t r a R e 35, L e i p z i g - 7 0 1 0 (GDR)
Leipzig,
SUMMARY O n t h e b a s i s o f s y s t e m a t i c a l l y d e v e l o p e d s u b u n i t c l u s t e r mode l s o f NaX z e o l i t e framework n o r m a l c o o r d i n a t e a n a l y s e s h a v e been c a r r i e d o u t by means o f WILSON'S GF m a t r i x method i n t e n d i n g t o u n d e r s t a n d t h e c o r r e s p o n d i n g e x p e r i m e n t a l v i b r a t i o n a l spect r a . An a p p r o x i m a t e a s s i g n m e n t o f v i b r a t i o n a l f r e q u e n c i e s o f t h e NaX z e o l i t e framework i s g i v e n o n t h e b a s i s o f c a l c u l a t e d e i g e n v a l u e s , p o t e n t i a l e n e r g y d i s t r i b u t i o n s and e i g e n v e c t o r s . The r e s u l t s a r e d i s c u s s e d i n c o n n e c t i o n w i t h t h e FLANIGEN c o n c e p t i o n o f i n t e r n a l and e x t e r n a l v i b r a t i o n s .
INTRODUCTION Z e o l i t e s have been w i d e l y used f o r i o n exchange, s e l e c t i v e a d s o r p t i o n and c a t a l y s i s .
Related t o t h a t r e s u l t s o f a great
number o f e m p i r i c a l s t u d i e s o f i n f r a r e d and Raman s p e c t r a have been p u b l i s h e d m a i n l y c o n c e r n i n g w i t h OH s t r e t c h i n g v i b r a t i o n s and more r e c e n t l y c o n s i d e r i n g t h e v i b r a t i o n s o f t h e z e o l i t e framework on t h e b a s i s o f FLANIGEN c o n c e p t i o n ( r e f .
1). I n t h e
most c a s e s t h e r e a r e d i f f i c u l t i e s w i t h r e s p e c t t o t h e i n t e r p r e t a t i o n o f v i b r a t i o n a l s p e c t r a because o f t h e c o m p l i c a t e d c r y s t a l s t r u c t u r e on one s i d e and t h e s t r o n g l y o v e r l a p p e d bands i n t h e z e o l i t e s p e c t r a on t h e o t h e r s i d e .
Beside t h a t i t i s i n s u f f i -
c i e n t l y known t h e i n f l u e n c e o f c a t i o n s , s t r u c t u r a l OH g r o u p s , a d s o r b e d w a t e r and o t h e r m o l e c u l e s on t h e v i b r a t i o n a l spectrum. F u r t h e r d i f f i c u l t i e s a r i s e w i t h i n a more t h e o r e t i c a l t r e a t ment o f v i b r a t i o n a l b e h a v i o u r o f z e o l i t e s c o n n e c t i n g w i t h t h e l a r g e u n i t c e l l s and t h e i n s u f f i c i e n t l y known f o r c e r e l a t i o n s between t h e l a t t i c e atoms and i o n s . The n o r m a l c o o r d i n a t e a n a l y s i s a s a r o u t i n e method c a n be u s e d t o i n t e r p r e t v i b r a t i o n a l spectra, but i s r e s t r i c t e d t o the s o l u t i o n o f d i r e c t eigenvalue p r o b l e m s because o f t h e l a c k o f e x p e r i m e n t a l d a t a f i t t i n g t h e model f o r c e c o n s t a n t s . So f a r some n o r m a l c o o r d i n a t e a n a l y s e s
512
have been c a r r i e d o u t on t h e b a s i s of s u b u n i t cluster models ( r e f s . 2 , 3 ) a s s u m i n g t h e d e c o u p l i n g p o s t u l a t e o f 6LACICI”IELL ( r e f . 3 ) and s i m p l i f i c a t i o n s w i t h i n t h e p s e u d o - l a t t i c e method ( r e f s . 4 - 6 ) , r e s p e c t i v e l y . R e c e n t l y t h e m e t h o d s of m o l e c u l a r m e c h a n i c s h a v e b e e n u s e d , t o o ( r e f s , 6-8). I n t h i s s t u d y , n o r m a l mode c a l c u l a t i o n s were d o n e f o r some s y s t e m a t i c a l l y s e l e c t e d s u b u n i t c l u s t e r models on t h e b a s i s of a simplified force f i e l d for understanding the vibrational spectra o f NaX z e o l i t e framework. SUBUNIT CLUSTER MODELS AND CALCULATIONS T h e s u b u n i t c l u s t e r m o d e l s o f NaX z e o l i t e f r a m e w o r k h a v e b e e n s e l e c t e d f o r t h e normal c o o r d i n a t e t r e a t m e n t s u s i n g a t e t r a h e d r a l s u r r o u n d i n g o f T ( T = S i , A l ) atoms a n d a s s u m i n g t h e v a l i d i t y o f LOEWENSTEIN r u l e ( r e f . 9 ) . G e o m e t r i c a l p a r a m e t e r s h a v e b e e n t a k e n from OLSON‘S X-ray d i f f r a c t i o n s t u d y o f h y d r a t e d N a X zeol i t e ( r e f . 10). The s e l e c t e d c l u s t e r m o d e l s t o g e t h e r w i t h t h e i r p o i n t g r o u p s y m m e t r i e s , t h e d i s t r i b u t i o n o f norrnal v i b r a t i o n s t o t h e i r r e d u c i b l e r e p r e s e n t a t i o n s a n d t h e number o f i n f r a r e d a n d Raman a c t i v e m o d e l v i b r a t i o n s a r e shown i n T a b l e 1. T h e r e s u l t of a n u c l e a r s i t e group a n a l y s i s f o r t h e complete u n i t c e l l o f NaX z e o l i t e b a s e d o n Fd3 s p a c e g r o u p symmetry a n d s u b t r a c t i n g t h e m o t i o n s o f Na’ c a t i o n s i s g i v e n i n t h e l a s t r o w . MARONI ( r e f . 1 1 ) a n d GODBER a n d OZIN ( r e f . 1 2 ) o b t a i n e d an o t h e r r e s u l t o n t h e b a s i s o f Fd3m s p a c e g r o u p s y m m e t r y . W i t h t h e e x c e p t i o n o f S i 0 4 ( A 1 0 3 ) 4 a n d A 1 0 4 ( S i 0 3 ) 4 A l g t h e number o f i n f r a r e d a n d Raman a c t i v e model v i b r a t i o n s i s a l w a y s smaller t h a n t h e c o r r e s p o n d i n g number o f o p t i c a l a c t i v e v i b r a t i o n s o f t h e u n i t c e l l . T h e n o r m a l mode c a l c u l a t i o n s were d o n e by m c a n s o f t h e GF m a t r i x m e t h o d o f WILSON ( r e f . 1 3 ) o n t h e b a s i s o f t h e JONES p r o g r a m ( r e f . 1 4 ) u s i n g i n t e r n a l c o o r d i n a t e s i n t h e u s u a l way w i t h e x c e p t i o n o f t o r s i o n a l o n e s w h i c h were c h o i c e d t o t a k e i n t o c o n s i d e r a t i o n o n l y o n e d i h e d r a l a n g l e b e n d i n g f o r o n e n o n - t e r m i n a l bond. A c c o r d i n g t o BLACKWELL ( r e f . 3 ) we h a v e u s e d t h e s o - c a l l e d bond l e n g t h s c a l e d f o r c e f i e l d o n t h e b a s i s o f t h e BADGER r u l e (BLSF-BR)
.
513
TABLE 1 D i s t r i b u t i o n o f n o r m a l v i b r a t i o n s o n t h e symmetry s p e c i e s o f d i f f e r e n t z e o l i t e c l u s t e r models C l u s t e r models; / p o i n t o r space group/
D i s t r i b u t i o n o f normal vibrations
S i 0 4 , A104
u(1R)
u(RA)
9
9
O3 S i -0 ( i) -A103
21
21
4R(hP),
42
42
44
44
40
40
21
21
4R(S)-
r=22a+22e
6 R ( h P ) ; /C3/ 06R;
r=20au t 2oa
/S6/
r=21a
Si04(A1)4 Si04 (A103)4 A104 P i 0 3 14 ( A 1 19 A124Si24096 ; /Fd3, (Th 4 ) /
9
+ 20eu+20e
9
r=57a
57
57
r=84a r=18A +18Au+18E +18Eu+ 9 9 54 Fa+53F,
84 53
84 90
( u ( 1 R ) number o f i n f r a r e d and u(RA) number o f Raman a c t i v e m o d e l v i b r a t i o n s ; i = 1 - 4 ; 4R(hP) 4 - r i n g i n t h e h e x a g o n a l p r i s m ; 4R(S) 4 - r i n g i n t h e s o d a l i t e u n i t : 6R(hP) 6 - r i n g i n t h e h e x a g o n a l p r i s m ; D6R d o u b l e - 6 - r i n g ) RESULTS AND DISCUSSION The r e s u l t s o f o u r c a l c u l a t i o n s f o r some s e l e c t e d s u b u n i t c l u s t e r models a r e r e p r e s e n t e d i n F i g .
1 and compared w i t h t h e
e x p e r i m e n t a l i n f r a r e d s p e c t r u m o f NaX z e o l i t e framework t a k e n f r o m MIECZNIKOWSKI and HANUZA ( r e f .
1 5 ) . Whereas t h e o b s e r v e d
spectrum i s g i v e n i n t r a n s m i s s i o n u n i t s t h e c a l c u l a t e d frequenc i e s a r e p l o t t e d a g a i n s t t h e number o f f r e q u e n c i e s p e r f r e q u e n c y interval i n arbitrary units.
I t c a n be seen t h a t an i n c r e a s e o f
s u b u n i t c l u s t e r s i z e l e a d s t o i n c r e a s i n g d e n s i t y o f bands p e r f r e q u e n c y i n t e r v a l . The A10 and SiO s t r e t c h i n g v i b r a t i o n s a b s o r b i n c l e a r l y d i f f e r e n t f r e q u e n c y r e g i o n s whereas OAlO and O S i O b e n d i n g ones appear i n t h e same r e g i o n what i s shown i n r e p r e s e n t a t i o n (b).
T h e r e f o r e t h e l a t t e r v i b r a t i o n s a r e d e n o t e d by OTO
bendings. T h i s p r i n c i p a l d i f f e r e n c e remains v a l i d a l s o i n t h e o t h e r s u b u n i t c l u s t e r s m o d i f i e d , however,
by c o m p l i c a t e d coup-
lings.
I f t e r m i n a l bonds e x i s t i n t h e c l u s t e r models as i n t h e c a s e s (b)-(e)
t h e c a l c u l a t e d h i g h f r e q u e n c y A10 s t r e t c h i n g v i b r a t i o n s
f a l l i n t h e e x p e r i m e n t a l f r e q u e n c y gap (780-850 cm").
Lacking
t e r m i n a l bonds ( f ) t h e above m e n t i o n e d v i b r a t i o n s a r e s h i f t e d t o
514 I
I
600
1200
1000
800
400
-
600 400 krn-11
515
l o w e r f r e q u e n c i e s by s t r o n g c o u p l i n g s .
T h i s r e s u l t i s supported
by FLANIGEN's i n f r a r e d s t u d i e s o f sodium a l u m o s i l i c a t e s i n d i f ferent crystallization states (ref.
16). Increasing the crystal-
l i z a t i o n degree and t h u s d e c r e a s i n g t h e number o f t e r m i n a l bonds i n t h e m i x t u r e l e a d s t o d i s a p p e a r a n c e o f an a b s o r p t i o n band a t 850 cm"
c h a r a c t e r i z i n g t e r m i n a l A 1 0 bonds.
The c a l c u l a t e d f r e q u e n c y r e g i o n s o f s t r e t c h i n g and OTO bending vibrations,
t o g e t h e r w i t h an a p p r o x i m a t e c l a s s i f i c a t i o n o f
t h e model v i b r a t i o n s a r e c o l l e c t e d i n t h e T a b l e s 2 and 3. The TOT b e n d i n g s and t o r s i o n a l v i b r a t i o n s a p p e a r i n g i n t h e f a r i n f r a r e d r e g i o n a r e n o t s u b j e c t o f t h i s s t u d y . V i b r a t i o n a l modes o f t h e s u b u n i t c l u s t e r models d e t e r m i n e d by t h e p o t e n t i a l e n e r g y d i s t r i b u t i o n and t h e e i g e n v e c t o r s show a c o m p l i c a t e d p i c t u r e . I t c a n be d i s t i n g u i s h e d between 3 t y p e s o f s t r e t c h i n g v i b r a -
tions: terminal,
bridging ( w i t h regard t o linkages connecting 2
t e t r a h e d r o n s ) and t e t r a h e d r a l v i b r e t i o n s ,
a l l o f t h e s e c a n be
f u r t h e r d i v i d e d i n s y m m e t r i c and a s y m m e t r i c ones.
I n addition t o
t h i s t h e p o t e n t i a l e n e r g y can be d i s t r i b u t e d i n a d i f f e r e n t way. T h i s statement i s v a l i d i n p r i n c i p a l f o r t h e bending v i b r a t i o n s , too, The b r i d g i n g and t h e t e t r a h e d r a l t y p e v i b r a t i o n s o f t h e subu n i t c l u s t e r models c a n be u s e d t o i n t e r p r e t t h e v i b r a t i o n a l s p e c t r a o f an a c t u a l z e o l i t e .
The a p p r o x i m a t e a s s i g n m e n t t h u s
o b t a i n e d i s shown i n T a b l e 4. So t h e h i g h f r e q u e n c y framework a b s o r p t i o n s can be a s s i g n e d t o a s y m m e t r i c b r i d g i n g s t r e t c h i n g vibrations.
The m a i n p a r t o f p o t e n t i a l e n e r g y o f t h e s e i s l o -
c a t e d i n t h e S i O bonds. Then f o l l o w a s y m m e t r i c b r i d g i n g s t r e t c h i n g v i b r a t i o n s w i t h dominant A 1 0 c h a r a c t e r ( S i O c h a r a c t e r less t h a n 10%). Symmetric b r i d g i n g and t e t r a h e d r a l v i b r a t i o n s a p p e a r a t l o w e r f r e q u e n c i e s . The d o u b l e r i n g and p o r e o p e n i n g modes m e n t i o n e d i n T a b l e 4 have been d e s c r i b e d i n d e t a i l e l s e w h e r e (ref.
Fig.
1 7 ) . The a b s o r p t i o n s between 510 cm"
1. Observed
and 408 cm"
c a n be
IR s p e c t r u m and c a l c u l a t e d v i b r a t i o n a l f r e q u e n -
c i e s o f NaX z e o l i t e framework ( T = t r a n s m i s s i o n , nf=number o f f r e quencies per frequency i n t e r v a l ; NaX z e o l i t e (Si/A1=1.12) (ref.
15),
a c c o r d i n g t o MIECZNIKOWSKI and HANUZA
( b ) S i 0 4 ( f u l l l i n e ) and A104 ( d o t t e d l i n e ) c l u s t e r ,
( c ) 03Si-0(1)-A103 prism,
( a ) e x p e r i m e n t a l IR s p e c t r u m o f
cluster,
( e ) double-6-ring
(d) 6 - r i n g c l u s t e r i n t h e hexagonal
cluster,
( f ) A104(Si0,)4(A1)9
cluster).
TABLE 2 C a l c u l a t e d f r e q u e n c y r e g i o n s and a p p r o x i m a t e a s s i g n m e n t s f o r t h e s t r e t c h i n g v i b r a t i o n s o f s e l e c t e d z e o l i t e c l u s t e r models Cluster
Si04 / A104
C a l c u l a t e d f r e q u e n c y r e g i o n and a p p r o x i m a t e a s s i g n m e n t
1011-978
828-805
(SiO)
Ztt,as(AIO)
V
t,as
03S i - C ( i ) - A 1 0 3
1086-1041 Y
b, a s
T,
6R(hP)
1C16-982
s
b, S
991
( S i o ,A10)
1095-1065
b,s
v,
564-561
(SiO.Al0)
1039-1027
Si C )
Y
T,s
5,705-690 ( s i o Ale) 8
% 744-682
vb, ( SiO ,A 10 )
1098-1045,936-901 vb, as ( Si0 , A 10 )
814-803
687-682 ( Si0 )
800-791 uT, as (
609-604 ( A10 )
“;-,
VT,
534-523 V
756 vb, as (A1O)
347,233
?,as
789-721,694-595 yb, a s (
785
5,a s (A1o)
vb,s(AIO,SiO)
(A10)
902-894
b.as ( S i 0 , A l O )
889,768
827 %,as (A1O)
866-864
669
(Al0,SiO)
Y ( A 10, t,s
820-803
4,
yt ,as (A10)
Y,as(SiO)
732 D6 R
(A10)
(A10,SiO)
1082-1052
t.S
t.s
239-231 Y
b, a s ( S i O . A l 0 ) ‘V
574 2r
845-811
5 ,a s ( S i O )
(Si0,AlO)
571-559 v (A10)
720 Vt,s(SiO)
T,s
(A10,SiO)
812,779 a (
*b,
569,365,260 vb, ( A 1 0 , S i O ) 584-539 VT,s(A1O)
(wave numbers i n cm-’; d i s p l a c e m e n t s o f t e r m i n a l , o f b r i d g i n g and o f t h e bonds o f a c o m p l e t e t e t r a h e d r o n a r e d e n o t e d by t , b and T, r e s p e c t i v e l y ; s and a s d e n o t e s y m m e t r i c a l ( t h e same s i g n ) and a s y m m e t r i c a l ( o p p o s i t e s i g n ) di s pl ac ement s , r e s p e c t i v e l y ; m c h a r a c t e r i z e s m ixed v i b r a t i o n a l modes; S i O a n d A10 d e n o t e t h e m a i n p a r t o f p o t e n t i a l e n e r g y w i t h i n t h e p o t e n t i a l energy d i s t r i b u t i o n )
TABLE 3
C a l c u l a t e d f r e q u e n c y r e g i o n s and a p p r o x i m a t e a s s i g n m e n t s f o r t h e OTO-bending v i b r a t i o n s o f s e l e c t e d z e o l i t e c l u s t e r models C a l c u l a t e d f requenc y r e g i o n and approx i mat e assignment
Si04
/
A104
t t393-381 , a s (Oslo 1
if
I
I
03Si-0 ( i)-A103
544-443 &tb.
304-376 Jt * a s ( O A l O 1 410-404 m '
363-356 Jt
381-376 dtt (OT0
362-357 ( O M 01
dt *
* (Oslo 1 276- 27 1 'tb, a s
6R(hP) 06 R
.
590-549
dbb (OT0) 219,136-114
524-503
dtb,
481-397
353-351.214-162
am
Jm
(OT0)
325-222 atb, a s
abb. a s
I ( t t * t b and bb c h a r a c t e r i z e a n g l e d e f o r m a t i o n s .
i f t h e a n g l e c o n s i s t s o f two t e r m i n a l ( t t ) .
o f a t e r m i n a l a n d a b r i d g i n g ( t b ) a n d of two b r i d g i n g bonds ( b b ) ; s and a s d e n o t e symmetr i c a l and a s y m m e t r i c a l d e f o r m a t i o n s , v i b r a t i o n a l nodes)
respectively;
OTO=OA10,
O S i O ; m c h a r a c t e r i z e s mixed
518
a s s i g n e d t o OTO b e n d i n g s .
I t s h o u l d be n o t e d t h a t so f a r an ex-
p l a n a t i o n o f t h e e x p e r i m e n t a l s p e c t r a c a n be o b t a i n e d o n l y by c o n s i d e r i n g t h e v i b r a t i o n a l b e h a v i o u r o f some d i f f e r e n t s u b u n i t c l u s t e r models,
TABLE 4
Approximate assignment o f observed v i b r a t i o n a l f r e q u e n c i e s o f NaX z e o l i t e ( S i / A l = l . l 2 ) cm-'
obs. IF! b,
Assignment a ) RA
1078
sh
975
ve
752
m
1082
m
985
m
800
w b,as
(A10)
700
m
ub, ( S i 0 , A l O )
612
vw
562
m
vl-s(AIO) d o u b l e r i n g mode
461
S
408
W
510
m
p o r e o p e n i n g mode
m
( s h s h o u l d e r , w weak, strong) ')notation
'bb,
$( OTO ) 375
364
vs
v w v e r y weak,
m medium,
s strong,
vs very
see t a b l e s 2 and 3
b ) t a k e n f r o m MIECZNIKOWSKI and HANUZA ( r e f .
15)
measurements
As a r e s u l t t h e FLANIGEN c o n c e p t i o n o f e x t e r n a l and i n t e r n a l t e t r a h e d r a l v i b r a t i o n s h a s t o be m o d i f i e d because o f t h e s t r o n g c o u p l i n g between t h e z e o l i t e framework v i b r a t i o n s .
Each bond has
b r i d g i n g and i n t e r n a l t e t r a h e d r a l c h a r a c t e r s i m u l t a n e o u s l y . m i l a r c c n c l u s i o n s have been drawn by WALTHER ( r e f s .
18,19)
Sifrom
c o r r e s p o n d i n g c a l c u l a t i o n s f o r ZSM-5 z e o l i t e s u b u n i t c l u s t e r models. The c o u p l i n g between t h e t e t r a h e d r o n s d o m i n a t e s t h e c o u p l i n g w i t h i n t h e t e t r a h e d r o n s . T h e r e f o r e we p r e f e r t o d i s -
519 t i n g u i s h between l o c a l i z e d and d e l o c a l i z e d v i b r a t i o n s ( s e e ref.
20) t h a n between i n t e r n a l and e x t e r n a l ones.
CONCLUDING REMARKS Z e o l i t e s r e p r e s e n t an e x t r a o r d i n a r y c h a l l e n g e f o r t h e v i b r a t i o n a l s p e c t r o s c o p y because o f t h e i r l a r g e u n i t c e l l s .
They a r e
c o n n e c t e d w i t h a l a r g e number o f v i b r a t i o n a l modes l y i n g c l o s e l y t o g e t h e r i n a s m a l l frequency region.
Vibrational calculations
o f z e o l i t e s u b u n i t c l u s t e r models c a n be a c c o u n t e d f o r an i n t e r mediate step i n t r e a t i n g t h e complete s p e c t r o s c o p i c a l l y s i g n i f i cant u n i t c e l l .
I f t h e c l u s t e r p r o b l e m w i l l be s o l v e d ,
q u e s t i o n f o r a p p r o p r i a t e f o r c e f i e l d s r e m a i n s open.
the
In a d d i t i o n
t o t h i s methods f o r t h e e s t i m a t i o n o f v i b r a t i o n a l i n t e n s i t i e s o f z e o l i t e s seem t o be n e c e s s a r y e x p l a i n i n g c o m p l e t e l y t h e e x p e r i m e n t a l s p e c t r a . W i t h r e s p e c t t o b o t h t h e problems mentioned above quantum c h e m i c a l methods have t o c o n t r i b u t e c o n s i d e r a b l y . REFERENCES E.M. F l a n i g e n , H. K h a t a m i and H.A. Szymanski, Adv. Chem. Ser. 101 (1971) 201 2 Y.S. Kong, M.S. 3hon and K.T. No, B u l l . Kor. Chem. SOC. 6 (1985) 57 C.S. B l a c k w e l l , 3. Phys. Chem. 83 (1979) 3251, 3257 3 4 K.T. No and M.S. Jhon, J. Phys. Chem. 8 7 (1983) 226 5 K.T. No, O.H. Bae and M.S. ahon, J. Phys. Chem. 90 (1986) 1772 K.T. No, B.H. Seo, 3.M. Park and M.S. Jhon, J. Phys. Chem. 6 9 2 (1988) 6783 7 K.T. No, Y.Y. Huh and M.S. Jhon, 3. Phys. Chem. 93 (1989) 6413 K.T. No, B.H. Seo and M.S. Jhon, Theor. Chim. A c t a 75 (1Y89) 8 307 K.T. No, H. Chon, T. Ree and M.S. Jhon, 3 . Phys. Chem, 85 9 (1981) 2065 10 D.H. Olson, 3. Phys. Chem. 74 (1970) 2758 11 V.A. M a r o n i , A p p l . S p e c t r o s c . 4 2 (1988) 487 12 J. Godber and G.A. Ozin, J. Phys. Chem. 9 2 (1988) 2841 jr., J.C. D e c i u s and P.C. Cross, M o l e c u l a r 13 E.B.Wilson, V i b r a t i o n s , M c G r a w - H i l l Book Company, Inc., New York, 1955 301188, Computer Programs f o r I n f r a r e d S p e c t r o p h o t o m e t r y 14 R.N. -Normal C o o r d i n a t e A n a l y s i s N.R.C.C. B u l l e t i n No. 15, Canada, 1976 15 A. M i e c z n i k o w s k i and J. Hanuza, Z e o l i t e s 5 (1985) 188 16 E.M. F l a n i g e n , S t r u c t u r a l A n a l y s i s by I n f r a r e d S p e c t r o s c o p y , ACS Monograph No. 171, Z e o l i t e C h e m i s t r y and C a t a l y s i s , 1976, chap. 2, pp. 80-117 17 E. G e i d e l , H. B o h l i g , Ch. Peuker and W. P i l z , Z. CIiem. 28 (1985) 155 18 P. W a l t h e r , 2. Chem. 26 (1986) 189, 222 19 P. W a l t h e r , Z. phys. Chem. L e i p z i g , 269 (1988) 809 20 R.A. v a n S a n t e n and D.L. Vogel, L a t t i c e Dynamics o f Z e o l i t e s , Advances i n S o l i d S t a t e C h e m i s t r y , V o l . 1, JAI P r e s s Inc., 1989, pp. 151-224 1
-
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
THE TOPOLOGICAL STRUCTURE REPRESENTATION
521
OF ZEOLITES
B. MULLER I n s t i t u t f u r P h y s i k a l i s c h e Chemie, Friedrich-Schiller-Universitat Jena. LessingstraRe 10, 0-6900 Jena, BRD SUMMARY I t i s shown t h a t z e o l i t e s can be c h a r a c t e r i z e d
and c l a s s i f i e d by means o f t o p o l o g i c a l s t r u c t u r e a n a l y s i s . The t o p o l o g y used i s a non-metrical geometry. The advantage o f u s i n g t h e s e t o p o l o g i c a l s t r u c t u r e d e s c r i p t i o n i s t h a t t h e general s t r u c t u r a l p r o p e r t i e s r e p r e s e n t i n g t h e frame f o r t h e p h y s i c s and c h e m i s t r y o f m a t e r i a l s can be d e c l a r e d w i t h s i m p l e numbers o f t h e t o p o l o g i c a l parameters Nk1. The l o c a l r e c i p r o c a l curves o f t h e c e l l u l a r c o n f i g u r a t i o n b31/l11, expressed i n t h e t o p o l o g i c a l parameters, a r e measures f o r both t h e compactness and t h e c a t a l y t i c a l p r o p e r t i e s o f z e o l i t e s . The general s t r u c t u r e f u n c t i o n N 2 1 = N21(NS2,NO',N12) o f t h e independent t o p o l o g i c a l parameters N k 1 i s deduced f r o m t h e t h e o r y of c e l l u l a r configurations. The d i s t r i b u t i o n f u n c t i o n s g21i and 9 3 2 1 o f t h e f a c e s and polyhedra, r e s p e c t i v e l y , which may b u i l d up t h e v a r i e t y o f z e o l i t e c o n f i g u r a t i o n s , a r e given. INTRODUCTION
The
simplest,
e 1ement s
of
topo 1ogy
obtains
but
spatial
a l s o t h e most e s s e n t i a l systems a r e o f
the
a
connective s t r u c t u r e
i n v a r i a n t a t continuos transformations
s non-metrical.
topo 1ogy ture
ana y s i s
consists
relations
topological
relations
l),
(ref.
(refs. uniquely
Atomic
mapped
i n the determination
condensed
into
which
are the
The problem o f s p a t i a l t o p o l o g i c a l s t r u c -
between those c o n s t r a i n e d t h e
2,3).
The
therefore,
of
both
t o p o l o g i c a l s t r u c t u r e elements and i n t h e corresponding relations
between
nature.
structure
intrinsic connective
configurations
systems such as z e o l i t e s
space-filling
cellular
may
be
configurations
(ref .4). One
k-celloid
structure make
it
element
of
t h e dimension k correponds
o f t h e concrete
zeolite.
inquely
These
p o s s i b l e f o r t h e t o p o l o g i c a l parameters o f
configurations
space t o u n i q u e l y r e p r e s e n t ,
parameters
Nk1
the
as w e l l as
and t h e i r d i s t r i b u t i o n
constrain
n o t o n l y t h e s p a t i a l c o n f i g u r a t i o n and
also
s p a t i a l e x i s t e n c e and s t a b i l i t y i n
the
nected system represented b y t h e z e o l i t e s ( r e f . 6 ) .
cellular classify, 5).
functions arrangement
every
one
homomorphisms
t h e t o p o l o g i c a l s t r u c t u r e o f atomic condensed system ( r e f . topological
to
The gkli but
cellular-con-
522
TOPOLOGICAL STRUCTURE CONFIGURATIONS The can
topological
be
s t r u c t u r e elements o f
zeolite
configurations
r e a l i z e d by k - c e l l o i d s w i t h d i f f e r e n t dimensions
0,1,2.3)
on a 4-dimensional c o n f i g u r a t i o n space.
celloids
represent one k-dimensional
(k-l)-celloids (k>O).
Each o f
f i g u r e bounded a t
=
(k
k
the
k-
least
k+l
These f i g u r e s can be r e a l i z e d as k-dimensio-
n a l polyhedra e x h i b i t i n g a t l e a s t k + l v e r t i c e s ( r e f . 5 ) . 0-celloids zeolites,
both
celloid, with
the
1.2.
Si-
...,N o ) :
I n our case
and t h e A l - t e t r a h e d r a
of
do
(dangling-bonds)
crystalline
represent
which i s a l s o r e f e r r e d t o as the v e r t i c e s .
single
atoms)
=
(p
mOp
o r two chemical bonds
n o t represent O - c e l l o i d s because o f
the
0-
Atoms o r i o n s (as
the
the
0-
general
pro-
p e r t i e s o f connected c e l l u l a r systems. 1-celloids neighbouring
=
(q
mlq
Si-
1,2,.
. . .N1 1:
The "bonds" e x i s t i n g
between
o r A l - t e t r a h e d r a by means o f t h e i n c l u s i o n
0-ion
represent t h e 7 - c e l l o i d s o f t h e z e o l i t e c o n f i g u r a t i o n s . 2 - c e l l o i d s nPr ( r
=
1,2,
...,Nz):
The 7 - c e l l o i d s which a r e
c a l l e d the edges o f the c e l l u l a r c o n f i g u r a t i o n may be b u i l t or rings.
cycles
Only t h e r i n g s which seperate two neighbouring polyhedra
are 2 - c e l l o i d s , tions
also
of
i.e.,
the 2-celloids depict c o r r e la t io n d i s t r i b u -
the third-order.
This i s a condition o f
the
homotopic
invariance o f c e l l u l a r c o n f i g u r a t i o n s ( r e f . 6 ) . 3-celloids either
dimensional formed
are
(s = 1 , 2 , , . . , N a ) :
nP.
The polyhedra
belonging
the same o r d i f f e r e n t classes cover compactly over c o n f i g u r a t i o n space. by t h e 2 - c e l l o i d s and
d i s t r i b u t i o n s o f the fourth-order, figurations
The 3 - c e l l o i d s represent
(the
spatial
to
the
3-
polyhedra) correlation
The polyhedra o f c e l l u l a r
do n o t agreed w i t h the conventional co-ordinate
conpoly-
hedra o f s t r u c t u r e chemistry ( r e f , 7 ) . CONNECTIVE STRUCTURE RELATIONS
The
simplest
O,l, . . . ,d: of
the
then, result
structure
p = 1,2,
relations
...,N k ) o f
i n the
only
of
the
then,
t h e border o f t h e 1 - c e l l o i d the
means mkp
mrq,
if
k-celloid
limboide connective s t r u c t u r e r e l a t i o n s can
mapped t o t h e connective r e l a t i o n m a t r i x ve s t r u c t u r e r e l a t i o n a m a t r i x element
Mk1
mklpO
g i v i n g each
=
k
(&p:
The k - c e l l o i d
i n t e r s e c t i o n between both i s
Thouse second-order
M
k - c e l l o i d s can be produced by
i n t e r s e c t i o n between a l l elements. and
set
is the &p.
be
connecti-
w i t h t h e values
523 1
i f mkp n
0
o t h e r wise
ml0
= rnkp
= I
mklpq
he mean v a l u e
which i s a l s o c a l l e d t h e t o p o l o g i c a l
Nk1
parame-
t e r c h a r a c t e r i z e s t h e number o f l i m b o i d e connective r e l a t i o n s o f typ cal
k - c e l l o i d i n regard t o i t s 7 - c e l l o i d s .
the
topological
the
typical
condensed
parameters
The m a t r i x
represents t h e g l o b a l as
Nk1
topological structure invariants o f
system.
The
a
d i s t r i b u t i o n f u n c t i o n gkli
a
Nk1
of
well
as
given
atomic
obtains
types
n k l i of t h e limboide connections between t h e k- and t h e
7-celloids
(ref. 7 ) . TOPOLOGICAL STRUCTURE CONSTRAINTS Three-dimensional three and
c e l l u l a r s t r u c t u r e s must be c h a r a c t e r i z e d
o f t h e f o u r independent t o p o l o g i c a l parameters (ref.
N32
by
No1,N12,N21,
The general g l o b a l s t r u c t u r e f u n c t i o n o f such a
7).
c o n f i g u r a t i o n i s given by equation ( 2 ) :
The
topological structure function ( 2 ) constrains not only
spatial
configuration
stence ted
and arrangement b u t a l s o t h e
and s t a b i l i t y i n every c e l l u l a r connected system
by
the zeolites.
theory
that
the
I t i s d e r i v e d from
t o p o l o g i c a l parameters
the
exi-
represen-
topological
the
o r d e r s o f t h e f a c t o r groups as w e l l as t o t h e o r b i t l e n g t h s o f
the
permutation group i n t h e c o n f i g u r a t i o n space (eqn.
N3
and
No a r e t h e d e n s i t y numbers o f t h e
respectively,
which
3-
related
group to
Nk1
are
the
spatial
(3) and r e f . 7 ) .
and
a r e i n v a r i a n t i n regard t o b o t h
0-celloids, homeomorphic
and permutative t r a n s f o r m a t i o n s . The
distribution
limboide celloids.
connections The
function of
the
g k l i gives the types k-celloids i n
respect
trizes
Mk1
between t h e d i s t r i b u t i o n f u n c t i o n 9211 and
Especially,
on
they y i e l d the following the
the
the
d i f f e r e n t r e l a t i o n s e x i s t between t h e
tions
7).
of
to
d i s t r i b u t i o n f u n c t i o n s g k l i a r e dependent
o t h e r as a d d i t i o n a l , (ref.
nkli
1each
submarela-
topological
524
Parameters N Z 1 and N 3 2 (eqns. ( 4 ) - ( 7 ) ,
and
7):
i=l
i=l
This
ref.
i s t h e p h y s i c a l and chemical reason why
different
atomic
bonding p r o p e r t i e s g i v e d i f f e r e n t s o r t s o f c o n f i g u r a t i o n s
evolutions.
and
The i n t e r a c t i o n o f l o c a l energy m i n i m a z a t i o n and topo-
logy produces complex t h r e e - d i m e n s i o n a l
of
c e l l u l a r configurations
zeol it e s . TOPOLOGICAL CURVATURE MEASURE The s p e c i f i c t o p o l o g i c a l c u r v a t u r e measures b ’ * k / v 1 1 1 both
the s t a b i l i t y ,
the zeolites ( r e f . trical
property
surface.
determine
t h e compactness and a d s o r p t i o n p r o p e r t i e s 7).
that
of
These measures have t h e i n t e r e s t i n g geomethey a r e t h e u n i t d i s t a n c e
of
a
parallel
A p a r a l l e l s u r f a c e i s formed by t r a n s f o r m i n g each p o i n t a
c o n s t a n t d i s t a n c e a l o n g t h e s u r f a c e normal a t t h e p o i n t . A z e o l i t e c r y s t a l c o n f i g u r a t i o n can be d e s c r i b e d as
consisting
o f a number o f d i f f e r e n t t y p e s o f p o l y h e d r a whose r e c i p r o c a l
unit,
c u r v a t u r e , i s o n l y g i v e n by t h e t o p o l o g i c a l parameters ( r e f . 7).
COMPUTING METHODS OF THE TOPOLOGICAL PARAMETERS
A
case those
complete e v a l u a t i o n o f a l l t o p o l o g i c a l parameters i s i n possible, in
if
t h e s e t o f atomic c o o r d i n a t e s i s know
c r y s t a l l i n e z e o l i t e structures.
The
algorithm
this
as
are
of
our
computational program NWS i s as f o l l o w s ( r e f . 8 ) : 1.
Generate t h e s e t o f atomic c o o r d i n a t e s i n t o a f i n i t e
2.
Determinate
structu-
r e model o f c r y s t a l l i n e z e o l i t e . the
c o n n e c t i v e m a t r i x Mo1 by means
of
metrical
525 distance
r e l a t i o n s between t h e g i v e n model
coordinate
poly-
hedra. 3.
Search
t h e c o n f i g u r a t i o n r i n g s and d i s c r i m i n e
faces ( r e f . Evalute
4.
between
their
9).
both the topological d i s t r i b u t i o n functions gk’i
and
t h e t o p o l o g i c a l parameters N k l . 5.
Correct
the
t o p o l o g i c a l p a r a m e t e r s Nkl w i t h
model s i z e e f f e c t ( r e f .
regard
to
the
o f most o f
the
10).
TOPOLOGICAL STRUCTURES OF ZEOLITES The t o p o l o g i c a l p a r a m e t e r s N 0 1 1 N 1 2 , N 2 1 ,
and
N32
z e o l i t e s a r e shown i n t a b l e 1 . TABLE 1
The t o p o l o g i c a l p a r a m e t e r s o f t h e z e o l i t e s
: Zeol it e
1
NZ1
NO1
N32
b 3 * 1 / ~ 1 * 1 N3/N0
1Afghanite
1 4 ; 4 1Bikitait.e 1 4 1Chabazite 1 4 1Deca dodecasil 1 4 1Dodecasil 1H 1 4 1 TMA-E( AB) 1 4 IEdlngtonIte 1 4 1 Ericnite 1 4 ; Faujasite 1 4 1Ferrierite 1 4 1Gismondine 1 4 1Gmenilite 1 4 1Heulandite 1 4 1 Levyne 1 4 1Liottite 1 4 1 Losod 1 4 1 L inde-A 1 4 1 L inde-N ; 4 1 L inde-W 1 4 1Mordenite 1 4 1Natrolite 1 4 1Phillipsite 1 4 1 RHO 1 4 1Sodalite 1 4 1Thomsonite 1 4 1 ZK-5 : 4 !Quartz 1 4 1Cristobalit.e 1 4 1 T r 1dymit e 1 4 !Random Voronoi 1 4 Li-A
Moreover,
3 3.5 4 3 3 3 3 4 3 3 3 3.5 3.4 5 3 3 3 3 3 3.28 3.72 4 3 3 3 3.5 3 9 6 5 3
5.08 5.6 6 5.2 5.14 5.14 5.2 6 5.2 5.25 5.4 5.6 5.66 6.67 5.14 5.2 5.25 5.22 5.19 5.41 5.63 5.82 5.33 5.33 5.14 5.6 5.2 6 6 6 5.23
10 8 16 14 14 15 8 15 16 20 10 12 6 14 16 15 15.33 14.8 11.33 8.25 7.33 18 18 14 10 15 3 4 5 15.5
3.0 3.0 3.5 3.0 3.0 3.25 3.0 3.25 3.5 4.5 3.0 3.5 3.0 3.0 3.5 3.25 3.3 3.2 3.0 2.68 2.67 4.0 4.0 3.0 3.0 3.25 1.75 2.0 2.25 3.39
.182 .25 .333 .143 .166 ,166 .154 .333 .154 .143 .111 .25 .2 .5 .166 .143 .154 .150 .156 .214 .32 .375 .125 .125 .166 .33 .154 2 1 .667 .148
132
P/1000 -~
2.89 4.75 6.73 2.09 2.92 3.07 2.37 5.53 2.40 1.82 1.96 3.85 2.92 8.5 2.52 2.17 2.43 1.93 2.37 3.42 5.50 6.67 1.95 1.78 2.86 6.56 2.26 53.0 23.2 15.11 ?
!
121
1 -1
~
2 1 1 2 3 2 3 2 3 3 2 1 3 2 2 3 2 3 5 3 4 2 1 2 1 2 4 1 1 1 ?
2 3 3 3 3 3 3 2 3 3 4 2 4 6 3 2 2 3 3 3 4 2 2 3 2 3 3 1 1 1
?
1 1 1 1 1 1 : 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 : 1 1 : 1 :
1
t h i s t a b l e 1 a l s o shows t h e t o p o l o g i c a l i n t r i n s i c r a t i o
between t h e numbers o f p o l y h e d r a end v e r t i c e s N 3 / N 0 ,
the
recipro-
526
cal topological u n i t curvature b3,l/v1,1,
t h e polyhedron
P / I O O o g i v e s t h e number o f polyhedron per 1000
of b o t h t h e p o l y h e d r a t y p e s In
order
132
and t h e
AS,
and f a c e s t y p e s
density, number
respectively.
121,
t o compare w i t h t h e r e l a t e d parameters
of
zeolites,
t h e t o p o l o g i c a l q u a n t i t i e s o f some S i O z m o d i f i c a t i o n s as w e l l as
a
random Voronoi c o n f i g u r a t i o n a r e a l s o shown i n t a b l e 1. Topological. c o n s t r a _ i n t s of_t h e ZeOl.ite c o n f i w r a t i o n s The
value
one c o n f i g u r a t i o n o b t a i n e d i n t a b l e 1.
with
four
This
respectively,
zeolite
the
for
vertex A10412-
o f the z e o l i t e configurations.
The t o p o l o g i c a l mean v a l u e parameters N O 1 , realize
four
typical and
i n c i d e n t edges r e p r e s e n t s t h e S i O 4 1 z -
tetrahedra, each
i s equal
o f t h e t o p o l o g i c a l parameter N O 1
each
c o n f i g u r a t i o n (and a l s o o f
N 1 2 , N Z 1 , and N 3 2 o f
each
the topological structure equation
SiOz-modification)
(eqn.
2),
z e o l i t e c o n f i g u r a t i o n s a r e t o p o l o c a l l y determined.
Each
i.e.,
the
topologi-
c a l parameter NR1 depends on each o t h e r . TABLE 2
P o s s i b l e face d i s t r i b u t i o n s F Z 1 1 o f a t y p i c a l p o l y h e d r o n w i t h N j 2 10 f o r t h e t o p o l o g i c a l parameters NO1 4, N1z = 3, and N21 = 4.8, r e s p e c t i v e l y _ _ 3 4 5 6 7 8 9 10 1 ;F21i\n21i 1
=
\
I
I
1
I -
1 1 ; 1 1 1 1
1 2 3 4 5 6 7
1
0
2
8
: 1
0
9
1 1 I
1 0 1 0 1 2
3 1 4 2
1
8
1
0
! f
9 10 11 12 13 14 15 16 17
5 3 1 6 4 2 0 3 1
6 7 4 5 6 2 3 4 0 1 2 3 0 1
0
0
1
: 1 1 1 1
:
I
I
I I
1
1
1
1 1 : 1
:
I
1 2 0 1 2 3 2 3 4
0
0 0
0 0 0 0
1 1 2 2 2
0 0
0
0 0
3 3
0
3 0
0
0
0
0 0 0
0
0
1
0
0
1 :
0 0 0
0 0 0
0
0
1 1 1 1 1 : 1
: I I
o f the topological constraints f o r c e l l u l a r
tion
spaces w i t h
only
be
affect
0
0 0
I
Because
0
0
0 0 0
0
1
0 0
0
0 0
1
1 1 : 1
0
0
1
0 0
0 0
0
0 0 0 0
0
0 0
0 0
0
4 4 4 4 5 5 6
0 0 0
0 0
0
0
0
NO1
=
i n t h e range 5 the
zeolite
f a c e s 9211 must e x i s t .
4,
<
configura-
t h e mean number o f edges p e r N21
<
7
(ref.
c o n f i g u r a t i o n so t h a t
7).
These
face
can
constraints
distributions
of
the
527 For example i n t a b l e 1, unique,
one
having
the c r y s t a l l i n e SiO2 modifications
o n l y one f a c e t y p e w i t h n 2 l i
t r i b u t i o n o f t h e faces 921, I n table 2,
=
t e r s No' The faces
an example o f t h e p o s s i b l e f a c e d i s t r i b u t i o n s
= 10 f o r t h e t o p o l o g i c a l
=
4, N 1 2
3 , and N Z 1
=
<
$11
>
6 than faces w i t h n21i
molecules can o n l y pass r i n g s w i t h n 2 1 i
>
Gismondine,
more
and
small
6.
NZ1,
determined
can o n l y be r e a l i z e d i n most cases by
o f s e v e r a l polyhedron types 1 3 2 A,
having
Atoms
6.
The d i s t r i b u t i o n o f faces f o r g i v e n mean v a l u e by equations ( 4 ) - ( 7 ) ,
Fzli
parame-
4 . 8 i s shown ( r e f . 7 ) .
polyhedron shape i s s i m i l a r t o t h a t o f a cage with
dis-
i s g i v e n by t h e e q u a t i o n s ( 4 ) - ( 7 ) .
a t y p i c a l polyhedron w i t h N3z
of
The
6.
are
Phillipsite,
>
and
means
Exclusively, the zeolites L i -
1.
S o d a l i t h e have o n l y one
single
polyhedron t y p e ( t a b l e 1 ) . The t o p o l o g i c a l s t r u c t u r e f u n c t i o n ( 2 ) as w e l l as o t h e r a d d i t i o nal
7 ) imply t h a t t h e v a l u e o f
topological constraints ( r e f .
t o p o l o g i c a l parameter must be N12 ons
with
N32
<
>
The z e o l i t e s w i t h
13.
the
3 f o r a l l cellular configuratiNl2
=
3 are
called
simple-
connected. Toml-ogicpl curvatures o f z e o l i t e c o n f i g u r a t i o n s The
mean
topological reciprocal u n i t
curvature
b311/vl
i 1
is
characteristic o f the adsorption properties o f the zeolites. The range
mean
<.
2.667
polyhedra,
reciprocal configuration curvature i s b311/v1,1
the
<
4.5 (table 1 ) .
<
range i s g r e a t e r 1.0
found
I n t h e case
bJ*l/vl.l
in
of
the
single
4.5 (table
11,
as f o r F a u j a s i t e . The g r e a t e r t h e polyhedron v a l u e o f
b3*l/vlpl,
l a r g e molecules can be adsorbed a t t h e polyhedra. configuration curvature The to
value
of
bJ.l/vl.l
the better
t h e more e a s i l y The s m a l l e r
a
force-field
i s g i v e n which can be bent and can break t h e
c a t a l y t i c a l p r o p e r t i e s of t h e z e o l i t e s e s s e n t i a l l y both
t h e mean r e c i p r o c a l u n i t c u r v a t u r e o f
the
the of
molecules. give
rise
configuration
and those o f t h e s i n g l e polyhedra. The
value
of
g r e a t e r than 8/3. zeolites.
bJ.l/vl
~1
for zeolite
configurations
must
be
Those i s a measure f o r t h e d i s c r i m i n a t i o n o f t h e
The m o d i f i c a t i o n s o f b o t h S i O z and f e l s p a r s have
values
f o r r e c i p r o c a l c u r v a t u r e s l e s s t h a n 8/3. The p o s s i b l e polyhedron d i s t r i b u t i o n n32i o f z e o l i t e
configura-
t i o n s are given i n r e f . 7 . The number o f polyhedra per v e r t e x N3/N0
i s also
characteristic
528
of
the c e l l u l a r zeolite configurations.
y i e l d s N3/N0 functional
<
0.5.
A
zeolite
I n t a b l e 1 t h e v a l u e s o f N3/N0
relations
between NJ/NO
and
configuration a r e shown.
bJ-l/v1.1
are
The
given
in
ref. 7. The polyhedron d e n s i t y P/1000 As arranges
( f o r t h e z e o l i t e s see t a b l e
t h e r e l a t i o n s h i p between t h e m e t r i c a l q u a n t i t i e s
of
1) the
z e o l i t e c o n f i g u r a t i o n s and t h e t o p o l o g i c a l parameters. The c o r r e l a t i o n between t h e polyhedron d e n s i t y P/1000 ( r e f . reciprocal
curvature
11) and t h e
b3~7/v1.1 i s b e t t e r than t h a t
t e t r a h e d r o n d e n s i t y T / l O O O and
b3
8
l /v1
I
between
mean the
( r e f . 7 1.
REFERENCES 1 H . S . M . Coxeter, Unvergangliche Geometrie, B i r k h b u s e r , Base1 1982 2 F. Yonezawa, T . Ninomiya ( E d s . ) , T o p o l o g i c a l D i s o r d e r i n Conden-
sed M a t t e r , S p r i n g e r , B e r l i n , 1983 N.J. T u r r o , Angew. Chem., 98 (1986) 872-879 4 8 . Mullet-. Exper. Tech. Phys., 36 (1988) 121-134 5 B. M u l l e r , The t o p o l o g i c a l c h a r a c t e r i z a t i o n o f amorphous s t r u c t u r e s , i n P h y s i c a l Research, Akademie-Verlag, B e r l i n 1990 6 D. Stoyan, W.S. K e n d a l l . R . Mecke, S t o c h a s t i c Geometry, Akademie-Verlag, B e r l i n 1987 7 B. M u l l e r , The t o p o l o g i c a l f o u n d a t i o n s o f s t r u c t u r e c h e m i s t r y , submitted f o r p u b l i c a t i o n Schubert, B. M i l l l e r , Computing Program NWS, Computer Centre 8 R. o f Friedrich-Schiller-Unioversity Jena, 1987 9 0. M u l l e r , R . Schubert. Exper. Tech. Phys. 36 (1988) 135-144 10 B. Moller, R . Schubert, Wiss. Z e i t s c h r . F r i e d r . - S c h i l l e r - U n i v . Jena, i n p r e s s 11 W.W. Meier, D.H. Olsen, A t l a s o f Z e o l i t e S t r u c t u r e Types, B u t t e r w o r t h s , London, 1987 3
C,. Ohlrnann et
nl. (Editors),Culalysk and Adsorption by Zeolites
529
1991 Elsevier Science PublishersR.V., Amsterdam
STUDIES OF SECONDARY SYNTHESIS ON MODIFIED PENTASIL ZEOLITES 1 t B. MEIER 1
W. RESCHETILOWSKI’, W.-D. EINICKE
E. BRUNNER2
and
ERN ST^
H. ‘Department of Chemistry, Karl Marx University, Liebigstr. 18, DDR-7010 Leipzig, German Democratic Republic ’Department of Physics, Karl Marx University, Linnbstr. S , DDR7010 Leipzig, German Democratic Republic
SUMMARY The secondary synthesis on modified pentasil zeolites and Its influence on the properties of these zeolites was investigated. It is shown by X-ray diffractiontZ7Al MAS NMR and nitrogen adsorption that a high degree of dealumination and a high pH value of the synthesis mixture led to a favourable insertion of aluminium into the ZSM-5 framework.
INTRODUCTION In connection with the industrlal application of zeolites a rapid development in the processes of chemical conversion, separation and cleaning can be observed. Unmistakable is the increasing importance of new zeolitic molecular sieves with adapted pore systems and defined surface-chemical properties. one way for the production of such zeolites is the isomorphous substitution of aluminium and/or silicon in the framework through other elements with the condltions of the hydrothermal crystallization. A high variation concerning the multitude of various elements was observed for the pentasil zeolites (ref. 1). However, the information about the state of these elements In the zeolitic framework is limited. The actual investigations have shown that such materlals are also available as a result of the secondary synthesis using aluminosilicates (refs. 2 , 3 ) . The secondary synthesis is a new and almost unknown procedure for the postsynthetic zeol te modificetion. This procedure allows the reversion of the dealuminination of zeolltes, observed during their use, by a s mple method (ref. 4 ) .
The aim of this paper is to show the systematical investigations of the secondary synthesis of modified pentasils to get informations concerning the influence of the conditions of synthesis on the structural, surface-chemical and adsorptional properties of the resulting zeolites. EXPERIMENTAL Materials and Methods The starting material ZSM-5 (Si/A1=50) synthesized with organic template (ref. 5) was calcinated f o r 6 hours at 700 " C t 800 Oc and 900 OC. One part of the these samples was transrered into the H-form, where 25 g of the freshly calcinated zeolite powder was ion exchanged wlth 1 1 of 0.5 M aqueous solution of ammonium nitrate ( 3 hrs, 80 "C) and deammoniated ( 4 hrs, 500 0
C)
.
For the secondary synthesis with aluminium insertion aluminium sulphate was contacted with the zeolite and an aqueous solution of 2 N NaOH where the atomic ratio of aluminium between the zeollte and the suspension was varied. The reaction mixture with pH 10 to 12 was stirred for 3 hours at 70 O C , filtered and washed carefully. The resulting products were dried and characterized by various analytical methods. The determination of the Si/A1 ratios (ref. 6) of the samples was carried out at a NMR pulse spectrometer designed at the KMU Leipzig at a resonance rrequency of 70 MHz. The chemical shltt against [A1(H20)6]3+ was obtained with an accuracy of 2 ppm. The identification of the crystalline phases and the crystallinity of the samples were determined by means or a goniometer HZG 4 in the 20 region from 5 to 70 '. For comparlson the crystallinity of the Leuna-ZSM-5 standard was used. The volumes of the unit cells of some zeolites were obtained from the difrerences of the angel positions of 5 references in the 20 range from in comparison to the references of a-A1203. 23.00 to 25.70 The adsorption properties were investigated by the measureo ments of the nitrogen isotherms at - 196 C on a Sartorius balance. The designation, the conditions of the secondary synthesis, the aluminium content per unit cell. the Sl/Al ratios, the crystallinity, the volume of the unit cell from X-ray data and the nitrogen adsorption capacity at 0.1 MPa are shown in table 1.
531
TABLE 1 Characterization of the ZSM-5 zeolites investigated Nr.
prepar. conditions T/OC
1 2 3 4 5 6 7 8
9 10
11 12 13 14* 15**
700 800 900 700 800 900 700 800 900 800
800 800 800 900 900
Al/u.c. Si/A1
pH Al-supply
-
-
-
-
12 12 12 12 12 12 10
1:l
11 10 11
1:l 1:l 1:8 1:8 1:8 1:l
1:l 1:8 1:8
-
-
12
1:l
cryst. %
1.96 1.29 0.91 1.86 2.91 2.91 12.31 9.80 10.79 1.71 1.86 0.98 4.00 0.40 7.27
48 73 104 52 32 32 6.8 8.8
7.9 55 51 97 23 239
12.2
V/u.c.
nm
84
-
88
-
48 76 77 93 38 68 64 91 85 62 67 82 93
5.3428
-
5.3933 5.3958
-
5.3743
-
5.3803
-
5.3743
mg g 94.1 124.5 51.0 98.0 111.4 128.4 45.7 129.6 129.0 124.7 116.3 70.4 118.7 132.1 134.6
* HZSM-5, proton form of starting material * * Sample was prepared by secondary synthesis from sample 14 RESULTS AND DISCUSSION x-ray diffraction The results of the qualitative analysis of crystalline phases have shown that the samples produced by our procedure have a ZSM-5 structure. For all samples a varying content of a-quartz was detected. A s can be seen from table 1 the crystallinities 0 maintained during the calcination of the samples at 700 C and 800 OC (samples 1 and 2, crystallinity of the starting material 85 % ) . The calcination at 900 OC (sample 3 ) led to a decrease of the crystallinity which is be due to a dissolution of the zeolitic framework connected with a further dealumlnation. In the case of the secondary synthesis with realumination and a low alumlnium supply (samples 4 , 5 , 10, 11) the crystallinity of the samples calcinated at 700 OC and 800 OC was remained almost unchanged. The increase in the content of aluminium in the synthesis mixture obviously led also to the increased insertion of aluminium into Si framework positlons. The silicon species obtained during the substitution procedure can combine to a-quartz and therefore they should be the reason for the decrease of the crystallinity (sample 7, 8, 12, 13). In the case of the realumination of a sample with a lower crystallinity calcinated at 900 OC, i t seems that as well a s the realumination a
reinsertion of silicon species occurs which led to the reconstruction of the lattice and the increase of the crystallinity (samples 6 and 15). A higher content of aluminium in the synthesis mixture (sample 9) has a negative influence on the crystallinity, because the increased insertion of aluminium prevents the insertion of silicon species. Due to the differences in the ionic radii of A1 (r=0.051 nm) and si (r=0.042 nm) the increased insertion of aluminium into the framework should change the lattice parameters which is connected with the lncrease of the volume of the unit cell. As can be seen from table 1 there is a relation between the Si/A1 ratios which are determined by 27Al MAS NMR and the volume of the unit cell. The higher the aluminium content of the framework the higher the volume of the unit cell. This behaviour seems to be indirect evidence for the realumination of the zeolite framework. 2 7 ~ 1MAS N M R Dealuminated samples The sharp signal in the 27Al MAS NMR spectra at 53 ppm was assigned to the tetrahedrically coordinated aluminium in framework positions. The higher the calcination temperature the lower the intensity of the signal which is due to the dealumination of the framework. In fig. 1 the change of the amount of aluminium
2.0 . u t
Fig. 1 Change of the amount of aluminium per unit cell in dependence on the calcination temperature (A:NaZSM-5, B:HZSM-5)
? 1.6A
6
1.2 -
0.8 0.1, -
700
800 - 9 0 0 T("C1
per unit cell is demonstrated by the dependence on the calcination temperature. It is shown that in the case of HZSM-5 zeolites the dealumination of the framework with higher temperatures becomes advantageous in comparison with NaZSM-5 zeolltes. Realuminated samples The results of the realumination of the calcinated NaZSM-5 zeolltes are shown in fig. 2. In all cases the pH of the aqueous
A
Fig. 2 Realumination in dependence on the calcination temperature and the Al-Supply at pH=12 ( theor. max. values A : 17.1 and €3: 3 . 8 Al/u.c.) I
700
800 -900 T ("C 1
suspension was 12 and the ratio of aluminium in the framework to aluminium in the suspension varied from 1:l to 1:8. From the sum of all aluminium atoms in the solid and in the suspension a theoretical value for the highest insertion possible for aluminium was calculated. While in the case of a 1:l ratio this value is 3.8 Al/u.c. for the 1:8 ratio the value increases to 17.1 All u.c.. From fig. 2 i t is shown that the aluminium insertion depends strongly on the aluminium supply of the suspension. For the 1:l ratio with calcination temperatures of 800 OC and 900 OC the realumination was possible up to 75 percent of the theoretical maximal value. The increase of the aluminium supply led to the highest increase of insertion which also about 70 percent of the theoretical value was reached. It seems that with the mentioned reaction conditions a barrier for the aluminium insertion can not be crossed. A purposeful variation of the reaction conditions has shown
that a constant pH value during the secondary synthesls is the main factor for a sucessful realumination. The influence of the pH value on the realumination is demonstrated in fig. 3 for NaZSM-5 zeolite calcinated at 800 OC. The higher the pH of the aqueous suspension the higher the amount of aluminium Inserted into the framework positions corresponding to the aluminium supPly.
10u t
< 8-
d
6L-
210
11
-
12
PH
Fig. 3 Realumination in dependence on the pH at Al-supply of 1:8 (A) and 1:l (6)
150- 50 d(PPm1
-50
Fig. 4 27~1 MAS NMRspectra of samples, prepared at pH=12 (A), pH= 11 (B) and pH=lO (C)
Prom the 27Al MAS NMR spectra of the realuminated zeolites in fig. 4 it becomes clear that in the case of a lower pH value the
content of the octahedrically coordinated extra-framework aluminium increased. The lncrease of the pH value led to a decrease of these aluminium species, which do not appear in the case of pH=12. Nitrogen Adsorption Now we should focus our attention on the influence of the chemical reactions of the solid on the pore system of the postsynthezised zeolites. The investigations have shown that in all
535
cases a typical and an almost identical nitrogen adsorption isotherm appeared. Differences in the limiting values of nitrogen adsorption caused by the different crystalline content of the zeolites as shown in table 1. Contrary to the adsorption investigation on realuminated pentasils dealuminated at 540 C' in a water stream (refs. 4 1 7 ) no formation of a secondary Pore system was found. Therefore the different aluminium distributions for the pentasils synthesized with and without organic templates should be the reason. It seems that for the pentasils synthesized without organic template a stronger aluminium and silicon gradient exist which can not be maintained after the dealumination and the following realumination. The reconstruction of the structural defects in such products is connected with the insertion of the tetrahedrically coordinated positions by silicon and the formation of a secondary pore system with a narrow distribution of the pore radi 1 . ACKNOWLEDGEMENTS The authors thank Dr. P. Kraak and K. Heimbach for supplying the X-ray data and Prof. D. Freude and Dr. K. Becker for helpful support. REFERENCES M . Tielen, M . Geelen and P.A. Jacobs, Proc. Intern. Symp. Zeolite Catal., szeged 1985, Acta Phys. Chem. Szeged., 1985, P. 1 G.W. Skeels and E.M. Flanigen, ACS Symposium series ( M . L . Ocelli and H. E. Robson, Eds.) 398 (1989) 420 B. Sulikowskl and J. Klinowski. ACS Symposioum Series ( M . L . Ocelli and H.E. Robson, Eds.) 398 (1989) 393 W. Reschetilowski, W.-D. Einicke, M . Juaek, R. SchOllner, D. Freude, M . Hunger and J. Klinowski, Appl. Catal. 56 (1989) L 15 DD-WP 207 186 (1984) D.R. Corbin, B.F. Burgess jr., A.J. Vega and R.D. Farlee, Anal. Chem. 59 (1987) 2722 W.-0. Einicke, W. Reschetilowski, M. Heuchel, M. v.Szombathely, M . Jusek, H.-R. Poosch, P. BrMuer, W. Schwieger and K.-H. Bergk, Chem. Techn. 42 (1990), 215 I
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
537
INCORPORATION OF S I L I C O N I N T O THE FRAMEWORK OF SAPO-5
S T U D I E D BY
NMR AND I R SPECTROSCOPY
BI Z j b r o w i u s
and J .
1
,
E.
Kornatowski
Loffler 4
1
,
G.
Finger
1
,
Sonntag',
E.
M.
Hunger
3
' C e n t r a l I n s t i t u t e o f P h y s i c a l Chemistry and 'Central I n s t i t u t e o f I n o r g a n i c Chemistry o f t h e Academy o f Sciences o f t h e GDR, Rudower Chaussee 5, B e r l i n , DDR-1199 (German Democratic R e p u b l i c ) 'Department o f Physics, K a r l - M a r x - U n i v e r s i t y L e i p z i g , L i n n s s t r . L e i p z i g , DDR-7010 (German Democratic R e p u b l i c )
5,
4 1 n s t i t u t e o f Chemistry, N i c o l a u s Copernicus U n i v e r s t y , Gagarina 7 , 87-100 Torun (Poland)
SUMMARY SAPO-5 was s y n t h e s i z e d from s t a r t i n g g e l s COVI - r i n g a broad range o f composition. The i n c o r p o r a t i o n o f s i l i c o n i n t o t h e A F I framework was s t u d i e d by m u l t i n u c l e a r MAS NMR and I R spectroscopy. I t was found t h a t t h e g e l composition determines t h e morphology o f t h e c r y s t a l s , b u t o n l y s l i g h t l y i n f l u e n c e s t h e amount o f s i l i c o n i n c o r p o r a t e d . An upper l i m i t f o r t h e i n c o r p o r a t i o n o f i s o l a t e d s i l i c o n atoms can be expected. INTRODUCTION
C r y s t a l l i n e s i l i c o a l u m i n o p h o s p h a t e m o l e c u l a r s i e v e s (SAPOs) a r e novel microporous s o l i d s which
can
be considered as
from t h e corresponding alurninophosphates tion (ref.
1).
derivatives
by isomorphous s u b s t i t u -
D i f f e r e n t mechanisms f o r t h i s s u b s t i t u t i o n a r e d i s -
cussed. Only i n t h e case o f an i n c o r p o r a t i o n o f monomeric s i l i c o n on phosphorus s i t e s ( i . e . to
four
adjacent
t h e s i l i c o n atoms a r e bonded v i a oxygen
aluminium
atoms),
the
calcination
of
the
as-
synthesized SAP0 generates one Bronsted a c i d i c s i t e per one s i l i con atom
incorporated.
Therefore,
not
only
the o v e r a l l
content,
b u t a l s o t h e t y p e o f s i l i c o n i n c o r p o r a t i o n determines t h e c a t a l y t i c a c t i v i t y o f SAPOs. The goal composition
o f our work was t o i n v e s t i g a t e t h e
on
the
final
composition
of
the
influence o f crystals
and,
gel in
p a r t i c u l a r , on t h e s i l i c o n i n c o r p o r a t i o n i n t o t h e framework o f t h e resulting differently
s i z e d and shaped SAPO-5
crystals.
Further-
more, we want t o show t h a t t h e combination o f NMR and I R s p e c t r o -
538 scopy i s an a p p r o p r i a t e t o o l t o o b t a i n more d e t a i l e d about t h e samples. ref.
2.
information
The p r e s e n t paper i s a d i r e c t c o n t i n u a t i o n o f
The samples were s y n t h e s i z e d f r o m s t a r t i n g g e l s w i t h i n a
broad range o f composition.
The s o l i d s o b t a i n e d a f t e r c r y s t a l l i z a -
t i o n were c h a r a c t e r i z e d by XRD
and scanning e l e c t r o n microscopy
and analyzed c h e m i c a l l y . To s t u d y t h e s i l i c o n i n c o r p o r a t i o n , s o l i d s t a t e 31P,
29Si,
27Al
and 'H
MAS NMR
as w e l l as I R spectroscopy
were appl l e d . EXPERIMENTAL Synthesis 0
sample were s y n t h e s i z e d h y d r o t h e r m a l l y a t 190 C s t a r t i n g
SAPO-5
from g e l s
b P205
o f d i f f e r e n t molar r a t i o s c TEA
.
. e H 2 0 ( c f . Table 2 ) .
d Si02
a A1 0 . 2 3 The g e l s were
o f t h e components
an aluminiumoxydehydrate s o l (2.35 mass% A 1 0 1, 2 3 phosphoric a c i d ( 8 5 mass% H PO 1, t r i e t h y l a m i n e (TEA) a c t i n g as 3 4 template, a s i l i c a s o l ( 4 0 mass% S i O ) , and b i d i s t i l l e d w a t e r . The 2 s t a r t i n g pH value o f t h e g e l was 3.3 !: 0 . 2 r i s i n g t o about 7.0 formed
after
applying
completion
of
the
crystallization
process.
c r y s t a l s were separated f r o m t h e mother l i q u o r and,
The
obtained
i f necessary,
from some o c c a s i o n a l l y formed s m a l l amount o f by-products,
washed 0
c a r e f u l l y and d r i e d . C a l c i n a t i o n was performed under a i r a t 750 C . A more d e t a i l e d d e s c r i p t i o n o f t h e s y n t h e s i s i s g i v e n i n r e f .
2.
Chemical a n a l y s i s
..____~___
The d e t e r m i n a t i o n o f
the
chemical
composition o f
the
SAPO-5
samples as g i v e n i n Table 2 was c a r r i e d o u t by wet chemical a n a l y s i s u s i n g t h e f o l l o w i n g procedures. After calcination a t with
sodium carbonate
i n t o hydrochloric acid.
the
lOOO"C,
powdered m a t e r i a l
( f i v e f o l d excess)
followed
For t h e d e t e r m i n a t i o n o f
by
was f u s e d dissolution
silicon,
poly-
e t h y l e n e o x i d e s o l u t i o n was used t o aggregate t h e s i l i c i c a c i d . The p r e c i p i t a t e was f i l t e r e d , washed and weighed a f t e r h e a t i n g a t 1100°C. dioxide.
Then h y d r o f l u o r i c
a c i d was
added
to
evaporate
silicon
The q u a n t i t y o f s i l i c o n was determined on t h e b a s i s o f
w e i g h t r e d u c t i o n . The f i l t r a t e was t a k e n t o q u a n t i f y t h e r e s i d u a l amount o f s i l i c o n d i o x i d e by means o f t h e molybdenum b l u e s p e c t r o photometry. ,
I
,
'..
539 The f i l t r a t e o f s i l i c i c a c i d was t a k e n t o o b t a i n t h e amounts o f A 1 0 and P 0 Aluminium was p r e c i p i t a t e d by 8 - h y d r o x y q u i n o l i n e 2 3 2 5' i n a weak a c e t i c a c i d s o l u t i o n (pH 6...7, 50'C). The oxyquino-
l a t e o f aluminium was f r i t t e d ,
washed w i t h l i t t l e c o l d water and
weighed as A1(CgH60N) a f t e r d r y i n g f o r 2 h a t 1 3 5 ' C . Phosphate 3 was p r e c i p i t a t e d as magnesium ammonium phosphate i n a two-step procedure. The p r e c i p i t a t e was f i l t e r e d , washed and t h e paper subsequently ashed. A f t e r h e a t i n g a t 1 1 O O " C ,
t h e r e s i d u e was weighed
as Mg P 0 2 2 7' 1-nst r ument a 1 The
27Al,
29Si
and 31P MAS
Bruker MSL 400 spectrometer.
NMR
s p e c t r a were
The experimental
NMR measurements a r e summarized i n Table 1 .
recorded on
a
conditions f o r the
A l l
s p e c t r a shown i n
t h i s paper were o b t a i n e d by s i n g l e p u l s e e x c i t a t i o n .
I n t h e case
o f 2 9 S i and 31P MAS NMR, high-power p r o t o n d e c o u p l i n g was a p p l i e d . The
chemical
shifts
reported
for
27Al
are
not
corrected
for
second-order quadrupole e f f e c t s . 'H
MAS NMR s p e c t r a were measured a t
room temperature u s i n g a
Bruker MSL 300 spectrometer and a home-made magic-angle-spinning equipment t h a t
a l l o w s t o s p i n sealed g l a s s ampoules.
The
total
c o n c e n t r a t i o n o f OH groups i n t h e a c t i v a t e d samples was determined by comparing t h e l i n e i n t e n s i t y w i t h t h a t o f a standard. TABLE 1 Experimental c o n d i t i o n s f o r MAS NMR measurements.
-
____-
nucleus resonance p u l s e flip repetino. o f MAS frequency d u r a t i o n angle t i o n t i m e scans f r e q u . /MH i! /us /S /kHz 1
10.0
400
2.5
TMS
T/l2
0.5
800
5.1
3+ Al(H20)6
2.5
n/4
5.0
4.2
TMS
3.0
n/2
215.0
4.8
H3P04(85%)
H
300.0
4.0
7r/2
2 7 ~ 1
104.2
0.6
79.5
161.9
29s1 31P
For t h e ' H M A S NMR measurements, a g l a s s tube o f
reference
5.5
,2000 64
t h e samples were p r e t r e a t e d i n
mm i n n e r diameter w i t h a 10 mm bed-depth.
S t a r t i n g a t room temperature,
t h e samples were heated under vacuum
540 w i t h a r a t e o f 10 K/h.
A f t e r keeping f o r 2 h a t 670 K ,
t h e samples
Pa f o r 20 h and s e a l e d
were evacuated up t o a pressure below 0 . 0 1 off. The
same
samples were
Fourier transform)
used
i n DRIFT
(diffuse
reflectance
IR
i n v e s t i g a t i o n s o f OH groups.
I n some cases an
anologous p r e t r e a t m e n t i n a s p e c i a l vacuum c e l l
( r e f . 3 ) was c a r -
r i e d o u t b e f o r e t h e I R measurements. performed w i t h a F o u r i e r
The D R I F T measurements were
spectrometer I R F 180 ( C e n t r e o f
Scien-
t i f i c I n s t r u m e n t a t i o n s o f t h e Academy o f Sciences o f t h e GDR) w i t h -1 a r e s o l u t i o n o f 4 crn . An empty g l a s s tube o r a rough copper s u r face
i n the
vacuum
cell
were
employed as
the
standard,
respec-
tively. RESULTS AND D I S C U S S I O N
means o f
By
XRD
i t was
proven t h a t
e x h i b i t indeed A F I s t r u c t u r e ( r e f . 4 ) .
Fig.
lA,
samples B and D i n F i g s .
The chemical compositions o f
samples a r e g i v e n
i n Table
g e l s cover a broad range,
investigated
in
ref.
sample A
2:
in
2D and 4 8 and sample C i n F i g . b o t h t h e g e l s and t h e c a l c i n e d
Although
2.
samples
The q u i t e d i f f e r e n t morpho-
logy o f t h e t h e c r y s t a l s i s i l l u s t r a t e d
4E.
all
the
t h e composition o f
compositions o f t h e SAPO-5
the
crystals
i s nearly c o n s t a n t . TABLE 2
Chemical compositions o f b o t h t h e g e l s and t h e c a l c i n e d and dehydrated samples. ge 1
samp 1 e
c a l c i n e d sample
A1 0 , : P 0 : S i O : 2 3 2 5 2
TEA:
H20
A1
:
P
Si
:
Al/(P+Si)
* A-
1.0
: 1 . 0 3 : 0.25:
1.55:
300
0.506:
0.458:
0.036
1.024
B
1.0
: 2.06:
0.25:
3.10:
300
0.505:
0.459:
0.036
1.020
C*
1.0
: 1.03:
1.00:
1.55:
300
0.505:
0.442:
0.053
1.021
D
1.0
: 2.06:
1.00:
3.10:
300
0.496:
0.465:
0.037
0.992
E*
1.0
: 1.10:
0.24:
1.12:
450
-
-
-
-
F
1.0
: 2.35:
1.00:
3.43:
450
-
-
-
~~~
~
__
_~_____
~~
*samples c o n t a i n i n g small amounts o f by-products which were separ a t e d b e f o r e t h e NMR and I R i n v e s t i g a t i o n s
541 31P
A-29.4
. . 100
0 -50
50
0
-20
PPm
-Ul PPm
-GO
F i g . 1 . 2 7 A l and 31P MAS NMR s p e c t r a o f SAPO-5 (sample A ) : ( a + d ) as-synthesized, (b+e) c a l c i n e d and dehydrated and ( c + f ) c a l c i n e d and rehydrated. S p i n n i n g sidebands a r e denoted by asterisks.
27Al
For
and 31P MAS NMR s p e c t r a o f sample A
both n u c l e i
we observe o n l y one
a r e shown
i n Fig.
l i n e i n the spectra o f
1.
the
c a l c i n e d and dehydrated sample c o n f i r m i n g t h e s t r i c t a l t e r n a t i o n o f phosphorus and aluminium a t T p o s i t i o n s o f t h e aluminophosphate framework
( r e f . 5 ) . The l i n e a t 3 5 . 4 ppm i n t h e 2 7 A l
has t o be a s c r i b e d
to
tetrahedrally
bonded v i a oxygen t o f o u r -30.2
aluminium atoms
whereas t h e
line at
ppm i n t h e 31P NMR spectrum corresponds t o phosphorus l i n k e d
w i t h f o u r A10 The
coordinated
phosphorus atoms,
NMR spectrum
4
tetrahedra ( r e f s . 5 , 6 ) .
substitution
of
a
part
of
the
framework
T
positions
by
s i l i c o n i s n o t r e f l e c t e d i n t h e s p e c t r a . The s i n g l e resonance l i n e o b t a i n e d f o r phosphorus c o r r o b o r a t e s t h a t P-0-Si l y t o occur ( r e f .
1).
bonds a r e u n l i k e -
The d i f f e r e n t environments o f t h e phosphorus
n u c l e i i n t h e case o f a s i l i c o n i n c o r p o r a t i o n on h y p o t h e t i c a l a l u minium
T
sites
should
The s p e c t r a i n F i g .
be
detectable
in
31P
MAS
NMR
spectra.
1 a l s o i n d i c a t e t h a t l a r g e changes i n t h e
542 framework
AFI
occur
during
These changes seem t o
sample
c a l c i n a t i o n and
be a common f e a t u r e
phate-based m a t e r i a l s ( r e f s . 6 - 9 ) .
of
rehydration.
many aluminophos-
Comparing t h e s p e c t r a i n F i g .
1
w i t h those o f A l P O - 5 , we found no i n f l u e n c e o f t h e s i l i c o n i n c o r 4 p o r a t i o n . The 2 7 A l and 31P MAS NMR s p e c t r a o f t h e d i f f e r e n t SAPO-5 samples are almost i d e n t i c a l and agree w i t h those o f t h e s i l i c o n f r e e A l P O -5. 4
Although o t h e r w i s e s t a t e d i n r e f .
10,
the i n t e n s i t y
o f t h e l i n e due t o o c t a h e d r a l l y c o o r d i n a t e d aluminium i n t h e 2 7 A l MAS NMR spectrum
i n Fig.
o f t h e c a l c i n e d and r e h y d r a t e d sample (-13.3
ppm
l c ) does n o t depend on t h e amount o f s i l i c o n i n c o r p o r a t e d .
N e i t h e r 2 7 A l nor 31P MAS NMR can be used t o d e t e c t o r t o q u a n t i f y t h e amount o f s i l i c o n i n c o r p o r a t e d i n t o an aluminophosphate framework. DRIFT
'95i M A S N M R
I
RL
-60
-80
-100
4
-120
ppm
F i g . 2. SAPO-5.
29Si
MAS NMR and D R I F T s p e c t r a o f c a l c i n e d and dehydrated D e n o t a t i o n as i n Table 2 . 29Si
MAS NMR i s t h e o n l y d i r e c t method t o i n v e s t i g a t e t h e t y p e
o f s i l i c o n incorporation.
The *'Si
NMR s p e c t r a t o g e t h e r w i t h t h e
corresponding d i f f u s e r e f l e c t a n c e I R F o u r i e r t r a n s f o r m s p e c t r a a r e shown i n F i g . 2 . and - 9 6
The 2 9 S i
NMR l i n e s w i t h
ppm a r e a s c r i b e d t o s i l i c o n atoms
maxima between - 9 1
ppm
bonded v i a oxygen t o
543 f o u r aluminium atoms ( r e f s . 7 , l l - 1 5 ) . ted
i n t h i s way
second
signal
should cause one
at
about
-110
Each s i l i c o n atom i n c o r p o r a Bronsted a c i d i c
ppm
can
be
seen
weak
site.
A
the
case
in
of
samples B, C and D. I t stems from s i l i c o n atoms bonded v i a oxygen
t o f o u r s i l i c o n . These pure s i l i c a u n i t s may be p r e s e n t i n s i l i c a structure or
i n an amorphous
islands within
the AFI
(refs.
Since we have found t h a t
12,13).
by-product
by-product
sometimes o b t a i n e d
i s i n our case an aluminophosphate
former e x p l a n a t i o n i s more l i k e l y . ref.
the
(vide infra),
the
I n l i n e w i t h the findings o f
1 3 , an i n c r e a s e o f t h e o v e r a l l s i 1 i c o n c o n t e n t i n c r e a s e s t h e
r e l a t i v e i n t e n s i t y o f t h e -110 ppm l i n e , i . e .
lowers t h e number o f
Bronsted a c i d i c s i t e s per s i l i c o n atom i n t h e sample. The h y d r o x y l s studied
generated by
b o t h I R and
by
’H
the
silicon
MAS NMR
reflectance I R Fourier transform
incorporation
spectroscopy.
spectra i n Fig.
3625
and
(SiOHAl
3520 cm-l
groups).
are
observed
due
Corresponding t o
propose t h a t these d i f f e r e n t
to
0-1 Bronsted
literature
be
The d i f f u s e present the
2
r e g i o n o f fundamental OH s t r e t c h i n g v i b r a t i o n s
can
-
Two bands a t acidic
(refs.
two bands c h a r a c t e r i z e
sites
16,171,
we
undisturbed
S i O H A l groups (3625 cm-’)
and S i O H A l groups i n t e r a c t i n g w i t h l a t -
t i c e oxygen ( 3 5 2 0 cm-’1.
The t h r e e
3745
and
cm-l
3675
are
caused
by
low
intensity
AlOH,
l i n e s a t 3785,
and
SiOH
POH
groups,
r e s p e c t i v e l y ( r e f . 17). 29si M A S NMR
O R I FT
- 95.6
I
3625 A ~~
-60
-80
-100
-12 0
PPm
4000
3520
3500
3000
;/crn-’
F i g . 3 . 2 9 S i MAS NMR and D R I F T s p e c t r a o f t h e samples E ( b r o k e n l i n e s ) and F ( s o l i d l i n e s ) .
544
As F i g .
3 shows,
the s e n s i t i v i t i e s o f
MAS NMR and D R I F T
29Si
a r e r a t h e r d i f f e r e n t . The samples d i f f e r s t r o n g l y i n t h e amount o f i n c o r p o r a t e d s i l i c o n due t o a non-optimized I n t h e case o f -95.6
there
sample F ,
ppm c h a r a c t e r i z i n g
i s only
synthesis
a weak
t h e monomerically
procedure. line at
NMR
29Si
incorporated s i l i c o n ,
b u t t h e v i b r a t i o n bands o f OH groups due t o t h i s t y p e o f s i l i c o n i n c o r p o r a t i o n are q u i t e intense.
i t can be concluded t h a t
Hence,
t h e D R I F T spectroscopy i s a v e r y s e n s i t i v e method t o p r o v e monomer i c a l l y incorporated s i l i c o n . 3.8
I
4.8
I
*
I
*
Fig.
IR
1
4.
* *
H MAS NMR spectrum o f SAPO-5
spectroscopy
reliable
always
quantitative
(sample A ) .
encounters
data
on
difficulties
the
number
of
in
obtaining
the
different
h y d r o x y l s . T h i s problem can be s o l v e d by t h e a p p l i c a t i o n o f ' H MAS NMR spectroscopy.
The
'ti
MAS NMR
spectrum o f
t h r e e l i n e s ( F i g . 4 ) . The l i n e s a t 4 . 8
consists o f
SAPO-5
and 3 . 8 ppm a r e caused by
b r i d g i n g S i O H A l groups and correspond t o t h e I R bands a t 3520 and 3 6 2 5 cm-l,
observed, terminal
respectively. which
Furthermore a broad band a t
was a s c r i b e d t o POH groups
S i O H groups
summarized i n Table 3 .
can
contribute
to
this
(ref. line.
1.9
ppm
is
11).
But a l s o
The
data are
Except f o r sample C , t h e number o f b r i d g i n g
545 hydroxyls
agrees
well
with
p r e s e n t i n t h e samples.
the
total
This f i n d i n g
number
of
silicon
atoms
i n d i c a t e s an almost e x c l u -
s i v e l y monomeric i n c o r p o r a t i o n o f t h e s i l i c o n and i s t h e r e f o r e i n accordance w i t h t h e s i l i c o n s p e c t r a shown i n F i g . 2 . pprn l i n e p r e s e n t i n t h e s i l i c o n s p e c t r a , except
for
sample
the
A,
total
From t h e -110
one would assume t h a t ,
number o f
s i l i c o n atoms
should
s l i g h t l y exceed t h e number o f b r i d g i n g h y d r o x y l s . D i s c u s s i n g these q u a n t i t a t i v e d e t a i l s , t h e f a c t t h a t we a r e d e a l i n g w i t h v e r y small amounts o f s i l i c o n and h y d r o x y l s has t o be considered. TABLE 3 Concentrations o f s i l i c o n atoms and h y d r o x y l groups. ~ _ _ _ _ ~ _ _ _ _ __~ ~ _________-__---_____-~ sample
s i l i c o n atoms /1O2O
g-1 -
~
c
b r i d g i n g OHb
/1O2O ._ .. __
g-l
n o n - a c i d i c OHC /1O2O
g-1
-
A
3.6
4.4
0.5
El
3.6
4.6
0.5
5.3
3.5
0.5
3.7
3.2
0.5
2.8
0.6
1.6
0.6
___ a b
a
__
from chemical a n a l y s i s 1 from t h e i n t e n s i t h e H NMR l i n e s a t 3.8 and 4 . 8 ppm ( i O . 5 . 1 0 g from t h e ' n t n s i t y o f t h e ' H NMR l i n e a t 1 . 9 ppm 26 (LO.l.10 g
-7,
Since some o f products
the
SAPO-5
samples
synthesized
contain
by-
( s p h e r u l i t i c p a r t i c l e s up t o about 20 pm i n d i a m e t e r ) ,
one c o u l d argue t h a t t h e s i l i c o n c o u l d be p r e f e r e n t i a l l y
incor-
porated n o t i n t o t h e A F I
There-
fore,
phase,
but
i n t o t h e by-product.
we have performed a s y n t h e s i s under c o n d i t i o n s t h a t f a v o u r
the formation o f
t h e by-product.
The s y n t h e s i s p r o d u c t o b t a i n e d
was t h e n separated i n t o t h e A F I phase and t h e by-product by s i m p l e d e c a n t a t i o n basing on t h e
large differences
i n the size o f
the
p a r t i c l e s . The 2 9 S i MAS NMR and D R I F T s p e c t r a a r e g i v e n i n F i g . 5 . Only t r a c e s o f s i g n a l s i n d i c a t i n g a s i l i c o n i n c o r p o r a t i o n a r e observed i n t h e s p e c t r a o f t h e by-product. a non-perfect
s e p a r a t i o n o f t h e phases.
They can be a t t r i b u t e d t o Hence,
we can exclude a
p r e f e r e n t i a l s i l i c o n i n c o r p o r a t i o n i n t o t h e by-products.
546 2 9 ~ i MAS NMR
DRIFT
~
... .----
,-
,I
- 95.4
I
3625 I
4000
1
13520
1
I
,
3800
3600
3LOO
L
3200
29 . Fig. 5 . S i MAS NMR and DRIFT s p e c t r a o f t h e by-product ( b r o k e n l i n e s ) and t h e SAPO-5 phase ( s o l i d l i n e s ) o b t a i n e d f r o m t h e same synthesis batch a f t e r c a l c i n a t i o n .
b
100
0
-100
PPm
F i g . 6 . 2 7 A l M A S NMR s p e c t r a o f t h e b y - p r o d u c t : ( a ) as-synthesized
and ( b ) c a l c i n e d and d e h y d r a t e d .
Concerning
the
chemical
nature o f
t h a t i t i s an aluminophosphate.
by-product,
we
assume
T h i s i s n o t o n l y supported by t h e
5, where bands due t o A l O H and POH a r e p r e -
I R spectrum i n F i g .
sent,
the
b u t a l s o by 2 7 A l
and 31P MAS NMR.
The l i n e p o s i t i o n i n t h e
2 7 A l MAS NMR spectrum o b t a i n e d f o r t h e u n c a l c i n e d by-product 6)
i s characteristic
of
aluminium i n o c t a h e d r a l
t i o n i n an aluminophosphate ( r e f .
(Fig.
oxygen c o o r d i n a -
1 8 ) . During c a l c i n a t i o n ,
t h e oxy-
gen c o o r d i n a t i o n i s changed i n t o a t e t r a h e d r a l one r e f l e c t e d by a characteristic
l i n e position a t 37.5
ppm. We f a i l e d t o a s s i g n t h e
by-product t o any known aluminophosphate s t r u c t u r e by XRD.
CONCLUSIONS
Morphology and s i z e of t h e SAPO-5
c r y s t a l s can be c o n t r o l l e d by
t h e composition o f t h e s t a r t i n g g e l . For t h e c r y s t a l l i z a t i o n procedure used, a predominant monomeric incorporation o f morphologies.
silicon
We1 1-shaped
i s achieved f o r q u i t e d i f f e r e n t c r y s t a l SAPO-5
crystals
containing
up
to
4%
s i l i c o n on h y p o t h e t i c a l phosphorus T s i t e s can be synthesized. An SAPO-5
upper
l i m i t
for
can be expected.
monomeric
incorporation
Above t h i s
s i l i c o n present i n s i l i c a - r i c h
limit,
only
of the
silicon
into
fraction
islands w i t h i n the structure
of
(and/
o r i n amorphous m a t e r i a l ) i s i n c r e a s i n g . The s t r u c t u r a l
changes caused by t e m p l a t e removal and by rehy-
d r a t i o n are n o t s i g n i f i c a n t l y
i n f l u e n c e d by t h e
incorporation o f
s i 1 icon. A sometimes o b t a i n e d by-product
o f t h e s y n t h e s i s was shown t o
be an aluminophosphate w i t h o u t s i l i c o n .
REFERENCES 1 E.M. F l a n i g e n , R.L. P a t t o n and S . T . Wilson, i n : P.J. Grobet, W.J. M o r t i e r , E . F . Vansant and G. S c h u l z - E c k l o f f ( E d s . ) , I n n o v a t i o n i n Z e o l i t e M a t e r i a l Science, E l s e v i e r , Amsterdam, 1988, pp. 13-28. 2 G. F i n g e r , J . Kornatowski, J . Richter-Mendau, K . Jancke, M. Bulow and M. Rozwadowski, t h i s volume. 3 C . Peuker, G. F i n g e r , E. L o f f l e r and W . P i l z , i n p r e p a r a t i o n . 4 W.M. Meier and D.H. Olson, A t l a s o f Z e o l i t e S t r u c t u r e Types, 2nd. edn., B u t t e r w o r t h s , London, 1987. 5 D. M u l l e r , E. Jahn, B. Fahlke, G. Ladwig and U . Haubenreisser, Z e o l i t e s , 5 (1985) 5 3 - 5 6 .
548 6 C . S . B l a c k w e l l and R.L. P a t t o n , J . Phys. Chem., 88 ( 1 9 8 4 ) 6135-6139. 7 C . S . B l a c k w e l l and R . L . P a t t o n , J . Phys. Chem., 92 ( 1 9 8 8 ) 3965-3970. 8 M. Goepper, F . Guth, L. D e l m o t t e , J . L . Guth and H . K e s s l e r , i n : P . A . Jacobs and R . A . van Santen ( E d s . ) , Z e o l i t e s : F a c t s , F i g u r e s , F u t u r e , E l s e v i e r , Amsterdam, 1989, pp. 857-866. 9 B . Z i b r o w i u s , U . Lohse, K . Szulzewsky, H. F i c h t n e r - S c h m i t t l e r , W . P r i t z k o w and J . Richter-Mendau, t h i s volume. 10 O . V . K i k h t y a n i n , V . M . M a s t i k h i n and K . G . I o n e , A p p l . C a t a l . , 42 ( 1 9 8 8 ) 1-13. 1 1 D. Freude, H. E r n s t , M. Hunger, H. P f e i f e r and E. Jahn, Chem. Phys. L e t t . , 143 ( 1 9 8 8 ) 477-480. 1 2 E. Brunner, H . E r n s t , D. Freude, M. Hunger and H . P f e i f e r , i n : P.J. Grobet e t a l . ( E d s . ) , I n n o v a t i o n i n Z e o l i t e M a t e r i a l s Science, E l s e v i e r , Amsterdam, 1988, pp. 1 5 5 - 1 6 6 . 1 3 J . A . Martens, M. Mertens, P.J. Grobet and P.A. Jacobs, i n : P.J. Grobet e t a l . ( E d s . ) , I n n o v a t i o n i n Z e o l i t e M a t e r i a l s Science, E l s e v i e r , Amsterdam, 1988, pp. 97-105. 14 G . F i n g e r , E . Jahn, D . Zeigan, B. Z i b r o w i u s , I<. Szulzewsky, J. Richter-Mendau and M . Biilow, B u l l . S O C . Chim. B e l g . , 98 ( 1 9 8 9 ) 291-295. 1 5 I . P . Appleyard, R . K . H a r r i s and F . R . F i t c h , Chem. L e t t . , ( 1 9 8 5 ) 1747-1750. 1 6 S.G. Hedge, P. Ratnasamy, L.M. Kustov and V . B . Kazansky, Z e o l i t e s , 8 (1988) 137-141. 1 7 C . H a l i k . J.A. L e r c h e r , H . Mayer, J . Chem. SOC., Faraday Trans. 1 , 84 ( 1 9 8 8 ) 4457-4469. 18 D. M u l l e r , I.Grunze, E . H a l l a s and G. Ladwig, Z . anorg. a l l g . Chem., 500 ( 1 9 8 3 ) 80-80.
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
549
ON THE SYNTHESIS AND STRUCTURE OF A1P04-14
B. ZIBROWIUS, U. W.
LOHSE, K .
PRITZKOW and J.
SZULZEWSKY, H.
FICHTNER-SCHMITTLER,
RICHTER-MENDAU
C e n t r a l I n s t i t u t e o f P h y s i c a l Chemistry, Academy o f Sciences o f t h e GDR, Rudower Chaussee 5, B e r l i n , DDR - 1199 (German Democratic R e p u b l i c )
SUMMARY A l P O -14 was prepared u s i n g p i p e r i d i n e as t e m p l a t i n g agent. The
4
XRD and MAS NMR r e s u l t s i n d i c a t e s i g n i f i c a n t s t r u c t u r a l changes o f
t h i s aluminophosphate d u r i n g template removal and r e h y d r a t i o n . The l a t t i c e parameters o f t h e as-synthesized as w e l l as t h e c a l c i n e d and dehydrated form o f A l P O -14 were determined f r o m X-ray powder 4 d i f f r a c t i o n d a t a . The l a t t i c e parameters o f t h e c a l c i n e d samples correspond t o t h e i d e a l i z e d geometry suggested f o r A1P04-14, b u t which has n o t y e t been a c t u a l l y observed. From 31P MAS NMR t h e e x i s t e n c e o f a t l e a s t t h r e e non-equivalent phosphorus s i t e s i n t h e c a l c i n e d form o f A l P O -14 i s d e r i v e d . 4 INTRODUCTION
The s y n t h e s i s and c h a r a c t e r i z a t i o n o f microporous c r y s t a l l i n e aluminophosphates
(A1P04-n
d u r i n g t h e l a s t few years.
(ref.
1) )
has a t t r a c t e d much i n t e r e s t
Because o f t h e i r s t r u c t u r a l and compo-
A l P O -based m o l e c u l a r s i e v e s can be t a i l o r e d 4 t o many d i f f e r e n t processes. Only some o f t h e AlPO t y p e m a t e r i a l s 4 have framework t o p o l o g i e s i d e n t i c a l w i t h those o f known z e o l i t e s .
sitional
diversity,
Among t h e huge number o f newly s y n t h e s i z e d A1P04 t y p e s ,
there are
some m a t e r i a l s which have
their
patterns,
been p a t e n t e d a c c o r d i n g t o
XRD
b u t l i t t l e i s known about t h e i r s t r u c t u r e s and proper-
ties. The goal (ref.
of
the
by
XRD,
measurements.
The
2)
present
solid
study
state
structure
was
NMR
type
of
to
as w e l l
characterize as
A1P04-14
thermoanalytical
as-synthesized
A1P04-14
is
presumed t o be t h a t o f a g a l l i u m phosphate s y n t h e s i z e d by means o f iso-propylamine
(refs.
3,4).
A
t u r e has n o t y e t been p u b l i s h e d .
r e f i n e m e n t o f t h e A1P04-14
struc-
550 EXPERIMENTAL Sample p r e p a r a t i o n As d i s t i n c t from
patent ( r e f .
t h e U.S.
2),
the synthesis o f
A l P O -14 was performed n e i t h e r w i t h t - b u t y l a m i n e n o r w i t h i s o 4 propylamine, b u t w i t h p i p e r i d i n e as t h e t e m p l a t i n g agent. The f o l -
l o w i n g procedure was used: 13 g o f water, (85 mass%)
6.6 g o f p h o s p h o r i c a c i d
g o f pseudoboehmite ( 7 4 mass% A 1 2 0 3 ,
and 2.8
H 0 ) were mixed by v i g o r o u s s t i r r i n g . 2
After
addition
2 6 mass%
of
4 g
of
p i p e r i d i n e t h e f i n a l c o m p o s i t i o n o f t h e r e a c t i o n m i x t u r e was
.
. 1.4 P 0
1 A1203
2 5
2.3
(CH ) NH 2 5
'
40 H20.
The pH v a l u e o f t h e g e l was found t o be 3.6...4.5. l i z a t i o n was performed i n s t a i n l e s s p o l y t e t r a f l u o r e t h y l e n e a t 473 K f o r
steel
The c r y s t a l -
autoclaves
90 h.
lined with
After centrifugation,
washing and d r y i n g t h e sample c o m p o s i t i o n o f
the
as-synthesized
sample was determined t o be a p p r o x i m a t e l y 1 2 A1P04
3 (CH215NH
'
8 H20.
'
Concerning t h e chemical c o m p o s i t i o n o f t h e c a l c i n e d A l P O
crys4 r a t i o was found t o be 1 : 1 b o t h i n t h e b u l k phase
tals, the A l / P
(chemical a n a l y s i s ) and i n t h e s u r f a c e l a y e r s (XPS). Instrumental Crystal
habit
and
size
were
ascertained
using
a
scanning
e l e c t r o n microscope TESLA BS300 w i t h p l a t i n u m coated specimens. The X-ray
powder
diffraction
G u i n i e r f i l m camera (CuK a
recorded
using
a
r a d i a t i o n ) , a high-temperature G u i n i e r
Crl
camera and
p a t t e r n s were
diffractometer
with
Bragg-Brentano
geometry
(CuK
CI
radiation). The
thermal
analyses
Derivatograph supplied r a t e d w i t h water a t room
(DTA and TG) by MOM,
temperature
1073 K a t a r a t e o f 5 K/min.
were
Budapest.
carried
out
using
a
Samples (200 mg) s a t u -
were heated
f r o m 293 K
For DTA measurements, 0 - A 1
to
0 powder 2 3
was used as t h e r e f e r e n c e m a t e r i a l . The magic angle s p i n n i n g (MAS) NMR s p e c t r a were Bruker MSL 400 spectrometer.
recorded on a
The 2 7 A l NMR s p e c t r a were o b t a i n e d a t
a resonance frequency o f 104.2 MHz and a s p i n n i n g r a t e o f 5 . 3 kHz. S i n g l e p u l s e e x c i t a t i o n (SPE) was used.
The
w i t h n/12 p u l s e s
o f 0.6 p s
length
l i n e positions are reported r e l a t i v e t o A l C l
3
in
551 aqueous
solution.
No
corrections
w i t h a s p i n n i n g r a t e o f 4.8 kHz. necessary
to
second
The 31P NMR s p e c t r a were
e f f e c t s are applied. were
for
Relaxation
avoid s a t u r a t i o n
effects
order
quadrupole
t a k e n a t 162.0 MHz d e l a y s o f up t o 90 s i n t h e case o f
SPE
spectra.
Chemical s h i f t s a r e r e p o r t e d r e l a t i v e t o 85 mass% H PO 3 4' were o b t a i n e d a t 100.6 MHz w i t h a s p i n n i n g The 1 3 C NMR s p e c t r a r a t e o f 4.8
kHz.
4.4 p s l e n g t h w i t h a r e l a x a t i o n d e l a y o f 5 c a l s h i f t s are r e p o r t e d r e l a t i v e t o TMS.
pulses o f
s were used.
13C chemi-
Durene was used as secon-
= 19.2 ppm ( r e f . 5 ) ) . I n b o t h t h e
dary standard (6(CH3)
31 P cross-pol a r i z a t i o n ( C P ) experiments,
were a p p l i e d . A l l
n/2
For t h e s i n g l e p u l s e e x c i t a t i o n ,
13C
and
proton n / 2 pulses o f 5 p s
13C and 31P NMR s p e c t r a were
recorded a p p l y i n g
high-power p r o t o n decoupling. RESULTS AND DISCUSSION
X-ray powder d i f f r a c t i o n The scanning e l e c t r o n micrograph i n F i g . crystal
morphology
of
the
aluminophosphate
1 shows t h e t y p i c a l obtained.
Usually,
c r y s t a l s o f t a b u l a r h a b i t a r e observed w i t h dimensions up t o ca. 15 pm i n l e n g t h and 3 p m i n w i d t h .
F i g . 1. Scanning e l e c t r o n micrograph o f t h e A1P04-14 c r y s t a l s . The X-ray
powder
alurninophosphate
pattern
of
the
investigated ( c f .
as-synthesized
Fig.
form o f
the
2) can be indexed on t h e
b a s i s o f a t r i c l i n i c u n i t c e l l w i t h t h e f o l l o w i n g parameters:
552
2Thefa /degre esF i g . 2 . X-ray powder d i f f r a c t i o n p a t t e r n o f t h e as-synthesized form o f A1P04-14.
a b c
= =
9.697(2) 6 9.947(2) 10.612(2) A
74.92(2)' 75.14(2)' 89.40(2)'.
CI
p y
A3
V = 953.6
These l a t t i c e parameters a r e c l o s e l y r e l a t e d t o those o f a g a l l i u m phosphate g i v e n i n r e f .
3.
T h i s g a l l i u m phosphate r e f e r r e d t o as
GaPO - 1 4 has been s t a t e d t o correspond t o A l P O - 1 4 because o f t h e 4 4 s i m i l a r i t y o f i t s powder p a t t e r n w i t h t h e d i f f r a c t i o n d a t a l i s t e d i n t h e U.S.
patent ( r e f .
2).
The d e v i a t i o n s
between t h e powder
p a t t e r n o f o u r as-synthesized sample ( c f . Table 1 ) and X-ray
data
g i v e n i n r e f . 2 may be caused by t e m p l a t e i n f l u e n c e . L a t t i c e parameters and c r y s t a l framework
symmetry
structure
type.
allow a clear
The
unambiguously i d e n t i f i e d as A l P O - 1 4 4 data. After AlPO -14 4
removal o f
t h e template,
on t h e b a s i s o f
the calcinated
produced a s i n g l e phase X-ray
dehydrated form, temperatures
i.e.
i n air.
can
be
our
i n the f u l l y
i n dry nitrogen o r
N o changes i n
the
i t s crystal
sample o f
diagram o n l y
a t room temperature
above 310 K
identification of
aluminophosphate o b t a i n e d
at
t h e XRD p a t t e r n
553 TABLE 1 . ____
2 Theta
X-ray
powder d i f f r a c t i o n d a t a o f as-synthesized AlPO -14.
4
present d a t a hkl
(degrees )
8.93 9.21 11.00 11.17 12.80 13.29 14.46 14.63 15.63 17.85 17.92 18.14
18.27 18.71 18.84 18.95 20.01 20.59 21.17 21.31 21.45 21.61 22.09 22.43 22.61 22.87 23.04 23.50 25.46 25.77 25.92 26.12 26.19 27.04 27.32 27.40 27.61 27.89 28.06 28.25 28.36 28.61 28.82 28.96 29.10 29.40 29.56 30.06
_.
001 010 01 1 101 1
io
111
oii 1o
i
iii ,ii 1 01 2 002,102 112 02 1 20 1 1 ii 200 121 2io 121 120 2i 1 210 172 102,202 122,021 21 2 20i 211 113 220 12i 013 103,217 003 221 22 1 30 1 030 123 213 131 300
131
032,311 212 3i 1 302,3io 312,130
I/Io
(%I 100 27 6 13 75 35 16 11 95 6 6 4 2 10 15 12 3 14 10 35 10 10 62 37 25 48 5 2 20 8 11 15 15 10 4 2 2 9 2 4 23 5 5 5 5 28 3 30
d a t a from r e f . 2 2 Theta I/Io ( degrees) (%I
9.2 9.4 11.2
shou 1 der
100
13.1 13.4 14.8
shou 1 der
15.8 18.0
23 12
18.8 19.2
1 1
-
-
18 17 3
20.9 21.6
5 10
22.2
22
22.7
36
23.4 23.7 25.2 26.1
1 1 2 20
27.1
9
27.7
2
28.5
5
29.5
12
30.2
8
554
were observed up t o a temperature o f 4 7 3 K .
I t i s supposed t h a t a t
room temperature t h e c r y s t a l s t r u c t u r e i s always i n f l u e n c e d by t h e presence o f water o r i g i n a t i n g f r o m atmospheric h u m i d i t y . diagram o f t h e c a l c i n e d and dehydrated sample can be
The X-ray
attributed t o
crystals with
the
following
C-centred
monoclinic
unit cell :
= 14.059(6) A
a b c
= 13.480(8) A 10.229(4)
v = 1850.8
I; = 107.30(4)'
a.
a3
These l a t t i c e parameters correspond t o t h e i d e a l i z e d geometry o f t h e A l P O -14 s t r u c t u r e t y p e as d e s c r i b e d i n r e f . 4, b u t which have
4
n o t y e t been a c t u a l l y observed. The m o n o c l i n i c c e l l o f t h e c a l c i n e d A l P O -14 i s r e l a t e d t o t h e
4
u n i t c e l l o f t h e as-synthesized transformation o f
axial
the
form.
T h i s may be seen a f t e r
l a t t i c e parameters o f
m o n o c l i n i c c e l l t o t h a t o f a p r i m i t i v e t r i c l i n i c one a c c o r d i n g -t
a
+ %mon ) / 2 , +btr =
-t
=
+
(amon
-
+bmon)/2
.)
and ctr
+
= c
. mon
(amon tr r e s u l t i n g reduced t r i c l i n i c c e l l has t h e f o l l o w i n g parameters: a b c
= 9.739 A = 9.739 A = 10.229 6
an
t h e C-centred to The
a = 77.60'
0 Y
77.60' 87.59'.
We suppose t h a t s i m i a r t o t h e d i s t o r t on o f t h e i d e a l geometry
o f t h e framework by t h e t e m p l a t e i n t h e as-synthesized a d s o r p t i o n o f water
form,
the
leads t o one o r even more r e l a t e d s t r u c t u r e s
w i t h reduced symmetry. T h i s assumption would be i n l i n e w i t h t h e r e s u l t s o f t h e s o l i d s t a t e NMR f o r t h e c a l c i n e d and f u l l y rehydrat e d sample. Thermoanal y s i s and 13C MAS NMR The
removal
measurements.
of
the
template
was
I n t h e DTA c u r v e shown
studied
by
i n Fig.
thermoanalytical
3 , two endothermic
peaks ( a t 4 2 0 K and 648 K i n a i r ) and a s t r o n g e x o t h e r m i c peak above 800 served. Fig.
K w i t h a s h o u l d e r on t h e h i g h temperature s i d e a r e ob-
Four w e i g h t l o s s s t e p s may be seen i n t h e TG c u r v e ( c f .
3). Two o f these a r e a s s o c i a t e d w i t h t h e endothermic peaks.
I n c o r r e l a t i o n w i t h t h e asymmetric e x o t h e r m i c peak,
a fast,
fol-
lowed by a slow decrease o f sample w e i g h t occurs. The endothermic e f f e c t s a r e
due t o t h e
desorption
of
water;
555 t h i s f o l l o w s from t h e comparison o f t h e t h e r m o a n a l y t i c a l d a t a w i t h those
of
the
elementary
analysis.
Water
can
be
eliminated
vacuum, and t h e weight l o s s a t 773 K amounts t o 9 %. t r e a t m e n t t h e endothermic peaks The minor peak a t 403 the handling o f
p r a c t i c a l l y vanish
in
A f t e r vacuum ( c f . F i g . 3).
K i s due t o t h e r e a d s o r p t i o n o f water d u r i n g
t h e sample i n a i r
(before the
thermoanalytical
measurement). 820
’.....
T/K
.---
F i g . 3. DTA and TO curves o f A PO - 1 4 : ( a ) i n a i r , 4 a f t e r e v a c u a t i o n a t 773 K and c ) i n argon. The removal o f p i p e r i d i n e i a i r by o x i d a t i o n ,
i
(b) i n a i r
p o s s i b l e o n l y by d e g r a d a t i o n
i n argon by c r a c k i n g .
The shape o f
-
in
t h e DTA/TG
curves c o u l d be e x p l a i n e d assuming e i t h e r two p i p e r i d i n e s p e c i e s t o be p r e s e n t o r a two-step
behaviour f o r t h e d e g r a d a t i o n .
From
t h e s y n t h e s i s c o n d i t i o n s (pH o f t h e g e l : ca. 3.6, a f t e r c r y s t a l l i zation:
ca.
+
6.51, t h e f o r m a t i o n o f p i p e r i d i n i u m i o n s ((CH 2 ) 5 NH 2
i s probable ( r e f . 6 ) . On t h e o t h e r hand,
t h e presence o f n e u t r a l
p i p e r i d i n e molecules was d e r i v e d f r o m t h e appearance o f wagging v i b r a t i o n band (720
ern-')
the
i n t h e I R s p e c t r a ( r e f . 7).
NH
556
23.0
'3c 45.8
a
F i g . 4. 13C MAS NMR s p e c t r a o f t h e as-synthesized A1P04-14 obtained (a)
A I
I
I
L
I
'
40
60
I
'
20
-
by
cross-polarization
(T
,
=
mix 0.25 ms, 2400 scans) and ( b ) by single pulse excitation (600 scans )
.
0
PPm
To o b t a i n more i n f o r m a t i o n on t h e n a t u r e o f t h e occluded p i p e ridine,
we a p p l i e d 13C MAS NMR as another
1 3 C NMR s p e c t r a o f
an as-synthesized as
well
as
s p e c t r o s c o p i c method.
sample o b t a i n e d by s i n g l e cross-polarization
(CP)
are
pulse
excitation
(SPE)
shown
i n Fig.
The l i n e shapes o b t a i n e d by t h e two d i f f e r e n t
4.
techniques a r e almost i d e n t i c a l i n d i c a t i n g t h a t t h e CP r e s u l t s a r e r e l i a b l e w i t h r e s p e c t t o t h e r e l a t i v e i n t e n s i t i e s and t h e number of
lines
observed.
A
prolonged
mixing
time
led
to
a
strong
decrease i n t h e s i g n a l i n t e n s i t y . Above 5 ms no CP s i g n a l was obof 1P I n the
served i n d i c a t i n g a v e r y s h o r t l o n g i t u d i n a l r e l a x a t i o n t i m e T the
protons
l i q u i d phase,
involved the
a,
in
the
transfer
of
nance l i n e s w i t h t h e i n t e n s i t y r a t i o o f 2 and 2 5 . 3 22.5
magnetization.
and y carbons o f p i p e r i d i n e e x h i b i t
ppm and those o f
piperidinium
: 2
reso-
: 1 a t 47.6,
ions
a t 45.2,
27.4
22.7,
and
ppm ( r e f . 8 ) . Although t h e l a r g e l i n e w i d t h o b t a i n e d f o r t h e
occluded
piperidine
s i b l e , t h e observed
renders
a
conclusive
interpretation
l i n e p o s i t i o n s a t 45.8 and
impos-
23.0 ppm s u p p o r t
t h e assumption t h a t p i p e r i d i n i u m i o n s a r e p r e s e n t . The e n c a p s u l a t i o n o f p i p e r i d i n i u m i o n s i n t h e s t r u c t u r e r a i s e s I n t h e case o f A l P O - 1 7 which 4 was a l s o s y n t h e s i z e d by means o f p i p e r i d i n e , i t has been suggested t h e q u e s t i o n o f charge compensation.
t h a t t h e p i p e r i d i n e has " r e a c t e d w i t h a water molecule t o y i e l d a p i p e r i d i n i u m i o n and a h y d r o x y l ,
which t h e n
interacts with
the
557 aluminophosphate analogue o f
framework"
A1P04-14,
(ref.
the
9).
For t h e g a l l i u m phosphate
formation
of
iso-propylammonium
ions
d u r i n g t h e c r y s t a l l i z a t i o n process was d e r i v e d f r o m XRD d a t a ( r e f . 3 ) . The charge compensation i s accomplished by h y d r o x y l i o n s . and 31P MAS NMR
27Al
F u r t h e r i n f o r m a t i o n on changes d u r i n g template s o l i d s t a t e NMR. given i n F i g . 5 . AlPO -14
4 lines
removal and
r e h y d r a t i o n was o b t a i n e d by
coordinated
A1P04-14
In this
spectrum,
aluminium
1 0 ) . The l i n e a t ca.
there
are
intense
(34.0 ppm) and o c t a h e d r a l l y
in
aluminophosphates,
respec-
10 ppm c o u l d be caused by o c t a -
h e d r a l l y c o o r d i n a t e d aluminium i n an aluminate,
i.e.
by a minor
amount o f unreacted pseudoboehmi t e remaining a f t e r s y n t h e s i s , more
probably
by
are
MAS NMR spectrum o f t h e as-synthesized
complex.
i n t h e range o f t e t r a h e d r a l l y
tively (ref.
and s t r u c t u r a l
and 31P MAS NMR s p e c t r a o f
The 2 7 A l
i s very
ppm)
(-9.0
27Al
t h e framework s t r u c t u r e
aluminium
in
fivefold
coordination
in
or the
aluminophosphate framework. The l a t t e r e x p l a n a t i o n i s supported by the
fact
gallium
that atoms
a
fivefold
oxygen
was
found
the
in
coordination above
of
mentioned
part
GaP04-14 ( r e f . 3 ) . A f i v e f o l d o r t r i g o n a l - b i p y r a m i d a l o f aluminium was a l s o found i n t h e as-synthesized other
aluminophosphate
f o u r non-equivalent
materials
(refs.
9,111.
of
the
as-synthesized coordination
forms o f some
Furthermore,
the
g a l l i u m s i t e s w i t h d i f f e r e n t oxygen c o o r d i n a -
t i o n found i n GaPO -14 (Ga(1) i n t r i g o n a l - b i p y r a m i d a l , Ga(2) and 4 Ga(3) i n t e t r a h e d r a l and Ga(4) i n o c t a h e d r a l c o o r d i n a t i o n ( r e f . 3 ) ) would correspond t o
t h e v e r y complex
2
7 MAS ~ ~NMR
spectrum
o b t a i n e d f o r t h e as-synthesized A l P O -14. 4 C a l c i n a t i o n t r a n s f o r m s t h e oxygen c o o r d i n a t i o n o f a l l aluminium atoms i n t o a t e t r a h e d r a l one. The non-symmetric l i n e shape i n F i g . 5b
i s caused by
c o u p l i n g as w e l l phic sites. 27Al
incomplete
of
second-order
as t h e presence o f n o n - e q u i v a l e n t
T h i s was proven by measurements a t
MAS NMR s p e c t r a taken
attributed t o Measurements
removal
second-order at
even
at
78.2
MHz were
lower
magnetic
fields
and i s o t r o p i c chemical 12).
s h i f t s as
crystallogra-
lower f i e l d .
t o o complex
quadrupolar c o u p l i n g
e x t r a c t i o n o f quadrupole c o u p l i n g c o n s t a n t s ,
quadrupolar
alone should
The
t o be
(ref. 7). allow
the
asymmetry parameters
i n t h e case o f
A1P04-17
(ref.
558 -30.5
3’ P
-21.9
..* -17.3
- L - - L - L
100
50
0 -50
0
-20
PPm
-40
-GO
PPm
F i g . 5 . 2 7 A l and 31P MAS NMR s p e c t r a o f A l P O -14: 4 ( a + d ) as-synthesized sample, (b+e) c a l c i n e d and dehydrated sample and ( c + f ) c a l c i n e d and f u l l y h y d r a t e d sample. The a s t e r i s k s denote s p i n n i n g sidebands. A s i s found
f o r many o t h e r s t r u c t u r e s ( c f .
r e f . 131, rehydra-
t i o n generates an o c t a h e d r a l c o o r d i n a t i o n o f some o f t h e aluminium atoms.
Furthermore, t h e l i n e a t ca.
10 ppm i s a g a i n p r e s e n t .
Be-
cause o f t h e l a c k o f s t r u c t u r a l d a t a f o r t h e r e h y d r a t e d sample,
a
d e f i n i t e assignment o f t h i s l i n e t o f i v e f o l d c o o r d i n a t e d aluminium i s not justified.
I n t h e e v e n t t h a t t h i s l i n e i s caused by unreac-
t e d pseudoboehmite, understood environment
as
an
i t s disappearance d u r i n g c a l c i n a t i o n can effect
produced
by
i n v i s i b l e t o NMR ( r e f s .
of
dehydration.
d e h y d r a t i o n can
A
highly
render
be
asymmetric
aluminium
atoms
14,151.
The d o w n f i e l d s h i f t o f t h e l i n e assigned t o t e t r a h e d r a l l y o r d i n a t e d aluminium i n t h e
rehydrated
sample
i s striking.
coA t
a
resonance frequency o f 78.2 MHz t h i s l i n e i s o b t a i n e d a t 42.8 ppm. T h i s f i n d i n g i n d i c a t e s t h a t t h e s h i f t t o lower f i e l d i s o n l y t o a small
degree caused by a
reduced quadrupolar
coupling.
Using a
559 c o r r e l a t i o n d e r i v e d f o r aluminophosphates by M u l l e r and co-workers (ref.
1 6 1 , t h i s s h i f t would correspond t o a r e d u c t i o n o f t h e mean
angle under r e h y d r a t i o n o f up t o 10'.
A1-0-P
T h i s may a t l e a s t be
taken as an e s t i m a t e o f t h e e x t e n t o f t h e s t r u c t u r a l changes. Large changes i n t h e A l P O -14 framework d u r i n g sample c a l c i n a 4 t i o n and r e h y d r a t i o n a r e n o t o n l y c o r r o b o r a t e d by t h e above g i v e n XRD resu t s ,
b u t a r e a l s o r e f l e c t e d by 31P MAS NMR s p e c t r a . Three
l i n e s i n a chemical s h i f t range ( - 2 2 ristic o
...-31
aluminophosphates a r e observed
ppm) which i s c h a r a c t e i n t h e case o f t h e as-
synthesized f o r m ( c f . F i g . 5d). The r e l a t i v e i n t e n s i t i e s o f these l i n e s vary s l i g h t l y from sample t o sample.
I n a d d i t i o n , t h e 31P
MAS NMR spectrum o f t h e as-synthesized samples e x h i b i t s a l i n e a t
-7.5
ppm.
The
systematic (ref.
highest value
study
of
the
chemical
shift
o f aluminophosphate m i n e r a l s
found
is 6
in a
-1 1.2
ppm
i s an aluminophosphate-hydrate.
17) f o r w a v e l l i t e which
A
s p e c i f i c i n t e r a c t i o n between t h e t e m p l a t e molecules and t h e phosphorus o f t h e framework i s u n l i k e l y t o be t h e reason f o r t h i s unusual
l i n e position,
aluminophosphates agent ( r e f .
18).
since t h i s
synthesized
l i n e was n o t observed f o r o t h e r
by
means
It i s interesting
of
the
same
templating
t o note t h a t there are four
non-equivalent phosphorus s i t e s i n t h e framework o f t h e GaP04 analogue o f as-synthesized A l P O -14 ( r e f . 3 ) . 4 I n o r d e r t o g e t more i n s i g h t i n t o t h e s t r u c t u r a l environment o f t h e phosphorus n u c l e i g i v i n g r i s e t o t h e d i f f e r e n t
l i n e s i n the
31P MAS NMR spectrum o f t h e as-synthesized sample, we a p p l i e d t h e
cross-polarization
(CP) technique.
I n F i g . 6 , 31P CP/MAS NMR spec-
t r a f o r d i f f e r e n t m i x i n g t i m e s a r e shown i n t h e a b s o l u t e i n t e n s i t y mode. Using s h o r t m i x i n g t i m e s , one o b t a i n s s p e c t r a t h a t a r e q u i t e similar
to
those
recorded
by
single
pulse
excitation.
Conse-
q u e n t l y , a l l phosphorus n u c l e i have p r o t o n s i n t h e i r neighbourhood from which t h e m a g n e t i z a t i o n can be t r a n s f e r r e d . t i m e exceeds about 1 ms, a l l b u t t h e l i n e a t -30.5 cally
reduced.
time o f
about
The l a t t e r 5
ms.
protons ( r e f s . for
17,191,
l i n e s a t -7.6,
-21.9
mixing
CP s i g n a l
intensity
is
o f the 1P we have t o conclude t h a t t h e p r o t o n r e s e r -
cross-polarization
t h e remaining n u c l e i .
the
reaches maximum i n t e n s i t y a t a m i x i n g
Since t h e decay o f
m a i n l y determined by t h e l o n g i t u d i n a l
voir
If
ppm a r e d r a s t i -
relaxation time T
t o phosphorus n u c l e i w i t h
and -27.8
resonance
ppm i s d i f f e r e n t f r o m t h a t one f o r
560
31P CP/MAS
20
1.5
-20
0
- 60
-40
Ppm
Fig. 6.
31P CP/MAS NMR s p e c t r a o f t h e a s - s y n t h e s i z e d A l P O - 1 4 4 u s i n g d i f f e r e n t m i x i n g t i m e s : ( a ) 0 . 5 ms, ( b ) 2 . 0 m s and
( c ) 4 . 0 ms. The
3 1 P MAS NMR spectrum
shows t h r e e w e l l 2.0
: 1.1.
o f t h e c a l c i n e d sample ( c f . F i g .
resolved l i n e s w i t h the i n t e n s i t y r a t i o o f
:
Hence i t f o l l o w s t h a t t h i s s t r u c t u r e has a t l e a s t t h r e e
c r y s t a l l o g r a p h i c non-equivalent four
5e) 1
non-equivalent
phosphorus
phosphorus s i t e s . sites
as
in
the
I n t h e case o f as-synthesized
GaPO - 1 4 ( r e f . 31, two o f them s h o u l d have a v e r y s i m i l a r geometry 4 l e a d i n g t o a c o i n c i d e n c e o f resonance l i n e s . A s expected f o r t h e dehydrated sample,
the cross-polarization
efficiency
for
all
the
t h r e e 31P resonance l i n e s i n F i g . 5e i s v e r y low. A broad l i n e a t about -18
ppm a d d i t i o n a l l y observed u s i n g CP i s a t t r i b u t e d t o POH
p r e s e n t as d e f e c t s i t e s o f t h e s t r u c t u r e o r amorphous p r e c u r s o r s .
i n minor amounts o f
561
After tained.
rehydration,
an
ill r e s o l v e d 31P
NMR
spectrum
i s ob-
Since t h e same r e l a t i o n between chemical s h i f t and mean angle was found
A1-0-P
f o r both 2 7 A l
and 31P NMR ( r e f .
161, the
d o w n f i e l d s h i f t o f t h e spectrum would correspond t o t h e
overall
d e r i v e d r e d u c t i o n o f mean A1-0-P was shown t h a t a l l n u c l e i
angle.
By c r o s s - p o l a r i z a t i o n
it
i n t e r a c t strongly w i t h the protons o f
adsorbed water molecules ( r e f .
7).
T h i s f a c t must be t a k e n i n t o
account f o r any q u a n t i t a t i v e i n t e r p r e t a t i o n o f t h e d o w n f i e l d s h i f t o f t h e 31P NMR spectrum d u r i n g r e h y d r a t i o n . The changes i n t h e 2 7 A l
and 3 1 P MAS NMR s p e c t r a caused by rehy-
d r a t i o n a r e completely r e v e r s i b l e . A l s o , t h e a d s o r p t i o n p r o p e r t i e s ( c a p a c i t i e s and k i n e t i c s ) a r e n o t a l t e r e d by t i o n cycles ( r e f .
7 ) . Consequently,
dehydration/rehydra-
we have t o conclude t h a t t h e
s t r u c t u r a l changes caused by a d s o r p t i o n o f water a r e indeed r e v e r s i b l e processes. The shape o f t h e observed water i s o t h e r m t o g e t h e r with
t h e very small uptake r a t e
physical
a d s o r p t i o n process.
( r e f . 20),
contradicts
The f i r s t water
a
pure
molecules adsorbed
are l i k e l y t o r e a c t w i t h t h e aluminophosphate and cause r e v e r s i b l e structural
changes.
can be expected.
The f o r m a t i o n o f
an aluminophosphate-hydrate
Combined XRD and s p e c t r o s c o p i c
investigations a t
d e f i n e d water l o a d i n g l e v e l s would be necessary t o e l u c i d a t e t h e i n t e r a c t i o n o f water w i t h t h e aluminophosphate framework. CONCLUSIONS Using p i p e r i d i n e as t e m p l a t i n g agent, small
crystallinity. guous
we have s y n t h e s i z e d t h e
pore molecular
s i e v e A l P O -14 as a pure phase w i t h h i g h 4 On t h e b a s i s o f t h e presented XRD d a t a , an unambi-
assignment
is
possible.
The
topological
equivalence
of
GaPO -14 and A l P O -14 i s confirmed. Furthermore, t h e l a t t i c e para4 4 meters o f t h e c a l c i n e d and dehydrated f o r m correspond t o t h e idealized
geometry
suggested
for
A l P O -14, 4
but
which
has
that
there
not
Yet
a c t u a l l y observed. From 31P
MAS
NMR
spectroscopy
l e a s t t h r e e non-equivalent
it follows
phosphorus s i t e s
are
at
i n t h e framework
of
t h e c a l c i n e d f o r m o f A1P04-14. The f o r m a t i o n o f
occluded
piperidinium
ions
during the
t h e s i s i s supposed on t h e b a s i s o f t h e r m o a n a l y t i c a l c o r r o b o r a t e d by 1 3 C MAS NMR.
syn-
d a t a and i s
562 The XRD p a t t e r n s as w e l l as t h e s o l i d s t a t e NMR s p e c t r a i n d i cate
significant
hydration. framework
structural
To e l u c i d a t e with
the
water
the
changes
during
interaction of
molecules
and
calcination the
their
and
re-
aluminophosphate influence
on
the
s t r u c t u r e i s a challenge f o r f u r t h e r i n v e s t i g a t i o n s . ACKNOWLEDGEMENT We a r e i n d e b t e d t o D r . M.
Hunger ( L e i p z i g ) f o r r e c o r d i n g
27Al MAS NMR s p e c t r a a t lower f i e l d . REFERENCES
1 S.T. E.M. 2 S.T.
Wilson, B.M. Lok, C.A. Messina, T.R. Cannan and F l a n i g e n , J. Am. Chem. SOC., 104 (1982) 1146-1147. Wilson, B.M. Lok, and E.M. F l a n i g e n , U.S. P a t e n t
4,310,440 (1982). 3 J.B. P a r i s e , J . Chem. SOC., Chem. Commun., (1985) 606-607. 4 J.M. Bennett, W . J . D y t r y c h , J . J . P l u t h , J.W. Richardson and J.V. Smith, Z e o l i t e s , 6 (1986) 349-361. 5 H . - 0 . K a l i n o w s k i , S. Berger and S. Braun, 13C NMR-Spektroskopie, Georg Thieme Verlag, S t u t t g a r t , 1984, p. 141. 6 E.M. F l a n i g e n , R . L . P a t t o n and S.T. Wilson, i n : P.J. Grobet, W.J. M o r t i e r , E.F. Vansant and G. S c h u l z - E c k l o f f ( E d s . ) , I n n o v a t i o n i n Z e o l i t e M a t e r i a l Science, E l s e v i e r , Amsterdam, 1988, pp. 13-28. 7 B. Z i b r o w i u s , U. Lohse, E. A l s d o r f and J . Richter-Mendau, J. Chem. SOC., Faraday Trans., i n p r e s s . 8 H . - 0 . K a l i n o w s k i , S. Berger and S. Braun, 13C NMR-Spektros k o p i e , Georg Thieme Verlag, S t u t t g a r t , 1984, p. 323. 9 J . J . P l u t h , J.V. Smith and J.M. B e n n e t t , A c t a C r y s t a l l o g r . ,
C42 (1986) 283-286. 10 D. M u l l e r , I.Grunze, E. H a l l a s and G. Ladwig, 2 . anorg. a l l g . Chem., 500 (1983) 80-88. 1 1 J.B. P a r i s e and C . S . Day, A c t a C r y s t a l l o g r . , C41 (1985) 515520. 12 U. Lohse, B. Z i b r o w i u s and J. Richter-Mendau, P o s t e r p r e s e n t e d a t ZEOCAT 90 13 M. Goepper, F. Guth, L. Delmotte, J . L . Guth and H. K e s s l e r , 14 15 16 17
i n : P.A. Jacobs and R . A . van Santen ( E d s . ) , Z e o l i t e s : F a c t s , F i g u r e s , F u t u r e , E l s e v i e r , Amsterdam, 1989, pp. 851-866, K.J.D. MacKenzie, I . W . M . Brown, R.H. Meinhold and M.E. Bowden, J. Am. Chem. SOC., 6 8 (1985) 293-297. G. Blumenthal, G. Wegner, D. M u l l e r , A. Samoson and G. Kranz, 2 . anorg. a l l g . Chem., 576 (1989) 43-53. D. M u l l e r , E. Jahn, G. Ladwig and U. Haubenreisser, Chem. Phys. L e t t . , 109 (1984) 332-336. W.F. Bleam, P.E. P f e f f e r and J.S. Frye, Phys. Chem. M i n e r a l s
16 (1989) 455-464. 18 B. Z i b r o w i u s and U. Lohse, u n p u b l i s h e d r e s u l t s . 19 T.M. Duncan and C . Dybowski, S u r f . S c i . Rep., 1 (1981) 157-250. 20 U. Lohse, M. Noack and E. Jahn, Ads. S c i . Techn., 3 (1986) 19-24.
G . Ohlrnann et al. (Editors),Catalysis and Adsorption by Zeolites 1991 Elsevier Science Publishers B.V., Amsterdam
563
Y-ZEOLITE TREATED WITH SiCl, VAPOUR. STRUCTURE AND PROPERTIES
G.M. TELBIZ’, A.I. PRILIPKO’ and I.V. MISHIN‘
’The L.V. Pisarzevski Institute of Physical Chemistry, Ukrainian Academy of Sciences, Nauki Pr., 31, 252028, Kiev, USSR
‘The N. D. Zelinksi Institute of Organic Chemistry, Academy of Sciences USSR, Leninski Pr., 47, 117913, MOSCOW, USSR
SUMMARY
A series of dealuminated HY zeolites was prepared by reacting NaY zeolites with SiC1, at elevated temperatures. High crystalline products were obtained with unusual distributions of Lewis and Bronsted acidity and with anticipated catalytic properties. Infrared and m.a.s.n.m.r. data demonstrated that the chemical properties of the A1 in the zeolite lattice were not homogenous and characterized various types of aluminium sites. A Dempsey model was used to interpret the data obtained.
INTRODUCTION
Aluminium-deficient zeolites are an important class of material because of their increased chemical and thermal stability, in comparison to normal HY zeolites. They have an appropriate combination of pore size and selectivity which is induced by their high silicon content. The dealumination can be obtained through the use of steam, by chelating agents (ref. 1) or by the treatment of the zeolites with silicon tetrachloride (Beyer et al. , 1980). The relationship between the strength of Bronsted acid sites and the distribution of the aluminium framework is a basic issue in acidity faujasite-like zeolites. In 1974, Dempsey (ref. 2) , in an at-
564
tempt t o i n t e r p r e t e x p e r i m e n t a l d a t a of Bartomeuf and Beamount, proposed a model i n which t h e a c i d s t r e n g t h of z e o l i t e s was r e l a t e d t o t h e environment of t h e A 1 atoms, i . e . , t h e s t r e n g t h o f t h e a c i d s i t e s was r e l a t e d t o t h e number of aluminium atoms w i t h 0 , 1, o r 2 d i a g o n a l n e i g h b o r a l u m i n i u m a t o m s i n t h e r i n g s o f t h e s o d a l i t e cage. Beagley ( r e f . 3 ) produced a review and d i s c u s s i o n of t h e r e l a t i o n s h i p between z e o l i t e a c i d i t y and aluminium d i s t r i b u t i o n i n f a u j a s i t e - l i k e z e o l i t e s . On t h e o t h e r hand, d e a l u m i n a t i o n of t h e z e o l i t e latticemaybeconnectedtotheappearanceofnon-skeletalaluminium s p e c i e s . I n d i r e c t d e d u c t i o n s have beenmade u s i n g , b u t many a s p e c t s remain t o be c l a r i f i e d . The p r e s e n t work is p a r t of t h e c o n t i n u i n g work i n t o t h e i n f l u e n c e of surfacestructureontheacidityandcatalyticpropertiesofzeolites. I n t h i s workwe, p a r t i c u l a r l y , p r e s e n t t h e r e s u l t s o f a d e t a i l e d examinationofdealuminatedHYzeoliteswhichwerepreparedbyreactingNaY zeoliteswithSiC1,vapor.
METHODS
NaY z e o l i t e s ( S i / A 1 = 2 . 4 5 ) w e r e d e a l u m i n a t e d b y r e a c t i o n w i t h a flowing S i C 1 , vapour, a c c o r d i n g t o a procedure described elsewhere ( r e f . 4 ) . Samples, having a range of S i / A 1 r a t i o s , were p r e p a r e d by r e a c t i n g 30 g of t h e p r e t r e a t m e n t z e o l i t e w i t h S i C l , / N , a t t e m p e r a t u r e s r a n g i n g from430-723 K f o r 4 0 m i n . The z e o l i t e s w e r e t h e n w a s h e d w i t h d e i o n i z e d w a t e r a n d ionexchanged inlMNH,NO,. A l l t h e z e o l i t e e x h i b i t e d g o o d c r y s t a l l i n i t y a s w a s p r o v e d b y X-
r a y d i f f r a c t i o n a n d t h e i r d e s o r p t i o n c a p a c i t i e s . The I . R . s p e c t r a of s k e l e t a l v i b r a t i o n s , OH-groups a n d p y r i d i n e a d s o r p t i o n w e r e r u n o n a Specord 1 . R . - 7 5 spectrophotometer u s i n g s e l f - s u p p o r t i n g z e o l i t e p l a t e s . The s p e c t r a d e p i c t e d c o r r e s p o n d e d t o a sample t h i c k n e s s of 3 mg/cm2. The samples were evacuated t o about lo-‘ Pa a t 7 2 3 K , p r i o r t o measuring t h e s p e c t r a . The “ S i and 27Al s p e c t r a w e r e r e c o r d e d on a Brucker CXR-200 s p e c t r o m e t e r w i t h t h e magic a n g l e s p i n n i n g . A 1 ( H 2 0 ) 2 * c o m p l e x w a s u s e d a s a n internalstandardandsamplesweremeasuredin theirhydratedstate.Atemperature-programmedsurfacemethod (TPSR) ( r e f . 5)wasusedtoinvestigatethebehaviourofmethanolpreadsorbed onzeolites.
565
To i d e n t i f y t h e p r o d u c t s formedduringtheexperiment, t h e m a s s range10-70m/ewasexaminedat15Kintervals. D e s o r p t i o n w a s c a r r i e d outinvacuosothatthedesorbingspecies c o u l d b e d i r e c t l y o b s e r v e d u s i n g a m o d i f i e d m a s s spectrometerM1-201. Methanolwas p u r i f i e d by r e p e a t e d l y f r e e z i n g and pumping. Methanol vapour a d s o r b e d a t room temperature.
RESULTBANDDISCUSSION
Structure
T h e Z e o l i t e p r e p a r e d w i th . S i C l , , h a d a f r a r n e w o r k S i / A l r a n g i n g from 4.3 t 0 4 2 . Itwasarguedthatmid-infrared s p e c t r o s c o p y m i g h t y i e l d
CM" F i g . 1. Mid. i . r . spectraofnormalHYanddealuminatedsamples: 1. HY-(2.5) ; 2 . DY-(4.3); 3. DY-(4.8); 4 . DY-(7.3) ; 5. D Y - ( 4 2 ) .
informationabouttheframeworkchangesduringthedealuminationprocess. The infraredspectraofdealuminatedYarecompared i n F i g . 1. F r o m t h e s p e c t r a it i s e v i d e n t t h a t a l l t h e f r a m e w o r k v i b r a t i o n a l modes a r e s h i f t e d upon d e a l u m i n a t i o n (&V= a b o u t 50 cm-‘) when t h e S i / A 1 r a t i o is i n c r e a s e d t h e medium bands a t 682 and 6 7 0 cm-’ a p p e a r . N o v i s i b l e d e s t r u c t i o n o f t h e frameworkduetodealumination isobserved. Dealumination w i t h S i C 1 4 l e a d s t o t h e f o r m a t i o n o f non-framework aluminium s p e c i e s i n t h e z e o l i t e p o r e s . T h i s is shown by t h e “ S i and 2 7 A l m.a.s.n.m.r. s p e c t r a of t h i s m a t e r i a l (see F i g s . 2 A , B ) . The A 1 m.a.s.n.m.r. spectrashowtwodistinctkindsofsignals- s h a r p p e a k s ( a t 5 6 ppm) d u e t o t e t r a h e d r a l l y c o o r d i n a t e d A 1 i n t h e z e o l i t e frame-
.
-90
-100
-110 PPM
F i g . 2 . S o l i d s t a t e 2 7 A l ( A ) and 29Si ( B) m.a.s.n.m.r. s p e c t r a of normal HYanddealuminatedsamples:l.HY-(2.5);2.DY-(4.3);3.~~-(4.8);4. DY-(7.3);
5. D Y - ( 4 2 ) .
567
work and r e l a t i v e l y s h a r p s i g n a l s ( o c t a h e d r a l l y A l ) a t 2 . 0 ppm. The
relativeamountsofthesealuminiumspeciesdependuponthecondition of d e a l u m i n a t i o n . An examination of “Si m.a.s.n.m.r. spectra faujasitesdealuminatedwithSiClcshowsthattheSi(3A1)s i g n a l i s a l r e a d y a b s e n t i n sample 2 . I n t h i s c a s e t h e S i ( 2 A l ) a n d S i ( l A 1 ) s i g n a l s a l s o d e c r e a s e a n d S i ( O A 1 ) predominates i n t h e s p e c t r a . OH-UrOUDS
I . R . spectra intheOH-stretchingregionofy zeolites,dehydrated
a n d d e a m m o n i a t e d a t 7 2 3 K, are r e p r e s e n t e d i n F i g . 3A.
\
3700 A
3500
CM-~
1700
1500
B
F i g . 3. I . R . s p e c t r a i n t h e regionoftheOH-stretchingnumber ( A ) : t o ( - ) a n d a f t e r (---) p y r i d i n e a d s o r p t i o n (B) O f n o r m a l H Y a n d d e a l u minated samples: 1. HY-(2.5) : 2 . DY-(4.3) : 3. DY-(4.8) : 4 . DY-(7.3) : 5. D Y - ( 4 2 ) .
I n c l o s e a g r e e m e n t w i t h t h e l i t e r a t u r e c o n c e r n i n g H f o r m s o f zeol i t e s o b t a i n e d through NH; exchange, I . R . bands are observed a t 3640 and 3540 cm-‘, a t t r i b u t e d t o t h e hydroxyl i n t h e l a r g e and s m a l l c a g e s , c o r r e s p o n d i n g l y . DifferencesintheI.R.spectraofhydroxyldea1uminatedwithSiCl4productsprovidedevidenceoftheessentialinfluence ofthemodificationprocedure onthepropertiesofthesegroups. The band a t 3 6 4 0 cm-’ s h i f t e d t o c a . 3630 cm-‘ f o r S i / A 1 = 4 . 8 and t o c a . 3 6 2 0 cm-’ f o r S i / A l > 4 . 8 . The i n t e n s i t y of t h e I . R . band changed w i t h increasingSi/Alratio.Newunspecifiedhydroxylvibrationsappeared a t 3610 cm-’ and 3510 cm*’. Slow d e c r e a s e of i n t e n s i t y of t h e I . R . bands accompanied by t h e appearance and r a p i d i n c r e a s e o f a band a t 3 7 2 0 cm-’ w e r e observed. T h e n a t u r e o f t h e a c i d s i t e s a n d t h e i r r e l a t i v e strengthswere s t u d i e d by means of i n f r a r e d s p e c t r a f o l l o w i n g p y r i d i n e a d s o r p t i o n . Pyridinewasadsorbedandthenevacuatedat423 K.Thebandsduetothe c h e m i s o r b e d p y r i d i n e a r e shown i n F i g . 3 . U n e x p e c t e d r e s u l t s f o r a l l dealuminated samples were o b t a i n e d . I n c o n t r a s t w i t h t h e l i t e r a t u r e d a t a f o r H forms of f a u j a s i t e - l i k e z e o l i t e s , a d s o r p t i o n of P y r i d i n e c a u s e s a s h a r p d e c r e a s e i n t h e i n t e n s i t y of t h e OH-bands a t 3 5 4 0 cm-’. Simultaneously, a s s e e n i n t h e s p e c t r a , bands a t t r i b u t e d t o t h e C-C s t r e t c h i n g v i b r a t i o n o f t h e p y r i d i n i u m i o n (1550 cm-’) a n d c o o r d i n a t i v e bonded p y r i d i n e complex ( 1 4 5 0 cm-’) were observed. Besides, thermal s t a b i l i t y t h e s e complexes i n c r e a s e w i t h i n c r e a s i n g S i / A 1 r a t i o . According t o Belanski thermal s t a b i l i t y can be used a s a measure o f t h e acidstrengthof zeolites. Moreover,basicstrengthhasbeenstudiedinzeolitesdealuminated w i t h S i C 1 , by u s i n g p y r r o l e a d s o r p t i o n . A s h i f t i n t h e NH s t r e t c h i n g bandofadsorbedpyrroleandthepresenceo€comb~nat~onbandsat2940 and 2850 cm-’ were observed ( r e f . 6 ) .
TPSRofmethanol InordertomoreclearlyunderstandthepropertiesofthedealuminatedYzeolitesweinvestigatedthetemperature-progra~edsurface r e a c t i o n of methanol preadsorbed a t ambient t e m p e r a t u r e . The TPSR spectrafortheproductsofthemethanol-zeoliteinteractionareshown i n Fig. 4A.
569
The r e s u l t s showthatmethanoldesorbing f r o m t h e samples g i v e s a
peakcenteredatapproximately400K, dimethyletherat500Kandole-
f i n s (especiallyetheneandpropene) a t 6 0 0 K . N o t e t h a t t h e m a x i m a o f t h e o l e f i n s ' d e s o r p t i o n s h i f t e d i n t h e h i g h e s t t e m p e r a t u r e r e g i o n as
theSi/Alratioincreased.Negligibleamountsofaromaticspeciesare observedat700K. I n Fig. 4 B t h e m a x ~ m a o f t h e r e l e a s e d p r o d u c t s a r e p l o t t e d a g a i n s t t h e molar f r a c t i o n of aluminium. With i n c r e a s i n g d e a l u m i n a t i o n t h e amounts of methanol and dimethyl e t h e r a r e d e c r e a s e d , w h i l e o l e f i n s mainlydesorbfromthemoderatelydealum~natedzeol~tes.Sincemetha-
B
A
C
a
+ x 5
-7
473
673
T, K
473
673
T, K
0
0.1
e
0.2 AL
F i g . 4 . TPSR s p e c t r a of methanol (A) DY-(4.3) : (B) DY-(7.3) :
(C)depedence ontheheightofthepeakofdesorbedproducts (&methanol ( m / e = 3 1 ) : (+) dimethyl e t h e r (m/e=45) : ( 0 ) o l e f i n s ( m / e = 4 1 ) on t h e aluminium mole f r a c t i o n normal HY and dealuminatedy.
570
n o l d o e s n o t evacuate e a s i l y i n a n u l t r a h i g h v a c u u m , w e b e l i e v e t h a t theseolefinsareformedbythereactionbackgroundgasespassingover thezeoliteathightemperaturesduringTPSR.~sit~infraredspectra of t h e methanol dealuminated samples r e i n f o r c e t h i s c o n c l u s i o n ( r e f .7). Thus, f r o m t h e a b o v e r e s u l t s , itcanbeconcludedthatinthecaseof S i C l , t r e a t m e n t , h i g h c r y s t a l l i n e p r o d u c t s are o b t a i n e d w i t h u n u s u a l distributionsofLewisandBronstedacidityandwithanticipatedcatal y t i c p r o p e r t i e s . I.R. and m.a.s.n.m.r. d a t a demonstrated t h a t t h e chemical p r o p e r t i e s of A 1 i n t h e z e o l i t e l a t t i c e are n o t homogenous and c h a r a c t e r i z e ( v a r i o u s ) t y p e s of aluminium sites. I n agreement w i t h Kubelkova e t a l . ( r e f . 8 ) , it seems r e a s o n a b l e t o assume t h a t t h e cationicaluminiumspeciesgenerallyprovidetheAlelectronaccepting centers.Thequestionremainsoftheoverlookedfactoftheinteraction of p y r i d i n e w i t h low f r e q u e n c e h y d r o x y l s a t 3 5 4 0 cm-’. I t seems t o be i m p o s s i b l e , when t a k i n g i n t o account t h e p e c u l i a r s t r u c t u r e o f t h e f a u ja s i t e . Ontheotherhand,itisevident,thatextra-framework-typealumina affectthepropertiesofYzeolites.Thedea1uminationprocessenhancedtheacidityofthe zeolitesbutthismay occureitherviathenonframeworkspeciesorbygeneratinga z e o l i t e i n w h i c h m o s t o f t h e a l u minahavenonearestneighborsalumina. I t i s a l s o c o n s i s t e n t w i t h t h e conceptofapreferredaluminiumdistributionsimilartotheDempsey model. Thebasicityofoxy-compoundscanalsobediscussed i n t e r m s o f t h e dataobtained.ComparedwithnormalHY,dealuminatedzeolitesexhibit anacido-basiccharacterandthusconsitutearelatedsystemwithnew p r o p e r t i e s . I n summary, w e b e l i e v e t h a t t h e r e s u l t s p r e s e n t e d h e r e g i v e a b e t t e r u n d e r s t a n d i n g o f t h e d u a l c h a r a c t e r of z e o l i t e s ( r e f . 9). F u r t h e r s t u d i e s on t h e n a t u r e of a c i d o - b a s i c s i t e s and t h e c a t a l y t i c propertiesofdealuminatedSiCl4Yzeolitesareinprogress.
AC KNOW LEDOMENT
TheauthorsexpresstheirappreciationtoDr.V.V. B r e i ( I n s t i t u t e ofSurfaceChemistryUkrainianAcademyofSciences,Kiev)f o r h i s k i n d presentationofthem.a.s.n.m.r.data.
571
REFERENCES
1. J. S h e r z e r , Octane-enhancing, z e o l i t e F C C c a t a l y s t s : S c i e n t i f i c a n d t e c h n i c a l a s p e c t s , Catal.Rev.Sci.Eng.,
31(3) (1989) 215-354.
2. E. Dempsey, Acid S t r e n g t h and aluminium s i t e r e a c t i v i t y o f Y zeo-
l i t e s , J . C a t a l . , 33(3) (1974) 497-499. 3. B. Beagly, J. Dwyer, F.R. F i t c h , R. M a n n a n d J . Walters, Aluminium d i s t r i b u t i o n and p r o p e r t i e s o f f a u j a s i t e s . B a s i s of models and z e o l i t e a c i d i t y , J . Phys.Chem., 88(4) (1984) 1744-1751.
4. I . V .
Beyer, A.L.
Mishin, G.K.
Klyachko, G.A.
Ashavskaja, V.D.
Nissenbaum and G. Borbely, R e c e i p t of h i g h s i l i c a f a u j a s i t e w i t h S i c l , , Kin. i k a t .
, 28(3) (1987) 706-711.
5. A . I . P r i l i p k o . G.M. T e l b i z , V.G. I l i n , V.P. Shabelnikov, E.N. Korol,
Somepeculiaritiesofdesorptionofthe z e o l i t e i n t h e T P S R c o n d i t i o n , Dokl. ANUkr. SSR, s e r B , 3 (1989) 55-57. 6. G.M.
T e l b i z , A.V. S v e t z , s u b m i t t e d f o r p u b l i c a t i o n .
7. G.M. T e l b i z , V . V . B r e i , i n p r e p a r a t i o n . 8. L. Kubelkova,S. Beran, A . M a l e c k a , . V . M a s t i k h i n , A c i d i t y o f m o d i -
fiedzeolite.EffectofnonskeletalAlformedbyhydrothe~a1treatment, d e a l u m i n a t i o n w i t h SiC1, and c a t i o n exchange w i t h A l l Zeol i t e s , 9(1) (1989) 12-17. 9. D. Bartomeuf, C o n j u g a t e a c i d - b a s e p a i r s
88(1) (1984) 42-45.
i n z e o l i t e s , J. Phys.Chem.
,
This Page Intentionally Left Blank
G. Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
CHARACTERIZATION AND NUCLEATION DIFFERENT ALUMINIUM CONTENT G. Golemme',
A.
Nastro',
OF
573
Na,TPA-ZSM-5
B.Nagy2, B. SubotiA',
J.
ZEOLITE WITH
F. Crea' and
R. Aiello'. lDipartimento di Rende (Italy).
Chimica,
UniversitA
della
1-87036
Calabria.
2Laboratoire de Catalyse, Facult& Universitaires Notre-Dame de la Paix. 61, Rue de Bruxelles, 0-5000 Namur (Belgium). 3Ruder BoskoviC Institute. P . O . (Yugoslavia).
Box 1016, 41001 Zagreb, Croatia
ABSTRACT The formation of zeolite ZSM-5 is studied from five different gels having various A1 content. The crystallization kinetics, the nucleation kinetics, the pH of mother 1 iqiiors, the physical state and the composition of the solid phases are examined by combined physico-chemical techniques: SEM, X-ray Diffraction, Atomic Absorption, Elemental analysis and Thermal Analysis (TG, DTG, DTA). INTRODUCTION Zeolite ZSM-5 (see ref. 1)
is widespread
in the petrochemical
industry as a catalyst for various processes. Because of its noteworthy properties this zeolite has been widely investigated, and the numerous
information gathered
makes ZSM-5 a good
test current theories on zeolite nucleation
model
to
(see refs. 2,3) and
growth (see refs. 4-12). The object of the present work is to characterize the products and derive
nucleation kinetic curves of
Na,TPA-ZSM-5 zeolite in five systems differing by the aluminium content in the starting gel. The products are characterized
in
different ways: the pH of the mother liquors is measured, and the sol id phases during the crystal 1 ization are analyzed and examined by TGA and chemical analysis methods. EXPERIMENTAL Zeolites ZSM-5 were synthesized hydrothermally in static conditions at 170f2'C under autogenous pressure from systems having molar
ratios
0.177,0.289,
5Na,0-8.8(TPA),0-xA1,03-100Si02-1250H20,
with
x=O. 125,
0.626, 0.987,for runs A, B, C , D and E, respective-
ly. Reagents were carefully admixed in the order: Al(OII), (Serva
'reinst'
grade)
with
composition
AI,OR.4.6iOFI,O
gel dry (TBA.
574 12OO'C) ; freshly
prepared
98.6%, A1 2ppm) ; n-Pr,NBr
30% NaOH soln.
(Baker Analyzed,
(Fluka, purum grade);
NaOH
distilled H,O; pre-
cipitated SiO, (BHD) having composition SiO,. 0.2951120. 0.0049NaZ0. 0.00116A1,03 (TGA, 1200'C, AAS). One part of the gel was saved for further characterization. The remaining part was divided in different portions, transferred to
50 cm3 PTFE autoclaves and immediately put into the preheated oven. The autoclaves were taken from the oven at various intervals of time, one by one, and
immediately quenched under cold water.
The pll of the mother liquors was measured using Lyphan strips and the solid phases were filtered and dried at 100°C overnight. The temperature variation inside the teflon lined steel autoclaves was followed, and time zero for the crystallization was assumed as the moment when the reacting gel reached a temperature of 160-165'C. Crystallinity was measured by X-ray diffractometry. The Na and the A1
content of the
solid
Absorption Spectroscopy (AAS)
phases were
.
determined
by
Atomic
Thermogravimetric measurements and
elemental analyses were carried out to determine the TPA'
content
of the crystalline phases. Granulometric distributions for each kinetic series were obtained from micrographs of the final, not sonicated samples (see ref. 13).
11.5
11.1
10.7
'
I
I
I
1
I
I
1'0
I
I
I
210
I
I
t/h Fig. 1. Variation of the pH of the mother liquors as a function of time.
RESULTS The pH values of mother
liquors immediately after the cooling in Fig. 1 f o r the fastest and the
of the autoclaves are shown
slowest kinetics, C and E respectively. The pH vs. time curves show a sharp increase when the amorphous material
is completely
consumed. The pI1 decreases during the nucleation, stabilizes during the growth of the crystals, and rises again when the crystallization over.
is
This
behaviour
is
consistcnt
with
multiple equilibria between the solid phases and
crystals)
and the
(a1umino)silicate
the
presence
(amorphous
species
of
polymer
in solution, of
the kind shown below:
I I
I I
-~i-o-Si-
+
0118
I I
2 -Si-OII
(amorph .)
+
(aq. 1
'O-Si-
I I
(aq. 1
I I 7 -si-o-si- + OH^ I I (cryst.)
When the gel i s present, its solubility determines the amount of soluble species; when,
instead, the
crystallization
process
is
over, the concentration of si 1 icate species drops suddenly because it is determined by the solubility of the zeolite, and the pll rises accordingly. The pll vs. time curves nicely recall the desupersaturation profiles derived by the theory of the solution mediated phase transformations (see ref. 1 4 ) . The ZSM-5 crystals show regular squared shapes f o r a low A1 content, and get rounded when its amount increases. The length/width ratio (crystallographic
axes
c
and a)
is always equal t o 1 ,
with
the exception of a small fraction of crystals in the series of synthesis B (9%) and A (2.5%) (ratio up t o 2.3). The maximum diameters of the crystals at the end of the crystallization processes range from 19 (C) t o 4 4 pm ( B )
.
In table 1 are listed the compositions of the solid phases from runs C and D . The DTA curve of the dried aged starting gel f o r run E is shown in Fig. 2. The
DTA curves
f o r the corresponding samples f o r runs
A to D are quite similar, thc only difference being the lowering of the temperature in correspondence of the second peak, due to a smaller A1 contcnt. The first pcak around 112'C is not accompanied by a weight loss, and must therefore be assigned to a phase transition. T h i s phase transition is reversible, and becomes exothermal upon cooling (see ref. 1 5 ) . This peak
was
assigned to [TPABr],
576 aggregates t r a p p e d
than
for p u r e
i n t h e gel, and a p p e a r s at l o w e r t e m p e r a t u r e s
TPABr.
The
peak
around
240'C
assigned
was
to
the
d e c o m p o s i t i o n of b o t h t h e m o n o m e r i c TPA a n d t h e aggregates, w h e r e as p u r e TPABr d e c o m p o s e s a t 280-29O'C.
TABLE 1 C o m p o s i t i o n of t h e s o l i d p h a s e s :
runs C and D
Run C amorph. t/hNa%
0 6.5 7.5 8.2 9
1.5 1.3 1.1 1.0
11
12 13.6 16 19.6 24.8
8
173 138 135 125
Run D
crystals
Naa
2.0 2.0 2.1 2.1 2.5
2.0
Ala
0.42 0.46 0.51 0.58 0.52
0.52
A1
226 207 186 166 182
183
amorph. TPAa'b
Na%
3
2.9
80
0.92 0.80
72 67
crystals
Naa
Ala
1.75
0.80
A1
TPAajb
3.36
3.54
1A8 0.89 2.0 2.3
tl$
1.02 1.14
119 107 93 83
3.66
a)Atoma o r c a t i o n s per u n i t c e l l b)From e l e m e n t a l a n a l y s i s c ) F r o m t h e r m o g r a v i m e t r y ( t h e l o s s of w a t e r d u e t o d e h y d r o x y l a t i o n is a l s o i n c l u d e d )
100
200
300
"C
P i g . 2 . DTA curve of t h e d r i c d s t a r t i n g g c l of r u n E aged 30 d a y s a t 26'C.
577
Fig. 3 represents the five kinetic curves of crystallization. All of
them show t h e typical sigmoid shape. T h e crystallization
rates in the five runs are in the o r d e r C>B>A>D>E.
z (t) 1.0
0.5
0 0
5
10
15
20
25
t/h Fig. 3 . Crystallization curvcs for the ZSM-5 samples.
The hystograms o f the granulornetric distributions in the final products of each run can be constructed together with the monotonic curvcs approximating them. These monotonic
curves and the
curves of the maximum average diametcrs of crystals are used t o derive
thc
be1 1
(sce r c f . 16)
The
shaped
curves
representing
nucleation
kinetics
(Fig. 4 ) .
linear growth
rates of crystals, and
also the times at
which nucleation is maximum in each run, are reported in Table 2.
Thc detailed quantitative analysis will be included
in a future
publication (see ref. 1 3 ) .
TABLE 2 Linear growth rate of crystals and maximum nucleation time
A (AL/2At)/(pm/h) tmax.""cl. /h
1.35 3.6
B
2.25 3.15
C 0.89 1.9
D 0.89
5.3
E 0.63 8.8
578
40
IC
0
ik 30
20
10
I 5
0
15
10
t/h F i g . 4 . Nucleation
rate
as
a
function
of
time
for
the
ZSM-5
samplcs.
DISCUSSION Previous studies (see ref. 8) showed that the conversion rate and the yield of silicalitc-1 arc favored by the amount of TPA ions up t,o TPA/Si m o l a r ratio equal to 0.08. Hence, an excess of ‘ W A beyond that, v a l u e i s not influent, any more on these parameters. The amount of organic ion in our samples (TPA/Si=0.088) is t h e r e f o r e optimal and does not. c n n R t , i t r i t . e PL 1 i m i t . i n r f f n * o * n r
579 cithcr for the yield or for the crystallization rate. The crystallization C>B>A>D>E.
The
rates
in the five runs are
in the order
relevant data t h a t can be found
in
literature
(see refs. 13,17-19) for similar studies also confirm that there is a n optimum amount of aluminium to speed up the crystallization
kinetics of ZSM-5. The linear growth rates of crystals (Table 2 ) are in the ordcr B>A>C,D>E.
The granulometric distributions, and
hencc the kinetic nucleation curves, are narrow (Fig. 4 ) and show
A1,0, content.
a tendency to broaden in the systems with higher
The results shown in Figs. 3 and 4 demonstrate that the faster crystallization rate found for ZSM-5
in run C is not a consequen-
ce of a higher crystallite growth rate, but is rather the effect of a faster nucleation.
I t is clear from Fig.
4
that nucleation of ZSM-5 in o u r systems
starts at time zero, o r even before. It must be pointed out that maximum crystalline dimensions are an averaEe of the diameters of the largest crystals found in each sample. S o , e.g., in the sample of run B at 20.5 h it is possible to find one crystal as long as 44 pm
(lcngth to width 1) whereas the average diameter of the
15 largest crystals is 10 pm less.
This fact is a strong evidence
for the autocatalytic theory (see ref. 3) according to which nuclcation starts immediately in the gel already at room temperature. The analyses of the solid phases during the syntheses show that both the gcl and the crystals of zeolite enrich in A1 with time. This means that the crystallisation of zeolite ZSM-5
proceeds via
the formation of an A1 poor core, and an aluminous outer shell. An exception to this behaviour seems to be run E .
The number of TPA'
ions per unit cell of ZSM-5
is nearly cons-
t a n t in all of the phascs of the crystallyzation process, ranging
from 3 . 1 to 3.8,that is close to the theoretical value of 4 . Elemental analyses always show a lower content of organic ion with may with
respect to the thermogravimetric measurements. This effect be due to the annealing of lattice defects
R =
11,
Na',
TPA')
during
thermogravimetric
which produccs water molecules. Nmr studies on
*'Si
(groups Si-OR, measurements, in ZSM-5
(see
ref. 20) showed that calcination at 440'C for 24 hours eliminated most of the latticc defects.
CONCLUSIONS It is shown that the nucleation starts when the crystallization
580 temperature
is reached
or
evcn
earlier, the gel still being at
temperature. The crystal1 ization rates as expressed by crystal1 inity pcr unit time are essentially linked to the nucleation rates. The analyses carried out on the solid phases show that the crystallization of Na,TPA-ZSM-5 in the five series examined proceeds by the formation of aluminium poor cores. The concentration of this element is higher on the external parts of t h e crystallitcs. room
ACKNOWLEDGEMENTS We gratefully acknowledge D r . Nicola La Rosa for kindly supplying calculation facilities. This work h a s been supported by CNR (Italian National Rcsearch Council), Progetto Finalizzato Chimica Fine e Sccondaria. REFERENCES 1 R.J. Argauer and G.R. Landolt, U . S . Pat. 3,702,886. 2 J.C. Jansen, C.W.R. Engelen and 11. van Rekkum, ACS Symp. Ser., 398 (1989) 257.-68 3 4
5 6 7 8
B. Subotib, ACS Symp. Ser., 398 (1989) 110-21. R . A . Van Santen, J. Keijsper, G . Ooms and A.G.T.G. Kortbcek, Stud. Surf. Sci. Catal., 26 (1986) 169-75. D.C. Hayhusrt, R. Aiello, J. B.Nagy, F. Crea, G . Giordano, A . Nastro and J .C. Lee, ACS Symp. Ser., 368 (1988) 277-91. G . Boxhoorn, 0. Sudmeijer and P . 1 I . G . van Kasteren, J. Chem. SOC.,Chem. Commun., (1983) 1416-8. J. B.Nagy, P. Bodart, E.G. Derouane and Z. Gabclica, Stud. Surf. Sci. Catal., 26 (1986) 231-8. F . Crea, A . Nastro, J. B.Nagy and R . Aiello, Zeolites, 8
(1988) 262-8. R . Aiello, F. Crea, A. Nastro and C. Pellegrino, Zeolites, 7 (1987) 549-53. 10 A. Araya and B.M. Lowe, Zeolites, 6 (1986) 111-8. 11 B.M. Lowc, Stud. Surf. Sci. Catal., 37 (1987) 1-12. 12 G . Bellussi, G . Perego, A. Carati, U . Cornaro and V. Fattore, Stud. Surf. Sci. Catal., 37 (1987) 37-44. 13 G. Golemmc, A. Nastro, J. B.Nagy, B. Subotik, F. Crca, and R .
9
Aiello, in preparation. 14 P.T. Cardew and R . J . Davey, Proc. R . SOC. London A, 398 (1985) 415-28. 15 Z. Gabelica, J. B.Nagy, P. Bodart, N. Dewaele and A. Nastro, Zeolites, 7 (1987) 67-72. 16 S . P . Zhdanov and N.N. Samulcvich, in: L.V.C. Rees (Ed.), Proc. 5th Intl. Conf. on Zeolites, lleyden, London, 1980, p p . 75-84. 17 V.N. Romannikov, V.M. Mastikhin, S . HoEevar and B . Driaj, Zeolites, 3 (1983) 311-20. 1 8 S . B . Kulkarn i , V.P. Shi ralkar, A.N. Kotasthane, R.B. Borade and P . Ratnasamy, Zeolites, 2 (1982) 313-8. 19 K.-J. Chao, T.C. Tasi and M.-S. Chen, J. Chem. SOC., Faraday Trans. I , 77 (1981) 547-55. 20 P. Bodart, J . B.Na y , 2. Gabalica and E;.G. Derouane, J. Chim. Phys., 83 (11-12) fi986) 777-90.
G . Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
NEW MOLECULAR SIEVE
J . Kornatowskil,
581
- VANADIUM SILICALITE KVS - 5
M. Sychev2. V. GoncharukZ, W.H.
Baur3
1Institute of Chemistry, N. Copernicus University, Gagarina 7, 87-100 Toruh, Poland 2Institute of Colloid and Water Chemistry, Ukrainian Academy of Sciences, Vernadsky Ave. 42, 252680 Kiev, USSR 31nstitut fur Kristallographie und Mineralogie der J.W. Goethe-Universitat, Senckenberganlage 30. D-6000 Frankfurt am Main 1, PRG
SUMMARY Vanadium silicalite KVS-5 has been grown under conditions of hydrothermal synthesis. Its structure corresponds to the MFI type. The product can be synthesized with a yield of 100% in form of slightly coloured large crystals up to 300 m. The incorporation of V5+ ions on T-sites in the zeolitic framework has been investigated by a variety of methods: XRD, MAS M,SSIMS, ESR, TGIDTA. IR. and XRF.
INTRODUCTION The catalytic properties of vanadium compounds and especially of its oxides have been investigated and used since the 1960's. These studies of vanadium catalysts were dealing mainly with the changes of oxidation state of vanadium during the catalytic process as well as with the role of support. Commonly. silica or alumina [l-31 have been used as supports for the vanadium compounds. No catalysts have been prepared with an exactly defined state of this metal except for the synthesis of heteropolyacids [4]. An introduction of vanadium into the ZSM-5 zeolite crystals has been already reported: Kucherov et al. 15-81 and later Sass et al. [9] introduced V4+ ions into extra-framework cationic positions by a solid state reaction. Inui e l al.
[lo] introduced vanadium compounds instead of the A1 component already into the reaction gel for synthesis of the zeolitic catalyst. The possibility of isomorphous substitution of a large number of elements with vanadium in three different oxidation states 111, IV, and V has been published by Xu Ruren et al.[ll]. Nevertheless, they have given neither synthesis conditions nor any proof or method for checking the actual incorporation of those elements into the crystal structure except that they report lattice constants. The differences in these lattice constants compared to ZSM-5 seem to be much too high considering the amounts of heteroatoms which could be possibly incorporated into the framework.
582
The aim of this work was to prepare a crystalline porous zeolite with isomorphously substituted Vt5 ions in the framework as a precursor of a catalyst for oxidation reactions and to check the product with a variety of methods which could show real substitution. EXPERIMENTAL The zeolite was synthesized in teflon-lined autoclaves by reacting silica s o l ,
alkali metal vanadate, sodium bicarbonate, tetrapropylammonium hydroxide (TPAOB) and water at 160-190.C
and autogeneous pressure for 5-9 days followiiig
the procedures given in refs. [12.13]. Reference samples were prepared by impregnation of silicalite with VOSO4 and by synthesis of VA1-ZSM-5 using analogous methods to KVS-5. The impregnation procedure was the following: siljcalite
was stirred for 24 hs at RT and pH
=
3 in an 0,1 M water solution of VOSO4 ( 2 0
ml/lg of zeolite), then washed with water, dried at 80'C calcined at 500.C 500'C
for 3 hs and finally
for 1 h in air. Calcination of KVS-5 samples was made at
for 48 hs in a slow air stream and 24 hs i n an 02 stream.
The products were characterized by XRD. SEM. MAS NMR, SSIMS, ESR, TG/DTA, and IR techniques. Apparatus used and condi.tionsof measurements: XRD: DRON 3 with PC, CuKa radiation, 0,5'/min SEM: NOVOSCAN 30. samples covered with Au
MAS NMR: BRUKER CXP 200 equipped with 4.8 T magnet ESR: TSN-254 and SE/X 2544 Radiopan
TG/DTA: MOM Q-1500, heating rate 1O0C/min, sample 400 mg IR: UR 20 SSIMS: LAS-3000 RIBER. range 0-200 amu. time 30 min. RESULTS AND DISCUSSION The experiments were successful in obtaining vanadium silicalite with a 100% yield and in form of slightly violet large crystals up to 300 vm in length. Fig. 1 shows the material of euhedral habitus of MFI topology from a sample with -200
long crystals. Crystal sizes distribution was usually within the
range of 210% and their aspect ratio length to width was about 2 - 2 . 5 . shows the crystals grown together with Al3'
F.ig.
2
ions. They were obtained only up to
-90 pm in length and with distinctly wider size distribution. The XRD pattern
was typical for an MFI topology for both cases. The KVS-5 crystals had lattice constants 20.087, 19.943, and 13.420 I for the a, b, and c axes, respectively. They were changing from one sample to another but it was impossible to get a relation between the lattice constants and content of vanadium as the differences were usually smaller than experimental error. A decrease of lattice
583
la
lb
Fig. 1.
Scanning electron micrographs of molecular sieve KVS-5.
584
2a
2b Fig. 2. Scanning electron micrographs of vanadoaluminosilicalite molecular sieve VAL-ZSM-5.
585
constants larger than the experimental error occurred after calcination which suggests a relaxation of the framework after removal of the template molecules. Thermogravimetry analysis (Fig. 3 ) shows that the calcination process proceeds with three thermal effects: endo- and exo-thermal effects at about 350
-
360'C overlapping one another and coinciding with a weight loss and the third
T PC
DTG 800 -.----L
-'\~
\
-
600
LOO
200
Fig. 3 .
Thermogravimetry analysis curves for KVS-5.
strong exothermal effect at -520.C not followed by a weight loss. The endo-effect. is connected with the template decomposition. Then the decomposition products leave the sample (weight loss) causing a relaxation of the framework (exo-effect and decrease of lattice constants). The exo-effect at -520.C can be due to combustion of the organic decomposition products outside the sample. The last effect is not observed when the calcination proceeds in a stream of gas (the decomposition products are taken away from the system). The possibility of a two step mechanism of the template removal was already reported [14]. All these effects are probably well resolved due to both the large dimensjons of the crystals and the high amount of sample hindering diffusion processes. The static SINS data (Fig. 4) show that KVS-5 samples are free from A1 and F e impurities (confirmed in ESR spectra) and the vanadium content (lines 51 and
67 for V(+) and VO(+), respectively, and no 181 line for V*O5(+)
[15]) can
586 72 LL
I
4. Static secondary ions mass spectroscopy data for vanadium s i l i c a l i t e Fig.
KVS-5.
Fig. 5.
Infrared spectra of vanadium s i l i c a l i t e KVS-5 before ( a ) and a f t e r (b) calcination.
l
1500
l
.
l
l
l
1000
.
l
l
l
l
500
l
cm-'
587
reach about 1%. This is i n agreement with t h e X-ray fluorescence a n a l y s i s res u l t s . For unknown reasons V atom cannot be d e t e c t e d by e l e c t r o n probe microa n a l y s i s of t h e samples. The I R s p e c t r a of KVS-5 (Fig. 5 ) show t h e weak band at -950 cm-l.
I t w a s ob-
served f o r t h e f i r s t time f o r T i s i l i c a l i t e [16] and its i n t e n s i t y was l a t e r r e l a t e d t o t h e T i content [17]. The band of o u r KVS-5 material i n c r e a s e s its i n t e n s i t y a f t e r c a l c i n a t i o n (Fig. 5b) which might suggest a n improvement of t h e ordering of t h e vanadium atoms i n t h e c r y s t a l .
3PPH
I
H
__c
200 G
I-----+
DPPt
F i g . 6. ESR s p e c t r a of KVS-5 samples with complete ( a ) and incomplete ( b ) inc o r p o r a t i o n of V i n t o t h e t e c t o s i l i c a t e framework. S p e c t r a l parameters: g~ = 1,961; ggg = 1.939; geff = 1.954; b = 8.78 mT; A,, = 19.25 mT; t h e broad l i n e i n spectrum ( b ) geff = 1,965, AH e 50 mT.
The ESR spectrum of t h e as-prepared KVS-5 sample (Fig. 6 ) shows t h a t vanadium occurs as V4+. The w e l l resolved hyperfine s t r u c t u r e shows a very high
d i s p e r s i o n of t h e V4+ ions. As a pure s i l i c a l i t e s t r u c t u r e would have no pref e r r e d p o s i t i o n s , a reason f o r such d i s p e r s i o n could be i n c o r p o r a t i o n of t h e
588 V atoms i n t o t h e framework. The p a r a m e t e r s of t h e s p e c t r u m are similar t o t h o s e
f o r V s p e c i e s i n n e a r l y s q u a r e p l a n a r environment [ 6 ] though t h e d i f f e r e n c e s
are large enough t o allow a l s o o t h e r d i s t o r t i o n s of t h e f o u r f o l d c o o r d i n a t i o n [ 9 ] . The v e r y broad l i n e i n F i g . 6b serves as a background f o r t h e h y p e r f i n e
s t r u c t u r e and c a n be a s s i g n e d t o c l u s t e r e d extra-framework V4+, as i t is i n t h e
s i l i c a l i t e impregnated w i t h VOSO4 ( F i g . 7 a ) .
Fig. 7. ESR s p e c t r a of a ) sample p r e p a r e d by impregnat ion of sit i c a l i t e w i t h VOSO4, b ) VAL-ZSM-5 a f t e r calcination.
I
I
3,lO
I
I
I
330
35 0
I
37 0 H [kGl
F i g . 8. 51V s o l i d s t a t e MAS NMR spectrum of vanadium s i l i c a l i t e KVS-5 a f t e r c a l c i n a t i o n (6 = 0 f o r VOC13. f r e q u e n c y 52,6 MHz, s p e c t r a l w i d t h 720 kHz. r o t a t i o n 3 , l kHz, D 1 = 0 , 5 vsec, D 0 = 0 , 3 s e c , p u l s a n g l e 22,5', 70.000 t r a n s i e n t s )
F i g . 9. 51V s o l i d s t a t e MAS NMR spectrum of s i l i c a l i t e impregnated w i t h VOSO4 ( c o n d i t i o n s s e e Fig. 8 , 100,000 transients).
589
After calcination, the vanadium ions represented by the hyperfine ESR spectrum are completely oxidized to V5+ and the ESR signal vanishes. The broad lines in Figs. 6b and 7a remain unchanged after calcination showing that the clustered V4+ is not oxidized. In the ESR spectrum of calcined VAl-ZSM-5. the hyperfine structure is not vanishing completely (Fig. 7b) which might suggest a hindering role of A1 in the oxidation of V4+. The 51V MAS NMR spectrum of calcined KVS-5 shows the band at -557 ppm (Fig. 8 ) corresponding to vanadium complexes with a tetrahedral oxygen environment [ I , 181 which might indicate again the incorporation of vanadium into the framework. Fig. 9 presents the spectrum for silicalite impregnated with VOSO4 (measured immediately after calcination). As V4+ should not give any NMR spectrum, the extremely broad band (Av
0
110 kHz) at about -620 ppm can be
identified as the band due to a small amount of V5+ ions broadened because of the influence of V4+ present in excess. A similar band at about -530 ppm with a small shoulder at --I000 ppm is observed for as-prepared KVS-5 except for a narrower width -17,5 kHz probably due to a much lower content of vanadium.
As it appears from 27Al MAS NMR spectrum of as-prepared VAL-ZSM-5 (Fig.10). 5
-
20% of the Al3+ ions are in octahedral coordination ( s e e the band at -0
pprn). This band vanishes completely after calcination and only the band for
tetrahedral A 1 at +50 ppm remains, thus proving the tendency of A1 for entering tetrahedral coordination during heat treatment [ 1 9 ] .
Fig. 10. 27Al MAS NMR spectrum of VA1-ZSM-5 before calcination ( 6 = 0 for A ~ ( H * O ) ~ ~in + aqueous solution, frequency 52.2 Mlz, spectral width 42 kHz, rotation 3 , 1 kHz, D 1 = 10 vsec, D 0 = 0,3 sec, 5.000 transients).
80
60
LO
20
0 ' -20
-LO
6
pprn
590
CONCLUSIONS The results show the possibility of preparation of vanadium silicalite KVS-5 containing approximately 1% vanadium. The vanadium atoms in the as-prepared material exist mainly as V4+ ions and after calcination they are oxidized to V5+. In both cases, they are tetrahedrally coordinated and they occur in a highly dispersed state. Depending on synthesis conditions, a part of the vanadium can occur as clustered extra-framework V4'
species which are difficult to
oxidize. After calcination, the vanadium atoms in KVS-5 appear
to
be in a more
uniform state than before. The work was partially supported by the Polish Ministry of National Education within the Project CPBP 01.06. REFERENCES 1 V.M. Mastikhin, K.l. Zamaraev. Z. Phys. Chem. Neue Folge, 152 (1987). 317. 2 B. Taouk, M. Guelton, J . Grimbolt. J.P. Bonnelle, J . Phys. Chem., 92 (1988). 6700. 3 K.J. Zhen, M.M. Khan, C.H. Mak, K.B. Lewis, G.A. Somorjai, J . Catal., 96 (1985). 501. 4 M.A. Fox et al., J . Am. Chem. SOC. 109 (19831, 6347. 5 A.V. Kucherov. A.A. Slinkin, Zeolites, (1986). 175. 6 A.V. Kucherov, A.A. Slinkin, Zeolites, I_ (1987), 38. 7 A.V. Kucherov. A.A. Slinkin, Zeolites, L (1987). 43. 8 A.V. Kucherov. A.A. Slinkin, Zeolites, 7 (1987). 583. 9 C . E . Sass, Xinhua Chen, L. Kevan, J . Chem. SOC., 86 (1990). 189. 10 T. Inui, A. Miyamoto. H. Matsuda, H. Nagata, Y. Makino, K. Fukuda. F. Okazumi, Proc. 7th Int. Zeolite Corif.. Tokyo, 1986, Y. Murakami et al. (Eds.). Kodansha, Tokyo, 1986. 859 and refs. 12-14 therein. 1 1 X u Ruren. Pang Wenqin, Proc. Tnt. Symp. Zeolites, Portoroz, 1985, B. Drzaj et al. (Eds.), Elsevier, Amsterdam, 1985, Stud. Surf. Sci. Catal.. 24 (1985). 27. 1 2 J . Kornatowski, M. Rozwadowski, Pol. Pat. Appl., P 281513, (1989). 13 J . Kornatowski. Zeolites, 8_ (1988), 77. 14 G . Debras, A. Gourgue, J.B. Nagy, G. De Clippeleir, Zeolites. 2 (1985), 377. 15 C. A. Altomare, G.S. Koermer, E. Martins, P.F. Schubert, S.L. Suib, W.S. Willis, Appl. Catal., 45 (1988). 291. 16 G . Perego, S. Bellussi, C. Corno, M. Taramasso, F. Buonorno, A. Esposito, Proc. 7th Int. Zeolite Conf., Tokyo, 1986, Y. Murakami et al. (Eds.), Kodansha, Tokyo, 1986, 859. 17 B. Kraushaar. J.H.C. van HooIf, Catal. Lett., L(1988). 81. 18 L.R. Le Coustumer. B. Taouk, M. Le Meur, E. Payen, M. Guelton. J. Grimblot, J. Phys. Chem.. 92 (1988), 1230. 19 J. Kornatowski. M. Rozwadowski, W. Schmitz, A. Cichowlas, Proc. 8th Int. Zeolite Conf., Amsterdam, 1989. Recent Research Reports Val., J.C. Jansen et al. (Eds.). L.. Moscou, Akzo Chemicals, Amsterdam, 1989, 79.
G. Ohlmann et 01. (Editors),Catalysis and Adsorption by Zeolites 01991 Elsevier Science Publishers B.V., Amsterdam
591
MULTINUCLEAR NMR STUDY OF THE CRYSTALLIZATION OF SAPO-37
N. Dumontl, T. lto2, J. B.Nagy1, Z. Gabelical and E.G. Derouanel Facultks Universitaires N.-D. de la Paix, Laboratoire de Catalyse, 61, rue de Bruxelles, B-5000-Namur (Belgium) Tamai Sangyo CO., LTD, Zenibako 3-chome, 524-11, Otaru 047-02 (Japan)
ABSTRACT A series of intermediate phases, isolated during the synthesis of SAPO-37, have been characterized by XRD, SEM, NMR of adsorbed Xe, and solid state 27Al- and 31P-NMR, in order to evidence the successive steps occurring during the crystallization process. SAPO-37 stems from a direct gel restructuration: large cavities form in the amorphous phase during the aging period at ambient temperature. Upon heating at 200°C, aluminum and phosphorus respectively incorporate in configurations Al(4P) and P(4AI) in the framework, at the expense of the amorphous phase, giving rise to the formation of well-defined SAPO-37 crystals, showing an octahedral morphology, and growing with time. The crystallinity reaches a maximum after 32h. For excessive synthesis times (149h), the appearing of structural defects observed by NMR illustrates the partial degradation of the SAPO-37 framework, whereas a side-phase, SAPO-40, possibly involving interconnected channels limited by 12T puckered rings, co-crystallizes from the mother liquid phase. INTRODUCTION The synthesis of SAPO-37, a silicoaluminophosphate involving a Faujasite topology [ref. 11 was first reported by Lok et al. [ref. 21. Because of its open pore structure, this material possesses potential interesting and attractive applications in catalysis or adsorption, this explaining the increasing attention that is now being devoted to the understanding of its synthesis and to its better characterization. We have recently examined the influence of various synthesis variables (aging time and temperature, agitation and crystallization time) on the nature and final properties of the various SAPO-type open structured phases obtained by two different methods currently yielding SAPO-37 [ref. 31. Our study allowed us to define the optimal conditions under which pure SAPO-37 could be obtained in high yield by using both the conventional "aqueous" synthesis [ref. 21 or
592
the new biphasic (water-hexanol) route [ref. 41. The short range structural configuration and coordination of Si, Al and P and their distribution in the SAPO-37 framework, as well as the mechanisms governing the potential framework substitution of Al or P by Si in the Faujasite structure, have been thoroughly investigated by high resolution solid state NMR (29Si, 27AI, and 31P) [refs. 3, 5-71. The same techniques [ref. 81, complemented by 129Xe-NMR [ref. 91 were used to follow the structural changes occurring when water and organic template molecules progressively escape from the SAPO-37 intracrystalline volume upon various thermal treatments. Finally, we have recently taken advantage of the multiple potentialities of the 129Xe-NMR method, linked to the great sensitivity of the xenon nucleus to its close environment [refs. 10, 111 to characterize a series of intermediate phases isolated during the hydrothermal synthesis of SAPO37 [ref. 121. In the present work, we confirm and complete these preliminary 129Xe-NMR data by investigating more in depth the successive stages of the crystallization (gel restructuration, crystal growth, formation of a side-phase) and by characterizing all the solid intermediates using complementary 27Al- and 31 P-NMR, in combination with XRD and SEM techniques.
EXPERIMENTAL The optimized procedure followed to synthesize the intermediate and final SAPO-37 phases was described in details in our previous papers [refs. 3, 121. Several aliquots of a hydrogel of molar composition 1.0 A1203 : 0.9 P2O5 : 0.4 Si02 : 0.86 (TPA)20 : 0.023 (TMA)20 : 50 H20, previously aged at 20°C for 48h, were heated at 200°C under stirring conditions for various periods of time. For the present study, we have selected the more representative samples among the intermediates isolated during the crystallization course. These specimens were obtained after 0, 5, 10, 32 and 149h of heating, and are referred to as Pn, where n denotes the crystallization time (Tablel). The as-synthesized materials were checked for nature and purity by X-ray powder diffraction (XRD) and scanning electron microscopy (SEM). The percentage of SAPO-37 in each phase was estimated by comparing the total surface area of the characteristic XRD peaks measured for this sample to the area of the same peaks recorded for the most crystalline SAPO-37 of the series, arbitrarily considered 100% crystalline (P32) (Tablel). The xenon adsorption isotherms were measured at 34% using a classical volumetric apparatus, following operating conditions described
593
TABLE 1 Nature, relative xenon adsorption capacities and relative crystallinities of the intermediate phases isolated during the crystallization course of ;APO-3; Sample a Pn
e f
9 h
129Xe-NMR (mean diameter)
Relative Xe adsorption capacities b
% of crystallinity C XRD 129Xe-NMR d
p10
p25
SAPO-37
p32
SAPO-37
octahedra (15.1 pm)
SAPO-37
100
100
100
SAPO-37
octahedra
SAPO-37
135 g
h
55
p149
d
SBA
amorphous amorphous SAPO-37
PO p5
C
Nature
m
+
amorphous cavities amorphous cavities octahedra SAPO-37 (7.7 w ) f octahedra (14.8 l m )
+
+
81
0 0 78
e e 83
f
95
f
18 18
SAPO-40 platelets SAPO-40 n denotes the crystallization time (h). Relative xenon adsorption capacities calculated from the isotherms with respect to the most crystalline SAPO-37 phase (P32) arbitrarily considered to have a 100% adsorption capacity. Weight percentage of SAPO-37 in each intermediate phase determined with respect to the most crystalline SAPO-37 (P32) considered to be 100% crystalline. Determined from the ratio between the slope of the straight line &=f(nXe/g) for each crystalline intermediate phase and the slope obtained for P32 [ref. 121 The crystallinities of Po and P5 could not be determined by 129Xe-NMR because these two phases do not contain well-defined SAPO-37 crystals. not investigated. Xe isotherms only allow to calculate the global Xe adsorption capacity of the mixture (SAPO-37 + SAPO-40). It is not possible to estimate the XRD crystallinity of SAPO-37 in the mixture Pi49 , as most of the XRD peaks of SAPO-37 and SAPO-40 overlap.
elsewhere [ref. 121. The 129Xe-NMR spectra of xenon adsorbed at different pressures on all the intermediate phases were recorded at t h e same temperature, with a Bruker CXP-200 spectrometer operating at 55.3 MHz. The 27Al- (105.3 MHz) and 31P- (162.0 MHz) high resolution solid state MAS N M R measurements were performed o n a Bruker MSL-400 spectrometer. The pulse length and recycle time were lps and 200ms for
594
27Al and 3ps ( d 2 ) and 30s for 31P, respectively. Typical measuring conditions for the 31P- CP-MAS NMR experiments were a 1H pulse length of 2.5ps, a contact time of l m s and a recycle time of 5s.
RESULTS AND DISCUSSION i ) XRD, SEM and 129Xe-NMR The structural and morphological characteristics of the SAPO-37 intermediate phases, determined by X-ray diffraction and scanning electron microscopy, are summarized in Table 1. These data are compared to those obtained previously by 129Xe-NMR [ref. 121. The relative xenon adsorption capacities of the calcined samples are compared to their relative crystallinities estimated by XRD and IzgXe-NMR, respectively. For Po and P5, we observe a broad 129Xe-NMR line, distinguishable from the one belonging to the gaseous xenon, and characterized by a quasi constant chemical shift value ( about 45 ppm), whatever the xenon pressure. This signal arises from the interactions of the xenon atoms with the walls of large cavities of about 25 A. We therefore conclude that, at these early stages of the crystallization, the gel already contains a preliminary void structure, responsible of the non-negligible adsorption capacity of these materials (about 20% of the maximum observed in the case of 100°/~ crystalline SAPO-37, P32). The degree of connection between these cages is small, but increases as the crystallization proceeds. These cavities do not show any regular ordering, which explains why no crystalline phase is detected in PO and P5, by XRD and SEM. P l o is found to contain already about 80% crystalline SAPO-37. The crystals show the typical octahedral morphology characterizing Faujasitetype materials (Fig. 1) [ref. 31. The crystal size increases with the crystallization time (Table 1). X-ray diffractograms indicate that P32 is a 100% crystalline SAPO-37, whereas, for longer synthesis times, a crystalline side-phase, SAPO-40, consisting of platelet-like crystallites, is found admixed with SAPO-37 (P149) (Fig. 1). In contrast to SAPO-37 that stems from a direct gel restructuration, SAPO-40 nucleates in the liquid-phase , probably because of the marked change in its composition observed as soon as 100% crystalline SAPO-37 is formed [ref. 121. The relative xenon adsorption capacities of P10 and P32 are in very good agreement with their relative crystallinities measured by XRD or calculated from the slopes of the straight lines 6 1 2 9 ~=f(nXe/g) ~ [ref. 121. We took advantage of this latter method to determine the weight percentage of SAPO-37 admixed with SAPO-40 in P14g (Table 1) [ref. 121, which is rather difficult to estimate by XRD, as most of their
595
Fig. 1. SEM micrographs of the various intermediate phases. (a) Po, (b) P5, (c) Pio,
(4 P25, (el P32, (f) PIN. diffractogram peaks overlap [ref. 21. Distinct 129Xe-NMR signals characterize these two SAP0 materials. The chemical shift of the high field resonance extrapolates to 53 ppm at zero xenon concentration for samples Plo, P32 and PI49 and confirms the presence of SAPO-37. For the low field signal, the corresponding parameter 1 3 ~is larger (95 ppm), which suggests that SAPO-40 has a pore structure narrower than SAPO-37, the cavities of which are connected by 12 T windows. n-hexane adsorption measurements confirm this assumption. Indeed, the porous volume of SAPO-40 was found to be about 60% of the void volume defined by the supercages of SAPO-37. We recently proposed a tridimensional structure for SAPO-40, consisting of interconnected straight channels limited by puckered 12 T rings [ref. 121. These structural properties, in line with the typical tetragonal crystal morphology, justify the behavior of this material upon adsorption. Indeed, whereas SAPO-40 shows a n-hexane adsorption capacity quasi identical to that of ZSM-5, xenon atoms diffuse
596
much more rapidly in the S A P 0 than through the ZSM-5 structure. Moreover, marked confinement effects, due to the puckered channels, give rise to a particularly high xenon sorption capacity explaining why, in spite of its narrower pore system, SAPO-40 adsorbs more xenon than SAPO-37 (Table 1) [ref. 121. For too long heating times, the SAPO-37 framework starts collapsing, as indicated by the broadening of the high field signal for the phase P14g.
and 3 1 P - N M R The progressive structuration of the gel into crystalline SAPO-37 is illustrated by the 27AI-NMR spectra presented in Fig. 2. As the octahedral aluminum resonates at higher fields (see below), we tentatively attribute the single line observed for samples Po and P5 at about +6 ppm (line P) to penta-coordinated aluminum in the amorphous phase. No signal is detected at lower fields, which confirms the absence of any crystalline material at these stages of the crystallization. The pure and well-crystalline SAPO-37 phases P l o , P25 and P32 exhibit an additional 27AI-NMR line at about +37 ppm (line T) due to tetrahedral aluminum in the configuration Al(4P) [refs. 5-61. From P i 0 to P32, the intensity of line P decreases, and that of line T increases, as the crystallization proceeds, i.8. as the amount of SAPO-37 increases. This reflects the progressive incorporation of aluminum into the tetrahedral positions of the SAPO-37 framework, at the expense of the amorphous phase. However, line P does not disappear completely, even for P32 which does not contain any amorphous phase detectable by XRD or SEM. We therefore conclude that an additional contribution to signal P, probably due to some (otherwise tetrahedral) aluminum atoms distorted by secondary coordinations with occluded molecules, adds on the contribution of the penta-coordinated aluminum belonging to the amorphous phase. Indeed, as shown in Fig. 3, this resonance is sensitive to the effects of dehydration and calcination. Upon heating to various temperatures (1 20, 230°C), the intensity ratio I(lineT)/l(line p) for sample P25 progressively increases. Moreover, line P tends to disappear when the sample is calcined under dry air up to 550°C. Saldarriaga et al. [ref. 51 and Blackwell et al. [ref. 61 have also suggested that such interactions with intracrystalline water molecules, OH- groups or organic templates could modify the geometry around some framework aluminum atoms and give rise to this additional signal appearing at a chemical shift value that characterizes aluminum in a coordination larger than 4. However, we observe that this "penta"coordination is not restored upon rehydration, which supports the
ii)
27AI-
597
T
100
0
-100
PPM
Fig. 3. 27AI-NMR spectra of phase p25.
100
0 PPM
-100
Fig. 2. 27AI-NMR spectra of the assynthesized intermediate phases. (a) Po, (b) P5, (c) Pie, ( 4 P25, (el p32, (f) p149.
(a) as-synthesized, (b) heated in a N p flow to 230°C, (c) heated in a N2 flow to 550°C then calcined in air at 550°C, ( d ) sample c left overnight in moist atmosphere,(e) sample c dispersed in water for 15 min then dried at 100°C, (f) sample e recalcined to 550°C.
598
hypothesis of a distortion of the Al tetrahedra by organic compounds, such as TMA+ or TPA+. Indeed, when calcined SAPO-37 is exposed overnight to atmospheric moisture or dispersed in water for 15 minutes and then dried at 100°C, another signal appears at about -11 ppm. It is attributed to octahedral aluminum stemming from a reversible coordination of some framework aluminum atoms with water molecules (Fig. 3). The hydration state of SAPO-37 also affects the chemical shift of line T: the higher the degree of hydration, the larger the chemical shift (calcined P25: +29 ppm, as-synthesized P25: +37 ppm, hydrated P25: +42 ppm). According to an NMR study reported by Muller et al. on a series of AIP04 phases [ref. 131, the 27AI chemical shift depends linearly on the AI-0-P bond angle. In the case of SAPO-37, such angular changes are probably induced by the presence of water molecules. The instability of SAPO-37, in its H-form, in the presence of water was well-recognized [ref. 81. We have also previously evidenced by 129XeNMR the formation of larger cavities and a partial pore blockage in SAPO37 leading to a dramatic decrease of the adsorption capacity [ref. 91. This loss of crystallinity, due to the partial structure collapse, is confirmed by XRD: the samples hydrated by contact with atmospheric moisture or by dispersion in water, become totally amorphous to X-rays. By contrast, the SEM micrographs reveal that the octahedral crystals still keep intact their outer shell, which suggests that moisture induces a true perimorphic transformation. However, as indicated by the persistency of line T, the tetrahedral geometry of some framework aluminum atoms is retained. The two main signals characterizing the crystalline SAPO-37 intermediate phases are also found in P14g (mixture of SAPO-37 and SAPO-40) (Fig. 2) but the ratio of their respective intensities is completely reversed: line P appears only as a shoulder. This inversion could be interpreted in terms of a decrease of the relative number of penta-coordinated aluminum atoms in the SAPO-37 framework but, if we balance the absolute intensity of this resonance by the percentage of SAPO-37 in the mixture, we find that this decrease is only apparent. The main cause of this ratio inversion is the large increase of the intensity of signal T. It is explained by the marked incorporation of aluminum into the tetrahedral positions of the SAPO-40 network, from the mother liquid phase. The 27AI-NMR spectrum of sample PI49 shows a broad shoulder at around -15 ppm, a ppm range that usually characterizes octahedrally coordinated aluminum. This shoulder noteworthly decreases when sample PI49 is calcined at 550°C, and nearly completely disappears after another
599
T
100
-100
0 PPM
Fig. 4. 27AI-NMR spectra of phase p i 49.
(a) as-synthesized, (b) heated in a N p flow to 550°C then calcined in air at 550°C, (c) sample b left overnight in moist atmosphere then recalcined to 550°C.
I . . . . I . . . . I . . . . I
0 I
-25
'
I
-35
I
PPM Fig. 6. Simulation of the 31P-NMR spectrum of as-synthesized phase p25.
-50
..
PPM
Fig. 5. 31P-NMR spectra of the assynthesized intermediate phases. (a) Po, (b) P5, ( 4 Pie, (4 P25, ( 4 p32, ( f ) p149.
600
rehydration-calcination cycle (Fig. 4). We therefore assign this line to structural defects (terminal AI(0H)” groups, ...) created during a long hydrothermal crystallization (149h) at a rather high temperature (200°C). Such defects easily recombine upon successive calcinations at 550°C. These findings go in line with the information obtained from 129Xe-NMR investigations of phase Pi49 in which SAPO-37 was found to have undergone a partial structural degradation. Finally, a weak line located at +19 ppm (P’) can be tentatively assigned to aluminum atoms in the SAPO-40 framework involving a secondary coordination with occluded organic molecules. As for SAPO-37, this line logically disappears upon calcination of PI49 at 550°C, a temperature at which all the organic molecules are released. Decoupled 31P-NMR spectra of non crystalline samples Po and P5 show a single but broad line located at about -20 ppm (Fig. 5). It obviously characterizes phosphorus atoms belonging to the amorphous phase. Its important width (about 2900 Hz) accounts for different types of phosphorus atoms located in various environments. They possibly could be terminal P(OH), groups. For sample P10, in which about 20% of amorphous phase is still present, this line still remains visible next to the resonance due to framework tetra-coordinated phosphorus (Fig. 5 ) . It becomes hardly detectable (and only by using cross polarization) for P25 and completely disappears in the case of the 100% crystalline P32. This progressive decrease in intensity confirms the attribution of this resonance to the presence of amorphous phase and quite well illustrates the progressive disappearing of this latter, as the crystallization proceeds. The main resonance at -31 ppm, due to SAPO-37 framework P(4AI) configurations [refs. 5-61 shows an asymmetrical character. An accurate decomposition of the spectrum (Fig. 6) suggests the presence of an additional small line at about -33 ppm. Its presence is not a simulation artifact, as evidenced by a series of calcination or ethanol washing experiments. Indeed, after such treatments, this shoulder completely disappears and the main line becomes symmetrical and narrow. We therefore assign the small shoulder at -33 ppm to slight deformations of P-O-AI angles produced by weak interactions of the framework atoms with the neighbouring organic molecules. The main line is broader by 100 Hz for PI49 with respect to that present in P32 (pure SAPO-37). Obviously such a broadening can be explained by the superposition of two very similar resonances of P(4AI) configurations belonging to both SAPO-37 and SAPO40 ordered structures. Blackwell et al. [ref. 61 have indeed observed small differences in position of the 3 1 P-NMR lines characterizing identical
601
P(4AI) configurations in various SAP0 frameworks. Finally, 31P-NMR confirms a partial collapse of the SAPO-37 lattice upon long heating times (149 h). Indeed, the presence of P-OH defects that appear upon partial degradation of the framework can be evidenced by a weak NMR line at -20 ppm, markedly enhanced under cross polarization. Internal P-OH terminal groups are indeed characterized by a small resonance in the -20 ppm range for a series of AIP04-n and SAPO-n partially degradated structures [ref. 141. CONCLUSION 129Xe-, 27Al- and 31 P-NMR techniques, combined with XRD and SEM, have been proved a useful tool to investigate the successive steps occurring during the crystallization process of SAPO-37. 129Xe-NM R was of particular interest to detect the presence, at the early stages of the crystallization, of preliminary void structures in the gel, which are not observable by XRD or SEM. It allowed to follow to progressive increase of the crystallinity of SAPO-37 and to evidence the partial degradation of its framework, as well as the formation of the side-phase SAPO-40, upon too long hydrothermal crystallization. The progressive incorporation of phosphorus and aluminum into tetrahedral positions in the SAPO-37 network, at the expense of the phosphorus and penta-coordinated aluminum belonging to the amorphous phase, was well illustrated by the 31P- and 27AI-NMR spectra of the successive intermediate phases. 27AI- and 31 P NMR also allowed to evidence the formation of structural P(OH), and AI(0H)n defects in SAPO-37, and the appearing of SAPO-40 after a long period of heating at 200°C, as well as the interactions of some framework Al and P atoms with occluded organic molecules. REFERENCES B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan and E.M. Flanigen, J. Am. Chem. SOC., 106 (1984) 6092-6093. B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan and E.M. Flanigen, U.S. Patent 4, 440, 871 (1984). L. Maistriau, N. Dumont, J. B.Nagy, 2. Gabelica and E.G. Derouane, Zeolites, 10 (1990) 243-250. E.G. Derouane and R. von Ballmoos, Eur. Patent 185, 525 (1986). L.S. Saldarriaga, C. Saldarriaga and M.E. Davis, J. Am. Cham. SOC, 109 (1987) 2686-2691. C.S. Blackwell and R.L. Patton, J. Phys. Chem., 92 (1988) 3965-3970. J.A. Martens, C. Janssens, P.J. Grobet, H.K. Beyer and P.A. Jacobs, Stud. Surf. Sci. Catal., 49A (1989) 215-225. M. Goepper, F. Guth, L. Delmotte, J.L. Guth and H. Kessler, Stud. Surf.
602
Sci. Catal., 498 (1989) 857-866. N. Dumont, T. Ito and E.G. Derouane, Appl. Catal., 54 (1989) Ll-L6. 9 1 0 J. Fraissard and T. Ito, Zeolites, 8 (1988) 350-361, and references therein. 1 1 T. Ito, J. Fraissard, J. B.Nagy, N. Dewaele, Z. Gabelica, A. Nastro and E.G. Derouane, Stud. Surf. Sci. Catal., 49A (1989) 579-588. 1 2 T. Ito, N. Dumont, J. B.Nagy, Z. Gabelica and E.G. Derouane, in: Proc. Intern. Symp. on Chemistry of Microporous Crystals, Tokyo, Japan, July 27-29, 1990, Kodansha/Elsevier, in press. 1 3 D. Muller, E. Jahn and G. Ladwig, Chem. Phys. Lett., 109 (1984) 332336. 1 4 L. Maistriau, Z. Gabelica, E.G. Derouane, E.T.C. Vogt and J. van Oene, Zeolites, in press.
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites
603
0 1991 Elsevier Science Publishers B.V., Amsterdam
METASTABILITY OF ZEOLITES I N TETRAETHYLAMMONIUM MEDIA
F r a n c e s c o D I R E N Z O I , A b d e r r a h m a n ALBIZANE1, Ma rie -A gn8s NICOLLE1, F r a n q o i s FAJULAI, F r a n c o i s FIGUERASl a n d T h i e r r y DES COURIERES2 1 L a b o r a t o i r e d e Chimie Organique Physique e t CinCtique Chimique A p p l i q u e e s , URA 418 d u CNRS, E c o l e N a t i o n a l e S u p e r i e u r e d e C h i m i e , 8 r u e d e 1 ’ E c o l e N o r m a l e , 34053 M o n t p e l l i e r C e d e x 1 ( F r a n c e ) * C e n t r e d e Recherche ELF-France,
Solaize
SUMMARY T h e i n f l u e n c e of o r g a n i c c o n c e n t r a t i o n , S i / A 1 r a t i o , s t i r r i n g and n a t u r e o f t h e s i l i c a s o u r c e on z e o l i t e s y n t h e s i s i n t h e p r e s e n c e o f t e t r a e t h y l a m m o n i u m h a v e b e e n s t u d i e d . Examples o f m e t a s t a b i l i t y o f z e o l i t e b e t a a n d ZSM-5 t o w a r d s ZSM-12 p r o p o s e t h e l a s t a s t h e more s t a b l e z e o l i t i c p h a s e i n t e t r a e t h y l a m m o n i u m m e d i a f e a t u r i n g h i g h e s t Si/A1 r a t i o . E v i d e n c e s of h e t e r o g e n e o u s n u c l e a t i o n o f z e o l i t e s o v e r amo r p h o u s o r c r y s t a l l i n e p h a s e s a r e discussed. 1NT RODUCT I ON
Many s t u d i e s h a v e d e a l t w i t h z e o l i t e s y n t h e s i s i n t h e p r e s e n c e o f t e t r a e t h y l a m m o n i u m , s i n c e t h e f i r s t r e p o r t of t h e s y n t h e s i s o f z e o l i t e b e t a ( r e f . 1 ) . Recent p a p e r s have been e s p e c i a l l y d e v o t e d t o t h e s y n t h e s i s o f z e o l i t e b e t a ( r e f s . 2 , 3 ) and t o t h e c o m p e t i t i v e f o r m a t i o n o f z e o l i t e b e t a a nd ZSM-PO
( r e f s . 4 , 5). T h e
open l i t e r a t u r e p r e p a r a t i o n s o f t h e s e z e o l i t e s a r e a l w a y s b a s e d on
te t r a a l k y l o r t h o s i l i c a t e s , w h i l e p a t e n t s y n t h e s e s p r o p o s e i n o r g a n i c s o u r c e s o f s i l i c a b o t h i n t h e case o f z e o l i t e b e t a ( r e f s .
1, 6 )
and i n t h e case of ZSM-12 ( r e f . 7 ) . The c r y s t a l l i z a t i o n s e l e c t i v i t y i n t e t r a e t h y l a m m o n i u m m e d i a when c o m m e r c i a l i n o r g a n i c s o u r c e s of s i l i c a a r e u s e d appears h en c e a s a p r o m i s i n g f i e l d of investigation. T h e n a t u r e of
t h e s o u r c e of s i l i c a i n f l u e n c e s t h e
c r y s t a l l i z a t i o n i n a m u l t i p l e way. Beyond t h e d i f f e r e n t d e g r e e o f polymerization of t h e s i l i c i c species i n s o l u t i o n , t h e s u r f a c e p r o p e r t i e s o f t h e a m o r p h o u s s o l i d p r e s e n t i n t h e s y n t h e s i s medium are deeply a f f e c t e d . A s a consequence, heterogeneous n u c l e a t i o n of
d i f f e r e n t p h a s e s can o c c u r and modify t h e s e l e c t i v i t y of t h e p r e p a r a t i o n . E v i d e n c e s of p r e f e r e n t i a l n u c l e a t i o n of z e o l i t e s o v e r
604
non-crystalline
s o l i d s i n o t h e r s y n t h e s i s media have been r e c e n t l y
p r o p o s e d ( r e f s . 8 - 1 1 ) . M o r e o v e r , t h e o r i e n t e d n u c l e a t i o n of a z e o l i t e framework o v e r a d i f f e r e n t z e o l i t i c s t r u c t u r e is a well known phenomenon ( r e f s . 1 2 - 1 4 ) .
I t seems t h e n e s p e c i a l l y
i n t e r e s t i n g t o pay a t t e n t i o n t o t h e i n t e r f a c e phenomena r e l a t e d t o t h e h e t e r o g e n e i t y o f t h e s y n t h e s i s medium. METHODS C r y s t a l l i z a t i o n e x p e r i m e n t s h a v e b e e n c a r r i e d o u t u s i n g two t y p e s o f e x p e r i m e n t a l d e v i c e s : 120 a1 s t a i n l e s s s t e e l a u t o c l a v e s f o r e x p e r i m e n t s i n s t a t i c c o n d i t i o n s and a 0.5 l i t r e s s t a i n l e s s s t e e l r e a c t o r f o r experiments under s t i r r i n g . T h i s l a s t autoclave wad e q u i p p e d w i t h a n a n c h o r - s h a p e d
s t i r r e r and a sampling o u t l e t
a l l o w i n g specimens t o be withdrawn w i t h o u t a l t e r i n g t h e s y n t h e s i s c o n d i t i o n s . The s y s t e m w a s d e s i g n e d s o a s t o a v o i d a c c u m u l a t i o n of
material i n s i d e s a m p l i n g p i p e and v a l v e s i n t h e c o u r s e of t h e r e a c t i o n ( r e f . 1 5 ) . The r e a c t o r was h e a t e d b y a n e l e c t r i c a l f u r n a c e r e g u l a t e d t h r o u g h a t h e r m o c o u p l e immersed i n t h e r e a c t i o n m e d i u m . T h e h e a t i n g r a t e was 3 "C/min
a n d t h e s t i r r i n g r a t e 150
rpm. S a m p l e s o f 1 U m l w e r e p e r i o d i c a l l y w i t h d r a w n f r o m t h e a u t o c l a v e . T h e s o l i d f r a c t i o n was r e c o v e r e d b y f i l t r a t i o n . The ::lcar
s o l u t i o n , c o n t a i n i n g t h e d i l u t e d m o t h e r l i q u o r , was
p r e s e r v e d f o r e l e m e n t a l a n a l y s i s . The s o l i d w a s washed w i t h d e i o n i z e d w a t e r up t o p H
Y a n d d r i e d a t 70
"C
i n a i r . The
temperature of a l l syntheses, both s t i r r e d and i n s t a t i c c o n d i t i o n s , was 1 5 0 ° C . T h e r e a g e n t s w e r e s i l i c a s o l ( C e c a s o l 3 0 , S i O z 2 5 % , Na 0 . 2 % , A 1 6 0 ppm, pH 8 . 8 , g r a i n s i z e 1 2 - 1 8 n m ) , p r e c i p i t a t e d s i l i c a ( Z e o s i l
175MP f r o m Rhane P o u l e n c , Na U . Y % , A 1 0.4%, Hz0 6 . 5 % , 2-2U
grain s i z e
u , p o r e volume 0 . 0 8 m l / g ) . s o d i u m a l u m i n a t e ( C a r l o E r b a R L E ) ,
t e t r a e t h y l a m m o n i u m (TEA!
h y d r o x i d e s o l u t i o n ( A l d r i c h ) , sodium
h y d r o x i d e ( P r o l a b o KP N o r m a p u r ) , d e i o n i z e d w a t e r . T h e r e a g e n t s were m i x e d u n d e r s t i r r i n g i n t h e o r d e r : a l k a l i n e s o l u t i o n , o r g a n i c
a g e n t , a l u m i n a t e , s i l i c a . T h e m i x t u r e was s t i r r e d f o r 4 h r s a t roam t empe r a t u r e b e f o r e t he b e g i n n i n g o f t h e s y n t lies i s . X - r a y p o w d e r d i f f r a c t i o n ( C t i R T h e t a 60 d i f f r a c t o m e t e r , C u K a r a d i a t i o n ) was u s e d t o i d e n t i f y t h e p h a s e s p r e s e n t i n t h e s o l i d f r a c t i . o n . C r y s t a l s i z e and h a b i t were d e t e r m i n e d b y s c a n n i n g e l e c t r o n m i c r o s c o p y ( C a m b r i d g e S l 0 0 i n s t r u m e n t , r e s o l v i n g power 7 nm).
605 RESULTS A N D DISCUSSION
C.om0p e t i Lian...be-ttwaen. an al.c.ine_,_. mx.deni..f_e.and..ze~l,i.t a .bet a S y n t h e s i s b a t c h e s o f s t o i c h i o m e t r y 8.10[xNa-(l-x)TEA]~Alz03~ l Y . O S i 0 2 . 6 3 0 H z O (OH-/SiOz 0.72), w h e r e x s p a n n e d f r o m 0 t o 1, h a v e b e e n h e a t e d a t 150°C f o r 72 h o u r s i n s t a t i c c o n d i t i o n s . P r e c i p i t a t e d s i l i c a was u s e d a s s i l i c a s o u r c e . T h e p h a s e s f o r m e d , t h e e l e m e n t a l composition o f t h e p r o d u c t and t h e y i e l d s of s i l i c o n and aluminum t h r o u g h o u t t h e p r e p a r a t i o n a r e r e p o r t e d i n t a b l e 1.
As t e t r a e t h y l a m m o n i u m c o n c e n t r a t i o n i n c r e a s e s t h e main p h a s e formed p a s s e s from a n a l c i m e t o z e o l i t e b e t a t h r o u g h m o r d e n i t e . The r o l e of t h e tetraethylammonium c a t i o n i n t h e formation of s i l i c a r i c h l o w - d e n s i t y c r y s t a l l i n e p h a s e s is e v i d e n t . From t a b l e 1 i t
emerges t h a t t h e aluminum i n c o r p o r a t i o n p r e s e n t s a minimum f o r i n t e r m e d i a t e v a l u e s o f t h e t e t r a e t h y l a m m o n i u m c o n c e n t r a t i o n . The phenomenon seems t o b e c o n s i s t e n t w i t h t h e f o r m a t i o n of m o r d e n i t e more s i l i c i c t h a n t h e c o r r e s p o n d i n g z e o l i t e b e t a . T h e m o r p h o l o g y o f some p r o d u c t s a r e d e p i c t e d i n f i g u r e 1. A t a TEA/SiOz r a t i o o f 0 . 5 0 ( f i g u r e l a ) r o u g h i r r e g u l a r g r a i n s o f z e o l i t e b e t a u p t o 1 . 5 cc l a r g e a r e f o r m e d , t o g e t h e r w i t h t a b u l a r c r y s t a l s of mordenite 12
CI
l o n g . Some o f t h e g r a i n s o f z e o l i t e
b e t a a p p e a r t o h a v e grown a r o u n d t h e e d g e s o f t h e m o r d e n i t e c r y s t a l s . Rare s p h e r e s o f a n a l c i m e ( n o t d e p i c t e d ) p r e s e n t a d i a m e t e r up t o 7
u . A t a TEA/SiOz r a t i o o f 0 . 1 6 ( f i g u r e
co-crystallization
lb)
of m o r d e n i t e and a n a l c i m e can be o b s e r v e d .
A n a l c i m e c r y s t a l s a r e l a r g e r ( u p t o 16
u)
a n d p r e s e n t t h e {211}
t r a p s z o h e d r o n h a b i t . When n o t e t r a e t h y l a m m o n i u m was p r e s e n t i n t h e s y n t h e s i s m i x t u r e ( f i g u r e l c ) t h e p r o d u c t i s composed by 2 5
u
l a r g e a n a l c i m e t r a p e z o h e d r a . The c r y s t a l e d g e s a r e e t c h e d by
c o r r o s i o n l i n e s , s u g g e s t i n g t h a t t h e s y s t e m h a s grown u n d e r s a t u r a t e d f o r a n a l c i m e . Analcime is s p r e a d of 2
CI
large
TABLE 1 Phase composition, e l e m e n t a l composition and material b a l a n c e a s a f u n c t i o n a f t h e tetraethvlaamonium c o n t e n t af-the s y n t h e s i s b a t c h .
T E AL
cristallinity %
mole f r a c t i o n - Al
beta
MUR
74
80
62 -
28
10
0.087
0.5U
52
48
-
-
0.185 0 252
u
0.16 0.00
Yield
(ele.ment/Si+Al>-
S i02
ANA
1UO
0.061
N
Na.
Si
A I-
0.083 0.069 0.022
0.027
0.94 0 61 0.35
0.85
0.000
0.160 0.179 0.305
0 25
0.37
U 75 0.78
606
F i g . 1. S y n t h e s e s a t d i . f f e r e n t t e t r a e t h y l a m m o n i u m c o n c e n t r a t i o n . F r o m t o p t o b o t t o m : TEA/Si02 0 . 5 0 ( a ) , 0 . 1 6 ( b ) a n d 0 . 0 0 ( c ) .
607
-\ 0.038
Q, 0.020
-a \
0.002
100
t (h)
200
F i g . 2 . N u c l e a t i o n l a g s o f z e o l i t e b e t a ( 0 ) a n d ZSH-12 ( 0 ) a s f u n c t i o n s o f t h e m o l e f r a c t i o n of a l u m i n u m .
f l a k e s o f g i s m o n d i n e , d e t e c t a b l e i n t r a c e amount f r o m t h e p o w d e r X-ray d i f f r a c t i o n s p e c t r a .
F ~ o u .~ e ~ l i tb.afa.t~-..Z.~n--_l.Z e Mixtures of s t o i c h i o m e t r y 0.08Na~0.30TEA-(x/2)Al~O~~(l-x)SiO~~
16Hz0 (OH-/SiOz
0.35) w i t h x r a n g i n g f r o m 0.038 t o 0 . 0 0 2 h a v e b e e n
h e a t e d a t 150°C i n a s t i r r e d r e a c t o r . S i l i c a s o l w a s t h e s o u r c e of s i l i c a . T h e i n d u c t i o n lags p r e c e d i n g t h e d e t e c t i o n of t h e c r y s t a l l i n e phases are r e p o r t e d i n f i g u r e 2. I n moderately s i l i c i c s y s t e m s ( S i / A l 2 5 - 5 0 ) z e o l i t e b e t a f o r m s b e f o r e ZSM-1%. T h e n u c l e a t i o n l a g o f ZSH-12 d e c r e a s e s when S i / A 1 i n c r e a s e s a n d a t h i g h e r S i / A 1 r a t i o ( 5 0 0 ) ZSM-12
i s t h e f i r s t and u n i q u e
c r y s t a l l i n e p h a s e f o r m e d . Aluminum seems n e e d e d t o f o r m t h e l a d d e r s o f 4-membered r i n g s t y p i c a l o f z e o l i t e b e t a ( r e f s . 16, 17), a l r e a d y p r o p o s e d a s a l u m i n u m l o c a t i o n ( r e f . 18). On t h e o t h e r hand, t h e p r e s e n c e of s i l i c o a l u m i n a t e s p e c i e s a p p e a r s t o i n h i b i t tlie f o r m a t i o n of t h e c r i s t o b a l i t e - l i k e w a l l s of i n t e r c o n n e c t e d 6-membered r i n g s t y p i c a l o f ZSM-12
( r e f . 19).
O n t h e o t h e r hand a l u m i n u m a l s o m o d i f i e s t h e p h y s i c s o f t h e
s y s t e m . I f t h e amount o f s o l i d f r a c t i o n r e c o v e r e d f r o m i n t e r m e d i a t e s a m p l i n g is c o n s i d e r e d , m a i n d i f f e r e n c e s a r e f o u n d a s a f u n c t i o n of composition.
I n t h e more a l u m i n i c s y s t e m a n
amorphous s o l i d is p r e s e n t t h r o u g h o u t t h e h y d r o t h e r m a l t r e a t m e n t . I n t h e c a s e o f t h e more s i l i c i c s t o i c h i o m e t r y t h e s y s t e m i n i t i a l l y c o n s i s t s o f a c l e a r s o l u t i o n . G e l l i n g , and r e c o v e r y of a n y s o l i d f r a c t i o n from s a m p l i n g , b e g i n o n l y a f t e r one d a y a t t h e s y n t h e s i s t e m p e r a t u r e . Anyway a n a m o r p h o u s s o l i d i s a l w a y s p r e s e n t when zeolite nucleates.
608
C ~ m p e tf;iQo...he.tween i z.e.Q.l.i.t.e..hzta...an.d._ZS.M.-...5 T h e n a t u r e o f t h e s o u r c e o f s i l i c a may i n f l u e n c e t h e p h a s e s e l e c t i v i t y o f t h e s y n t h e s i s . Two g e l s o f s t o i c h i o m e t r y
1.2Na20.9.2TEA~0~A120~~78SiOz~128O (O HH zO -/Si02
0.24) h a v e b e e n
p r e p a r e d f r o m p r e c i p i t a t e d s i l i c a o r s i l i c a s o l . Under hydrothermal treatment i n a non-stirred
a u t o c l a v e t h e experiment
with p r e c i p i t a t e d silica has given o r i g i n t o z e o l i t e b e t a . In t h e e x p e r i m e n t w i t h s i l i c a s o l u n d e r t h e same h y d r o t h e r m a l c o n d i t i o n s
ZSH-5 h a s b e e n f o r m e d . T h i s s e l e c t i v i t y e f f e c t h a s n o t b e e n o b s e r v e d when t h e s y n t h e s i s h a s b e e n c a r r i e d o u t u n d e r s t i r r i n g . Two g e l s o f stoichiometry 2.2Na~0-8.8TEA~0~A1~0~~51Si0~~ (OH-/Si02 1040Hz0 0.36) have been p r e p a r e d from p r e c i p i t a t e d s i l i c a or s i l i c a s o l . Under h y d r o t h e r m a l t r e a t m e n t b o t h g e l s g a v e o r i g i n t o z e o l i t e b e t a . The c r y s t a l l i z a t i o n from s i l i c a s o l f e a t u r e d a l o n g e r i n d u c t i o n t i m e (22 h o u r s v e r s u s 15) a n d b i g g e r c r y s t a l s o f z e o l i t e ( 1 . 5 v e r s u s 0.8 p ) .
T h e n u c l e a t i o n o f z e o l i t e b e t a is h e n c e e a s i e r when t h e s y n t h e s i s g e l is p r e p a r e d f r o m p r e c i p i t a t e d s i l i c a , b o t h i n s t i r r e d a n d s t a t i c c o n d i t i o n s . On t h e c o n t r a r y , s t i r r i n g seems t o h i n d e r t h e f o r m a t i o n o f ZSH-5.
T h i s e f f e c t a p p e a r s i n agreement
w i t h t h e l o c a t i o n o f ZSH-5 e m b r y o s a t t h e e x t e r n a l s u r f a c e o f t h e a m o r p h o u s g r a i n s ( r e f . 8), w h e r e s t i r r i n g may a f f e c t t h e e p i t a x i a l cond i t i o n s .
F r ~ mZ.SMk5 t Q ZSM-12 I n t h e p r e v i o u s l y c i t e d e x p e r i m e n t o f s t o i c h i o m e t r y 1.2Na2O.
Y . Z T E A 2 0 ~ A 1 ~ 0 s ~ 7 8 S i 0 2 ~ 1 2 8f0eHa~t O u r i n g s i l i c a s o l , ZSH-5 p r o v e d u n s t a b l e on l o n g e r s y n t h e s i s t i m e . The p h a s e c o m p o s i t i o n o f t h e s y s t e m is r e p r e s e n t e d i n t a b l e 2 a s a f u n c t i o n o f t h e s y n t h e s i s
t i m e . A f t e r t e n d a y s i n h y d r o t h e r m a l c o n d i t i o n s ZSH-12 n u c l e a t e s
a n d ZSM-5 b e g i n s t o d i s s o l v e . A f t e r t w e n t y d a y s ZSM-12 is t h e main c r y s t a l l i n e p h a s e p r e s e n t . The e v o l u t i o n o f t h e s o l i d c o m p o s i t i o n i s a l s o r e p o r t e d i n t a b l e 2 . The o r g a n i c c o n t e n t f e a t u r e s a
minimum i n c o r r e s p o n d e n c e w i t h t h e h i g h e r y i e l d o f ZSH-5. The e l e c t r o n m i c r o g r a p h s o f t h e s o l i d a t d i f f e r e n t s y n t h e s i s t i m e s
a r e r e p o r t e d i n f i g u r e 3 . F i g u r e 3a r e p r e s e n t s t h e z e o l i t e f o r m e d a f t e r 10 d a y s i n h y d r o t h e r m a l c o n d i t i o n s . T h e X-ray d i f f r a c t i o n p a t t e r n o f t h i s s a m p l e i n d i c a t e s t h e p r e s e n c e o f t r a c e s of ZSM-12 besides well-crystallized
ZSM-5. The s a m p l e c o n s i s t s o f 35
c h a r a c t e r i s t i c e u h e d r a l c r y s t a l s o f ZSM-5.
p
long
609
Fig. 3. Synthesis of ZSM-5 and ZSM-12. Crystallization time, from top to bottom: ( a ) 10, (b) 15, ( c ) 2 0 d a y s .
610
TABLE 2 P h a s e c o m p o s i t i o n , e l e m e n t a l c o m p o s i t i o n and material b a l a n c e d u r i n g the time (days) ._
3 10 15 20
4-~xma!i.ii~n.nf _zSK:3xd-_zSHzL2, Y-isI d
crisfa.lLi~fy--% m o l e f r a c t i o n L e l e m e n t / S i + A l L
ZSMz5 __ - ZSK=l2..
-Al-..-.-.--N--_ 0.026 0.022 0.024 0.025
amorphous 70 3 64 14 39 40
---N& 0.053 0.041 0.045 0.053
0.027 0.007 0.007 0.006
_Si-Al0.57 0.83 0.94 0.91
0.59 0.73 0.89 0.89
E x t e n d e d t w i n n i n g i m p l i e s t h a t t h e l a t e r a l f a c e s of t h e p r i s m s a r e e s s e n t i a l l y ( 1 0 0 ) f a c e s . T h e c r y s t a l s p r e s e n t a l s o some l e s s common f e a t u r e s . T h e i r l a t e r a l f a c e s a r e s p r e a d o f r e c t a n g u l a r e t c h i n g f i g u r e s , c l e a r l y showing t h a t t h e s y n t h e s i s system h a s grown u n d e r s a t u r a t e d f o r ZSM-5.
This e f f e c t should be accounted
f o r by t h e p r e s e n c e o f a l e s s s o l u b l e c r y s t a l l i n e p h a s e . I n d e e d a d e n s e p o p u l a t i o n o f s u b m i c r o n p r i s m a t i c c r y s t a l s are embedded i n t h e ( 1 0 0 ) f a c e s of ZSM-5.
T h e smaller p r i s m s a p p e a r t o h a v e g r o w n
up f r o m t h e s u r f a c e p r e c i s e l y o r i e n t e d w i t h r e g a r d s t o t h e ZSM-5 l a t t i c e . T h e i r main a x i s l a y s i n t h e (010) p l a n e a n d f o r m s a 53” angle w i t h the
direction
o f ZSM-5.
Also t h e n o r m a l t o o n e
f a c e o f t h e smaller p r i s m s l a y s i n t h e ( 0 1 0 ) p l a n e o f ZSH-5. I t c a n b e o b s e r v e d t h a t t h e o r i e n t a t i o n of t h e smaller c r y s t a l s
c o u l d c o r r e s p o n d t o 234-12 c r y s t a l s g r o w i n g i n t h e d i r e c t i o n f r o m a ( 2 0 1 ) f a c e l a y i n g o n t h e ZSM-5 ( 1 0 0 ) f a c e . T h i s o r i e n t a t i o n a c c e p t e d , t h e ZSM-12
< 0 1 0 > d i r e c t i o n s h o u l d c o r r e s p o n d t o t h e ZSM-
5 < 0 1 U > d i r e c t i o n . An e p i t a x i a l r e l a t i o n c a n b e f o u n d b e t w e e n t h e b parameters:
5 . 0 and 1 9 . 9 A , r e s p e c t i v e l y . O r i e n t e d n u c l e a t i o n of
a z e o l i t e over a n o t h e r z e o l i t e f e a t u r i n g e p i t a x y i n only one d i r e c t i o n h a s a l r e a d y b e e n o b s e r v e d , f o r i n s t a n c e i n t h e case o f o f f r e t i t e a n d m a z z i t e ( r e f . 14). F i g u r e 3b d e p i c t s t h e s o l i d f o r m e d a f t e r 1 5 d a y s o f s y n t h e s i s . The s u r f a c e o f t h e ZSM-5 c r y s t a l s is d e e p l y e t c h e d a n d a h o p p e r p i t i s open i n t h e ( 0 1 0 ) f a c e s . The s u r f a c e o f t h e l e s s - e r o d e d c r y s t a l s is s p r e a d o f s l i g h t l y f l a t t e n e d s p h e r e s 3
LI
large, with a
r o u g h s u r f a c e . X-ray d i f f r a c t i o n a l l o w s t o i d e n t i f y t h e s p h e r i c a l c r y s t a l s a s ZSM-12.
Some c h a r a c t e r i s t i c p i t s i n t h e ZSM-5 s u r f a c e
seem t o c o r r e s p o n d t o t h e o r i g i n a l l o c a t i o n o f d e t a c h e d s p h e r o i d a l c r y s t a l s . The i n war d g r o w t h o f a c r y s t a l a f t e r i t s n u c l e a t i o n on
611 a n h e t e r o g e n e o u s s u r f a c e is a we1 -known p h e n o m e n o n . I t h a s b e e n carefully described, for instance
i n t h e case o f ZSM-5 n u c l e a t i o n
o v e r amorphous s i l i c a ( r e f . 8 ) . A c h a n g e o f m o r p h o l o g y o f ZSM- 2 f r o m e u h e d r a l p r i s m s t o h i g h -
i n d e x s u r f a c e s p h e r e s i s s u c h a n u n u s u a l phenomenon t h a t i t d e s e r v e s a comment. I t c o u l d b e e x p l a i n e d o n l y b y a s h a r p r i s e o f s u p e r s a t u r a t i o n , as i f a p r e v i o u s l y confined source of s o l u b l e s p e c i e s had b e e n made a c c e s s i b l e t o t h e s o l u t i o n . T h e h o p p e r p i t s i n t h e (010) f a c e s o f t h e ZSM-5 c r y s t a l s p r o p o s e t h e c o r e o f t h e ZSM-5 c r y s t a l s a s a p o s s i b l e s o u r c e o f more s o l u b l e s p e c i e s . U n d e r t h i s h y p o t h e s i s , once t h e o u t e r l a y e r of (100) t w i n n i n g d i s s o l v e d , t h e r a p i d d i s s o l u t i o n o f t h e ( 0 1 0 ) f a c e s would i n c r e a s e t h e r a t e o f g r o w t h o f ZSM-12 b e y o n d t h e t h r e s h o l d o f d e v e l o p m e n t o f h i g h index f a c e s from t h e e d g e s of t h e f l a t f a c e s . The f i g u r e 3 c , r e p r e s e n t i n g t h e s o l i d a f t e r 20 d a y s o f s y n t h e s i s , c l e a r l y shows t h e growth of t h e f l a t t e n e d s p h e r e s of ZSM-12,
t h e s h r i n k i n g o f t h e ZSM-5 c r y s t a l s a n d t h e d e e p e n i n g of
t h e hopper p i t s i n t h e (010) f a c e s of t h e l a t t e r .
CONCLUSIONS C h e m i c a l and p h y s i c a l f a c t o r s a f f e c t t h e s e l e c t i v i t y o f formation of t h e c r y s t a l l i n e p h a s e s t y p i c a l of t h e tetraethylammonium m e d i a . The i s s u e of b o r d e r l i n e s i t u a t i o n s i n w h i c h s e v e r a l p h a s e s c a n b e f o r m e d may b e d i r e c t e d b y t h e n a t u r e of s o l i d s u r f a c e s p r e s e n t i n t h e s y n t h e s i s s y s t e m .
ACKNOWLEDGMENTS Many t h a n k s a r e d u e t o t h e S e r v i c e C e n t r a l d ' A n a l y s e o f CNRS i n S o l a i z e f o r e l e m e n t a l a n a l y s i s and t o George Nabias f o r e l e c t r o n microscopy e x p e r i m e n t s . REFERENCES 1
2
E . L . W a d l i n g e r , G . T . Kerr a n d a n d E . J . R o s i n s k i , C a t a l y t i c c o m p o s i t i o n o f a c r y s t a l l i n e z e o l i t e , US P a t . 3 , 3 0 8 , 0 6 9 ( 1 9 6 7 ) . J . P e r e z - P a r i e n t e , J . A . M a r t e n s and P.A. J a c o b s , C r y s t a l l i z a t i o n m e c h a n i s m o f z e o l i t e b e t a f r o m ( T E A ) z O , NazO and K2O c o n t a i n i n g a l u m i n o s i l i c a t e g e l s , A p p l i e d C a t a l . , 31 ( 1987) 35-64.
3
J . P e r e z - P a r i e n t e , J . A . M a r t e n s and P . A . J a c o b s , F a c t o r s a f f e c t i n g t h e s y n t h e s i s e f f i c i e n c y o f z e o l i t e BETA f r o m a l u m i n o s i l i c a t e g e l s c o n t a i n i n g a l k a l i and t e t r a e t h y l a m m o n i u m i o n s , Z e o l i t e s , 8 (1988) 46-53.
612
U e w a e l e , L . M a i s t r i a u , J . B . Nagy a n d E . G . Derouane, D i r e c t i n g p a r a m e t e r s i n t h e s y n t h e s i s of z e o l i t e s ZSM-20 and b e t a , ACS Symp. S e r i e s , 3 9 8 ( 1 9 8 9 ) 5 1 8 - 5 4 3 . D.E.W. V a u g h a n , M . M . J . T r e a c y , J . M . Newsam, K.G. S t r o h m a i e r a n d W.J. M o r t i e r , S y n t h e s i s a n d c h a r a c t e r i z a t i o n o f z e o l i t e SZM-20, ibidem 544-559. Y . F . C h u , P r o c e s s f o r ZSM-11 p r o d u c t i o n , US P a t . 4 , 8 4 7 , 0 5 5 (1989). E . J . K o s i n s k i and M.K. R u b i n , C r y s t a l l i n e z e o l i t e ZSM-12, US P a t . 3,832,449 (1974). J . C . J a n s e n , C . W . R . E n g e l e n a n d H . v a n Bekkum, C r y s t a l g r o w t h r e g u l a t i o n a n d m o r p h o l o g y o f z e o l i t e s i n g l e c r y s t a l s o f t h e MFI t y p e , ACS Symp. S e r . , 398 ( 1 9 8 9 ) 2 5 7 - 2 7 3 . J . B.Nagy, Ph. B o d a r t , H . C o l l e t t e , C . F e r n a n d e z , Z . G a b e l i c a , A . N a s t r o and R . A i e l l o , C h a r a c t e r i z a t i o n of c r y s t a l l i n e a n d a m o r p h o u s p h a s e s d u r i n g t h e s y n t h e s i s o f (TPA,M)-ZSM-5 Z e o l i t e s ( M = L i , Na, K ) . J . Chem. S O C . F a r a d a y T r a n s . 1 , 85 ( 1 9 8 9 ) 2749-2765. F . D i R e n z o , F . Remoue, P . M a s s i a n i , F . F a j u l a , F . F i g u e r a s a n d T . Des C o u r i e r e s , C r y s t a l l i z a t i o n k i n e t i c s o f z e o l i t e TON, Zeolites, submitted. C . S . Cundy, p e r s o n a l c o m m u n i c a t i o n . I.Y. Chan a n d S . I . Z o n e s , A n a l y t i c a l e l e c t r o n m i c r o s c o p y (AEM) o f C u b i c P z e o l i t e t o Nu-3 z e o l i t e t r a n s f o r m a t i o n , Z e o l i t e s , 5 ( 1 9 8 9 ) 3-11. E . d e Vos B u r c h a r t , J . C . J a n s e n a n d H . v a n Bekkum, O r d e r e d o v e r g r o w t h o f z e o l i t e X o n t o c r y s t a l s o f z e o l i t e A, Z e o l i t e s , 9 ( 1 9 8 9 ) 432-435. F . F a j u l a , F . F i g u e r a s , C . Gueguen a n d R . D u t a r t r e , B i n a r y z e o l i t i c s y s t e m s , t h e i r s y n t h e s i s a n d t h e i r u t i l i z a t i o n , US P a t . 4,847,224 (1989). F . F a j u l a , S . N i c o l a s , F . D i R e n z o , C . G u e g u e n and F . F i g u e r a s , K i n e t i c s and m e c h a n i s m o f c r y s t a l g r o w t h o f z e o l i t e o m e g a , ACS Symp. S e r . , 398 ( 1 9 8 9 ) 4 9 3 - 5 0 5 . R . B . L a P i e r r e , A . C . Kohrman J r . , J . L . S c h l e n k e r , J . D . Wood, M . K . Kubin and W . J . R o h r b a u g h , T h e f r a m e w o r k t o p o l o g y o f ZSM-12: A h i g h - s i l i c a z e o l i t e , Z e o l i t e s , 5 ( 1 9 8 5 ) 3 4 6 - 3 4 8 . J . M . Newsam, M.M.J. T r e a c y , W . T . K o e t s i e r a n d C . H . d e G r u y t e r , S t r u c t u r a l c h a r a c t e r i z a t i o n of z e o l i t e b e t a , Proc. ti. S O C . ( L o n d o n ) A, 42il ( 1 9 8 8 ) 375-4135. N.A. B r i s c o e , J . L . C a s c i , J . A . D a n i e l s , D.W. J o h n s o n , M.D. S h a n n o n a n d A. S t e w a r t , Some a s p e c t s o f t h e s y n t h e s i s , c h a r a c t e r i z a t i o n a n d p r o p e r t i e s o f z e o l i t e Nu-2, S t u d . S u r f a c e S c i . C a t a l y s i s 45A, E l s e v i e r , A m s t e r d a m 1 9 8 9 , 1 5 1 . J . B . H i g g i n s , K.B. L a P i e r r e , J . L . S c h l e n k e r , A . C . Rohrman J r . , J . D . Wood, G . T . Kerr a n d W . J . R o h r b a u g h , T h e f r a m e w o r k topology of z e o l i t e b e t a , Z e o l i t e s , (1988) 446. Z . Gabelica, N .
10
11 12 13 14
15 16
17 18
19
G. Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
613
SYNTHESIS OF FERRIERITES WITH HIGH GALLIUM CONTENT M.A. Camblor*, J.A. Martens, P.J. Grobet and P.A. Jacobs Laboratorium voor Oppervlaktechemie, Kathol ieke Universitei t Leuven, Kardinaal Mercierlaan 92, 8-3030 Leuven, Belgium
* on leave from Instituto de Catalisis y Petroleoquimica, C.S.I.S, Madrid, Spain.
SUMMARY Ferrierite samples with Ga/GatAl compositions ranging from 0 t o 90% are synthesized hydrothermally in the presence o f piperidine. The incorporation of aluminium and gallium in the zeolite is monitored with 27Al, 29Si and 71Ga MAS - NMR . INTRODUCTION The isomorphic substitution of A1 and Si by other elements in the framework of zeolites is of considerable interest. For example, gallium zeolites possess surface chemical properties which are different from those o f aluminosil icate zeolites and have special catalytic properties, e.g. in LPG aromatization (refs. 1 and 2). A number of (Si, Ga)-analogs o f (Si,Al)-zeolites have been made by direct synthesis (see e.g. the references cited in ref. 3). In this paper we report on the synthesis of aluminosilicate ferrierite samples with extensive isomorphous substitution by Ga. We developed a new synthesis method according to which alumino-galloferrierites can be obtained with Si/(GatAl) ratio’s ranging from 7.1 t o 3.4. The gallium content of alumino-galloferrierites can be considerably higher than the aluminium content of aluminosilicate ferrierites, for which the lower limit of Si/A1 is ca. 6 (ref. 3). EXPERIMENTAL Synthesis mixtures with the following molar compositions are used: 23.3 Na20 : T2O3 : 46.5 Si02 : 18.6 CSHIONH : 976 H20 where T stands for Ga and Al. The Ga/(GatAl) fractions used were 0.0, 0.5, 0.75 and 0.9. The gels were prepared as follows. In a first solution NaOH (Merck) was added t o waterglass (Merck) with stirring and then piperidine (Aldrich) was added. The second solution, consisting o f A12(S04)3.18H20
614
(U.C.6)
and Ga(N03).xH20 (Janssen) solved i n water, was added t o t h e f i r s t
solution
with
agitation.
The
synthesis
mixtures
were
transferred
into
s t a i n l e s s steel autoclaves o f 120 m l ; a g i t a t i o n was performed by r o t a t i n g t h e autoclaves a t 50 rpm. crystallization
C r y s t a l l i n e phases were obtained a f t e r ca.
22 h o f
a t 473 K.
The phase p u r i t y o f t h e f e r r i e r i t e products was v e r i f i e d using XRD patterns, recorded w i t h a Siemens instrument, which was equipped w i t h a Mc Brawn p o s i t i o n s e n s i t i v e detector. The chemical compositions were determined by I C P . The MAS NMR spectra were obtained w i t h a Bruker 400 MSL spectrometer.
The
experimental conditions used were as f o l l o w s :
I
Parameter NMR frequency (MHz) Pulse angle ( * ) Pulse l e n g t h ( p s ) R e p e t i t i o n time ( s ) Spinning r a t e (kHz) Number o f scans Reference f o r chemical s h i f t The
29Si
spectra
were
I
29Si
79.5 45 4 3 3 18,000 TMS
deconvoluted
1
27Al 104.2 20 0.6 0.1 5 3000 AlC13
according
71Ga 122.0 20 0.6 0.1 15
150,000 Ga(N03) 2 to
a
procedure
described
previously (ref. 4). RESULTS AND DISCIUSSION The XRD p a t t e r n s o f t h e c r y s t a l l i z a t i o n products are shown i n Fig.1. The XRD l i n e s which are c h a r a c t e r i s t i c f o r t h e f e r r i e r i t e f a m i l y o f m a t e r i a l s are present. Traces o f quartz may be present i n samples S3 and S4. The T-atom composition o f synthesis g e l s and c r y s t a l l i z a t i o n products are given i n Table 1. I t can be seen t h a t the Ga/GatAl
r a t i o o f the f e r r i e r i t e product r e f l e c t s
t h a t o f the synthesis mixture. Scanning e l e c t r o n micrographs o f t h e samples are shown i n Fig.2. Sample S 1 c o n s i s t s o f elongated c r y s t a l s w i t h a l e n g t h o f up t o 20 have t h e shape o f hexagonal
prisms.
pm.
Some c r y s t a l s
I n sample S2 p l a t e l e t and r o d - t y p e
morphologies are observed. The rods reach a l e n g t h o f up t o 40 Fm. Samples S3 and S4 e x h i b i t the p l a t e l e t - t y p e morphology. I t seems t h a t t h e morphology changes from r o d - l i k e t o p l a t e l e t - l i k e when t h e g a l l i u m content i s increased.
615
s1 5
10
15
20
25
30
35
40
45
50
55
50
55
50
55
50
55
2e
5
10
15
20
25
30
35
40
45
2 e
5
10
15
20
25
30
35
40
45
2e
5
10
15
20
25
30
35
40
45
2e Fig. 1. The XRD patterns o f samples S1-S4.
616
Fig. 2 . The scanning electron micrographs o f the samples Sl-S4.
617
TABLE 1 T-atom c o mp os it i o n o f s y n t h e s i s m i x t u r e and f e r r i e r i t e p r o d u c t s Sample
S y nt h es i s q e l Ga/(GatAl )
C r y s t a l 1 i z a t i o n Droduct Ga/(GatAl ) S i / ( G a t A l ) Si/(GatAl ) ICP
ICP
Ga/U.C.
Al/U.C.
MAS NMR
s1
0
0
7.6
7.1
0
4.4
s2
0.5
0.4
6.4
5.9
2.1
3.2
s3
0.75
0.7
4.1
4.4
4.7
2.0
s4
0.9
0.9
4.1
3.4
7.4
0.8
D i r e c t evidence f o r t h e A1 and Ga i n c o r p o r a t i o n i n t h e f e r r i e r i t e s t r u c t u r e i s f o und i n t h e 27Al and 71Ga MAS-NMR s p e c t r a ,
shown i n F igs. 3
and 4,
r e s p e c t i v e l y . The 2 7 A l resonance a t 54 ppm ( F ig. 3) and t h e 71Ga MAS-NMR s i g n a l a t 152 ppm (F ig. 4 ) a r e c h a r a c t e r i s t i c o f a t e t r a h e d r a l environment. The "Si
MAS-NMR
s p e c t r a o f t h e samples a r e shown i n Fig.5.
chemical s h i f t v a l u e s o f t h e "Si spectrum o f S4,
l i n e s summarized i n T able 2.
and t h e The "Si
which c o n t a i n s t h e l a r g e s t amount o f g a l l i u m (T able
l),
c l e a r l y d i s p l a y s f i v e l i n e s c o r r e s p o n d i n g t o t h e f i v e p o s s i b l e Si(n(A1,Ga)) s t r u c t u r a l s urro u n d i n g s , w i t h 0 < n < 4. I n t h e 29Si s p e c t r a o f t h e S2 and S3 samples,
which c o n t a i n i m p o r t a n t amounts o f b o t h aluminium and g a l l i u m ,
no
d i r e c t d i s c r i m i n a t i o n between Si(nA1) and S i (nGa) s i g n a l s c o u l d be made, s i n c e no
extra
splittings
in
the
Si(n(A1,Ga))
lines
are
observed
(Fig.5).
Nev ert heles s , t h e n = 0 l i n e s have a tendency t o s h i f t t o l o w e r ppm values when h i g h e r Ga l o a d i n g s a r e achieved ( Ta b l e 2); t h e Sil(Ga,Al) g a l l i u m content
and SiZ(Ga,Al)
t h e chemical s h i f t v a l u e s o f
l i n e s , f o r example, move u p f i e l d w i t h i n c r e a s i n g
.
T h i s i n f l u e n c e o f t h e presence o f Ga on t h e 2 9 S i chemical s h i f t was a l s o observed i n o t h e r g a l l i u m s i l i c a t e s ( r e f s . 5 and 6 ) . Probably,
a homogeneous d i s t r i b u t i o n o f A1 and Ga i n t h e f e r r i e r i t e framework can be assumed. The Si/GatAl
r a t i o s o f t h e samples were c a l c u l a t e d f r o m t h e i n t e n s i t i e s o f
t h e i n d i v i d u a l S i (n(A1 ,Ga)) 1 i n e s o f t h e d e convolut ed 29Si s p e c t r a (T able 1). The r e s u l t s a r e i n agreement w i t h t h e r e s u l t s f rom ICP.
618
Fig. 3. The 2 7 A l MAS NMR spectra o f samples Sl-S4. The *’s indicate the position o f the spinning side bands.
B
52 51
Fig. 4 . The 71Ga MAS NMR spectra o f samples Sl-S4.
619
SiO(AI,Ga)
I Sil(A1,Ga)
I
Fig. 5. The "Si
MAS NMR spectra o f samples Sl-S4.
620
TABLE 2 2 9 S i chemical s h i f t s of S i (n(A1 ,Ga)) 1 i n e s i n f e r r i e r i t e . 29Si chemical s h i f t s ( m m l
Sample
Si4(A1 ,Ga) Si3(A1 ,Ga) Si2(A1 ,Ga) S i l ( A 1 ,Ga) SiO(A1 ,Ga)
s1
/
/
-99.1
-105.8
-112.4
s2
/
/
-97.4
-105.1
-112.3
s3
/
-90.6
-96.5
-105.2
-112.9
s4
-85.0
-90.5
-96.5
-104.6
-112.6
I f one assumes t h e complete i n c o r p o r a t i o n o f b o t h Ga and A1 i n t h e z e o l i t e as suggested by t h e absence o f s i g n a l s due t o o c t a h e d r a l c o o r d i n a t i o n i n t h e 71Ga and 27Al s p e c t r a ( F i g s . 3 and 4 ) , one can c a l c u l a t e t h e amount o f Ga and A1 atoms p e r u n i t c e l l f r o m a c o m b i n a t i o n o f t h e Ga/(GatAl) and S i / ( G a t A l ) r a t i o s , measured b y I C P and MAS NMR, r e s p e c t i v e l y and c o n s i d e r i n g a u n i t c e l l o f 36 T-atoms ( T a b l e 1). S4 c o n t a i n s 7.4 g a l l i u m atoms p e r u n i t c e l l . T h i s i s a h i g h e r g a l l i u m c o n t e n t compared t o t h e 4.7 g a l l i u m atoms p e r u n i t c e l l i n an a l u m i n i u m - f r e e g a l l o f e r r i e r i t e , r e p o r t e d by Sulikowski e t a l . ( r e f . 3). A l u m i n o s i l i c a t e f e r r i e r i t e s c o n t a i n t y p i c a l l y 6 o r l e s s aluminium atoms p e r u n i t c e l l ( r e f . 7 ) . It seems t h a t i n t h e presence o f g a l l i u m , t h e u n i t c e l l can c o n t a i n up t o 8 t r i v a l e n t atoms (T able 1 ) . The y i e l d o f z e o l i t e , c a l c u l a t e d as t h e w e i g h t o f t h e p r o d u c t over t h e w e i g h t o f framework,
Si02,
Ga2O3 and A1203 i n t h e s y n t h e s i s
mixture,
i s p l o t t e d against
the
Ga/GatAl f r a c t i o n i n Fi g . 6 . I t shows t h a t t h e replacement o f aluminium w i t h g a l l i u m reduces t h e p r o d u c t y i e l d c o n s i d e r a b l y . When no aluminium was p r e s e n t i n t h e s y n t h e s i s m i x t u r e (Ga/(GatAl) = l . O ) , t h e c r y s t a l l i z a t i o n o f f e r r i e r i t e f a i l e d . T h i s r e s u l t suggests t h a t t r a c e s o f A1 a r e necessary, p r o b a b l y a t t h e n u c l e a t i o n s t a ge o f t h e s y n t h e s i s , and t h a t A1 can be r e p l a c e d by Ga i n t h e c r y s t a l gro w t h process. The A1 c o n t e n t p e r u n i t c e l l i s p r o p o r t i o n a l t o t h e a b s o l u t e i n t e n s i t y o f t h e 2 7 A l resonance, as shown i n F i g u r e 7 . The 71Ga a b s o l u t e l i n e i n t e n s i t i e s do n o t show p r o p o r t i o n a l i t y
w i t h t h e Ga c o n t e n t o f t h e u n i t c e l l
Q u a n t i f i c a t i o n o f t h e g a l l i u m c o n t e n t w i t h 71Ga
(Fig.7).
MAS-NMR i s n o t p o s s i b l e
d e s p i t e t h e use o f a resonance magnetic f i e l d o f 9.4 T, a s t r o n g R F - f i e l d o f 10 mT, a s h o r t RF-pulse o f 0.6
ps
and a h i g h s p i n n i n g f requency o f 15 kHz. The
71Ga MAS NMR l i n e s a r e s t i l l b r o a d ( F i g . 4 ) and, p r o b a b l y , s t r o n g l y a f f e c t e d by
621
\
a
M
0
'2 3 0 d 0 20
.-
c1
6
N
Irl m
10
cd c1
m h
6
o 0.25
0.00
1 .oo
0.75
0.50
Ga/(Ga+Al) in initial mixture Fig. 6. The yield o f alumino-galloferrierite zeolite as function of the Ga/(GatAl) fraction in the synthesis mixtures o f samples S1-S4.
120 100 9 80
cd 60
v
n
40 20 0
0
1
2
3
4
5
6
7
8
Al or Ga / (u.c.) Fig. 7. The 27Al and 71Ga NMR line intensities versus their content per unit cell in the samples Sl-S4.
622
s p i n n i n g s i d e bands due t o t h e i n c o m p l e t e r e d u c t i o n o f t h e second o r d e r qu adru polar e f f e c t s . CONCLUSIONS The f e r r i e r i t e
structure
seems
to
be p a r t i c u l a r l y
susceptible t o
Ga
i n c o r p o r a t i o n . The l o w e s t Si/Ga r a t i o o f alumino g a l l o f e r r i e r i t e s o b t a i n e d i n t h i s work i s ca. 4, whereas t h e S i / A l r a t i o s f ound w i t h a l u m i n o s i l i c a t e f e r r i e r i t e s a r e equal t o o r l a r g e r t h a n 6.
Traces o f aluminium seem t o be
necessary i n o r d e r t o c r y s t a l l i z e f e r r i e r i t e s w i t h h i g h g a l l i u m c o n t e n t s . The g a l l i u m c o n t e n t o f alumino g a l l o f e r r e i r i t e s c o u l d n o t be q u a n t i f i e d w i t h 'lGa MAS-NMR,
even when u s i n g a s p i n n i n g f r e q u e n cy o f 15 kHz.
ACKNOWLEDGMENTS The a u t h o r s acknowledge t h e Fl e m i s h NFWO and t h e B e l g i a n M i n i s t e r y f o r Science P o l i c y f o r a r e s e a r c h g r a n t . J.A.M. and P.J.G acknowledge t h e F lemish NFWO f o r a re s e a r c h p o s i t i o n as Research A s s o c i a t e and S e n i o r Research A s s oc iat e , r e s p e c t i v e l y . REFERENCES
1 J.R. Mowry, R.F. Anderson and J.A. Johnson, O i l Gas J., 83 (1985) 128. 2
H. Kitagawa, H. Sendoda and Y. Ono, J. C a t a l . ,
3
B. S u l i k o w s k i and J. K l i n o w s k i , J. Chem. SOC., Chem. Commun., 1289 (1989).
4
P.J.
5
D.E.
6
S.
7
SOC. Jpn., 58 (1985) 52. P.A. Jacobs and J.A. Martens, Stud. S u r f . S c i . C a t a l . 33 (1987) 217.
Grobet, H.
Geerts, M. T i e l e n , J.A.
101 (1986) 12.
Martens and P.A.
Jacobs,
Stud.
S u r f . S c i . C a t a l . , 46 (1989) 721. Vaughan,
Dwyer (Eds.),
M.T.
M e l c h i o r and A.J.
Jacobson,
i n : G.D.
St uck and F.G.
I n t r a z e o l i t e Chemistry, ACS Symp. Ser. 218 (1983) 231.
Hayashi, K. Suzuki, S. Shin, K. Hayamizu and 0. Yamamoto, B u l l . Chem.
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 01991 Elsevier Science Publishers B.V., Amsterdam
S Y N T H E 5 I S O F ARTIFICIAL %EC>LIYE-LIKF M.Ii.EPBA?iX\',
623
N1O~If:'lAIPJITE
L!.n4.C-AP.'l~f~h.(-:V, C.B.TP.GIYEV
Irlstitute of Inorganic a n d Fhysica.1 Ckemistry Gf t h e A zerbaijan S S R Academy of S c i e n c e s , 370143 Baku., Narimanov p r o s p e c t 29, USSF
suR.I MAR 1T h e p e p e r clenlsi wit11 kiyclrotl-:errv. e.1 t l cLricfi.>r mation of artificial g k s s e s into zeolite-.like silica.t€?s,systr?r~leticallS. studied.Syr.thetic mowntainite,teing a n a n a l o g u e of prirrary n r i r e r a l , w a s p r o d u c e d for t l i e first tir!ic.Ffft:(.t:- (.I vc:riijt .c I . ~c t c r s , i . e . t e r ~perclture,coricentration,g r a s s grading,nature o f s c l ver,t, solid plie.se/liquid p h a s e ratic, on p h a s e forniaticiri w a s c o x s i d e r e d . Zeolite-.like n a t u r e of sodiuri hydros.ilica.te w e s p r o v e d by radiography., t h e r m o g r a ~ i m e t r i c u l , chenlical &nd electrorir;,icro..copic methc.-ds of irivesti
-
g,i+tiCIr..
I i i T K ODUCTIOP I Due to uriique s t r u c t u r a l p r o p e r t i e s a n d wide r a n g e of c h a - a c t e r i u t i c s ze o lites a r e u s e d iri cateil)rtic, a d s o r p t i o r
p r o c e s s e s a n d in vm-ious bran-
c h e s of ratiorial econoniy. To meet gron-irig dec;ands gi cal
of nioderri t e c h r o l o -
p r o c e s s e s it i s n e c e s s a r y tG s t u d y a n d select n o v e l nELtLral Emcl
syrithetic alunmiriosilicate
s,yster,is which c a n b e u s e d for
z eo lites witk? predeterniinec!
production
of
crysteillirte s t r u c t u r e , compositior. e n d proper-
ties. In t h e F r o c e s s of z e ol i t e s s y n t h e s i s
tt?s with
natural a n d
s y n t h e t i c alur.?iriosilica-
c r) stallirie arid a ni orphous s t r u c t u r e ( t h e latter being p r e s e n t e d
by gels arid g$rsses)
were used
2.5
an iritial s t c c k . High reactivity of
g l a s s e s b ein g ar. a m orphous material arid r i e c e s c i w of reproductiorl o f rlatural p r o c e s s . e s with
a c t i ve
Farticipatiori
of n i i r . e r d s formation in labora-
tory, w ad e r e s e e . r c k e s to s t u d y thorotighly ra.w n.aterial.However,ccmplexi-
t-,
a n d uristability
of chemical c o n i p o s i t i o ~ ,i t s
clu.sioris, admixtLires iri
urivaria.Dilit4.,cryste.llic in-
natureil g l a s s e s turr-ed r e s e a r c h e s attention
to
sy n th etic glasses. A nuntber of w o r k s
has b e e n d e v o t e d
from v a r i o u s syr~tl-ieticg l a s s e s . O n e of si l icate, e.g. r h ode si t e -
hydrosi l i c a t e of
h a v e p r c d L . c e c i two syr:thetic forms-K-
to crystellizatiori of zeolites
t h e s e i s s y n t h e s i s of zeolite-like alkali a n d alkali-earth a n d Na-rhodesits.It
metetle.WE
must Ire r.otec!
that z.ccordir.g to its, s t r u c t u r e mountainite a s well a s r h o d e s i t e could b e r e g a r d e d as n:irierel of c.elhq:elite grc'up.
624 TABLE
1.
Ccmparison of minerals
N a m e of mirt e r a1
charecteristics
~
-
Chemical composition
-_____
Lattice p r r a m e t e r s , A b
a
-------
--
Space group
C
- - - - - ---
I
( H 2 0 ) 2] 4 H 2 0
23,8
6,54
7,05
Pnin2
Delheyelite K,Na F a 2 ( E i A1)801g] F, C1 2
6,56
24,6
7,lO
Pmn2
6,65
23,84
7,07
Pmri2
3 3,51
13,51
13,51 Pniri2
BaH2 F a 2 S i e O l 9 ( H 2 0 ) 8H20 14,OE
13,08
23,52 CniCnc
23,91
13,09 B m a t
&odesite
KNa[ Ca2Si8019
Hydrcdelhayelite
KF
4
C2 ( S I N ) 7039 (H 2 0 ) 44H20 2
K 2 N a 2 k a 2 S i g ) 03 9 ( O H ) 2] 4 H 2 0
Mountainte W..ecdonaldite
Moriteregia.nite K N a 2 [ Y S i80 1 9
]
;I
10K20
0 r . e c a n s e e analogy- between
14,OI
peranletere a n d chenlice.1 p r o p e r t i e s
-
of r e L - e r a l minerals and mountairiite. T h i s a n d o g y i s d u e to t h e s t r u c t u ral l i k e n e s s of t h e s a i d mir.ernls. Mirierals of t h i s eigkt-
nierrber c a v i t i e s
* 4A0 ,
g r o u p h a v e wide
filled with l a r g e ce8tior.s a n d w&er molecu-
les: K + cations a n d water rr.olecules in delhayelite, hydrodelha.yelite,m~un-
tairiite a.rid
r h o d e s i t e ; R a 2 + a n d water n i o l e c d e s in macdor:aldite. I t
found t h a t b a s i c delhayelite dimensiocal by C a -
/
s t r u t u r o l L.r,its of
ref.21
and
rhodesite
/
r e f . l / , a n d also h y d r o
rrountairiite a r e C.ct-octnhedron
r a d i c a l ( two-
was
l e v e l 1cttice) .Colurr.ns a l o n g tk.e r x i s fornlod
octe.hedr0r.s a r e c o n n e c t e d by
wollastonitr c h a i n whick i s arrar.-
g e d aroL1r.d t h e columns s o that two silica-
o x y g e n t e t r a h e d r o n s in t h e
s a i d chair1 a r e corijugated witti r i k of Cz-
o c t a . h e l r c n s andtk.e third
c o n n e c t e d with tkle adjacer?t columns. Wollastonite c h a i n s additional
diorthogbGps fornmir g two-
. Togetk.er
positiort [sic;olg]
-
c o h m r - e.nd two-
le\ el
a r e tied
is
by
lattice cef t h e chemical com-
t h e y form tt-ireedinier,sic,r.e~lstrL:cture
with
N.
zeolite-like the
cavities. Cations of Ir.etals a n d water n.olecules a r e p l ~ c e c ?in
c h a n n e l s . Water rrolecules
frame tetre.hedrons. Water
a r e conrNected Icy t-,ydroSerr h0r.d witti
p&rtially t & k e s p e r t in Ca-
atoms coordinaticr.
(fis. 3.). ME: T H 0 D S Res,u.lts of
s y s t e m a t i c kvestigatiori of
tic gle,ss with chemical
crystallization p r o c e s s of syrlthe-
composition, m a s s %: S i 0 2
71,6
; A1203
3,2 ;
625
Fig.1. Stt cictcire of R h o d e f i t e .
Fe203
0,3 ;
CaO
MgO
6,4 ;
3,6 ; N a 2 0
pres er . t p a p e r . Wk.en treetiris g l a s s mourtainitc v * a s
pt.ciduced for
tl:esis were: g l a s s NaOE ; g l a s s
/
14,4
a r e s i v e n in t h e
powder by NaOW solution s ynthetic
t h e first time. OptirrJuni conditions of syri-
fraction 1 , O O nies,h. ccncentratior,-
solutior. ratior. = 1:Fs
;
1 n solutior. of 0
t e mperature cf s y r , t h e s i s 170 C ;
crysta.llizatior time is, 2 0 dr?ys. E x p e r i a e r t s w e r e c a r r i e d out a t 60dr o p of f
0
2 C. In l ow-
25OCC with t h e terxperature 0
~ to p 90 C ) ql;artz
t e m pe ra t ure experinierlto (
g k s s h b e s witk. i nne r volume of 3 0 c m 3 w e r e u s e d . At higker tempereh i r e s "Mori" a u t o c l a v e s rr,ade of 45 M h - F T s t a i n l e s s steel of differert of 3 ('7.5, 1 9 . 6 , 13.0, 9 G cm ) w e r e used.Hydrotk~erma1crystalliza-
volumes
tior, ar.d
r ecr y s t a l l i z a t i on experimerlts
w e r e c a r r i e d out in a u t o c l a v e s ,
without building t h e t e m pe ra t ure gra di e nt (&?'=O)
and
without
s tirring
the r eactio n mixture. After finishing t he
experinNer,t
n a c e a n d water quer.ched. P r o d u c t s
e u t o c l a b e was t a k e n out of t h e furw e r e filtered, w a s h e d u p to pEI=e,
dried a.t 80 c C , horrogenized a n d further StLdied. Qualitative a n d quantitative ar.alysis of s a m p l e s w a s c a r r i e d out by powder r~ etl.od orr DEC)b.j-3,0 Irrier a n d o u ter s t a n d a r d s
diffractor. e t e r
( C.LLK d
-
radia.tion, Ni-filter).
K a C I , S i 0 2 ( q u a r t s , ) as well
tes w e r e applied. Diffractrometry w a s kV, c u r r e n t 1 4 mA, velocities 4 ar.d
c a r r i e d out
a s p u r e zeoli-
a t a n o d e b-oltage 3 2
2 grad/rr.ir. Survey w a s d o n e in t L J e
626 r a c g e of reflection a n g l e s 8 = 2 + o w a s of 0.1 .
-
70
0
. Angles m e a s u r e m e n t
accuracy
-
Thermogravimetrical a n a l y s i s of t h e s a m p l e s was c a r r i e d out in dynamic mode with t h e u s e of “(3-
Derivatograph 1500-D” of
MOM. T e m p e r a t u r e r a n g e w a s within 20
0
-
Eiurigariari firm
1000°C.
0
S u r v e y conditi0ns:heating r a t e i s 10 /min ; pa.per t r a v e l s p e e d is miri ; sensitivity of DTA, D T G a n d TG i s 500
rniV;
2,s
nim/
ceramic cubicles were
u s e d ; standa.rd i s A1203. Chemical s i l i c a t e a n a l y s i s of th.e nitial material a n d of transformation
c a r r i e d out with t h e u s e of
WBS
- 18 s r e c t r o m e t e r ( W e s t G e r m a n y ) . M e e s u r e m e n t mode: P d - ar.ode ; v o l t a g e -- -- ~-p o s u r e period i s 1 0 0 s ; sensitivity’ limit w e r e p r e p a r e d a s ftrllbws: t h e
prrJducts
niultichannel X-ray
Of
CPW
i s 2 5 kV; c u r r e n t i s 7 0 niA; ex-2 i s 1 0 . S a m p l e s for ariG1~~sis
s u b s t a n c e beir.5 a n a l y s e d w a s fLsed with
f1v.x Li B 0 ( r a t i o 3. : 1 0 ) a t 1 2 5 0 2 4 7 up to 300 m e s h a n d p r e s s u r i z e d at
0
C. Ohttiiried g l e i s s
2 0 ton / c m
2
WES
powdered
with time l a g of 1. minu-
te. t h e u s e of e l e c -
Electronmiroscopic investigation w a s c a r r i e d ol;t with tron
s c a r i n i r g m i c r o s c o p e B S - 3 0 1 ‘TESLA. T h i s vetkiod a l l c ~ w sto irives-
t i s a t e s p z c e a n d time relatioriship of microtlrysteils. It that t h e s e relationships a r e more complicated
hits
been shown
thar. t h o s e obteiir e d by ro-
entyenonl etry.
RE; SUL’I’S Diffredonietry
d a t a for synthetic rrokntainite
d a t a for i t s n a t u r a l a n a l o g u e /ref.
(Table 2 )
.?I.
TARLE: 2. Ciffrectometr)- d a t a f a r s y n t h e t i c moLnteir.ite ar.d i t s thermally t r e s t e d ( a t 370
C
) fvrr.;.
C
Mourlttrinite --d
exp
A0
-J
I
13,18 6,58
6,OO 5, 59 4,72 4,18 3,76 3,62 3,41 3,22 3,04 2,92
90 90 16 14 62 44 16 11 25 13 21 100
dexp
P.
0
J
9
11, 24 6,38 5, 7 8 4, 3 6 4, 06 3,9 2 3, 06 2, e9 2, 7 7
46
6 52 8
1c 5 21 26 16
627 2, 2, 2, 1, 1,
Mounta.inite
84
36
70
18 23 32 22
34 96 74
i s crystallized iri mor1ocliriic
p a r a m e t e r s of w.it cell: a= 13,53 A VolLme of the unit
cell i s 2 4 2 8 A
3
0
,
systec. with
t h e fcillovuing
b = 13,O.S A@, c = 13,53 A
.
0
.
To determine relative s t a t i l i t y of newly fcjrnied p h a s e s orid tc: estotr-
lish their
ste.bili@ limits influence of v a r i o u s p m a n l e t e r s of syntkiesis ori
p r o d u c t s crystetllizi’.tiori in this s y s t e m w a s studied. Influence cf temperature. To c r e a t e fo\-orc?..ble conditions for t h e transformiition of t h e ir.itial raw material into
rriouritainite a n d a t t h e sane time
to s u p p r e s s q u a r t z formation optinla1 temperature region w a s s e a r c k e d . I n spite of coniplexity of tt:e
prot,len: c a u s e d by pi.opensib of c . q s t a l l i z a t i c n
of both niir e r a l s to high-teniper&tLire conditions, coritrolliriy hydrothernial crystallizatior; we singled out wide t e m p e r a t ~ i r eregiorr of
nicmntairiite c r y s t a l s 150
-
195
0
C
. Syr:tk.esis
of riiaxirnurri yield
of nlol;r.t&inite from
higher temperatures g a v e r.0 r e s d t . TemperatLre r i s e r e s u l t e d in c h a n g e of reaction direction, witti
fornation of pec tolite. Higher teniperatu-es
bL.t low cor.centrations of solutions a r e more quarts
rather
peratures
prefers.ble fcr
ec‘ucing
than pectolite. Attempts to s y n t h e s i z e mouritiiinite s t teni-
lower than 150
0
C were unsuccessful.
Influence of concentration. A s a therrnd solution sodium bydroxide w a s
-- - __ - - - -__
- - --
-
c k o s e n . It w a s found t k t c h i w g e
CJf
ccirocc~ntrfiticn had
greatly inf1uer.c-
ed pl-lase formatior.. I n c r e s s e o f tkie solution pH prcrniotes solubility gle.ss.es, h r e a k i n g down their silicaI~itiori as.r.ccintiorl of p11i:c-ez. Q z
+
o x y g e r lattice. In 0,3 N
Gf
NaOH
SO-
F c i s crystiilliaed arid in t h e c o u r s e
of r i s i n g solu.tior, cor!cer.tra.tior. from 0 , 3 N to l,ON abovesaicl essocieticin i s su.bstituted by n,ountainite ( T a b l e 3 ) . I n c r e a s e in coficer.trettion of
NaGHcclutior, u p to 1 K l e a d s to destruction of rhodesite. F u r t h e r r i s i n g of cor.certratiori of N a O E solution fron
lr’! to 2N a n d more w l f a v o u r a t l y in-
f l u e n c e s the p r o c e s s of nlourrta.inite cry-stalliz.eticr. Mear:wkiIf? syt>thc?tic mounteinite i s b e i n s reple.ced by- pectolite. 1nfluer.ce of s o l v e n t
riatL
re. In our e x p e r i n er.ts direction of t h e crystalli-
- - - -- -- -- - - - - - - - -
za,tior p r o c e s s w a s determiried f:y ttie s.ol\.ent nature. At t h e i d e c t i c a l concentratior.s of h’a-
lization and
c a t i o n s but with ce.rlr*onate solL.tion acting, c r y s t a l -
of mountainite d e c e l e r a t e s ar.d
quertz i s o b s e r v e d , t h e latter
forniaticln of k,oth n.,c)uritairiite
o n e predomina.tiny. At low cor,centra-
628 q u a r t s i s formed, a t ccncentraticir: of 1 N asso-
tiori of P\’a2C03 solution ciation of cluartz urider
a n d r h o d e c i t e apFeara.. Mour.ta.inite i s not crq’stdlized
the action of soliunl c a r b o n a t e solution.
I r f l u e n c e of solid p h a s e /liquid p h a s e ratio. Glass/solu.tior; ratio g o v e r n s pot only cor.ipositior of mother liquor, but e k o conipcjsitic,r: a.nd structL r e of t h e products. In t h e c o u r s e cf interaction of alkali s o l u t i o n s with s i l i c a rich g l a s s ratio glass/ solution g r e a t h , i n f l u e n c e s t h e p r o c e s s of p r o d c c ts
( T a b l e 3 ). Obtained exprriniental dwta confirm obser-
crysterllizatior
vation of the p r o c e s s of silicalution
/
r i c h z e o l i t e s formation. Thus., a t
l o w so-
gJnss ratio boundery k y e r i s s a t i a t e d witL S i 0 2 a n d alkali catinrls
n e c e s s a r y for silica-rick
z e o l i t e s formatior.. At high solution / g l a s s ratio,
i. e. a t soluticri excess, s i o ,
L.
-
diffuses from t h e t o u n d a r y l a y e r of s o h .
tion to more f r e e s p c c e , a n d pec.tolite c r rkJodesite r a t h e r tk.an mountainite i s formed. TABLE 3. Ccnditions a n d crq.stallizetiori p r o d u c t s of silica-
r i c h g l a s s i r NaOH ar.d
N a 2 C 0 3 so1utior.s
---- ---
_ _ - --_
--
___ ___----------c___---_I-
i m
Ccr.centration of NaOH,N
Concentration Solid F h h s e of N a 2 C 0 ,hT liquid p h a s e ratio
--1
0,3
2 3. 4. 5. 6. 7. 8. 9. 10. 11. 3 2. 13. 14. 1 5 16. 17. 18. 19
0,3 0,3
Note:
hln Rd Qz Pc
--
-
-
1 1
1:5 1:5
mountainitc rk.odecite quartz pectolite
C
---
O,-? 0: 5
085 0,5 0,5 1 1 1 3 1 1 1 2 - 6
o
--
I _ ~
25 1:10 1:15 1:5 1:1 5 1:15 1:5 1:5 1:3 1:5 1:5 1:1 5 1:5 1:10 1:5 1:5 1:5
03
TeniFerature, F r o d L c t s
170 170 200 170 170 PO0 200 170 170 150
1eo 200
200
170 170 170 170 170 200
-
-
-
~
Mn Rd
45: Rd
+ +
Rd Mn
Rd Rd
92 Mn amorphous a n 1 CI r p k 10LI s hlrr + P c Pc Pc Pc Pc
Qz Qz -t Rd amorphous on.orplious,
629 It could b e s e e n from T a b l e 3 that solid p h a s e respor.ding
/
to 1:10 l e a d s to d i s a p p e a r a n c e of
of pectolite. At 1 N
liquid p h a s e ratio corn ountainite a n d i-..olation
concentration of NaOH solution cryste.llization of so-
diun. r h o d e s i t e i s o b s e r v e d . A s to lower r a t i o n s solid pklase/ liquid p h a s e
-
e.g. 1:3
-
s u c h proportions a r e unefficient for p h e s e formation in tk-.is
system. Thermogravimetrical investigations s h o w that dehydration of mountainite
,
i s continuous
whick c o r r e s p o n d s to
hydratior c h a r a c t e r i s t i c of zeolites. DI'A c u r v e of eristic t h r e e endo- a n d tvro exo-effects.
t h e 3--5t
synthetic
type of d e
nlom-lta.inite den;onst 0
Ecclotkrermal p e a k s a t 1 7 0
-
-
and.
320° C w e r e indticed Icy dehydration of water which got lost r e v e r s i b l y 0
at 340 C. Water losses constitute 3.2,A %. Week exotk.ermal band a t 3 7 0 ° ~ a s c r i k e d to destruction. Diffraction d a t a of t h e n e w p r o d u c t
could b e
0
a p p e a r i n g a t 370 C a r e given iri T a b l e 2. Smee.rir.g of endo-therrral 0
a t 7 2 0 OC could b e ascrik,ed to fLsior, exo-
effect a t 8 3 5 C-
peak
to produc-
tion of pseudowollastonite. Chemical conipositiori of synthetic mour.ta.ir.ite i s ( m a s s 43; CaO
-
1 4 , l l ; hIg0
-
2,61; K 2 0
-
/ref. 31. Electron microccoFic investigction h a s synthetic mountainite c o u l c
b e not
-
11,38; H 2 0- 12,8. niounteinite a r e fibrous formaticrs
3,79; N a 2 0
It i s known that c r y s t a l s of nature.1
%): Si02- 55,
only
shown that crystetls of
fibrous, but also needle-like
ar.d in the form of thin pellets. When studying ior.tior, with N H
4.
+
'
e x c h a n g e ability of synthetic mountainite in rela2+ K a n d C.a c a t i o n s singnifica.ct c h a n g e s in i n t e n s i t i e s of
reflections d = 13,18; 6,58; 3,22 A
0
were
observed.
R e v e r s i b l e chare.cter of dehydration, ion-exchange
ability, p r e s e n c e
of specific cavities, c h a r a c t e r r i s t i c for z e o l i t e s , in tkje s t r u c t u r e cor.firr.
that rr.otir.tainite b e l o c g s to s i l i c e t e s
having
z e o l i t e s piroperti es.
R E F E R E N CE S
1. H.P... H e s s e " Verfeinerung d e r s t r u c u r d e s R h o d e s i t s H K 2 C a 4 S i 1 6 O Z 8 . 1 0 H 2 0 . " Zeifur Kristallograpl-y 1 9 8 7 , V 178, N 1-2, p.99-98. 2. Rd. D. P o r f n a n , M. I. Chiragov, Noviye darmiye c mineralack S S S R , 1 9 7 9 , NO. 28, p. 172-175
.
3. J. P. C-ard, H. F. W. Taq-lor, "An irivcstigation of tvro r e w minerals: r h o d e s i t e a n d mountainite" Mireral. Was., 1957, V. 31, p. 611-623 4. R. H. Mariner, R C S u r d a n o , "Alkaliri a n d forrt1e.tic.r c.f zeolites in s a l i n e alkaline lakes". S c i e n c e , 1 9 7 0 , V . 70, p. 977-582
This Page Intentionally Left Blank
G . Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
THE D I S T K I B U ' r ~ O N OF IRON I N Z S M - 5 TYPE
AND FERRI SI L I CATES:
IRON
63 1
CONTAINING
ZEOLITES
ACIDIC AND CATALYTIC PROPERTIES
G . VOREECK, M. RICHTER, R. FRICKE, B. PARLITZ, E. SCHREIER, K . SZULZEWSKY and E . ZIBROWIUS lnstitute of Physical Chemistry, Academy o f Sciences, 1199 Berlin, SUMMAKY
The state of iron in microporous ferrisilicates with MFI structure as well as in template free synthesized FeZSM-5 zeolites has been examined by XRD, IR, ESR, Si MAS NMR, ammonia TPD, and catalysis. The percentage of iron in the ferrisilicate framework could reliably be determined by two independent methods, ammonia TPD and Si MAS N M R . Catalytically, the Fe related acid sites are weaker than those of the parent ZSM-5 but. nevertheless, they isomerize m-xylene and convert ethylbenzene. Nonframework iron dehydrogenates ethylbenzene to styrene, where the selectivity correlates with the amount of nonframework iron. INTRODUCTION
One possibility to modify the acidic and catalytic properties of zeolites and zeolite-like metallosilicates (1) consists in varying their chemical composition at the stage of synthesis. Advantageously, this a1 lows the tailored design of active molecular sieve components via substitution of A1 or Si by several elements (e. 8 . E, Ga, Cr, V, Co, and others) ( 1 - 15). Iron containing zeolites or pure ferrisilicates with MFI structure are of special interest as catalysts for the conversion of several aromatic compounds, above a l l C 8 nlkylaromatics. Numerous investigations have been performed in order t o confirm the incorporation of iron into the molecular sieve framework and to reveal the nature of iron in both framework and nonframework positions (1,16 - 2 6 ) . Incontestably, the location and the state of iron are essentially for their acidic, catalytic and shape-selective properties. But the knowledge about the exact distribution of iron between the lattice and channels is still incomplete. Therefore, this contribution emphasizes the quantitative determination of iron throughout the matrix of aluminum free ZSM-5 type ferrisilicates. The consequences for the aciditiy and catalytic activity of the samples are discussed and the results are related to reference samples such as iron
632
containing
ZSM-5
zeolites,
where
is
aluminum
only
partially
r e p l a c e d by i r o n , a n d a p u r e ZSM-5 z e o l i t e . EXPERI MENTAL
Samale w&am.Lu?n . . . . . with ( i ) Perrlslllcates ( F e S i 1 ) . T h e A 1 f r e e f e r r i s i l i c a t e s Z S M - 5 s t r u c t u r e were s y n t h e s i z e d b y m X i n g a n a c j.d s o l u t i o n of i r o n s u l f a t e w i t h h a l f t h e v o l u m e of a s o d i u m s i l i c a t e s o l u t i o n a t room t e m p e r a t u r e . An a q u e o u s s o l u t on o f tetrapropylammonium bromi de and t h e r e m a i n d e r o f t h e d i l u t e d so d iu m s i l i c a t e s o l u t i o n were a d d e d t o t h e r e s u l t i n g g e l ( a f t e r a n a g i n g p e r i o d o f 1 h ) . T h e mixture w 3 s s t i r r e d f o r 1 h, then placed i n a t e f l o n coated
-
s t a i n l e s s s t e e l a u t o c l a v e an d k e p t u n d e r a u t o g e n o u s p r e s s u r e a t 4 4 3
K f o r 48 h . The r e s u l t i n g w h i t e o r p a l e y e l l o w s o l i d was washed a n d d r i e d a t 383 K . T h e t e m p l a t e was removed b y
filtered, calcination
a t 773 K f o r 7 h i n a i r f o l l o w e d b y a t h r e e f o l d e x c h a n g e o f Na'. by NH4+ i . o n s a t 353 K for 2 h . AAS was u s e d f o r a n a l y s i s ( T a b l e l a ) . ( i i j
E.e
. . contalnlna Z S M - 5 0 .
Ferric
a l u m i n a t e , s i l i c a s o l a n d s o d i u m h y d r o x i d e were materials f o r t h e s y n t h e s i s . S i l i c a s o l solution containing the aluminate, t h e
nitrate,
added
was
sodium
as
used to
sodium starting
a
hydroxide
diluted a nd
the
f e r r i c n i t r a t e . T h e m o l a r r a t i o s o f t h e c o m p o n e n t s were: 2-8 N a g O
*
*
(Al2O3+FezO3j * 3 0 - 9 0 S i Oz 6 0 0 - 2 , 5 0 0 HzO. The r a t i o Fe 203 : A 1 2 0 3 was v a r i e d b e t w e e n 0 . 1 a n d 0 . 8 . Th e c r y s t a l l i z a t i o n o f t h e z e o l i t e s
was p e r f o r m e d h y d r o t h e r m a l l y u n d e r a u t o g e n e o u s p r e s s u r e w i t h i n 8
-
24 h a t 4 5 3 - 4 8 3 K . T h e a d d i t i o n of s e e d c r y s t a l s ( a b o u t 3 w t . % of
was f o u n d t o b e n e c e s s a r y f o r p r o m o t i n g the c r y s t a l l i z a t i o n p r o c e s s . Ho wev er , n o o r g a n i c t e m p l a t e was n e e d e d . ferrisilicate)
The r e m o v a l o f Na+
cations
in
the
as-synthesized
products
was
a c c o m p l i s h e d b y i o n - e x c h a n g e ( T a b l e l b ) . T h e f i n a l ammonium f o r m o f t h e z e o l i t e s c o n t a i n e d less t h a n 0 . 1 w t . % N a p O . T h e d e s i g n a t i o n t h e samples is t o
be
read
as
HFeSil
for
the
H
form
of
of the
f e r r i s i l i c a t e s r e s p . HFeZSM-5 f o r t h e i r o n c o n t a i n i n g z e o l i t e s . The a n a l y t i c a l (Si0z/Fez03)z
r a t i o s are g i v e n i n p a r e n t h e s e s .
( i i i ) Reference samples.A c o m m e r c i a l
ZSM - 5 p r o d u c t
(HS 3 0 , a silicalite i m p r e g n a t e d w i t h a F e( N0 3 ) 3 s o l u t i o n t o r e a c h a n i r o n c o n t e n t o f 1
C h e m i e - AG
Bitterfeld)
i n t h e H form a s
well
w t . % a f t e r c a l c i n a t i o n ( 7 7 5 K , 2 h ) were u s e d a s lb).
as
reference
(Table
633
TABLE 1 List of samples and characterization data a) ferrisilicates a
NH3
Overall HPeSil HFeSil HFeSil HFeSil HFeSil HFeSil HFeSil HFeSil HFeSil
Htpd
(58)
91
0.44
0.25
657
130
45
(74)
97
0.45
0.25
653
129
58
(76)
97
0.47
0.26
648
126
61
(78)
84
0.45
0.31
655
104
75
(94)
98
0.40
0.25
653
132
72
(155)
91
0.29
64 5
172
89
(180)
95
0.19 -
-
-
-
(259)
86
0.14
0,12
274
93
(508)
68
0.10
0.06
520
97
b ) iron containing zeolites
Sample
T~(KP
desorbed
(
sio2) Fe203 E
HFeZSM-5 ( 1 5 2 ) HFeZSM-5 ( 1 1 8 ) HFeZSM-5 (100) HFeZSM-5 ( 0 )
(
sio2 Me203
fe203 Z A1203
1
cXRDe Z
NH3 desorbedb d Overall Htp 0.57
705
90
40
0.31 0.52
87
1.04
0.54
695
41
0.70
80
0.99
0.53
680
40
0.00
100
1.30
0.84
720
36
1.07
Tm(K)'
a: Crystallinity parameter, estimated from n-hexane adsorption, b: Desorbed ammonia in inmol/g, c: Temperature of the peak maximum (high temperature peak), d : High temperature peak, e: Crystallinity parameter derived from XRD, Subscripts X. and f: Analytical r e s p . framework values, Me: sum o f Fe and Al, %Fef: percentage of framework iron. SamDle characterization The XRD data were obtained with a HZG 4 diffractoineter by using qi-filtered Cu K, radiation. IR spectra were recorded with a Specord M 85 (Zeiss Jena) using the KBr pellet technique. ESH measurements were performed with a ZWG ERS - 220 X - band
634
s p e c t r o m e t e r a t 7 7 an d
2 9 5 K. T h e 2 9 S i HAS N M K s p e c t r a were a B r u k e r MSL 400 s p e c t r o m e t e r a t a r e s o n a n c e f r e q u e n c y o f 7 9 MHz u s i n g s i n g e - p u l s e e x c i t a t i o n w i t h high-power p r o t o n d e c o u p l i n g . NH3-TPD m e a s u r e m e n t s were c a r r i e d o u t a t n o r m a l 3 p r e s s u r e i n a f l o w r e a c t o r w i t h He a s c a r r i e r g a s ( f l o w r a t e 1 cm s - l j c o n t a i n i n g 3 Vo1.X o f NH3. T h e h e a t i n g r a t e a m o u n t e d t o 12 K measured
min-l.
with
Sample w e i g h t s of
200
mg
were
used
(mesh
Pa, T
=
size
25-60).
Details are g i v e n e l s e w h e r e ( 2 9 ) . B d s o r D t i q n and catalvsis A f t e r a c t i v a t i o n in VUCUO ( p < =
570 K , t > =
5
t h e f e r r i s i l i c a t e s were c h a r a c t e r i z e d b y i s o t h e r m a l a d s o r p t i o n d e s o r p t i o n of n-hexane (T = 293 K , saturation pressure) using
a
<=
0.02
McBain
p/p
balance.
<=
S
The
h) a nd
0.9,
ps crystallinity
parameter CAds was e s t i m a t e d b y r e f e r r i n g the experimental s a t u r a t i o n v a l u e a t p/p, = 0 . 5 t o t h e c a l c u l a t e d m i c r o p o r e v o l u m e 3 o f a n i d e a l MFI s t r u c t u r e ( 0 . 1 9 cm / g > . A d s o r p t i o n d a t a f o r b e n z e n e a n d water o f t h e template f r e e
synthesized
Fe-ZSM-5
samples
are
reported i n Ref. ( 2 9 ) . Both t h e
conversion
of
ethylbenzene
of glass r e a c t o r ( 2 cm i n t e r n a l d i a m e t e r ) a t a t m o s p h e r i c p r e s s u r e a n d a s t a n d a r d f l o w r a t e o f 1 0 1 h-' h y d r o g e n , l o a d e d w i t h 1 V o 1 . X of t h e o r g a n i c s u b s t r a t e . I n e i t h e r case 1 8 : of t h e b i n d e r - f r e e zeolite p o w d e r i n i t s H f o r m was u s e d ( m e s h s i z e 2 5 t o 6 0 ) . D e t a i l s c a n b e f o u n d i n Ref. ( 2 9 ) .
m - x y l e n e were c a r r i e d o u t i n a
and
fixed-bed,
the
isomerization
continuous
flow
RESULTS AND DISCUSSION
. .
C r v s t a l l l n l t v andclualltativecDnfirmatlon . Q€
Lhe i n c o r t o r a t i Q n Q €
irpn
A c c o r d i n g t o t h e XRD p a t t e r n a l l s a m p l e s a r e
well
crystallized
case were d e t e c t e d . N e v e r t h e l e s s , t h e c r y s t a l l i n i t y C X R D o f t h e samples i s h i g h , b u t i t obviously d e t e r i o r a t e s with increasing iron content ( c f . Table l b ) . In c a s e of t h e f e r r i s i l i c a t e s w h i t e and p u r e c r y s t a l l i z a t i o n 150 p r o d u c t s were o b t a i n e d f o r a ( S i 0 2 / F e 2 0 3 ) x r a n g e o f a b o u t 50i n t h e s y n t h e s i s g e l . H i g h e r r a t i o s l e a d t o t h e a p p e a r a n c e o f an a l i e n p h a s e ( q u a r t z ) . Lower r a t i o s r e s u l t i n p r o d u c t s o f a s l i g h t l y y e l l o w c o l o u r i n d i c a t i n g t h a t a p a r t of t h e f e r r i c i o n s r e m a i n e d o u t s i d e t h e f r a m e w o r k . T h e l a r g e l y p u r e f o r m a t i o n o f t h e ZSM-5 a n d show t h e t y p i c a l r e f l e c t i o n s of t h e MFI s t r u c t u r e . I n t h e o f t h e FeZSM-5
zeolites
some
ZSM-11
admixtures
635
s t r u c t u r e i s c o n f i r m e d by t h e r e s u l t s o f n - h e x a n e
adsorption
T a b l e 1 ) . T h e v a l u e s o f t h e r e a l m i c r o p o r e volum e f o r m o s t samples were f o u n d
of
the
theoretically calculated pore v o l u m e o f a n i d e a l MFI s t r u c t u r e ( 0 . 1 9 cm 3/ g ) . T h e h i g h v a l u e s o f t.0
be near
(cf.
the
t h e c r y s t a l l i n i t y parameter C A c o n f i r m
the
successful
synthesis.
The low Cads v a l u e o f H F e S i l ( 5 0 8 ) o b v i o u s l y comes a b o u t b y t h e low a d s o r p t i o n c a p a c i t y of t h e a l i e n quartz phase. Figure 1 d e m o n s t r a t e s t h e m o d i f i c a t i o n of t h e u n i t c e l l volume w i t h t h e i r o n c o n t e n t . A t i n c r e a s i n g i r o n c o n c e n t r a t i o n ( d e c r e a s i n g (Si02/Fe203)=
r a t i o s ) t h e u n i t c e l l is expanded d u e t o t h e
different
length
of
t h e S i - 0 and F e - 0 b o n d s . This r e s u l t p r o v e s t h e i n c o r p o r a t i o n o f a t
l e a s t a c e r t a i n p a r t o f i r o n atoms i n t o t h e l a t t i c e . S z o s t a k e t al. found similar r e s u l t s f o r a n a l o g o u s f e r r i s i l i c a t e s .
(21)
100
0
200
Si02/Fe203
F i g . 1. Expansion of t h e u n i t c e l l volume V u . c
of
ferrisilicates
w i t h i n c r e a s i n g i r o n c o n t e n t ( d e c r e a s i n g (Si0,/Fe2O3>,
-
I
1
2
ratio).
3 Fe/UC
F i g . 2 . D e p e n d e n c e o f t h e wavenumber v f o r t h e T-0-T
vibration
on
t h e i r o n c o n t e n t per u n i t c e l l F e / u . c . IR
spectroscopy
provides
further
evidence
for
the
iron
636
i n c o r p o r a t i o n . A s shown i n F i g u r e 2 t h e a b s o r p t i o n llUll cm-l
at
band
w h i c h is u s u a l l y a s s i g n e d t o t h e a s y m m e t r i c T - 0 - T
about lattice
v i b r a t i o n s h i f t s t o l o w e r wav en u mb er s w i t h i n c r e a s i n g i r o n c o n t e n t . 'This s h i f t o r i g i n a t e s f r o m t h e d i f f e r e n t a t o m i c w e i g h t s of s i l i c o n a n d i r o n . The e x t e n t o f t h e
number
of
iron
4 : o t a s t h a n e eL
atoms
el.
shift
isomorphously found
(2'1)
an
a nd
corresponds
substituted
analogous
for
radii to
the
silicon.
diminution
of
the
u a v e n u m b e r c o m p a r i n g a HZSM-5 a n d a FeZSM-5 z e o l i t e . Finally, the incorporation of iron i n t o t h e
is
framework
also
c o n f i r m e d by ESR s p e c t r o s c o p y . A l t h o u g h t h e d e t a i l e d i n t e r p r e t a t i o n a f t h e ESR s p e c t r a i s s t i l l a m a t t e r o f d e b a t e ( 2 3 , 2 5 , 2 6 , 2 8 )
=
is a f a r - r e a c h i n g a g r e e m e n t t h a t t h e b r o a d s i g n a l a t g
the signal a t g
=
indicate
4.3
octahedrally
c o o r d i n a t e d i r o n s p e c i e s i n n o n f r a m e w o r k and
a nd
there
2.0
a nd
tetrahedrally
franiework
positions,
r e s p . . b o t h t h e FeZSM-5 z e o l i t e s an d t h e F e S i l s a m p l e s
show
two s i g n a l s d e m o n s t r a t i n g t h e s u b s t i t u t i o n o f i r o n f o r
silicon
these in
t h e frameworks a s well as t h e l o c a t i o n of i r o n i n t h e c h a n n e l s .
d e t e r l n i n a t i o n p f u i r n n distrlb .vtion between franlework and nonframmd !GLQSiLiQns Mi3 T L F i g u r e 3b s h o w s t h e N H 3 TPD s p e c t r a o b t a i n e d f o r H P e S i l - ' i v r ~ l e sw i t h
different
overall
iron
contents.
~ i e ~ ~ . ~ ammonia ~ l : ~ ~a d r e s u mn i ar i zed i n T a b l e l a . '
rramework r a t i o
is
based
on
The
amounts
of
The e s t i m a t i o n o f
the
a m ount
of
the
ammonia
t h e a r e a of t h e h i g h t e m p e r a t u r e peak ( c f . T a b l e la). R e a s o n a b l y , it. is as s u med t h a t o n e ammonia is a d s o r b e d a t o n e ijr-iisted s i t e c r e a t e d b y t h e i n c o r p o r a t i o n of iron into the r e p r e s e n t e d by
f e r r i s i l i c a t e f r a m e w o r k . 'The r e s u l t i n g d a t a a r e c o n t a i n e d i n l a . As expec-:t.ed t h e p a r t o f i r o n s u b s t i t u t e d f o r s i l i c o n high a t low u v e r a l l
iron
contents
and
decreases
h i g h e r i r o n c o n c e n t r a t i o n s . Ho wev er , a s m a l l
a m ount
Table
atoms
markedly of
is
with
iron
can
a l w a y s tgc f o u n d i n n o n f r a m e w o r k p o s i t i o n s , e v e n a t v e r y low o v e r a l l iron contents.
Figure
4
illustrates
the
distribution
for
the
MAS NMR s p e c t r o s c o p y
has
t e r r i s i l i c a t e samples. S o l i d P t a t a WE become
a
routine
Th e "Si
-&OSCODY,
met h o d
for
the
determination
of
Si02/AlZ03
f r a m e w o r k r a t i o s i n a l u m o s i l i c a t e s (31). T h e d e t e r m i n a t i o n
-
Sic)?/Pe 0
2'3
f r a m e w o r k r a t i o i n f e r r i s i l i c a t e s is e x p l a i n e d
of F i g u r e 5 which shows a
typical
"Si
MAS
NMR
of by
spectrum
the help of
a
f e r r i s i l i c a t e s a m p l e . Two s i g n a l s c a n b e d i s t i n g u i s h e d : a n i n t e n s e a s y m m e t r i c l i n e a t a h o u t -113 ppm w h i c h i s c a u s e d b y s i l i c o n a t o m s
637 655
Fig. 3. NH3 TPD profiles of zeolites and ferrisilicates a) comparison of (1) HZSM-5(0), (2) HFeZSM-5(100), (4) HFeSil(94) b ) Influence o f the iron content on the NH3 TPD profiles o f the ferrisilicates HFeSil(78) (3), 94 (4), 155 ( 5 ) , 259 ( 6 ) . 508 (7). Note that a) and b) were recorded at different magnification (cf. curve 4 )
0 2 4 6 8 10 12 11 16 18 Overall iron content (Fe203/Si02) x lo3
Fig. 4. Distribution of iron versus the overall iron content of the ferrisilicate series according to ammonia TPD. 4
bonded via oxygen to four other silicon atoms (Q ( O F e ) ) ( 3 1 ) and a shoulder at about -104 ppm which is assigned to silicon atoms 4 surrounded by three other silicon atoms and one iron atom (Q (1Fe)) (20). An evaluation of the Si02/Fe203 framework ratio is based on the relative intensities of these two signals but has to account 3 for the fact that silanol groups in silicalite (Q (OFe)) exhibit a resonance line at the same position o f -104 ppm as the
638
I I . l , l l l l , l,
-90
,
-1 00
I
,
I , , , , I , , , I, , , , , I , , , , , , , , I , , ,
-110
-1 20
-130
q -
MAS N M K spectrum of sample HFeSil(76).
Fig. 5. &'Si
configuration Q4 (1Fe) (31,32). Therefore, the amount of silanol groups was determined by 'H MAS NMR (33). A value of ( 8 ~ 2 ) 1019 SiOH g-l was found for all samples. After substraction of the "Si MAS N M K line intensity at -104 ppm the Si/Fe resp. Si02/Fe203 framework ratio could be calculated by applying the equation proposed in Kef. (32) for the determination of Si/A1 ratios in alumosilicates. The results for some ferrisilicates are given in Table 2. TABLE 2 Quantitative "Si Sample
MAS NMR data f o r some ferrisilicates
Chemical analysis (Si02/Fe203)T Fe/u.c.
"si
MAS NMH
(Si02/Fe203)f
Fe/u.c.
%Fef ~
HFeSil (79) HFeSil (108) HFeSil (146)
79 108 146
2.57 1.74 1.30
140 150 182
1.35 1.26 1.04
57 72 80
639
are
The f r a m e w o r k r a t i o s o b t a i n e d b y N H 3 TPD a n d NMR F i g u r e 6 . The a g r e e m e n t of t h e r e s u l t s a l l o w s t h e both
methods
can
alternatively
be
used
for
compared
conclusion the
in that
quantitative
d e t e r m i n a t i o n of Si02/Fe203 r a t i o s i n f e r r i s i l i c a t e frameworks.
Id0 200 300 ( Si 02 I Fe 203 1
F i g . 6 . C o m p a r i s o n b e t w e e n t h e d a t a f o r t h e SiO / F e 0 f r a m e w o r k 2 2 3 r a t i o o b t a i n e d f r o m N H 3 TPD ( X ) an d "Si MAS NMR ( 0 )
UCiiLx F i g u r e 3a s h o w s t y p i c a l TPD s p e c t r a of a n HZSM-5 z e o l i t e l ) , a n HFeZSM-5 z e o l i t e ( c u r v e 2 ) an d a
TPD s p e c t r a o f f e r r i s i l i c a t e s p r e s e n t e d i n F i g . 3b ( c u r v e s
with
ferrisilicate
different
3-7).
iron
(curve
(curve contents
Characteristically,
both
4).
are the
o v e r a l l c o n c e n t r a t i o n o f a c i d s i t e s an d t h e a m ount o f B r a n s t e d a c i d
s i t e s d e c r e a s e d i s t i n c t l y f o r samples
with
comparable
r a t i o s when t u r n i n g f r o m HZSM-5 t o t h e H F e S i l
samples
Si02/Fe203 (Table
S i m u l t a n e o u s l y , t h e maximum p o s i t i o n f o r t h e h i g h t e m p e r a t u r e s h i f t s t o i o w e r v a l u e s ( F i g . 3 a ) . W i t h i n t h e HFeZSM-5
1). peak
series
Ccf.
T a b l e l b ) t h e d i m i n u t i o n of t h e o v e r a l l a c i d i t y a n d o f
the
a m ount
of
is
small.
Brprnsted
centres
with
increasing
N e v e r t h e l e s s , t h e maximum p o s i t i o n o f
iron the
content
high
temperature
€leak
c h a n g e s t o l o w e r v a l u e s . F o r t h e H F e S i l samples t h e o v e r a l l a c i d i t y
as w e l l a s t h e amount o f Brprnsted a c i d s i t e s d e p e n d on t h e (Si02/FeL03)t r a t i o , although the a c i d i t y does not change r e m a r k a b l y w i t h i n t h e r a n g e f r o m 5 8 t o 9 4 . ( c f . F i g . 3 b a nd T a h l r 1 ) . A small s h i f t i n t h e maximum p o s i t i o n o f t h e p e a k i s a ss u m e d t o l i e i n t h e m a r g i n of e r r o r . s u b s t i t u t e d f o r aluminium b o t h t h e
number
a nd
high Thus,
temperature if
strength
iron of
s i t e s , e s p e c i a l l y t h a t of Brmnsted a c i d c e n t r e s are i n f l u e n c e d .
is acid
640
b t a l v t i c erooerties The activity of the samples shown in Figure 7 follows the same sequence as the concentration of strong acid centres revealed by NH3 TPD, U L Z . HZSM-5 > HFeZSM-5 > HFeSil. The iron-impregnated silicalite does not possess isomerization activity, but converts ethylbenzene at T > 675 K, yet to a minor extent (5 X at 775 K). Within the ferrisilicates series the sample HFeSil(58) with the highest Fe content shows t,he highest activity. It is, however, low in comparison to the parent ZSM-5 (especially for the ethylbenzene conversion). The lower activity may primarily be attributed to the lower concentration of framework acid sites (theoretically 3 . 2 Fe 3+ per unit cell in case of the ferrisilicates compared to approximately 4.6 A13+ per unit cell in case of ZSM-5).
-s
60
- 50 5 40
._
Ln
& 30
5 20 10
600
700
T(KI
800
600
700
800
T(K)
Fig. 7. Catalytic properties of the catalysts. (a) conversion of ethylbenzene (b) m-xylene isomerization ( A , HZSM-5 (Si02/A1203 = 4 U ) , ( 0 ) HFeZSM-5(152), ( 0 ) HFeZSM-5(100), ( ) HFeSil(58), (x) HFeSil(SU8), ( V ) iron impregnated silicalite (1 wt.X Fe). The dashed line indicates equilibrium conditions.
As regards the m-xylene isomerizatinn sample HFeSil(58) compares better to the HFeZSM-5(100) resp. to the parent HZSM-5 zeolite. However, it has t o be taken into account that the m-xylene isomerization is a reversible reaction with an equilibrium conversion of about 50 X . That is, higher conversions are only possible due to side reactions ( e .6. disproportionation,
64 1
I
10
20
40
30
m- Xylene Conversion
Fig.
Shape
8.
selectivity
isomerization of HFeZSM-5(100),
(ratio
m-xylene.
(YO1 of
p-/o-xylene)
HZSM-5
(m)
(A) H F e S i 1 ( 5 0 8 ) ,
50
(Si02/A 1203
=
for
the
401,
(0)
( X ) H F e S i 1 ( 2 5 9 ) , (+) H F e S i l ( S 8 ) .
The d a s h e d l i n e i n d i c a t e s e q u i l i b r i u m c o n d i t i o n s a t 5 7 3 K. d e a l k y l a t i o n ) . Moreover,
the
isomerization
of
comparison
to
the
ZSM-5
as
is
reflected
on
m-xylene
f e r r i s i l i c a t e s b e n e f i t s f r o m t h e low mass t r a n s f e r
the
resistances
by
the
in
low
shape
with
shape
s e l e c t i v i t y (see b e l o w ) .
. . shaee s e l e c t i v i t s During selective
the
p r e d i c t e d by m-xylene
isomerization
properties
maintain
thermodynamics.
of
m-xylene p/o-isomer
Th e
isomer
catalysts ratios ratio
c o n v e r s i o n ( 3 0 ) . I t is e v i d e n t ( F i g . 8 )
higher depends
that
sample sh o w s t h e s t r o n g e s t s h a p e s e l e c t i v i t y f o l l o w e d b y
than on
the
the
HZSM-5
HFeZSM-5.
selective e f f e c t s . decreasing iron c o n t e n t s . This c o n t r a d i c t s c u r r e n t i d e a s t h a t f e r r i s i l i c a t e s should r e v e a l h i g h e r s h a p e s e l e c t i v i t y d u e t o t h e nonframework i r o n which i n e v i t a b l y is p r e s e n t a f t e r s y n t h e s i s a nd w h i c h r e n d e r s mure d i f f i c u l t t h e mass t r a n s p o r t i n s i d e t h e p o r e s y s t e m . I n t h e p r e s e n t c a s e a l l i n v e s t i g a t e d s a m p l e s h a v e ZSM-5 s t r u c t u r e w h e r e t h e p o r e d i a m e t e r s h o u l d b e t h e same. Th e e n l a r g e m e n t o f t h e u n i t c e l l dimensions as observed f o r t h e f e r r i s i l i c a t e s should b e n o t great e n o u g h a s t o e x p l a i n t h e d i f f e r e n t s h a p e s e l e c t i v i t i e s . The c o k e d e p o s i t i o n c a n n o t b e t h e r e a s o n s i n c e more c o k e i s f o r m e d on t h e f e r r i s i l i c a t e s a s a l r e a d y i n f e r r e d f r o m v i s u a l i n s p e c t i o n . REM p h o t o g r a p h s s h o w , h o w e v e r , t h a t t h e ZSM-5 h a s t h e l a r g e s t c r y s t a l s i z e f o l l o w e d by t h e F e a n a l o g u e s . Large c r y s t a l s d o n o t o n l y The p u r e f e r r i s i l i c a t e s e x e r t o n l y weak
shape
The d a t a seem t o i n d i c a t e a s l i g h t i n c r e a s e
with
642
prolong
t h e d i f f u s i o n pathways
but
they
also
a lower enhance t h e
exhibit
p e r c e n t a g e of o u t e r s u r f a c e area. Both f a c t o r s s h o u l d
s i z e of t h e f e r r i s i l i c a t e c r y s t a l s Sample HFeSi1(508), f o r example, d e p e n d on t h e F e c o n t e n t (35). r e s e m b l e s i n i t s s h a p e t h e p a r e n t ZSM-5 but, nevertheless, the shape s e l e c t i v i t y i s s t i l l low. T h i s p o i n t s t o a f u r t h e r e f f e c t superimposing t h e i n f l u e n c e of t h e c r y s t a l s i z e . Presumably, the s h a p e s e l c t i v i . t y . Form
and
f r a m e w o r k i r o n a t l o w c o n c e n t r a t i o n i s o v e r w h e l m i n g l y b e l o c a t e d on t h e o u t e r s u r f a c e what a l l o w s t h e
approximative
establishment
of
equilibrium regarding the isomers.
DehvdruaenationQ€ethvlben,ene 7 indeDendenceQnucontentnf
nonframeworkksDecies Since t h e iron impregnated s i l i c a l i t e has
no
Fe
in
p o s i t i o n s t h e a c t i v i t y observed f o r t h e c o n v e r s i o n of originates
from
nonframework
Fe
species.
dehydrogenate ethylbenzene t o s t y r e n e ( F i g .
framework
ethylbenzene
These The
9).
b e t w e e n t h e s t y r e n e s e l e c t i v i t y an d t h e n o n f r a m e w o r k
species
relationship iron
content
is n o n l i n e a r . O b v i o u s l y , n o t o n l y t h e c o n c e n t r a t i o n o f n o n f r a m e w o r k Fe b u t
also
its
dispersity
determines
the
f o r m a t i o n . A similar r e l a t i o n s h i p between -:bvlbenzene
an d t . h e o v e r a l l
iron
extent
styrene
content
is
of
styrene
formation
reported
for
P i g . 9 . S t y r e n e s e l e c t i v i t y enhancement w i t h i n c r e a s i n g c o n t e n t
nonframework i r o n i n t h e temperatures.
(0)
673 K ,
ferrisilicate ( 0 )
773 K.
series
for
two
from the
of
reaction
643
HFeZSM-5 s e r i e s
But
(29).
contrary
to
ferrisilicates
the
s t y r e n e s e l e c t i v i t y was f o u n d t o b e c o n s i d e r a b l y
lower.
course, occurs due t o t h e A 1 r e l a t e d a c i d sites p r e v a i l i n g HFeZSM-5 s a m p l e s w h i c h ,
above
all,
lead
a
to
the
This, in
of the
dealkylation
of
ethylbenzene. B u t as a whole t h i s c a t a l y t i c e f f e c t
of
opens t h e p o s s s i b i l i t y t o prove
existence
the
nonframework of
species
Fe
species,
such
a d d i t i o n a l l y t o o t h e r c h a r a c t e r i z a t i o n t e c h n i q u e s , b y m e a ns o f
the
c a t a l y t i c dehydrogenation of ethylbenzene. CONCLUSIONS
Microporous f e r r i s i l i c a t e s , synthesized with
Si02/Fe203
o f 58 - 5 0 8 , i n c o r p o r a t e i r o n t o a p e r c e n t a g e > 80 %
ratios
at
only
low
i r o n c o n c e n t r a t i o n s an d w i t h i n c r e a s i n g a m o u n t s o f q u a r t z ( u p t o 30 X).
At
higher
iron
i n c r e a s e s u p t o 50 % .
content
part
the
of
species
nonframework
I r o n i n framework p o s i t i o n s e n l a r g e s t h e u n i t
c e l l v o l u m e and s h i f t s
the
T-0-T
vibration
frequency
to
lower
v a l u e s . Si0H:Pe B r e n s t e d s i t e s are weaker t h a n Si0H:Al
sites.
c a t a l y t i c a c t i v i t y f o r t h e t e s t r e a c t i o n s is
related
t h e c o n c e n t r a t i o n of
Brensted
sites.
directly
N onfra m e w ork
iron
dehydrogenate ethylbenzene t o s t y r e n e t h e formation t h u s be t a k e n a s Template-free
evidence
synthesized
for
the
FeZSM-5
presence zeolites
of
of
to
species
which
these
show
The
can
species.
intermediate
p r o p e r t i e s c o m p ar ed t o t h e p a r e n t ZSM-5 a nd t h e f e r r i s i l i c a t e s . T h e A 1 r e l a t e d a c i d s i t e s of
FeZSM-5
zeolites
decide
the
catalytic
p r o p e r t i e s predominantly. ACKNOW LEDGEMENT
The a u t h o r s g r a t e f u l l y ack n o wl ed g e t h e
supply
of
FeZSM-5
and
s i l i c a l i t e samples b y Dr. U . H P d i c k e ( C h e m i e AG B i t t e r f e l d ) a n d
by
Dr. B . F a h l k e ( I n s t . I n o r g . C h e m . ) , resp., a nd s u p p o r t
M.
H u n g e r ( U n i v e r s i t y of L e i p z i g , 'H
(REM),
MAS N H R ) , Mr.
J.
by
Dr.
Richter-Mendau
D r s . H . K o s s l i c k ( S y n t h e s i s ) and J . Janchen ( A d s o r p t i o n ) .
REFERENCES
1
2 3 4
5 6
R . S z o s t a k , Molecular s i e v e s : P r i n c i p l e s of S y n t h e s i s and I d e n t i f i c a t i o n (Van N o s t r a n d R e i n h o l d New Y o r k , 1989), 205. B . U n g e r , T h e s i s , TH " C a r l S c h o r l e m m e r " M e r s e b u r g , 1 9 8 8 . S . R . E l y , M . H . K l o t z . Belg. P a t . 8 8 0 8 5 8 ( 1 9 8 0 ) . M . Taramasso, G. Perego, B . N o t a r i , P r o c . 5 t h I n t e r n . Conf. Z e o l . , London 1980, 4 0 . M. T i e l e n , M . G e e l e n , P . A . J a c o b s , P r o c . I n t e r n . Symp. Z e o l . C a t a l . , S i o f o k 1 9 8 5 . 1. C . T . W . Ch u , C . D . C h a n g , J . P h y s . Chem. 89 ( 1 9 8 5 ) 1 5 6 9 .
644
7 8 Y
I0 11 12 13
14 15 lb:
17 18
19
20 21 22 23 24 25 26 27 28 29 30 31
32 33
34 35
S . tlayashi, K . Suzuki, 5. Shin, K .
H a y a m i z u . 0 . Yamamoto, B u l l .
Chem. S O C . J p n . 5 8 ( 1 9 8 5 ) 5 2 . G . P . H a n d r e c k , T . D . S m i t h , J . Chem. S O C . F a r a d a y T r a n s . 1 , 85 (1989) 3215. T . J . G r i c u s K o f k e , R . J . G o r t e , G. T . K o k o t a i l o , A p p l . C a t a l . 54 ( 1 9 8 9 ) 1 7 7 . L . M a r o s i , J . S t a b e n o w , M . S c h w a r z m a n n , DE 2 8 3 1 6 3 0 ( 1 9 8 0 ) . S. J . M i l l e r , DE 3 0 3 1 1 0 2 ( 1 9 8 1 ) . R . B . B o r a d e , A . B. H a l g a r i , T . S . R . P r a s a d a R a o , P r o c . 7 t h Nat. Symp. " A d v a n c e s i n C a t a l y s i s , S c i e n c e and Technology" ( J o h n Wiley, New York, 1985) 385. A . B . H a l g a r i , R . B . E o r a d e , T . S . R . P r a s a d a Ra o, P r o c . 2nd Indo-Soviet. Conf. Catal., I n d i a 1986, 94. L . M a r o s i , J . S t a b e n o w , M . S ch war z m a nn, DE 2 8 3 1 6 3 1 ( 1 9 8 0 ) . A . J . D a b r o w s k i , K. M o s t o w i c z , 9 . C z e r w i n s k a , M . S o l i k D a b r o w s k a , B u l l . P o l i s h Acad . S c i . , Chem. 36 ( 1 9 8 8 ) 2 4 3 . P . R a t n a s a m y , R. B . B o r a d e , 5 . S i v a s a n k e r , V . P . S h i r a l k a r , S. G . H e g d e , P r o c . I n t e r n . Symp. Z e o l . C a t a l . , S i o f o k 1 9 8 5 , 137. H . B. E o r a d e , Z e o l i t e s 7 ( 1 9 8 7 ) 3 9 8 . G. C a l i s , P. F r e n k e n , E . d e B o e r , A . S w o l f s , M . A. H e f n i , Zeolites 7 (1987) 319. T . I n u i , H . N a g a t a , 0 . Yamase, H . M a t s u d a , T . K u r o d a , M . Y o s h i k o w a , T . T a k e g u c h i , A . Mi y amoto, A p p l . C a t a l . 2 4 ( 1 9 8 6 ) 257. W . J . B a l l , J . Dwy er , A . A . G a r f o r t h , W . J . Smith, Proc. 7th I n t e r n . C o n f . Z e o l . , To k y o 1 8 8 6 , 1 3 7 . K . S z o s t a k , V . N a i r , T . L . T h o mas , J . Chem. S O C . F a r a d a y T r a n s . 1, 8 3 ( 1 9 8 7 ) 4 8 7 . A . Meagher, V. N a i r , R . S z o s t a k , Z e o l i t e s 8 ( 1 9 8 8 ) 3. L . M . Kustov, V . B . Kazansky, P . Ratnasamy, Z e o l i t e s 7 (1987) 79. G. P . H a n d r e c k , T . D . S m i t h , J . Chem. S O C . F a r a d a y T r a n s 1. 85 (1989) 3195. G . V . Kharlamov, V . N. Romanikov. V. J . Kuznetzov, V . F. A n u f r i e n k o , K i n . i R a t . 30 ( 1 9 8 9 ) 1 1 8 2 . D . H . Lin, G . Coudurier, J . C . Vedrine, Proc. 8 t h I n t e r n . Conf. Z e o l . , Amsterdam 1 9 8 9 , 1 4 3 1 . A . N . K o t a s t h a n e , V . P . S h i r a l k a r , S . G . H e gde , 5 . B . K u l k a r n i , Z e o l i t e s 6 ( 1 9 8 6 ) 253. L . M. K u s t o v , V . B . Kazansky, Proc. 3rd Indo-Soviet C o n f . C a t a l . , Baku 1 9 8 8 , 1 2 8 , 2 8 . M . R i c h t e r , R. F r i c k e , B . P a r l i t z , U . H a d i c k e , G . O hlm a nn, Z . Phys. Chemie, ( L e i p z i g ) , i n press. M . R i c h t e r , W . F i e b i g , H.-G. J e r s c h k e a i t z , G . L i s c h k e , G . ohlmann, Z e o l i t e s , 9 (1989) 238. G . E n g e l h a r d t , D . Michel, High R e s o l u t i o n Solid-state NMR of S i l i c a t e s an d Z e o l i t e s , ( W i l e y , C h i c h e s t e r . 1 9 8 7 ) . G . E n g e l h a r d t , B . F a h l k e , M . M a e g i , E . L i p p m a a . Z . P h y s . Ch m. ( L e i p z i g ) , 266 ( 1 9 8 5 ) 239. D. F r e u d e , M . H u n g e r , H . P f e i f e r , Z . P h y s . Chem. (NF) 1 5 2 ( 1 9 8 7 ) 171. A. R. Halgeri, T. S. R. P r a s a d a Ra o, P r o c . I n t e r n . Symp. Acid-Base C a t a l . , Sapporo 1988, 319 J . Richter-Mendau, M. R i c h t e r , G. V o r b e c k e t a l . , t o be published.
645
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
ZEOLITE ZSM-57: SYNTHESIS, CHARACTERIZATION AND SHAPE SELECTIVE PROPERTIES S.ERNST and J. WEITKAMP Institute of Chemical Technology I, University of Stuttgart, Pfaffenwaldring 55, D-7000 Stuttgart 80 (Federal Republic of Germany)
SUMMARY Zeolite ZSM-57 can be synthesized from sodium-containing alumosilicate els using hexaethyl-C5-diquatas organic template and silica sol as silica source. ZSM-57 can e readily distinguished from the structurally closely related zeolites with the FER framework topology by its X-ray powder attern and mid-infrared spectrum. From thermogravimetric analysis of as-synthesized ZSM-!7 it follows that a plying an air calcination at 550 "C for several hours is sufficient to remove the organic temp ate from the pores of the zeolite. The results of two selected test reactions for probing the effective pore width of zeolites, viz. ethylbenzene dis ro ortionation and n-decane isomerization su est that the pore width is lar er for ZSb-f7 as compared to ZSM-35 (ferrierite) and 8 M - 5 . This is in agreement wit8; their crystallographic structures and offers new opportunities in shape selective catalysis and molecular sieve separation.
%
P
INTRODUCTION ZSM-57 is a medium pore, high silica zeolite with a structure very similar to ferrierite (refs. 1,2). Both zeolites possess channel systems consisting of intersecting linear eightmembered and ten-membered ring pores. However, the ZSM-57 framework contains fourmembered rings which are not present in ferrierite (refs. 1,2). Interestingly, the structure of ZSM-57 appeared in a systematic derivation of hypothetical structures based on the ferrierite net (ref. 3). Although the apertures of the eight-ring channels in ZSM-57 and ferrierite are of a comparable size (0.33 x 0.48 nm versus 0.35 x 0.48 nm), the ten-ring channels are significantly larger in the former zeolite (0.51 x 0.58 nm versus 0.42 x 0.54 nm) (refs. 1,2), and they are even slightly larger than the sinusoidal (0.51 x 0.54 nm) and the linear (0.54 x 0.56 nm) channels in zeolite Z S M J (ref. 4). Hence, from their crystallographic structures differences in the shape selective properties of these zeolites can be expected. The hydrothermal synthesis of zeolite ZSM-57 was first described by Valyocsik et al. (ref. 5 ) from sodium containing alumosilicate gels using N,N,N,N,N,N - hexaethylpentamethylenediammoniumdibromide (hexaethyl-C5-diquat) as organic template. Here we report on our attempts to synthesize zeolite ZSM-57, and on its characterization by X-ray powder diffraction (XRD), scanning electron microscopy (SEM), thermogravimetry (TGA), midinfrared- (IR-) spectroscopy and selected catalytic test reactions. For comparison, data obtained with a material of the FER structure type, viz. ZSM-35, are also included.
-
646
EXPERIMENTAL The template required for the synthesis of zeolite ZSM-57 was prepared by refluxing 1J-dibromopentane and a surplus of triethylamine in ethanol as solvent for several hours (ref. 5). After cooling the mixture with an ice bath, a white solid precipitated which was recovered by filtration, washed with diethylether und then used for the synthesis experiments without further purification. As silica sources, either sodium waterglass (Merck; 28.5 wt.-% Si02, 8.8 wt.-% Na20, ca. 62.7 wt.-% H20) or colloidal silica sol (Ludox HS-40, DuPont; 40 wt.-% Si02 in water) were used. All synthesis experiments were conducted at 160 "C and the crystallization time was restricted to 6 days. After this time the autoclaves were quenched in cold water and the solid products recovered by filtration, washed with distilled water and dried at 120 "C. Characterization of the as-synthesized materials occurred by X-ray powder diffraction (Cub-radiation; 40 kV, 30 mA) and IR-spectroscopy in the region of the lattice vibrations using the KBr pellet technique. The product of an optimized synthesis was further subjected to scanning electron microscopy and thermogravimetric analysis. HZSM-57 was obtained by calcining the as-synthesized material for 16 hours at 540 "C in air, followed by an ion exchange with a 1 n aqueous solution of NH&l and a second air calcination for 16 hours at 350 "C. The bifunctional form (HZSM-57 loaded with 0.27 wt.-% of palladium) was prepared via an optimized procedure which has been described previously (ref. 6). For a characterization of the shape selective properties of ZSM-57, two established test reactions, viz. the disproportionation of ethylbenzene (ref. 7) and the bifunctional conversion of n-decane (ref. 8) were selected. Both reactions were performed in a flow-type apparatus with fixed bed reactor under atmospheric pressure. The partial pressures of the feed hydrocarbons ethylbenzene (E-Bz) or n-decane (n-De) amounted to pHC= 1.3 kPa. Nitrogen or hydrogen was used as the carrier gas with ethylbenzene or n-decane as feed, respectively. Product analysis was achieved by automatic sampling and on-line analysis via temperature programmed capillary glc. For a comparison of zeolite ZSM-57 with materials of the ferrierite structure, zeolite ZSM-35 was synthesized according to a published method using 1,Zdiaminoethane as template (ref. 9). RESULTS AND DISCUSSION In the first synthesis experiments, sodium waterglass was used as the silica source. The synthesis gel was prepared by subsequently adding to the sodium waterglass, aqueous solutions of the alumina source (usually Al(NO3)3 * 9H20, in some cases Al2(SO& * 18 H20) and the template, and concentrated (98 wt.-%) sulfuric acid. The gel composition was systematically varied within the following molar ratios (calculated as described in ref. 10): Si02/Al,03 = 30 to 120, H20/Si02 = 20 to 60, Na+/Si02 = 0.6 to 0.8, OH-/Si02 = 0.1 to 0.3 and R/Si02 = 0.05 to 0.2. No ZSM-57 at all crystallized from these gel compositions under our synthesis conditions. Instead, in some cases a not yet identified crystalline component, which reversibly adsorbs/desorbs ca. 7 wt.-% of water, was obtained. Even if seeds of uncalcined ZSM-35 were added to the gel, the system could not be forced to produce
647
a ferrierite material or the structurally related zeolite ZSM-57. However, if sodium waterglass is replaced by colloidal silica sol as the silica source, pure ZSM-57 was obtained after 6 days at 160 "Cfrom a gel with the composition: Si02/A1203 = 60, H20/Si02 = 40, Na+/Si02 = 0.6,OH-/Si02 = 0.1 and R/Si02= 0.1. Note that in this case no sulfuric acid was required to adjust the pH, instead a concentrated aqueous solution of NaOH was added. This example demonstrates that, like in numerous other zeolite syntheses, the nature (reactivity) of the silica source plays an important role in directing the structure of the crystallizing material. The X-ray powder pattern of as-synthesized zeolite ZSM-57 is shown in Figure 1 (upper part). The agreement of line positions and relative intensities with literature data (ref. 5) is very good. However, ZSM-57 and ZSM-35 can be readily distinguished on the basis of their X-ray powder patterns: The most significant differences are the presence of an additional reflexion of zeolite ZSM-57 around 7.5 degrees 2 8 and the absence of a peak of medium I
I
I
I
I
I
I
I
ZSM-57
>
k v)
5
10
15
20
25
30
35
40
45
I
I
I
I
I
I
I
I
10
15
20
30
35
40
45
50
Z
w IZ
-
5
25
50
ANGLE 28 , d e g Fig. 1. X-ray powder patterns of as-synthesized zeolites ZSM-57 and ZSM-35 (Cubradiation; 40 kV, 30 mA).
648
intensity at ca. 14 degrees 2 8 . Indeed, it was stated in the original patent (ref. 5) that due to the latter difference, ZSM-57 can be "readily distinguished from ferrierite-type zeolites, such as ZSM-35". Mid-infrared spectra of ZSM-57 and ZSM-35 are depicted in Figure 2. The pattern for ZSM-35 essentially agrees with those published in the literature (refs. 11,12). The pattern for ZSM-57 is reported here for the first time. Both zeolites have two characteristic absorption bands around 1230 cm-1. These two bands can be assigned to the vibrations of five-membered oxygen rings (refs. 12,13), which is in agreement with the crystallographic structures of both materials. Slight shifts in the frequency of each IR-band may be explained by the unique framework structure of every zeolite (ref. 14). One significant difference between both zeolites appears in the region around 600 cm-1, which is known to be sensitive to structural changes (ref. 15). Whereas ZSM-35 shows a single band in this range, ZSM-57 possesses a doublet. This additional band may be tentatively assigned to the presence of four-membered rings in the structure of the latter zeolite (ref. 1,2) which are absent in the ferrierite framework. Hence, in addition to X-ray powder diffraction, IR-spectroscopy also offers a means to distinguish between both zeolites. 60
55 50 45 40 35 30
25 20 15 10 5 1400
1200
1000
800
600
1200
1000
800
600
40 0
60 55 50 45 40 35 30
25 1400
WAVE NUMBER , c m
-1
Fig. 2. Mid-infrared spectra of ZSM-57 and ZSM-35.
400
649
Fig. 3. shows a scanning electron micrograph of a typical preparation of zeolite ZSM-57. The sample consists of small platelet-like crystallites with a diameter of ca. 0.5 to 1 p m and a thickness of ca. 0.1 to 0.2pm. Frequently, these platelets are intergrown.
Fig. 3. Scanning electron micrograph of zeolite ZSM-57. The white scale bar corresponds to 10pm. 100
M
95
I--
I
c,
LLJ
B
90
85
80
'
0
70
140 210 280 350 420 490 5 6 0 630 7 0 0 770
Temperature,
O C
Fig. 4. Weight-loss of as-synthesized ZSM-57 upon heating from room temperature to 770 "C (rate: 5 K/min) in an air flow ($air = 50 cm3/min). Full line: Sample weight in dependence of temperature, dotted line: first derivate of that curve.
650
As-synthesized ZSM-57 was further characterized by thermogravimetric analysis in order to obtain guidelines for the removal of the organic template occluded in the pores during synthesis. For this purpose, the sample was heated in a flow of dry air (Vab = 50 cm3/min) from room temperature to ca. 770 "C with a heating rate of 5 K/min. The weight-loss curve and its first derivative are plotted in Fig. 4. The main weight-loss occurs between temperatures of 350 "C and 550 "C and amounts to ca. 11 wt.-%. It can be attributed to the desorption of water and the combustion of the organic template occluded inside the pores. There is a further weight-loss corresponding to ca. 3 wt.-% at temperatures above 550 "C which can be tentatively assigned to dehydroxylation of Bronsted-acid sites at these high temperatures. However, combustion of an additional, more strongly bound template species may also be envisaged. As a whole, it follows from the TGA-experiments that an air calcination at ca. 550 "C for several hours should be sufficient to remove the organic material from the pores. The catalytic and shape selective properties of HZSM-57 were characterized by the disproportionation of ethylbenzene to benzene and diethylbenzenes. This test reaction was initially proposed by Karge et al. (refs. 16,17) to characterize the Bronsted-acidity (number and/or strength of acid sites) in zeolites. Later the method was shown to be also useful for probing the shape selective properties of zeolites (ref. 7). From the time on stream behaviour of the catalyst four criteria can be derived which enable to characterize the effective pore width of the investigated material: (i) the presence or absence of an induction period, (ii) the ratio of yields of diethylbenzenes and benzene, (iii) the selectivities for ortho-, meta- and para-diethylbenzene and, (iv) the rate of catalyst deactivation. The results with HZSM-57 are plotted in Figure 5. There is no induction period, viz. conversion of ethylbenzene ( X E - ~ ~ ) decreases continuously with time on stream, and the ratio of yields of diethylbenzenes (YDE-BJ and benzene (YBz)amounts to ca. 0.77. Both features are typical for medium pore (ten-membered ring) zeolites (ref. 7). By contrast, the selectivities for the three diethylbenzene isomers resemble more closely to what could be expected with a large pore zeolite. However, there may be a marked contribution of the non-shape selective outer surface of the small zeolite crystallites of the sample used in this case. Again typical for medium pore zeolites is the increase of the selectivity for the slim p-diethylbenzene at the expense of the bulkier m- and o-diethylbenzenes with time on stream and hence coke content. This effect has been referred to as "coke selectivation" in the patent literature (ref. 18). As a whole HZSM-57 behaves in the disproportionation of ethylbenzene like a medium pore zeolite. This is in agreement with its proposed framework structure. For a more accurate ranking of ZSM-57 among other members of the group of ten-membered ring zeolites, the Modified Constraint Index (CI*) of 0.27 Pd/HZSM-57 was determined. It is defined as the ratio of selectivities for the formation of 2-methylnonane and 5-methylnonane in the isomerization of n-decane at an isomer yield of 5 Yo (ref. 8). A Modified Constraint Index of 4.3 was determined for ZSM-57. This is a typical value for a medium pore zeolite. As compared to the corresponding CI*-values for ZSM-5 and ZSM-35 (6.8 and 8.1, respectively,
651
M
L
i n
10
-1
W -
>.
75
12
-
T W/FE-,z
= 250 "C = 200 gh/mol= 0.2 g
8
t
IY
50
0 X
6
Z
0 -
cn of W >
25
4
2
0
2
4
6
8
1
0
0
2
4
6
8
1
0
TIME ON STREAM, h Fig. 5. Disproportionation of ethylbenzene in HZSM-57; P E - B= ~ 1.3 kPa, PN2 = 100 @a. ref. 19), this indicates a larger effective pore width of ZSM-57 compared to ZSM-5 or ZSM-35, which is in good agreement with the structures of these zeolites. CONCLUSIONS Provided that the right silica source is used, ZSM-57 can be readily synthesized with the hexaethyl-C5-diquat cation as organic template. It can be readily distinguished from the structurally very closely related zeolites with the FER structure-type, e. g. ZSM-35, on the basis of their X-ray powder patterns and mid-infrared spectra. Thermogravimetric analysis revealed that, like with many ofther high-silica zeolites, the template occluded inside the pores during synthesis can be readily burned off by air calcination for several hours at 550 "C. According to the results of selected catalytic test reactions the effective pore width of zeolite ZSM-57 is larger than that of ZSM-35 and ZSM-5. Hence, this new high-silica zeolite offers the opportunity to close the gap in the effective pore widths between ZSMJ and the twelvemembered ring zeolite with the most restricted pore system, ZSM-12. This probably offers new possibilities for shape selective catalysis in zeolites and molecular sieve separations. ACKNOWLEDGEMENTS The authors gratefully acknowledge financial support by Deutsche Forschungsgemeinschaft, Fonds der Chemischen Industrie and Max Buchner-Forschungsstiftung.
652
REFERENCES 1 J.L. Schlenker, J.B. Higgins and E.W. Valyocsik, in: J.C. Jansen, L. Moscou and M.F.M. Post (Eds.), Zeolites for the Nineties; Recent Research Reports, 8th Intern. Zeolite Conference, Amsterdam, July 10-14, 1989, pp. 287-288. 2 J.L. Schlenker, J.B. Hi ins and E.W. Valyocsik, Zeolites 10 (1990) 293-296. 3 R. Gramlich-Meier, Z%istallographie 177 (1986) 237-245 4 D.H. Olson, G.T. Kokotailo, S.L. Lawton and W.M. Meier,J. Phys. Chem. 85 (1981) 2238-2243. 5 E.W. Val ocsik and N.M. Page, Europ. Patent Appl. 174 121, Mar. 12, 1986, assigned to
Mobil OirCom. 6 J. Weitkamp, %’.Gerhardt and P.A. Jacobs, in: Proc. Intern. Symp. on Zeolite Catalysis, Si6fok, Hungary, May 13-16, 1985, pp. 261-270. 7 J. Weitkamp, S. Ernst, P.A. Jacobs and H.G. Karge, Erdol, Kohle-Erdgas-Petrochem. 39 (1986) 13-18. 8 J.A. Martens, M. Tielen, P.A. Jacobs and J. Weitkamp; Zeolites 4 (1984) 98-107. 9 P.A. Jacobs and J.A. Martens, Synthesis of Hi h-Silica Aluminosilicate Zeolites, Elsevier Science Publishers, Amsterdam, Oxford, New$ork, Tokyo, 1987, p. 8. 10 L.D. Rollmann and E.W. Valvocsik. Inoreanic Svnthesis 22 f 1982) 61-68. 11 Reference 9 12 K. Suzuki, $ . k ~ ~ ~ m m S. iShin, , K. Fujisawa, H. Watanabe, K. Saito and K. Noguchi, Zeolites 6 (19861 290-298. 13 G. Coudurier, C. Nacchache and J.C. Vedrine, J. Chem. SOC.,Chem. Commun. (1982) 1413-1415. 14 J.C. Jansen, F.J. van der Gaag and H. van Bekkum, Zeolites 4 (1984) 369-372. 15 E.M. Flanigen, in: J.A. Rabo (Ed.), Zeolite Chemistry and Catalysis; ACS Monograph 171, American Chemical Society, Washington, D.C., 1976, pp. 80-1 17. 16 H.G. Karge, J. Ladebeck, Z. Sarbak and K. Hatada, Zeolites 2 (1982) 94-102. 17 H.G. Karge, K. Hatada, Y. Zhang and R. Fiedorow, Zeolites 3 (1983) 13-21. 18 R.M. Dessau, US Patent 4 444 986, Apr. 24, 1984, assigned to Mobil Oil Corp. 19 P.A. Jacobs and J.A. Martens, in: Y. Murakami, A. Iijima and J.W. Ward (Eds.), New Developments in Zeolite Science and Technology; Proc. 7th Intern. Zeolite Conference, Kodansha/ Elsevier, Tokyo/Amsterdam, 1986, pp. 23-32.
G . Ohlmann et al. (Editors),Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science PublishersB.V., Amsterdam
653
NEW DATA ON THE STRUCTURE AND PROPERTIES OF ACIDIC SITES IN HZSM-5 ZEOLITES: IR-SPECTROSCOPIC STUDIES AND NONEMPIRICAL QUANTUM CHEMICAL CALCULATIONS
1.N.SENCHENYA and V.Yu.BOROVKOV N.D.Zelinsky Institute of Organic Chemistry, Academy of Sciences of the USSR, Leninsky prospect,47. Moscow B-334 117913 (USSR)
SUMMARY The quantum chemical analysis of the influence of geometry on properties of zeolite acidic OH-groups and on the strength of the A1-0 bond in SiOAl bridges is given. It is shown that bridge hydroxyls possessing high values of SiOAl angle (-180O) are able to form a strained hydrogen bond with the neighboring lattice oxygen atom attached to aluminum resulting in the appearance of HZSM-5 zeolites of the broad absorption band at 3250-3300 cm-l in IR spectra The increase of SiOAl angle in bridged hydroxyl and alkoxide leads to continuous weakening of A1-0 bond. Therefore in the presence of basic molecules the A1-0 bond in the straitened SiOAl bridges of alkoxides could be broken with the formation of trigonal A1 cation coordinated with the base being a strong Lewis acid. Such unusual manifestation of Lewis acidity of alkoxylated HZSM-5 zeolite is proved spectroscopically. The new hypothetical mechanism of the methanol conversion with the participation of such Lewis acidic sites is proposed.
.
INTRODUCTION Decationated ZSM-5 zeolites are active in the production of high quality gasoline from non-oil row materials (methanol, light olefins, paraffins etc). From the scientific point of view they are convenient model objects for the study of the influence of different structural factors on properties of their acidic sites. Indeed the structure of the lattice and pores for these catalyst are well established by X-ray analysis (ref.1). The high Si/Al ratio in HZSM-5 zeolites provides rather homogeneous chemical surrounding of their active sites (ref.2). Moreover among other zeolites they have a higher thermal stability and practically do not undergo dealumination during thermovacuum treatment. The present report deals with the theoretical analysis of two novel phenomena detected recently by ourselves using IR spectroscopy: 1. The existence in HZSM-5 zeolites of the considerable amount of acidic OH-
groups hydrogen bonded with the basic oxygen atom of the lattice (ref.3). 2. An unusual manifestation of Lewis acidity in the presence of adsorbed
sic molecules for zeolites in which a fraction of OH-groups
ba-
is substituted by
alkoxide fragments (ref.4). In addition, on the basis of considered theoretical and experimental data the new mechanisms of acid catalyzed reactions over HZSM-5 zeolites are discussed.
654
METHODS IR spectra of powdered samples were measured in diffuse reflected light as described in ref.5. Before study HZSM-5 zeolites (Si/Al = 17 and 351 were pretreated in vacuum for 2-3 h at 50OoC. The partial substitution of groups of zeolites for alkoxide fragments was performed by small quantities of propene on the zeolites at 300'K.
acidic
OH-
chemisorption of
MAS 13C NMR
spectra were
measured using CXP-300 "Brucker" spectrometer. Quantum chemical calculations were carried out using nonempirical SCF method with a "Gaussian-80''program (ref.6). Five different basis sets ranged from mi** nimal STO-3G to 6-31G were used (ref.71. Zeolite lattice was simulated by clu-
(I),
sters of following composition: H3A10HSiH3
(OH)3A10HSi(OH)3
and
(11)
OH(H)A1(OH)20Si(OH)ZOHAl~OH)3 (1111, (OH)3SiOSi(OH)20HA1(OH)3 (IY). Calculations for cluster 11-IY were made only with STO-3G basis set. All
calculations were
carried out with complete geometry optimization using gradient technique. RESULTS AND DISCUSSION
The literature data suggest that acidic OH-groups of HZSM-5 zeolites are nonhomogeneous (ref.8-9). Thus the substitution of only 20 % of acidic hydroxyls for methoxide fragments by zeolite treatment with methanol vapor is accompanied by the complete suppression of zeolite activity
in ethylene oligomerisation
(ref.8). At the same time the remained 80 % of OH-groups are "free" as they are still able to adsorbe benzene molecules with the larger kinetic diameter than that of ethylene. The detailed analysis of the band shape in IR spectra of
aci-
dic OH-groups perturbed by hydrogen bonding with adsorbed cyclohexane and benzene molecules showed the existence at least of 5 types of hydroxyls with different acidic strength in HZSM-5 zeolites. It was also found (ref. 9 ) that poisoning no more than 10
%
of the sites (counted as lattice aluminum ions) was
suf-
ficient to eliminate the activity of zeolites. Quantum chemical interpretation of the broad IR band at 3250-3300 cm-' The typical spectrum of OH-groups measured at
room
temperature in diffuse reflected
light is shown in Fig.1. It consists of two narrow lines at 3740 and
3610 cm-l
corresponding to terminal silanols and isolated acidic hydroxyls respectively. Additionally the broad absorption band with the maximum at
3250-3300 cm-l
present. This band is always detected in the spectra of HZSM-5 zeolites but
is the
position of its maximum could be changed a little for the samples of different origin. The broad absorption band displays an unusual behavior. Under cooling of
the
sample its maximum is shifted to higher wave numbers as in complexes with hydrogen bond. Adsorption of weakly basic molecules (C -C paraffins, CC1 , cyclohe-
655
xane, C02, etc) on zeolite leads to vanishing of both 3610 and 3250-3300 cm-l bands and to the development of a single band at 3480-3530 cm-' characteristic of acidic OH-groups forming H-complexes with these adsorbed molecules. The inteis nsity of the broad 3250-3300 cm-l band in the IR spectrum measured at 500'C greatly reduced but can be completely restored after sample cooled to room temperature. On the basis of these observations this band was assigned in ref.3 to two different states of the same bridged hydroxyls such as isolated OH-groups and those forming the strained hydrogen bond with the nearest basic oxygen atom attached to aluminum.
, 3610 Fig.1. IR diffuse reflectance spectrum of HZSM-5 zeolite (Si/A1=35) pretreated in vacuum at 70 K for 4 h.
I
2400
I
2800
1
3200
I
3600
v, cm-1
We tried to elucidate theoretically what type of hydroxyls is able to form such hydrogen bond. It is known that zeolites of higher protic acidity have a range of T-0-T bond angles higher than those of lower protic acidity. For instance, in ZSM-5 these values are changed from 133 to 177' (ref.1). and those in mordenites are varied in the range of 143-180' (ref.10). At the same time in Y zeolites the T-0-T angle variation is much smaller, from 137 to 143'.
It is reasonable to propose that the increase of SiOAl angle in the bridged hydroxyl will promote the formation of strained hydrogen bonding of OH-group with the nearest basic lattice oxygen atom due to the decrease of the distance between H and 0 atoms. To verify this assumption we performed corresponding calculations for clusters simulating the lattice fragments with different SiOAl angle values in the bridged hydroxyl. The obtained results demonstrate that the straighten SiOAl angle leads to the significant elongation of the T-0 and 0-H bonds as well as to the increase of the OH-group acidity (ref.11).
Fig.2 shown
the geometry of cluster I 1 with the SiOAl bond angle close to 180'.
There is al-
so the noticeable flattening of the A103 fragment. As a result the distance between the proton of the bridged OH-group and the basic oxygen atom in such clus-
656
ter becomes equal to 1.56
1 which is characteristic of a strong hydrogen bond.
The incline of the OH-group towards basic site equals to 12'. Thus the broad 3250-3300 cm-' band in I R spectra of HZSM-5 zeolites indeed could be assigned to the bridged hydroxyls with the
large SiOAl angle which to
forms rather strong and strained hydrogen bond with the oxygen atom attached
aluminurn.
Fig.2
Completely optimized geometry of cluster (LSiOA1 = 178O).
I1
The main difference in the catalytic action of HZSM-5 and other zeolites is usually explained by the higher protic acidity of the pentasil. Taking into account that the acidic strength of the bridged hydroxyls should rise with
the
increase of SiOAl angle we can assume that the hydroxyls characterized by
the
broad 3250-3300 cm-l I R band would have the highest
acidity.
worthwhile to pay special attention to the analysis of broad
It is therefore I R bands of OH-
groups of high silica zeolites. Lewis acidity of HZSM-5 zeolites The nature of Lewis acidic sites (L-sites) and their role in catalysis by
zeolites is widely discussed in the
literature
(ref.12). Now the existence of several types of Lewis sites well established. Among them there are the extralattice aluminum containing species and the lattice trigonal aluminum or silicon cations appeared as a result of zeolite dehydroxylation at high pretreatment temperatures. In our opinion these sites, however, could be considered as the active sites in the catalysis by
zeolites with
caution. The reasons for this are as follows: 1. HZSM-5 zeolites hardly undergo dealumination during their activation under conventional conditions and therefore the concentration of extralattice aluminum is usually small. Moreover these sites are believed to participate in the coke
formation which is not characteristic of HZSM-5 zeolites.
2. The formation of lattice L-sites are known to proceed at temperatures above 5OO0C (ref.12). So their appearance under usual
activation conditions also
has a low probability. Moreover during catalytic reactions these L-sites could be easily poisoned by moisture traces of the feedstock.
657
In this connection we would discuss the idea about two different ways of the interaction between basic molecules and the acidic OH-groups in zeolites proposed by Uytterhoeven et a1 in 1965 (ref.13). The first one is the usual reaction of the base protonation: HB+ H
The second route is associated with the break of the Al”’0 bond in the Si-0-A1 bridge accompanied by the formation of complex between basic molecule and the trigonal A1 cation being the strong Lewis acid:
H
H
To estimate the probabilities of these two reaction paths we performed the quantum chemical calculations of the A1-0 bond energies in model clusters I-IY. These values were estimated as the energy of the reaction H
which is equal to the difference between total energy of the initial cluster and / the sum of total energies of 3i-OH and A ~ Kseparated fragments having equilibrium geometries. The energies of complete A1-0 bond break for cluster I calculated with different basis sets are listed in Table 1. TABLE 1 Basis set dependence of A1-0 bond energy (AE, kJ/mole) for cluster I 9
Basis set
STO-3G
3-21G
6-31G
MF’2/6-31G
AE
176.5
156.5
79.5
91.0 9
According to the estimation at the MF’2/6-31G level the energy of Si-0 and 0H bonds breaking in the molecule H3SiOH which respectively correspond to enthalpies of reactions H3SiOH - H3SiO’ + H’ and H3SiOH - H3Si‘ + HO’ is approximately the same and equal to 500-550 kJ/mole. Therefore the strength of the A1-0 bond in the SiOAl bridge is much smaller than that of the Si-0 and 0-H bonds.
It is clear that minimal cluster model I is only a very primitive model of the zeolite lattice. Therefore we also calculated the BE values of the A1-0 bond in more extended clusters 11-IY. The transition from cluster I to I 1 is accompanied by the decrease of the A1-0 bond energy from 176.5 to
136.5 kJ/mole.
The
658
substitution of one H atom in cluster I1 for
real silicon-oxygen tetrahedron
does not lead to significant changes in cluster geometry whereas similar substitution of the same H atom for aluminum-oxygen tetrahedron is accompanied by noticeable shortening of the A1-0 bond. This evidences the increase of the A1-0 bond strength with the decrease of the Si/Al ratio in the zeolite lattice. The direct calculation indicates that the A1-0 bond strength in cluster I 1 1
is the
same as in cluster 11. The existence of aluminum-oxygen tetrahedron in the neighborhood of bridged hydroxyl (cluster I Y ) increases the A1-0
bond
strength
by the factor of 1 . 5 (197 kJ/mole). Quantum chemical analysis (clusters I and 1 1 ) of the influence of zeolite lattice geometry and substitution of the acidic proton for sodium cation or methoxy fragment on the A1-0 bond strength led us to the following conclusions: 1.
The change of the SiOAl angle from its equilibrium value to -180’ leads to
a significant decrease of the A1-0 bond strength and to the flattening of
Al-
atom surrounding. 2. The substitution of the bridged proton for Na’ is accompanied by
shorte-
ning of the Si-0 and A1-0 bonds along with sharp strengthening of the A1-0 bond as compared to that in the cluster containing bridged hydroxyl. 3. The energies of the A1-0 bond in the
methoxylated cluster I
calculated
with STO-3C and 3-21G basis sets are noticeable lower than those in the cluster containing OH-group, by 20 and 50 kJ/mole, respectively. On the contrary the decrease of the A1-0 bond energy in the methoxylated cluster I 1 is much smaller
(*
5 kJ/mole) due to formation of the intermolecular hydrogen bond between the hyd-
rogen atom of the methoxy group and the oxygen atom attached to aluminum. The weakness of the Al-0 bond strength in high-silica zeolites could be manifested in the course of chemical reactions as well as during interaction of
the
acidic hydroxyls with adsorbed basic molecules. Let us demonstrate this using the interaction of the bridged OH-group with the ammonia molecule as an example. The nonempirical STO-3G calculations indicate that the interaction of ammonia with the SiOAl bridge, which leads to the breaking of the A1-0 bond
(scheme 2 )
proposed by Uytterhoeven is allowed energetically. Nevertheless the ammonia protonation (scheme 1) which in practice usually takes place is energetically more favorable. There is, however, an opportunity to switch off the protonation channel by substitution of the acidic proton for an alkoxy fragment. As was mentioned above such a substitution also results in weakening of the Al-0 bond. Therefore the interaction of the NH3 molecule with the alkoxy group by the route
R
R 0
>Si’ \AIL \
0 + B
-+ >Si’
is even more probable.
B’.’Alf
(4)
659
IR spectroscopic studies of the basic molecule adsorption (NH3. CD3CN) on HZSM-5 zeolites support this assumption (ref.4). Fig. 3 displays the IR spectra of NH3 and CD CN molecules adsorbed both on 3 fresh HZSM-5 zeolite and on the sample with propene preliminary chemisorbed at 300 K. There are only two bands at 1460 and 2295 cm-l in IR spectra of the initial H-form which correspond respectively to bending 6N-H and stretching v
CeN vibrations of ammonium ions and CD3CN molecules coordinated with the acidic OHgroups. Additional bands at 1610 and 2370 cm-' appear in the zeolites containing chemisorbed C3H6. They are characteristic of ammonia (aNH = 1610 cm-'1 and deuteroacetonitrile ( uC=N = 2370 cm-'1 molecules adsorbed on L-sites.
I
2400
I
2300
I
I
u.cm
-1
1700
I
1600
I
1500
I ,
-1
v,cm
Fig.3. IR spectra of NH3 (A)and CD3CN (B) adsorbed on fresh HZSM-5 zeolite (1) and on the sample with propene preliminary chemisorbed at 300 K (21. Line ( - -1 corresponds to the background. The process of such an L-sites formation is a reversible one. They disappear after removal of chemisorbed hydrocarbons by evacuation of the sample at higher temperatures but are restored again after repeated chemisorption of the olefin.
It is worthwhile to emphasize that in this case no additional coordinatively unsaturated L-sites was found from IR spectra of adsorbed weak basic molecules (H2 and CO). The conclusion was made that the formation of L-sites in the presence of strongly basic molecules in zeolites Containing chemisorbed olefins proceeds according scheme (4).where the oligomer hydrocarbon chains play a role of the alkoxide R. The formation of alkoxy structures is evidenced by the appearance of the intense line at 70 ppm corresponding t o resonance of C-atom in the fragment in MAS 13C NMR spectrum of chemisorbed 13C-enriched ethylene (Fig.4 ) . The energy calculated for reaction (4) (B = NH3) indicates that this process
'0-bH
/
I
is energetically favorable: AE = 33.5 kJ/mole for cluster I (3-21C1 and AE = 13
660
kJ/mole for cluster I1 (STO-3G). For the acetonitrile molecule the energetics of such reaction is less advantageous. It, however, became much more favorable f o r bridged alkoxy groups with the larger SiOAl bond angle. Indeed, the increase of this angle by 50'
from its equilibrium meaning leads to essential enhancement of
system total energy and, as consequence, to sharp lowering the energy of A1-0 bond breaking in the presence of strong bases. At the same time for zeolites with the low Si/A1 ratio the energy of the A1-0 bond is higher, than the energy of the interaction between the basic molecule and the L-site. This explains why L-sites are not formed in methoxylated zeolites of such type with strong basic molecules adsorbed.
29
Fig.4.
MAS 13c NMR spectrum of chemisorbed I3C enriched ethylene on HZSM-5 at 370 K.
75
50
25
0
The theoretical results considered above demonstrated that the formation of L-sites in the presence of strong bases occurs only in the moieties of the
zeo-
lite framework having large values of angles in SiOAl bridges. As only protons localized on such SiOAl bridges are able to form a strained hydrogen bond with neighbor lattice oxygen atom, the substitution of OH-groups characterized by the broad IR band at 3250-3300 cm-l f o r alkoxides results in the manifestation of the unusual L-sites in the presence of adsorbed basic molecules.
It is reasonable to suppose that such L-sites could also appear during chemical reactions with the participation of basic molecules, for instance, methanol. In this case they can play a role of active sites for methanol conversion. L-sites and chemical reactions. Let us analyze this assumption in more detail. Indeed, the first step of the CH OH molecule interaction with acidic OH-groups 3
is the formation of bridge methoxy fragments. .H
The next methanol molecule is able to break A1-0
bond
and
consequently
to
661
transfer the bridge methoxide into the terminal one, as well as to dissociate on the pair consisting of the formed trigonal A1 cation and the basic oxygen atom attached to aluminum. This process could proceed via two energetically favorable ways with the formation of either terminal A1-0 group and bridged methoxide or A1-OCH3 fragment and bridged hydroxyl. The second structure is more preferable in the presence of the methanol excess due to the formation o f the additional strong hydrogen bond between the acidic OH-group and CH OH molecule. 3
In the first case the production of methane and formaldehyde could be accomplished from the formed structure by the following synchronous mechanism (scheme 6 ) . It includes the attachment of the basic OH-group to bridge methoxide accompanied by hydride ion abstraction. The interaction of hydride ion with the nelghboring terminal methoxide results in the break of the 0-C bond and formation of the methane molecule. Simultaneously the restoration of initial SiOAl bridge and the formation of an unstable bridged hydroxymethoxide group take place.
Abstraction of the proton from the latter complex being accompanied by
the 0-C
bond breaking produces the formaldehyde molecule and regenerates the acidic OH-group.
initial
When second type structure is formed via CH30H dissociation on the A1-0 pair the next adsorbed methanol molecule is activated by the acidic OH-group. Its .further conversion could be presented by the scheme:
H I
H3C'
662
The proton transfer from bridged hydroxyl to the adsorbed CH30H molecule accompanied by the nucleophylic attack of terminal methoxide by
is
the protonated
methanol molecule. As a result the abstraction of hydrogen atom from terminal methoxlde to neighboring oxygen and the formation of
the acidic OH-group and
terminal ethoxide occur. In this case the regeneration of
the
initial SiOAl
bridge unit and acidic hydroxyl proceeds throughout the desorption of methanol and ethylene molecules. The schemes similar to (6) and (7) could be also suggest for dimethylether conversion. The realization of scheme (7) would be accomplished only
in the excess of
methanol. In the lack of the CH OH molecule the structure containing the A1-OCH 3 3 group and acidic OH-group would be isomerized into a more energetically preferable one, which leads to methane and formaldehyde formation. It is worthwhile to mention that the schemes presented here allow to explain the difference observed in the products of methanol conversion at low (methane and
formaldehy-
de) and high (hydrocarbons) pressures of alcohol (ref.161. REFERENCES
K.J.Chao, J.C.Lin, Y.Wang, G.H.Lee, Single crystal structure refinement of TPA ZSM-5 zeolites, Zeolites 6(1) (1986)pp. 35-38. V.B.Kazansky, Theory of Broensted acidity of crystalline and amorphous aluminosllicates: quantum chemical cluster model and IR spectra Kinet. Catal. 23(6) (1982) pp.1334-1348. V.L.Zholobenko, L.M.Kustov, V.Yu.Borovkov, V.B.Kazansky. A new type of acidic hydroxyl groups in ZSM-5 zeolite and in mordenite according to diffuse reflectance i.r. spectroscopy, Zeolites 8 ( 5 ) (1988)175-178. A.S.Medin, V.Yu.Borovkov. V.B.Kazansky, A.G.Pelmenshchikov,G.M.Zhidomirov, On the unusual mechanism of Lewis acidity manifestation in HZSM-5 zeolites, Zeolites 1990 (in press) V.B.Kazansky, Diffuse reflectance IR spectroscopy and its new potentialities in studying chemisorbed species and the structure of surface oxide catalysts, Izv. AN SSSR 1 (1984) pp.40-51. J.S.Binkley, R.A.Whiteside,R.Krishnan, R.Seeger, D.J.DeFrees. H.B.Schlege1, S.Topio1, L.R.Kahn, J.A.Pople, QCPE, 13 (1981) 507. W.J.Hehre, L.Radom. P.v.R.Schleyer, J.A.Pople, Ab initio molecular orbital theory, Willey Intersci., New York etc., 1986. A.S.Medin, Ph.Diss., Moscow, 1990. E.A.Lombardo. G.A.Si11.W.K.Hal1. The assav of acid site on zeolites as measured by ammonia poisoning, J. Catal., 119(2) 1989 pp.426-40. 10 J.L.Schlenker, J.J.Pluth, J.V.Smith,Position of cations and molecules in zeolites with the mordenite framework. IX. Dehydrated H-mordenite via acid exchange, Mat. Res. Bull., 14(7) (1979)pp.849-56. 11 I.N.Senchenya, V.B.Kazansky, S.Beran, Quantum chemical study of the effect of the structural characteristics of zeolites on the properties of their bridging OH groups, J. Phys. Chem., 90(20) (19861 4857-4859. 12 V.B.Kazansky, On the nature of Lewis acidic sites in high silica zeolites and the mechanism of their dehydroxylation, Catalysis Today,3 (1988) 367-72. 13 L.B.Uytterhoeven, L.G.Crystner,W.K.Hal1, Studies of the hydrogen held by solids, J. Phys. Chem., 69(6) (1965) pp.2117-2126. 14 L.Kubelkowa, J.Novakova, P.Jiru, Reaction of small amounts of methanol on HEM-5, HY and modified Y zeolites, In: Structure and Reactivity of Modified Zeolites, Stud. Surf. Sci. Catal. 18 (1984) pp.217-224.
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 01991 Elsevier Science Publishers B.V., Amsterdam
663
FTIR IN-SITU INVESTIGATION OF ZEOLITE ACTIVATION
R. SALZER', B. EHRHARDT', J. DRESSLER', K.-H. STEINBERG' and P. KLAEBOE~ 'Department of Chemistry, University of Leipzig, Talstr. 35, Leipzig 7010, GDR *Department of Chemistry, University of Oslo, N-0315 Oslo, Norway
ABSTRACT Activation of powdered, undiluted zeolites was studied in-situ by FTIR diffuse reflectance (DRIFT) spectroscopy. From the DRIFT spectra we constructed contour plots, which display the process quasi continuously. At temperatures below 2OO0C, the diffusion of water and ammonia towards the cation coordination sphere is seen. The maximum loss of a particular NH species was observed at 500°C. The concentration of OH species increases most strongly at 430"C, after the on-set of an almost uniform decomposition of all NH species. The activation of 0.93 NH,' erionite was completed at around 600°C. INTRODUCTION Diffuse
Reflectance Infrared Fourier-Transform (DRIFT-) spectroscopy has a high potential for in-situ studies of heterogeneous catalytic systems. Even very slow processes can be monitored over long periods (ref. 1). Most samples may be investigated without further pretreatment in a wide range of temperatures and pressures (ref. 2). Lateral resolution in the millimeter range can be achieved. We used DRIFT spectroscopy to investigate the activation process of zeolites i n - s i t u under atmospheric pressure between room temperature and 640°C. NH,' exchanged erionites were selected as test compounds. The DRIFT results were compared to the results of a thorough study of erionites by conventional infrared transmission technique (ref. 3). Here we describe the experimental technique and present the first results.
664
SAMPLING SYSTEM
DRIFT spectra are given in Kubelka-Munk units f (R) (ref. 4 ) , which are directly related to the absorbance coefficient k of the sample and its scattering coefficient s via ( 1 ) f(R) = (1 - R/R')' / 2R/R' = k / S . R is the single-beam reflectance spectrum of the sample, R' the single-beam reflectance spectrum of a non-absorbing standard. The Kubelka-Munk equation (1) was derived for the ideal case of purely diffuse reflectance. For real samples, diffuse reflectance is always accompanied by specular reflectance, which upsets the ordinate scale and deteriorates the detection limits. The only way to prevent contributions of specular reflectance from reaching the detector is to select a DRIFT attachment of a particular optical design, the so-called off-axis geometry (ref. 5), where the incoming beam and the normal to the sample surface do not share a common plane (or axis) with the reflected beam (Fig. 1). A heatable sample stage for DRIFT spectroscopy, which is commercially available (HVC-DRP chamber of Harrick Scientific Corp., Ossining/NY), was tested over a wide temperature range. The vacuum chamber body, which encloses the sample stage, was omitted because the investigation was carried out at ambient pressure. Any change in the sample position during the measurement series, including realignment, was avoided.
reflected beam incoming beam sample to power supply heater thermal insulator adjustment screw base
Fig. 1. Scheme investigations.
of
the
sample
heater
for
the
DRIFT
665
S i n g l e beam s p e c t r a of t h e t e s t sample ( c o r r e s p o n d i n g t o R i n e q n . 1) a r e summarized i n F i g . 2 . The a r e a below a p a r t i c u l a r c u r v e r e f e r s t o t h e t o t a l amount of modulated r a d i a t i o n r e a c h i n g t h e d e t e c t o r . Upon h e a t i n g t h e HVC-DFW sample s t a g e , t h e s i n g l e beam s p e c t r a become i n c r e a s i n g l y f l a t ( F i g . 2 a )
. The
observed r e d u c t i o n
i n e n e r g y t h r o u g h p u t i s more s e v e r e a t s h o r t e r w a v e l e n g t h s (compare the
lowering
of
t h e c u r v e maxima
4 0 0 0 a n d 2200 cm-I).
at
This
s u g g e s t s a g e o m e t r i c a l e f f e c t dominating t h e energy d r o p d u r i n g h e a t i n g , p r o b a b l y b e due t o t h e r e l a t i v e l y l o n g h e a t i n g zone of t h e sample s t a g e , c h a n g i n g t h e sample p o s i t i o n r e l a t i v e t o t h e
HVC-DRP
reflectance mirrors. 1
1
i 0,5
0
/p
25 *C
I
a)
i
~
0.
, 6000
0 2000
4000
+-
cm-'
0
6000
4000
2000
+-
cm-'
F i g . 2 . S i n g l e beam s p e c t r a of 0 . 9 3 NH,' e r i o n i t e a t d i f f e r e n t t e m p e r a t u r e s . a ) Sample s t a g e of HVC-DRP chamber. b ) Oven a s i n F i g . 1. The o b s e r v e d e n e r g y l o s s d u r i n g h e a t i n g r e d u c e s t h e S / N r a t i o ,
causes
background
ordinate
non
fluctuations
linearities.
in
These
the
spectra
effects
and
can
introduces be
reduced
c o n s i d e r a b l y , when t h e r e f e r e n c e s p e c t r u m i s measured a t t h e same t e m p e r a t u r e a s t h e sample spectrum. The b e s t
results
according t o Fig.
1,
c o u l d be
observed
using a
sample
stage
whose t h e r m a l e x p a n s i o n was minimized by
i n t r o d u c i n g t h e h e a t e r d i r e c t l y i n t o t h e sample b u l k ( F i g . 2 b ) .
666
EXPERIMENTAL
A Bruker FTIR spectrometer IFS 88 was used for all tests of the heatable sample stage of the HVC-DRP reaction chamber. The sample stage was inserted into a diffuse reflectance attachment D R A 3 (Harrick Scientific Corp. , Ossining/NY, USA) . All the other FTIR spectra were scanned on an IRF 180 (ZWG, Akademie der Wissenschaften der DDR, Berlin), equipped with a the off-axis diffuse reflectance attachment of the same manufacturer. 128 interferograms were co-added before Fourier transformation. The thermal insulator of the heatable sample stage was made of glass ceramics (Fig. 1). A micro thermocouple was used for temperature measurement. The alignment of the thermocouple to the uppermost layer of the powder is important for exact measurement. Samples were heated by 2 K min-', and during the spectral measurements the temperature was kept constant. The synthesis and ion exchange of the erionites used in this study have already been described (ref. 3). The samples were investigated as received. Finely ground KBr was used as reference. No features, e.g. peaks due to atmospheric C02, have been omitted from the spectra. RESULTS AND DISCUSSION The OH and NH stretching bands are selected in order to monitor the activation of undiluted 0.93 NH,' erionite in the open
atmosphere (Fig. 3). All the band locations and intensity ratios correspond completely to the transmission spectra (ref. 3). The graduate loss of physisorbed species is indicated by the overall reduction in the size of the band complex. The almost complete desorption of NH species up to 550°C is seen by the disappearance of the corresponding NH stretching bands below 3 3 0 0 cm-'. Simultaneously, the band appearing above 3600 cm-! indicates the formation of OH groups. A detailed investigation of the whole activation process (Fig. 3) is difficult because of the complicated shape of the bands. For this purpose we applied a complex evaluation procedure to the spectra. It is important to note, that no individual assumptions are necessary in course of the following computations.
667 A s a first step,
Fig. 3. DRIFT spectra of 0.93 NH,' erionite during activation at the open atmosphere.
all the constant (background) features apparent during heating are removed by s u b t r a c t i n g consecutive spectra. It depends upon the available software, w h e t h e r t h e temperature intervals have to be constant or not. A high S/N ratio and a constant ordinate accuracy in the single beam s p e c t r a a r e prerequisites. The ordinate accuracy of DRIFT spectra is mainly determined by the scattering coefficient s in eqn. 1. It varies remarkably for different sample preparations, even for samples from the same batch. One must only consecutive series of
evaluate spectra, which belong to one measurements. The D R I F T difference spectra in Fig. 4 enhance all the changes taking place during activation, which have been discussed regarding Fig. 3. The removal of all the constant features did not simplify the shapes significantly. In the following, we shall regard the complete set of curves in Fig. 4 as a landscape. The height of the hills corresponds to the increase of a particular XH species per temperature step, the
668
Fig. 4 . DRIFT d i f f e r e n c e 0 . 9 3 NH4’ e r i o n i t e
spectra
for
the
activation
of
d e e p n e s s of t h e v a l l e y s t o t h e d e c r e a s e i n c o n c e n t r a t i o n . A t e v e r y v i e w i n g a n g l e mountains i n t h e f o r e g r o u n d h i d e f e a t u r e s b e h i n d , whereas a map, t h e s o - c a l l e d c o n t o u r d i a g r a m , p r o v i d e s immediate a c c e s s t o a l l t h e d e t a i l s of t h e measured s e r i e s . Contour d i a g r a m s f o r t h e DRIFT d i f f e r e n c e s p e c t r a i n F i g . 4 a r e g i v e n i n F i g s . 5 and 6 . To e n s u r e c l e a r n e s s of t h e monochrome d i s p l a y , t h e r e g i o n s of n e g a t i v e r e f l e c t a n c e c h a n g e s ( d e c r e a s i n g c o n c e n t r a t i o n s ) d u r i n g h e a t i n g a r e h i g h l i g h t e d i n F i g . 5, r e g i o n s of p o s i t i v e r e f l e c t a n c e c h a n g e s ( c o r r e s p o n d i n g t o i n c r e a s i n g concentrations) i n Fig. 6 . A t t e m p e r a t u r e s below 200°C t h e I R r e f l e c t a n c e d e c r e a s e s i n
t h e r e g i o n 3700
-
2800 cm-’ due t o d e s o r p t i o n o f w a t e r and ammonia
( F i g . 5 ) . The r e f l e c t a n c e l o s s e s s t a r t most
pronounced
around
3450 cm-’ (OH s p e c i e s ) and 3 1 0 0 cm-’ ( N H s p e c i e s ) . Above 150°C t h r e e bands f o r d e s o r b i n g NH c a n be s e e n ( 3 3 0 0 , below 3100, below
2900 cm-I). Up t o 200°C a s i m u l t a n e o u s i n c r e a s e i n r e f l e c t a n c e i s o b s e r v e d between 2800 and 2200 cm-’. The i n c r e a s e d r e f l e c t a n c e i s e a s i l y s e e n
669
640 O C
530
420
310
200 130
Fig. 5. Contour plot of DRIFT difference spectra in Fig. 4. The bright area indicates decreasing concentrations.
640 O C
530
420
310
200
130 Cm-'
Fig. 6. Contour plot of DRIFT difference spectra in Fig. 4. The bright area indicates increasing concentrations.
670
in Fig. 6, whereas it could only be guessed in Figs. 3 and 4. The increased reflectance amounts to ca. 10% compared to the decrease mentioned above. It points to diffusion towards the cation coordination sphere. The assignment of the bands to water and ammonia is under study. At 26OoC, a minimum of reflectance changes across the investigated spectral region is indicated in Figs. 5 and 6 . Above this temperature the decomposition of the NH,+ cations starts. Three bands are observed (ca. 3 2 5 0 , 3 0 3 0 and 2 7 8 0 cm-') . The decomposition is first indicated for the band having the highest frequency. This band a l s o shows the maximum loss of NH,' ions at 500°C. The on-set of an almost uniform decomposition of all the NH,' species is indicated by the spread of the contour line at 400°C (Fig. 5). The formation of new OH groups is first observed for species absorbing above 3 6 0 0 cm-' (Fig. 6 ) . This band has been assigned to strong Brernsted acid sites (ref. 3 ) . An OH species, absorbing at 3 5 6 0 cm-' and located in the six-membered ring units (ref. 3 ) , is formed at somewhat higher temperatures. The stretching frequencies of both OH species are red shifted with increasing temperature. The concentration of OH groups increases most strongly around 430"C, after the beginning of the uniform decomposition of all NH,' species. At ambient pressure the activation of 0 . 9 3 NH,' erionite is completed at around 650°C. ACKNOWLEDGEMENT
This research was supported by The Norwegian Research Council for Science and the Humanities (NAVF). REFERENCES 1 R. Salzer, K.-H. Ste nberg, P. Klaeboe, B. Schrader, Zeolites submitted. 2 S. A. Johnson, R.-M Rimkus, T. C. Diebold and V. A. Maroni Appl. Spectr. 42 1 9 8 8 ) 1 3 6 9 3 A. Kogelbauer, J. A . Lercher, K.-H. Steinberg, F. Roessner, A. Soellner and R V. Dmitriev: Zeolites 9 ( 1 9 8 9 ) 2 2 4 . 4 P. Kubelka and F. Munk: 2. Techn. Phys. 1 2 ( 1 9 3 1 ) 5 9 3 . 5 D. M. Hembree and H. R. Smyrl: Appl. Spectr. 4 3 ( 1 9 8 9 ) 2 6 7
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
671
IR SPECTRA OF CO ADSORBED AT LOW TEMPERATURE (77 K) ON TITANIUMSILICALITE, €I-ZBMS AND SILICALITE
A. ZECCHINA~,G. SPOTO~,s. BORDIGA~,M. PA DO VAN^, G. LEO FAN TI^ and G. PETRINI~ 'Dipartimento di Chimica Inorganica, Chimica Fisica e Chimica dei Materiali, Via P. Giuria 7, 10125 Torino (Italy)
'Montedipe, Unit& di Ricerca, Via San Pietro (MI) (Italy)
50,
28100 Bollate
ABSTRACT
The IR band at 960 cm'l observed on TS is due to a local stretching mode of a [Ti04] unit in the silicalite framework. Framework Ti does not show Lewis acidity as probed by CO adsorption at 77 K. TiOH and SiOH have undistinguishable IR properties. At 77 K they form with CO very weak 1:l OH...CO adducts. Small amount of extralattice Ti in form of Ti02 can be probed by CO adsorption at 77 K. INTRODUCTION
The substitution of Ti and A1 for Si in zeolites of the pentasil family (Silicalite) leads to Ti-Silicalite and ZSMS which are two important catalysts for oxidation with HZOZ (refs. 1-3) and acid catalyzed reactions (ref. 4 ) . The characterization of Titanium-Silicalite (TS) by means of physical methods has been recently published (ref. 5). However, a few problems are still open concerning: i) the presence of extralattice Ti (as Ti02 microparticles); ii) the interpretation of the vibrational spectrum of the solid; iii) the crystallinity of TS with respect to pure S and ZSMS; iv) the presence of titanols in the channel8 and cavities. The CO molecule is an efficient probe of Lewis and Broensted acidity and can be used to explore the acidity of OH groups in the channels (TiOH and SiOH in TS, (Si,Al)OH, SiOH and AlOH in H-ZSM5). On dehydroxylated samples coordinatively unsaturated Ti4+ and A13+ ions in lattice and extralattice positions can be revealed as well, because they can form Lewis adducts with CO. In this contribution we report on the IR spectra of CO adsorbed at 77 K on TS, s and ZSMS. The vibrational spectra of s,
TS and Na-ZSM5 are also compared and discussed. EXPERIMENTAL
Silicalite and Titanium-Silicalite have been synthesized in Montedipe laboratories following the method described in ref. 2; Na-ZSM5 and H-ZSM5 (external surface area 60 m’9-l) have been provided by the same laboratory. The ER spectra have been obtained on a Bruker IFS 113V FTIR spectrometer using a specially designed silica cell permanently attached to a vacuum manifold and allowing in situ outgassing procedures at temperatures in the 273-1073 K interval and gas dosing at 7 7 K. The samples were either in form of thin pellets or of films deposited on KBr or Si plates. RESULTS AND DISCUSSION IR modes associated with framework Ti
The spectra recorded at 7 7 K of S, TS and ZSM5 (sodium form; Si/A1=35) films outgassed at 573 K are compared in Fig.1. The major difference between S and ZSM5 on one side and TS on the other side is represented by the presence in the spectrum of TS of a (finger print) peak at 960 cm-l. In order to give a correct assignment of this band the following considerations have to be made: a) it cannot be assigned to an optical mode of TiOZ microparticles entrapped into the channels or at their intersection because neither rutile nor anatase show strong IR bands at this frequency; b) despite the similarity with the IR manifestations of homogeneous titanyl containing analogs , the 960 cm-’ peak cannot be attributed to the stretching mode of internal (Ti=O)’+ because: i) it is not perturbed by filling the pores and channels with CO (results not reported), which, being a weak Lewis base, is expected to interact with the positive centres and to perturb them; ii) it does not have the expected counterpart in the UV-Vis diffuse reflectance spectrum (absorption in the 25000-35000 cm’l range) (ref. 6); iii) it cannot be removed by reductive treatments in H2 and CO even at very high temperature or under plasma discharge conditions (Ha). On the basis of the previous **negativet* evidences, the hypothesis has been made that the 960 cm-l band is due to a local mode of the Titanium-doped pentasilic structure (essentially a stretching mode of a [Si04] unit perturbed by an adjacent Ti) (ref. 5 ) .
673
F'LI
0.5
A
/ I
1
I0
YRVENUHBER C H - I
Fig.1. IR absorbance spectra (recorded at 77 K) of: a) Silicalite, b) Na-ZSM5 (Si/A1=35) and c) Ti-Silicalite. In the inset: W-Vis reflectance spectra of a) Silicalite and c) Ti-Silicalite.
A full assignment of the IR spectra of s, TS and ZSM5 requires a detailed unit cell vibrational analysis and is outside the scope of this contribution. However, an explanation of the presence of the extra-absorption in the TS spectrum can be equally given on the basis of the following qualitative considerations. The vibrational representation of the stretching modes of an llisolatedll tetrahedral [ S i O , ] unit is:
rstret.= T2 *I (TZ: IR active; A1: IR inactive). Packing of the [Sio,] (by sharing the corners) to form the zeolitic structure has two consequences: i) the local *'site1@symmetry of each [SiO,] primary +
674
building unit is lowered to C2 (ref. 7) so that the triply degenerated T2 mode is splitted into three, one weak (A) and two strong (B), components and ii) due to the high number of [SiO,] units forming the unit cell (where 9 6 tetrahedral units are present) broad bands are expected because of the further splitting of each component into a maximum of 96 sub-components. On these basis, the spectrum of S in the stretching region can be interpreted as follows: i) the two clear absorption at 1235 (mw) and 1200-1050 (vs) cm-l, are essentially associated with the A and B modes of the primary unit broadened by unit cell splitting effects; ii) the complex (vw) band centered at =770 cm-l derives from the A1 (IR inactive) mode (the fine structure observed at 77 K being associated with the unit cell sub-components). The bands at lower frequency belong to skeletal modes having bending character and will not be considered here for sake of brevity. We only mention that the peak at ~ 5 5 0cm-l is absent on amorphous silica and is considered as a gfcrystallinityfg (structure sensitive) band (ref. 5). As far as TS is concerned, some 1-2% of the [SiO,] building blocks of pure silicalite framework are substituted for Igheavierff [TiO,] units. The substitution does not dramatically change the overall IR spectrum other than for the appearence of localized fgimpuritygg modes associated with the [TiO,] dopant units. In the stretching region four additional local modes are, in principle, expected, three (A+2B) deriving from the T2 degenerate vibrations of the free [TiO,] unit and one (A) from the A1. These modes should be shifted to lower frequencies with respect to those of Silicalite, because of the mass effect. Actually only the band deriving from the strongest absorption at 1120 cm-l (presumably of B simmetry) has enough intensity and is enough shifted to show up clearly in the gap between the 1250-1050 and 820-750 cm-l absorptions. The previous analysis predicts that, due to the small difference in the mass of A1 and Sir the impurity modes associated with [AlO,] should be not osservable at all. This explains the great similarity between the Silicalite and Na-ZSM5 spectra. The assignment of the 9 6 0 cm" peak to local modes of [TiO,] units is in agreement with: i) the characteristic reflectance spectrum of TS (inset of Fig.1) where an high intensity band at 48000 cm-', with charge transfer character, is indicative of the presence of [TiO,] units (ref. 5 and references therein) ; ii)
675
the IR spectrum of Si02/Ti02 glasses where a similar, although broader, band is observed (refs. 8,9). Of course these considerations, based only on the mass effect, are an heavy approximation. In fact, due to the higher ionicity of the Ti-0 bond, the Ti-0 and the Si-0 stretching constants are not the same. This assignment is slightly different with respect to that given in ref. 5, where a larger ionicity of the Ti-0 bond was assumed. It can be easily demonstrated that they transform the one into the other when the ionicity of the Ti-0 bond is gradually changed from pure covalent to ionic. Ir spectra of CO adsorbed on Silicalite, Ti-Silicalite and HZSM5: assianment and comarison The spectra of CO adsorbed at 77 K on S and TS samples outgassed at 573 K under vacuo are reported in Fig.2a (3800-3000 cm-l region) , and Fig.2b (2250-2050 cm-l) and Fig.3a and 3b respectively. Similar spectra have been obtained for samples outgassed at lower and higher temperature. They are not described in detail for sake of brevity. The following can be commented. Silicalite The IR spectrum in the OH stretching region (Fig.2a) is characterized by two main absorptions at 3750-3700 cm-' (composit with narrow components at 3750 and 3730 cm-l and a broader one at These two bands 3710 cm-l) and at 3460 c m ' l (0))112=130 cm") essentially correspond to free (isolated and terminal, external and internal) and hydrogen bonded silanols respectively. Upon CO dosage the bands due to free silanols (both isolated and terminal) are eroded, while two new peaks are formed at 3640 and 3585 .''nrc The presence of a clear isosbestic point indicates that the free OH (isolated and terminal) are transformed into hydrogen bonded species (because of formation of 1:l OH...CO adducts) as illustrated in the following scheme:
-
OH.. .CO
I
.
OH...OH...OH...CO
(isolated)
Si Upon CO adsorption the peak at ic changes, in agreement with small but clear shift at lower reinforcement of the hydrogen
I
l
si
Si
l
(terminal)
si
3460 cm" does not undergo dramatthe given assignment. However, a frequency is indicative of a small bond between adjacent OH groups.
676
t
a
a 2
Fig.2. IR spectra of CO adsorbed at 77 K (increasing doses) on Silicalite outgassed at 573 K. a) OH stretching region: the evolution of'the bands is indicated by the arrows; Aindicate the isosbestic points. b) CO region. The spectrum of CO adsorbed on H-ZSM5 (at 77 K and maximum coverage) outgassed at the same temperature is reported for comparison.
This can be easily explained on the basis of the previous scheme: in fact the formation of the 1:l adducts involving terminal OH groups and CO induces the polarization in the terminal 0-H bond and this in turn reinforces the hydrogen bonding between the vicinal groups of the chain. Similar effects have been observed also in homogeneous conditions (ref. 10). The previous assignment is confirmed by the following other experiments: i) by outgassing Silicalite at 973 K the peaks at 3460 and 3710 cm-' simultaneously disappear; ii) the adsorption of CO on samples containing only isolated OH both on silica and silicalite gives the 3750 cm-' peak only. In the part b of the figure the parallel spectra in the CO stretching region is illustrated. Two main absorptions are clearly observed at 2159 and 2137 cm-'. The first peak, with frequency slightly higher than that of the CO gas (a7=16cm-l) , is due to
677
the CO stretch of the OH...CO (1:l) adducts (both isolated and terminal). The second peak (with frequency nearly coincident with that of (CO)liq) is due to CO molecules physically adsorbed (or in a liquid-like state) in the channels and pores. In the same figure (2b) the spectrum of CO on H-ZSM5 (Si02/A1203=28) outgassed at the same temperature is reported for sake of comparison. We can immediately notice that the two type of spectra are very similar but for the peak at 2170 cm-' present only on H-ZSM5: this peak, characterized by a blue shift of 27 cm" with respect to CO gas, is associated with Broensted acid-CO complexes. This comparison shows that CO is a good probe of the strong Broensted acidity of H-ZSM5 and that SiOH groups are much less acidic in agreement with ref. 11. Titanium-Silicalite An identical experiment carried out on TS outgassed at 573 K gives the spectra reported in Fig.3a and 3b (OH and CO stretching region respectively). The great similarity with the spectra of Fig.2 is immediately evident. This observation has two main consequences: i) titanols (either on framework or extra-framework Ti) cannot be easily distinguished from silanols by IR spectroscopy (especially when the Titanium content is small) and ii) titanols (either framework or extraframework) have Broensted acidity not substantially different from silanols. In conclusion, the presence of Ti does not substantially modify the OH distribution and their acidity. In these conditions we can ask wether TiOH groups are really present (specially if we consider that if Ti is totally framework and if the Ti-0-Si bridges are not at least partially hydrolyzed TiOH should be absent). For the time being we cannot answer this question. However, we shall come back again on this problem later on, on the basis of further data. As known from indipendent experiments on the Ti02/C0 system (ref. 12) , the adsorption of CO on microparticles of Ti02 outgassed at temperature t 873 K, gives surface Ti4+cus-C0 (cus: coordinatively unsaturated) species characterized by a well defined high intensity peak at 2179 cm-'. The presence, or the absence, of this peak on properly outgassed TS samples can be so used as a text of the presence or the absence of extraframework Ti in form of Ti02 microparticles. The results of this experiments on a TS sample outgassed in vacuo at 873 K are reported in Fig.4a and 4b. In the same figure the spectrum of CO (CO-stretch-
678
1
t TO'
a
Ii1
n
b
0.5
-z
r 0 3
Y
u
z U
m w 0 VI
U m
3888
3688
3400
YRVCNUHBtR CM-I
3288
380022
8 YRVt"ljHBtR
CH-I
Fig. 3. IR spectra of CO adsorbed at 77 K (increasing doses) on Ti-Sil icalite outgassed at 573 K. a) OH stretching region. The arrows and A have the same meaning as in Fig.2. b) CO region.
ing region) adsorbed at 77 K on TiOZ (anatase) outgassed at the same temperature is reported for sake of comparison. From these figures we can conclude that: i) the bands at 3450 and 3700 cm" disappear simultaneously upon outgassing, in agreement with the given assignment ; ii) isolated silanols (either external or internal) are the main species present on the sample and give the expected OH...CO 1:l complexes characterized by peaks at 3650 cm" (OH) and 2155 cm'l (CO); iii) very low intensity band at ~ 2 1 8 0cm-l reveals that Ti4+cus-C0 adducts are indeed present on the sample outgassed at high temperature. However, comparison with the spectrum of CO adsorbed on Ti02 clearly indicates that the amount of extralattice Ti under form of Ti02 microparticles is negligible. The last observation confirm that the major part of Ti is in framework position and does not have relevant Lewis acidity. We can now go back to the question concerning the presence or
679
a
I!
h '
I
3600
3400 YRVENUHBER
3200
CM-I
31
12250
2200 2150 2100 YRVENUMBER C H - I
2
0
Fig.4. IR spectra of CO adsorbed at 77 K (increasing doses) on Ti-Silicalite outgassed at 873 K. a) OH stretching region. The arrows and A h a v e the same meaning as in the previous figures. b) CO region. The spectrum of CO adsorbed at 77 K (maximum coverage) on pure Ti02 outgassed at the same temperature is reported for comparison.
absence of TiOH groups. First of all we can now say with confidence that TioH or Ti02 microparticles do not play any relevant role in the spectrum of TS (OH stretching region), and secondly that TioH (if present) must comes only from partial hydrolysis of polar Si-0-Ti bridges, as already hypothesized in ref. 5. In order to prove or reject the hypothesis that Si-0-Ti bridges exposed in the channels and cavities are preferential sites for H20 adsorption, the following comparative experiment has been designed. Silicalite and Ti-Silicalite samples were outgassed at 973 K in high vacuum for 4 hrs to eliminate all the hydrogen bonded hydroxyl groups. After this treatment the IR spectra in the OH stretching region were substantially undistinguishable (with only one peak left at 3720 cm-' similar to that shown in Fig.4a). H20 vapuor was then dosed (10 torr) on the sample and
680
left to stand for 15 minuts. After this dehydration step the unreacted H20 was pumped off at 573 K and the hydration state of the surface monitored by recording the IR spectrum in the OH stretching region. The result was as follows: after the thermal treatment Silicalite shows hydrofobic character, as H20 dosage does not restore the original hydroxyl concentration; on the contrary, Ti-Silicalite is slightly more hydrophylic and can be partially rehydrated. We think that this behaviour could be a consequence of the presence of Si-0-Ti bridges which, being more polar, can easily undergo H20 attack giving silanols and titanols in vicinal position. CONCLUSIONS
The IR band observed in the spectrum of TS at 960 cm-', associated with framework Ti, is due to a local impurity stretching mode of the [Ti04] unit in the Silicalite lattice. The small fraction of extralattice Ti in form of Tio2 microparticles does not contribute to the IR spectrum of TS. TiOH and SiOH cannot be distinguished by IR spectroscopy and have very similar acidity (as tested by CO adsorption at 77 K). Moreover, comparison with the spectra of OH...CO adducts on H-ZSM5 shows that their acidity is much lower. Lewis acidity associated with Ti4+ ions is also pratically absent even on samples outgassed at 973 K. REFERENCE8
W. Holderich, M. Messe and F. Naumann, Angew. Chem. Int. Ed. Engl., 27 (1988) 26. C. Neri, A. Esposito, B. Anfossi and F. Buonomo, Eur. Pat. 100 119.
C. Neri, M. Taramasso and F. Buonomo, U.K. Pat. 102 665. J.A. Rabo, Catal. Rev. Sci. Technol., 24 (1982) 202. M.R. Boccuti, K.M. Rao, A. Zecchina, G. Leofanti and G. Petrinil Spectroscopic characterization of Silicalite and Titanium-Silicalite, in: C. Morterra, A. Zecchina and G. Costa (Eds.), Structure and Reactivity of Surfaces, Elsevier, Amsterdam, 1989, pp. 133-144. P. Comba and A . Merbach, Inorg. Chem., 26 (1987) 1325. A. Miecznikowski and J. Hamuza, Zeolites, 7 (1987) 249. M.F. Best and R.A. Condrate, J. Mat. Sci. Letters, 4 (1985) 994. B.G. Varshal, V.N. Denisov, B.N. Maurin, G . A . Paulova, V.B. Podobedov and K.E. Stebin, Opt. Spectrosc. (USSR), 47 (1979) 344. 10 G.C. Pimentel and A.L. McClellan, The hydrogen bond, W.H. Freeman and Co., San Francisco and London, 1960. 11 V.B. Kazanskii, Kinet. Catal., 28 (1987) 482. 12 G. Spoto, C. Morterra, L. Marchese, L. Orio and A. Zecchina, Vacuum, in press.
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
681
THE PROPERTIES OF BORALITES OF V A R I O U S BORON CONTENTS
J . DATKA. A.
CICHCCKI and Z. PIWOWARSKA
F a c u l t y of C h e m i s t r y . J a g i e l l o n i a n U n i v e r s i t y . Karasia 3, 30-060 Krakdw C Poland>
SUMMARY The p r o p e r t i e s of Bog u n i t s and a c i d p r o p e r t i e s of
boralites
w i t h v a r i o u s boron c o n t e n t s w e r e s t u d i e d b y I R s p e c t r o s c o p y . The band of a n t i s y m m e t r i c s t r e t c h i n g v i b r a t i o n s of 9-0 i s s p l i t i n t o t w o m a x i m a . I t s e e m s t o b e d u e t o t h e removal of d e g e n e r a t i o n of t h i s vibration. The i n t e n s i t y of t h e B03 doublet increases l i n e a r l y w i t h t h e b o r o n c o n t e n t b o t h i n N a - and H - b o r a l i t e s . The e x t i n c t i o n c o e f f i c i e n t of 9-0 v i b r a t i o n i s d i s t i n c t l y l h i g h e r i n t h e . . c a s e of H-boralites. The c o n c e n t r a t i o n of 3720 c m hydroxyls CBronsted a c i d s i t e s 3 i n c r e a s e s l i n e a r l y w i t h b o r o n c o n t e n t i n H - b o r a l i t e s and i s close t o t h e t h e o r e t i c a l v a l u e s . S m a l l amounts of L e w i s a c i d sites a r e p r e s e n t i n H-boralites. their acid s t r e n g t h b e i n g h i g h e r t h a n i n HZSM-5. I NTRODUCTI ON Isomorphically studied
substituted
extensively
because
zeolites of
their
have
recently
interesting
been
chenucal
p r o p e r t i e s and p o t e n t i a l i n d u s t r i a l a p p l i c a t i o n [ f o r a r e v i e w see ref. the
C1>1. small
zeolites.
Boron s u b s t i t u t e d z e o l i t e s . s i z e of draw
t r i -coordinated
boron
the
atom
greatest
show
boralites,
properties
attention.
The
i n d e h y d r a t e d b o r a l i tes , t h e B03
t h e spectrum Cref.
2.31. The p r o p e r t i e s of B03
which owing t o
not
observed
boron
atom
in
is
band a p p e a r s i n
units i n boralites
of v a r i o u s b o r o n c o n t e n t s w e r e s t u d i e d b y I R s p e c t r o s c o p y a n d t h e
r e s u l t s a r e described i n t h e present paper. Our p r e v i o u s I R s t u d y C r e f . 3,4> h a s shown t h a t f o u r k i n d s of -1 OH g r o u p s exist i n H - b o r a l i t e s . Only 3720 cm h y d r o x y l s were f o u n d t o b e B r z n s t e d a c i d sites. lower
than
in
zeolites
Cref.
Their
2-51.
B r z n s t e d a c i d s i t e s i n H-boralites
of
acid
The
strength
concentration
of
much the
v a r i o u s boron c o n t e n t s w a s
s t u d i e d by I R s p e c t r o s c o p y and t h e r e s u l t s a r e a l s o t h i s paper.
is
presented i n
682 EXPERIMENTAL Five
samples
of
MFI
were
boralites
synthesized
in
the
N a 0-TPABr-B 0 -SiO -H 0 s y s t e m as d e s c r i b e d i n d e t a i l i n r e f . 7 . 2 2 3 2 2 C o n c e n t r a t e d s i l i c a sol, s o l i d o r t h o b o r i c a c i d H3B03. sodium hydroxide,
and
distilled
tetrepropylammonium
water
bromide
Individual
synthesis
C=Si02/B203
in
the
were
differed
reaction
used
as
TPBABr only
in
mixture).
substrates
as
a
template
the
which
ratio
molar
varied
was
r a n g e 5-100. C r y s t a l l i z a t i o n w a s c a r r i e d o u t i n steel lined
with
at
PTFE.
438
K.
for
days,
7
without
p r e c i p i t a t e w a s washed w i t h d i s t i l l e d w a t e r o r d e r t o decompose o r g a n i c species
mml.
p r e p a r e d are d e n o t e d as Na-boralites. the
NH4-form
at
solution
by
triple
XRD
studies
have
without
autoclaves
stirring.
The In
for
The
that
in
ammonium
compositions analysis)
all
crystalline
Boralites
4 h.
thus
They w e r e n e x t t r a n s f o r m e d
from chemical
shown
RM
the
a n d d r i e d i n air.
exchange
temperature.
room
NH - b o r a l i t e s ( o b t a i n e d 4 T a b l e 1.
crystalline
ion
n
in
boralites were c a l c i n a t e d i n
a i r a t 773 K i n a t h i n l a y e r C l - 2
into
and
substance.
the
chloride and
Na-
of
are presented
were
samples
admixtures.
in
highly
They
had
the
boron
loss
from
or thor hombi c MFI s t r u c t u r 8 . The N a / N H 4 boralites. paper
This
Cref.
exchange
resulted
problem is d i s c u s s e d
The c a l c i n a t i o n
81.
boron
further H-BOR-5.
ion
loss
H-BOR-4.
Cref.
H-BOR-3.
8).
0.69. 0.85 a n d 1 . 0 5 B/u. c. r e s p . for
The
our
previous
w e r e f o u n d t o be 0.88, 0.59,
upon c a l c i n a t i o n i n a i r a t 773 K .
71.
For I R s t u d i e s NH - b o r a l i t e s C4-7
i n
h C s a m p l e numbers a r e t h e s a m e as i n o u r p r e v i o u s p a p e r ,
4
ref.
i n detail
NH - b o r a l i t e s r e s u l t s i n a 4 b o r o n c o n t e n t s i n H-BOR-8.
of
H-BOR-2
some
in
m g cm-').
4 The w a f e r s
vacuum at. 773 K , during
this
for
activation
1 h.
was
were p r e s s e d
w e r e activated It
was
the
u n d e r c o n d i t i o n s described i n r e f .
assumed
same
as
into thin
wafers
in situ i n I R c e l l that
during
the the
boron
in
loss
calcination
8.
T h e s p e c t r a w e r e r e c o r d e d u s i n g a SPECORD 75 I R s p e c t r o m e t e r CCarl
Zeiss J e n a l
working on l i n e w i t h
a K S R 4100 minicomputer.
Pyr i d i ne C PCCh-G1 i w i c e l u s e d w a s o f a n a l y t i c a l g r a d e .
683 TABLE 1 Compositions of N a -
and NH -boralites 4
*
Sample
composition ~~
Na-BOR-8
6’
30H0.10CCB02’0. 3OcAlo2’0. 1OcSi02’Q5.
NH4 -BOR-8
03CNH4’0.37ccB02’0.2QcA102’0. 1lcsi02’S5. 6’
Na-BOR-5 NH 4 -BOR-5
2’
80CCB02’0. 8OcSi02’Q5.
04CNH4’0.69[cB02’0. 63CA102’0.10csi02395.2’
N a -BOR -4
N a lOO[CBo2’l. .
NH4-BOR-4
0’
0OcSi02’Q5.
Na0..02CNH4’0. 84CCB02’0. 78cA10220.0ScSi02’Q5. 0’
N a -BOR -3
N a l30H0. . 10CcB0231. 4OCSio2’Q4.
NH4-BOR-3
6’
01CNH4’0.98[cB02’0. 99csi022Q4. 63
Na-BOR-2
N a l60H1. , 0OCCB02’2.
4’
6OcSi02’Q3.
OlCNH4’1. 36[cB02’1. 37csi022Q3.4’
NH4-BOR-2
*
~~
s a m p l e numbers a r e t h e same a s i n p r e v i o u s p a p e r C r e f .
73
RESULTS AND DISCUSSION BO -3
vi b r a t i ons I R spectra
cm
-1
B-0
of
dehydrated
b o r a l i t e s show a band about 1400
, c h a r a c t e r i s t i c of v3 a n t i s y m m e t r i c s t r e t c h i n g v i b r a t i o n s of T h i s band i s p r e s e n t b o t h i n t h e s p e c t r a of N a -
i n B03 u n i t s .
and H - b o r a l i t e s Cref. 53
that
Cref.
boron
The MAS NMR s t u d i e s h a v e also p r o v e d
2.32.
is
tri-coordinated
in
dehydrated
boralites.
Because of i t s s m a l l s i z e , boron i s n o t s i t u a t e d i n t h e c e n t r e of TO,
t e t r a h e d r o n b u t on ‘ o n e of
its f a c e s .
Our
previous s t u d y has
shown t h a t t h e B03 band i s s p l i t i n t o t w o maxima: 1380 a n d 1 4 0 5 -1
cm
.
The o r i g i n of t h i s s p l i t w i l l be now d i s c u s s e d . interpretations existence
of
will
two
be
kinds
considered. of
boron
One
sites
of in
c h a r a c t e r i z e d by d i f f e r e n t B-0 f o r c e c o n s t a n t . d i f f e r e n t f o r c e s of
them
Two p o s s i b l e assumes
boralite
the
lattice
I t c o u l d be d u e t o
i n t e r a c t i o n between t h e b o r o n a n d t h e f o u r t h
684 o x y g e n i n 0 B---.O u n i t s . A n o t h e r i n t e r p r e t a t i o n a s s u m e s t h e removal 3 o f d e g e n e r a t i o n o f t h e v3 €3-0 v i b r a t i o n i n B03 u n i t s . This c o u l d be d u e t o t h e f a c t t h a t B03 u n i t s a r e n o t i s o l a t e d b u t e a c h o x y g e n
i s bonded that
the
t o a s i l i c o n a t o m of v3
vibration
band
I t s h o u l d be n o t e d
t h e lattice.
in
the
spectra
of
many
crystalline
i n o r g a n i c b o r a t e s i s a l s o s p l i t C r e f . 91. In
to
order
get
information
more
s p l i t t i n g , t h e s p e c t r a of
on
a n d H-boralites
Na-
c o n t e n t s C a c t i v a t e d a t 773 KI w e r e r e c o r d e d . if in
the
lattice.
boralite
one
of
them
p r e f e r e n t i a l l y a t l o w boron c o n t e n t s .
of
the
w i t h v a r i o u s boron
I t w a s supposed t h a t
t w o k i n d s o f b o r o n sites d i f f e r i n g i n B-0 the
origin
force c o n s t a t e x i s t be
should
I n t h i s case,
occupied
the intensity
r a t i o of
b o t h m a x i m a 1380 a n d 1 4 0 5 c m - l
content.
On t h e o t h e r h a n d i f t h e s p l i t t i n g t h e v 3 b a n d w a s d u e t o
should change w i t h boron
t h e d e g e n e r a t i o n removal,
t h e i n t e n s i t y r a t i o is expected n o t
depend
The s p e c t r a of H-boral i tes c o n t a i n i n g :
on boron c o n t e n t . 0.59,
0.28,
1 . 0 5 B/u. c .
0 . 6 9 and
i n t e n s i t y r a t i o of
It
suggests
that
reason f o r t h e s p l i t t i n g of units.
Of
course t h e
i n
Fig.
The
1.
bands p r a c t i c a l l y does n o t
T h e s a m e w a s a l s o o b s e r v e d i n t h e case of
depend on b o r o n c o n t e n t .
Na-boralites.
are presented
1380 a n d 1 4 0 5 c m - l
to
the
degeneration
t h e v3 b a n d o f
possibility
d i f f e r i n g i n B-0 f o r c e c o n s t a n t
that but
is
the
€3-0 s t r e t c h i n g i n
removal
B03
t w o kinds
without
of
boron
sites
preference i n their
o c c u p a t i o n c a n n o t b e t o t a l 1y e x c l u d e d . Fig.
2
shows
the
plots
of
intensities
d o u b l e t as a f u n c t i o n of boron c o n t e n t i n N a t h e case o f
H-boralites.
t h e loss of
of
1380,
1405 cm-l
a n d H-boralites.
In
boron d u r i n g t h e a c t i v a t i o n
o f NH -form w a s t a k e n i n t o a c c o u n t . B o t h p l o t s a r e l i n e a r . I n t h e 4 case of H - b o r a l i t e s t h e s l o p e i s much h i g h e r t h a n i n N a - b o r a l i t e s .
M o s t probably t h i s is due t o t h e higher e x t i n c t i o n c o e f f i c i e n t of t h e v3 B - 0 s t r e t c h i n g Cand h e n c e t o h i g h e r 8-0 bond p o l a r i z a t i o n > . We
can
however
Na-boralites
exclude a possibility
not
are n o t
i n B03
units
and i t
that
some
B atoms i n
could r e s u l t
in a low
i n t e n s i t y o f t h e B 0 3 band.
Our p r e v i o u s s t u d y C r e f . boron
content
results
in
the
101 h a s shown t h a t t h e i n c r e a s e i n contraction
of
the
unit
cell
of
b o r a l i t e s ( b o r o n i s s m a l l e r t h a n s i l i c o n > . T h e force c o n s t a n t s o f Si-0 s t r e t c h i n g i n
content.
boralites
w e r e f o u n d t o be i n d e p e n d e n t
of
B
T h i s l a s t e f f e c t i s d i s c u s s e d i n t e r m s of t h e c o l l e c t i v e
685
Fig. 1. The B-0 band in t h e spectra of *H-boralites of various B contents: a - 0.28, b - 0.59, c - 0.69, d - 1 . 0 B1u.c.
686
7 c
I
5 -
er, \
E \
In 0
.
2
3 -
-
0
(0
$ a
1 2
1
3 D I U.C.
F i g . 2. - The i n t e n s i t i e s of B - 0 bands i n t h e s p e c t r a of: N a - b o r a l i t e s Col and H - b o r a l i t e s Col as a f u n c t i o n of B c o n t e n t .
-P
/
0 .-
theoret.
U
Cl 0
a
c
v)
C :0
115;
-,,
LewisL
1.o (B+Al-Na)/u.C. F i g . 3. - The c o n c e n t r a t i o n s of Brgnsted a c i d sites COD, t h e o r e t i c a l values of c o n c e n t r a t i o n of Brgnsted sites Co> and t h e i n t e n s i t y of PyL band i n H - B o r a l i t e s a s a f u n c t i o n of boron c o n t e n t Co3
model
zeolites
of
Cthe
of
electronegativities
and
Si
are
B
practically the same>. Aci d proper ti es of H-bor a1 i tes N a - b o r a l i t e c o n t a i n s o n l y 3740 c m to
analogous
those
in
zeolites.
d e c o m p o s i t i o n of NH - f o r m CB-OH>.
OH g r o u p s
3460 cm-'
The
CtermFnal Si-OH> ion
NafiH4
Cref.
3.4):
CSi-OH.--03.
3720 cm-'
Only 3720
cm
groups,
exchange
r e s u l t s i n t h e a p p e a r a n c e of
4
k i n d s of
-1
and
t h r e e new
CSi-OH---B>. 3680 cm-' -1 hydroxyls w e r e found
t o be B r o n s t e d a c i d sites. T h e i r a c i d s t r e n g t h
i s much l o w e r t h a n
i n z e o l i t e s C r e f . 2-5) e v e n though t h e e l e c t r o n e g a t i v i t y of
boron
t h a n t h a t of aluminum C 2 . 9 3 a n d 2,22 r e s p . > . I t c a n be
is higher
e x p l a i n e d by w e a k i n t e r a c t i o n between t h e OH g r o u p a n d s m a l l b o r o n a t o m i n t h e Si-OH-.--B u n i t s .
The c o n c e n t r a t i o n of B r o n s t e d a c i d sites C3720 cm-' in
H-boralites
various
of
pyridine sorption.
boron
contents
P y r i d i n e molecules
+
react
determined
with
BrGnsted
sites f o r m i n g p y r i d i n i u m i o n s CPyH 3 f o r which 1 5 4 5 cm-' is
characteristic.
H-boralites band.
portion
Small
of
Si -OH-.-B>
was
pyridine
was
by acid
I R band sorbed
in
a t 430 K up t o t h e c o n s t a n t i n t e n s i t y of t h e 1 5 4 5 cm-'
The c o n c e n t r a t i o n of B r z n s t e d a c i d sites w a s c a l c u l a t e d f r o m t h e 1 5 4 5 cm-'
t h e m a x i m a l i n t e n s i t y of coefficient
determined
c o n c e n t r a t i o n of i n Fig.
in
a
previous
band a n d i t s e x t i n c t i o n study
Cref.
B r z n s t e d a c i d sites i n H-boralites
3 a s a f u n c t i o n of
The
is p r e s e n t e d
boron c o n t e n t C t h e loss of
t a k e n i n t o a c c o u n t > . The t h e o r e t i c a l v a l u e s :
11>.
boron is
t h e c o n c e n t r a t i o n s of
B + A1 - N a ( o u r b o r a l i t e s c o n t a i n e d small amounts of Al> are also p r e s e n t e d i n t h e same f i g u r e . The c o n c e n t r a t i o n of BrGnsted sites in
H-boralite
increases
linearly
with
the
e x p e r i m e n t a l v a l u e s are close t o t h e o r e t i c a l
boron
content.
The
ones. Small d e f i c i t
of B r z n s t e d sites may be d u e t o s o m e d e h y d r o x y l a t i o n p a r t i c u l a r l y -1 band of p y r i d i n e bonded t o Lewis a c i d sites s i n c e a w e a k 1460 c m
CPyL3 i s p r e s e n t i n t h e s p e c t r u m of p y r i d i n e s o r b e d i n b o r a l i t e s . The i n t e n s i t y of
t h e 1 4 6 0 cm-'
PyL band ( F i g .
2) i s l o w a n d
p r a c t i c a l l y t h e same i n a l l t h e b o r a l i t e s s t u d i e d i n d i c a t i n g t h a t t h e d e g r e e of content.
d e h y d r o x y l a t i o n i s small a n d i n d e p e n d e n t
The f r e q u e n c y of
h i g h e r t h a n i n t h e case of
PyL band
of
i n b o r a l i t e s C1460 cm-'>
234-5 z e o l i t e s C1450
ern-',
ref.
boron is
11).
T h i s means t h a t t h e e l e c t r o a c c e p t o r p r o p e r t i e s of Lewis a c i d sites
688 i n H - b o r a l i t e s are s t r o n g e r t h a n i n H-ZSM-5.
Cref 3. 12.131. I t i s
p r o b a b l y d u e t o s m a l l s i z e and h i g h e l e c t r o n e g a t i v i t y of b o r o n .
The
authors
thank
Professor
Jerzy
Haber
of
Institute
C a t a l y s i s a n d S u r f a c e C h e m i s t r y of t h e P o l i s h Academy of
of
Sciences
i n Krakdw f o r a g r a n t t h a t made t h e p r e s e n t w o r k p o s s i b l e .
REFERENCES 1 2 3
4
5
6 7 8 9
10 I1 12 13
M. T i e l e n . M. G a l l e n a n d P. A. J a c o b s , P r o c . I n t . Symp. Z e o l i t e C a t a l y s i s , , S i o f o k , Hungary, May 1985. p. 1. G. C o u d u r i e r , J . C. V e d r i n e . P u r e Apl. Chem. , 58. C1986>, 1389. J. D a t k a and Z . Piwowarska. J . Chem. Soc. , F a r a d a y T r a n s . I , 85, C1989>, 47. J . D a t k a and Z. Piwowarska. J . Chem. Soc., F a r a d a y T r a n s . I . 85, C19893, 837. K. F.M.G. J . S c h o l l e . A.P.M. Kentgens. W . S . Veeman, P. F r e n k e n a n d G. P. M. van d e r Velden. J . Phys. Chem. , 88. C19843, 5. P. Ratnasamy, S.G. Hegde a n d A. J . Chandwalkar, J . C a t a l . , 102, 1986. 467. A. C i c h o c k i , M. M i c h a l i k , M. Buf a n d J . P a r a s i e w i c z Kaczmarska, Zeolites. 10. (19893. i n p r i n t . A. C i c h o c k i . M. M i c h a l i k . M. BUS. W. Lasocha a n d Z. Saw&owicz, Zeolites. 10. Cl989>, i n p r i n t . C.W. Weir and R . A . S c h r o e d e r . J o u r n a l of R e s e a r c h of t h e N a t i o n a l Bureau of S t a n d a r d s A. 6 8 A C 5 > , 465. 1964. A. Cichocki. J. Datka, M. M i c h a l i k . A. O l e c h , a n d Z . Piwowarska. J . Chem. Soc., F a r a d a y T r a n s a c t i o n s 1 . 86C4). c19903. 753. J . Datka and E. TuPnik, J . C a t a l . , 102. C19863. 43. J . W . Ward, J. C a t a l . , 10. C19681, 34. J . W. Ward, J . C o l l . I n t e r f a c e Ski. , 28, ClQ683. 269.
G . Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
689
AN ELECTROSTATIC MODEL TO PREDICT THE INFRA RED CHARACTERISTICS OF ZEOLITE
HYDROXYL GROUPS AFTER ADSORPTION OF AROMATICS P a t r i c k J . O'Malley, Department o f Chemistry, UMIST, Manchester M60 l Q D U.K SUMMARY
The p e r t u r b a t i o n o f t h e h i g h frequency (h.f.1, h y d r o x y l s t r e t c h i n g i n f r a r e d band o f a HY z e o l i t e i s i n v e s t i g a t e d a f t e r a d s o r p t i o n o f benzene, t o l u e n e , o-xylene, p-xylene, m-xylene and cumene. A s h i f t o f t h e band t o a lower frequency accompanied by an i n c r e a s e i n i n t e n s i t y i s noted. The f r e q u e n c y s h i f t v a r i e s i n t h e o r d e r , o-xylene > m-xylene = U t i l i s i n g a t h e o r e t i c a l model f o r t o l u e n e = cumene > benzene = p-xylene. t h e i n t e r a c t i o n which t a k e s account o f o n l y t h e d i e l e c t r i c p r o p e r t i e s o f t h e h y d r o x y l group environment i t i s demonstrated t h a t t h e v i b r a t i o n a l p r o p e r t i e s p r e d i c t e d by such a model i . e . f r e q u e n c y s h i f t and i n t e n s i t y change a r e i n c l o s e agreement w i t h e x p e r i m e n t a l l y determined q u a n t i t i e s . INTRODUCTION
The i n t e r a c t i o n o f aromatic molecules w i t h t h e h y d r o x y l groups o f z e o l i t e s r e s u l t s i n a decreased frequency and an i n c r e a s e d i n t e n s i t y f o r t h e h y d r o x y l i n f r a r e d band.
The e x t e n t o f p e r t u r b a t i o n o f t h e h y d r o x y l
band i s o f t e n r e l a t e d t o t h e a c i d s t r e n g t h (1,21. I n t h i s s t u d y a s e r i e s o f aromatic molecules i . e . o-xylene,
p-xylene,
benzene, t o l u e n e ,
m-xylene and cumene a r e absorbed on a 100% exchanged
HY sample and t h e i r e f f e c t on t h e h i g h f r e q u e n c y ( h . f . )
monitored. h.f.
h y d r o x y l group i s
Both t h e decrease i n frequency and i n c r e a s e d i n t e n s i t y o f t h e
band a r e m o n i t o r e d and an e l e c t r o s t a t i c model, due o r i g i n a l l y t o
Coggeshall ( 3 1 , i s u t i l i s e d t o p r e d i c t t h e e x p e r i m e n t a l l y observed i n f r a red properties.
EX PER I MENTAL D e t a i l s o f t h e sample p r e p a r a t i o n and i n f r a r e d s t u d i e s a r e as described previously (4).
100% exchange o f Na by NH4 (NH4Y-100) was
achieved by t r e a t i n g t h e sample w i t h c o n c e n t r a t e d NH&l c o n d i t i o n s f o r s e v e r a l weeks. HY-100.
under r e f l u x
Subsequent c a l c i n a t i o n g i v e s r i s e t o
For benzene, t o l u e n e and o-xylene s a t u r a t i o n o f t h e a d s o r p t i o n
s i t e s was achieved r a p i d l y , however f o r p-xylene, m-xylene and cumene t h e o r g a n i c vapour had t o be l e f t i n c o n t a c t w i t h t h e z e o l i t e d i s c f o r 1 hour f o r reasonable coverages t o be a t t a i n e d .
690
:I
\
::
z
s
p
v1
B bp
\J
3600
3400
3200
~/c,'~
F i g 1 . E f f e c t o f a d s o r D t i o n on t h e hydroxyl s t r e t c h i n g ' region of HY-100. ( a ) HY-100, ( b ) HY-1001 benzene, ( c ) HY-lOO/toluene, ( d ) HY-lOO/o-xylene.
3600
3400 3200
p/c In-i F i g . 2. E f f e c t of a d s o r p t i o n on t h e hydroxyl s t r e t c h i n g x g i o n o f HY-100. ( a ) HY-100, ( b ) rlY-lOO/ p-xylene, ( c ) HY-lOO/m-xylene, ( d ) HY-lOO/cumene.
691 RESULTS AND DISCUSSION The e f f e c t o f a d s o r p t i o n o f benzene, t o l u e n e and o-xylene on t h e h y d r o x y l s t r e t c h i n g bands o f HY-100 i s i l l u s t r a t e d i n f i g . 1 w h i l e f i g . 2 shows t h e e f f e c t o f a d s o r p t i o n o f p-xylene,
m-xylene and cumene.
In all
cases, a d s o r p t i o n leads t o a s h i f t t o lower f r e q u e n c i e s o f t h e h . f . band and a l s o causes a c o n s i d e r a b l e broadening and i n c r e a s e i n i n t e n s i t y of t h i s band.
I n a l l cases, a d s o r p t i o n does n o t l e a d t o p e r t u r b a t i o n o f t h e
1 . f . band which i s s i m i l a r t o t h a t found by p r e v i o u s workers f o r a d s o r p t i o n o f r a r e gases and alkane molecules (5).
T h i s can be e x p l a i n e d
by t h e i n a c c e s s i b l e p o s i t i o n s o f t h e h y d r o x y l groups g i v i n g r i s e t o t h i s band.
The magnitude o f t h e s h i f t o f t h e h . f . band t o lower f r e q u e n c i e s
decreases i n t h e o r d e r o-xylene > m-xylene = t o l u e n e = cumene > benzene = p-xylene (see Table 1 ) . The i n c r e a s e i n i n t e n s i t y and s h i f t t o l o w e r f r e q u e n c i e s on a d s o r p t i o n can be a t t r i b u t e d as r e s u l t i n g f r o m e l e c t r o s t a t i c i n t e r a c t i o n s o f t h e OH d i p o l e w i t h t h e adsorbed molecules.
T h i s r e s u l t s i n a displacement o f t h e
e l e c t r i c a l charge i n t h e s e molecules, and as a r e s u l t , t o an i n c r e a s e i n t h e d i p o l e moment as w e l l as a decrease i n t h e f o r c e c o n s t a n t o f t h e OH bond.
Coggeshall (3) has g i v e n a t h e o r e t i c a l s o l u t i o n t o t h e problem of
an o s c i l l a t i n g d i p o l e p e r t u r b e d by an e l e c t r o s t a t i c f i e l d .
This theory
was subsequently extended by White e t a l . (5) and used q u i t e s u c c e s s f u l l y i n p r e d i c t i n g t h e s h i f t s and changes i n i n t e n s i t y undergone b y t h e h i g h frequency OH s t r e t c h i n g band when t h e OH o s c i l l a t o r i n t h e supercage was surrounded by p o l a r i s a b l e adsorbed molecules.
I n order t o deal w i t h t h e
problem o f an OH group p e r t u r b e d by an e l e c t r o s t a t i c f i e l d Coggeshall has adopted a s i m p l i f i e d f o r c e f i e l d s i t u a t i o n and assumed t h a t t h e h y d r o x y l group can be c o n s i d e r e d as a d i a t o m i c group possessing a d i p o l e moment immersed i n a c o n s t a n t e l e c t r o s t a t i c f i e l d .
T h i s w i l l r e s u l t i n an
i n t e r a c t i o n energy between t h e d i p o l e and t h e e l e c t r i c f i e l d which w i l l depend on t h e d i p o l e moment, t h e angle o f o r i e n t a t i o n and t h e e l e c t r i c f i e l d strength.
Since a l l t h e v i b r a t i o n a l energy terms w i l l be s h i f t e d b y
t h e same amount, t h e r e w i l l be no change i n t h e t r a n s i t i o n e n e r g i e s between t h e v i b r a t i o n a l s t a t e s as a r e s u l t o f t h i s i n t e r a c t i o n energy term.
There w i l l , however, be a change i n t h e p o t e n t i a l energy o f t h e
d i a t o m i c group due t o p o l a r i s a t i o n o f t h e OH bond by t h e e l e c t r i c f i e l d . T h i s w i l l be g i v e n by -qE z where q i s t h e charge unbalance f o r t h e group, P i s t h e e l e c t r i c f i e l d component p a r a l l e l t o t h e d i r e c t i o n o f t h e P valence bond and z i s t h e i n t e r n u c l e a r displacement f r o m e q u i l i b r i u m .
E
F o l l o w i n g r e f . (3) t h i s leads t o an e x p r e s s i o n f o r t h e p o l a r i s a t i o n f o r c e
Q,
N W
TABLE 1 Observed and t h e o r e t i c a l p e r t u r b a t i o n s o f h . f . z e o l i t e hydroxyi band
Adsorb a t e
Benzene To1 uene 0- Xy 1e ne m-Xy 1e ne p-Xylene Cumene
t
'obs.
2.274 2.379 2.568 2.374 2.270 2.380
310 350 380 350 300 340
Ap2 pl
Elql
5.33 6.88 8.82
3.729 4.070 4.330 -
-
-
-
2
6 -
42
P1
41
1.093 1 .lo6 1.128 -
-
2.31 2.57 2.95 -
1.61 1.58 1.47 -
5.33 6.64 8.73 -
-
-
-
-
E
M't
P E
2.37 2.54 2.91 2.53 2.35 2.54
1.49 1.51 1.55 1.51 1.48 1.51
(
",
5.61 6.45 8.47 -
-
''talc.
300 330 410 330 295 330
693 qE as qE P = [I - $(v/uo) - v / u o ~ + l / 1 6 ( u / v o ~ ) D
P
(1)
w D ) i
where
a
=
I f t h e wave number o f t h e f r e e OH v i b r a t i o n i s 3750 cm-’ d i s s o c i a t i o n energy ( D ) o f t h e OH bond i s 461.077 kJmol-’,
and t h e then t h e
Equation 1 P a l s o enables us t o c a l c u l a t e t h e v a r i a t i o n i n t h e p o l a r i s a t i o n f o r c e (Eq)
p o l a r i s a t i o n f o r c e (qE
)
can be deduced f r o m t h e r a t i o v/uo.
on t h e hydrogen atom o f t h e OH o s c i l l a t o r b r o u g h t about by t h e p h y s i c a l a d s o r p t i o n of gases and vapours and i t can be expressed as
where E2 and E l a r e t h e f i e l d s a t t h e d i p o l e , i n t h e presence and t h e absence of adsorbed molecules r e s p e c t i v e l y and u2 and v1 a r e t h e c o r r e s p o n d i n g wave numbers. F o l l o w i n g t h e methods adopted by White e t a l . (5) a model f o r t h e i n t e r a c t i o n o f t h e OH group w i t h adsorbed a r o m a t i c s can be proposed which t r e a t s t h e h y d r o x y l group as a p o l a r i s a b l e d i p o l e which i s p l a c e d a t t h e c e n t e r o f an e l l i p s o i d a l c a v i t y .
T h i s i n t u r n i s immersed i n a c o n t i n o u s
d i e l e c t r i c of d i e l e c t r i c c o n s t a n t
which i s a c t e d on by a homogeneous
electrostatic f i e l d E parallel t o the a axis of the e l l i p s o i d .
Because of
t h e p a r t i c u l a r geometry o f t h e OH group, t h e e l l i p s o i d adopted was c o n s i d e r a b l y f l a t t e n e d i n t h e d i r e c t i o n o f t h e f i e l d and had dimensions o f i n t h e d i r e c t i o n o f t h e f i e l d and a l e n g t h of 2.486
0.994
perpendicular t o t h e f i e l d .
a
The former v a l u e i s a p p r o x i m a t e l y t h e l e n g t h
o f t h e O H bond w h i l e t h e l a t t e r i s r o u g h l y t h e d i a m e t e r o f t h e oxygen atom i n silicates.
The i n c r e a s e b o t h i n t h e d i p o l e and i n t h e e l e c t r i c f i e l d
a t t h e d i p o l e i n t h e presence o f t h e d i e l e c t r i c a r e subsequently g i v e n ( 5 ) by MS/M =
Eh/E
=
(3)
(1.397+<)/(3.057-0.66&)
(4)
/(0.583+0.4176)
where M and
E a r e t h e d i p o l e moment and t h e e l e c t r i c f i e l d r e s p e c t i v e l y i n
t h e absence o f t h e d i e l e c t r i c and
MX
and Eh a r e t h e d i p o l e moment and t h e
e l e c t r i c f i e l d i n t h e presence o f t h e d i e l e c t r i c .
T h i s model f i t s t h e
r e p r e s e n t a t i o n o f t h e phenomena o r i g i n a t i n g f r o m t h e a d s o r p t i o n o f
694
molecules h a v i n g d i e l e c t r i c c o n s t a n t 12 i n t h e v i c i n i t y o f t h e OH d i p o l e . B e f o r e a d s o r p t i o n , t h i s d i p o l e i s surrounded by a vacuum w i t h t = 1.0 and M*/M and Eh/E a l s o equal t o 1.0.
The i n c r e a s e d i n t e n s i t y o f t h e p e r t u r b e d band can a l s o be s i m u l a t e d . The r a t i o o f t h e absorbance o f t h e p e r t u r b e d ( A bands can be shown ( 3 , 5 ) t o be
A /A P
= U
P
)
and u n p e r t u r b e d (Au) OH
v / v (q /qUl2r O
where
P
2
( ‘ = [C(v/v,) - 31 ( C - 2 ) c [v/vo) - 2 I L (C - 3)
-
S i n c e f does n o t change v e r y much i n t h e u s u a l u/uo range, t h e i n c r e a s e i n i n t e n s i t y observed f o r t h e d h i f t e d h i g h f r e q u e n c y OH band i s t h u s almost s o l e l y due t o an i n c r e a s e of t h e e f f e c t i v e charge gained by t h e hydrogen atom.
E q u a t i o n 5 enables us t o c a l c u l a t e t h e r a t i o s o f t h e
e f f e c t i v e charges o f t h e
OH o s c i l l a t o r p e r t u r b e d by t h e e l e c t r o s t a t i c
f i e l d e i t h e r i n t h e absence o r t h e presence o f adsorbed molecules. and A
I f Ap2
a r e t h e absorbance o f t h e h i g h f r e q u e n c y OH band i n t h e presence
Pl o r absence o f adsorbed molecules r e s p e c t i v e l y , t h e n
From e q u a t i o n s 2 and 6, t h e r a t i o s qp2/qp,
and E2/E1 can be c a l c u l a t e d
f r o m t h e r e l a t i v e i n t e n s i t i e s and t h e observed f r e q u e n c y s h i f t s and t h e s e can be compared w i t h t h e t h e o r e t i c a l M”/M and Eh/E v a l u e s .
Also, f r o m t h e
t h e o r e t i c a l M*/M and Eh/E values, t h e r a t i o
may be c a l c u l a t e d f o r any g i v e n adsorbate m o l e c u l e .
3640 cm-l band f r o m e q u a t i o n ( 1 ) i s 0.66 x lO-’N,
S i n c e qplEl
f o r the
t h e q u a n t i t y qp2E2 can
be c a l c u l a t e d and t h u s t h e f r e q u e n c y s h i f t f o r a g i v e n adsorbate can be predicted.
The i n c r e a s e i n i n t e n s i t y o f t h e s h i f t e d band may a l s o b e
p r e d i c t e d as b e i n g r o u g h l y
Table 1 g i v e s t h e observed f r e q u e n c y s h i f t s a f t e r a d s o r p t i o n o f benzene, t o l u e n e , o-xylene, m-xylene,
p-xylene and cumene.
As was
695 mentioned i n t h e e x p e r i m e n t a l s e c t i o n , i t was found necessary t o l e a v e p-xylene, m-xylene and cumene i n c o n t a c t w i t h t h e z e o l i t e f o r 1 hour i n o r d e r t o e f f e c t an a p p r e c i a b l e p e r t u r b a t i o n o f t h e h . f .
band.
Thus, as
comparison o f i n t e n s i t i e s depends on s a t u r a t i o n o f t h e z e o l i t e supercage w i t h t h e adsorbate molecules, and a l s o because t h e t y p e o f s p e c t r a o b t a i n e d d i d n o t l e n d t o a c c u r a t e i n t e n s i t y measurements, i t was decided t o c a l c u l a t e t h e change i n i n t e n s i t y undergone by t h e h . f . benzene, t o l u e n e and o-xylene a d s o r p t i o n o n l y .
band f o r
Also i n c l u d e d i n t h i s
and t h e t a b l e , where a p p r o p r i a t e , a r e t h e q2/q1, E2/E1, (q2/ql) 2 values f o r t h e v a r i o u s adsorbates M”/M and (M“/M)
t h e o r e t i c a l Eh/E,
c a l c u l a t e d u s i n g e q u a t i o n s 1, 2, 3, 4 and6.
It i s c l e a r f r o m
t h e t a b l e t h a t t h e c a l c u l a t e d and t h e o r e t i c a l frequency and r e l a t i v e i n t e n s i t y changes agree q u i t e w e l l , t h u s g i v i n g s u p p o r t t o t h e c o n c l u s i o n t h a t i t i s t h e d i e l e c t r i c p r o p e r t i e s i n t h e v i c i n i t y o f t h e OH group t h a t a r e r e s p o n s i b l e f o r t h e p e r t u r b a t i o n of t h e h . f . band. REFERENCES 1 2 3 4 5
P.A. Jacobs, C a t a l . Rev. S c i . Eng., 24, 1982, 215-225. J. Dwyer and P.J. O’Malley i n S . K a l i g u i n e (Ed.), ’Keynotes i n Energy R e l a t e d C a t a l y s i s , E l s e v i e r , Amsterdam, 1988, pp. 5-55. N.D. Coggeshall, J. Chem. Phys., 18, (1950) 978-985. P.J. D’Malley, Ph.D. Thesis, N a t i o n a l U n i v e r s i t y o f I r e l a n d , 1981. J.L. White, A.N. J e l l i , J.M. Andre and J.J. F r i p i a t , Trans. Faraday SOC., 63, 1967, 461.
This Page Intentionally Left Blank
697
G. Ohlmann et al. (Editors), Catalysis and Adsorption by Zeolites 0 1991 Elsevier Science Publishers B.V., Amsterdam
THE A C I D I T Y OF ZEOLITES ON THE
THE INFLUENCE OF CARBENIUM I O N S 'I.K i r i c s i , 'H.
FORMATION
OF
UNSATURATED
F o r s t e r , 'G. Tasi and ' P . F e j e s
' A p p l i e d Chemistry Department, J o z s e f A t t i l a U n i v e r s i t y , H-6723 Szeged R e r r i c h Bela t e r l . , Hungary, ' I n s t i t u t e o f P h y s i c a l Chemistry, U n i v e r s i t y o f Hamburg, Bundesstrasse 45, D-2000 Hamburg 13, Federal R e p u b l i c o f Germany.
SUMMARY The i n f l u e n c e o f t h e a c i d i t y o f z e o l i t e s on t h e f o r m a t i o n o f u n s a t u r a t e d carbenium i o n s from propene and a l l e n e was i n v e s t i g a t e d by combined U V - V I S - I R spectroscopy. The h i g h e r t h e B/L a c i d i t y r a t i o t h e l o w e r t h e frequency o f t h e a l k e n y l carbenium i o n s and t h e l e s s i n t e n s e t h e g e n e r a t i o n o f monoenyl i o n s from b o t h a l l e n e and propene o v e r z e o l i t e s HNaY-FAU and HNa-ZSM-5. INTRODUCTION The
various
a d s o r p t i o n o f o l e f i n s on z e o l i t e s r e s u l t s i n t h e f o r m a t i o n o f
s u r f a c e species.
G e n e r a l l y t h e s t r e n g t h o f t h e i n t e r a c t i o n between t h e a c t i v e
s i t e s and t h e o l e f i n molecules can be c h a r a c t e r i z e d by t h e frequency s h i f t
of
t h e C=C double bond fundamental. On e s s e n t i a l l y n o n a c i d i c z e o l i t e s ,
e.g. z e o l i t e A exchanged w i t h a l k a l i n e ,
a l k a l i n e e a r t h o r t r a n s i t i o n metal i o n s ,
t h e s h i f t o f t h e C=C s t r e t c h
t o be p r o p o r t i o n a l t o t h e " p o l a r i z i n g power",
proved
i . e . the charge/radius r a t i o o f
t h e c a t i o n s f o r butenes ( r e f . l ) , propene ( r e f . 2) and a l l e n e ( r e f . 3 ) . On
acidic
z e o l i t e s the adsorption o f o l e f i n s leads t o t h e
s a t u r a t e d as w e l l as u n s a t u r a t e d carbenium i o n s ,
generation
b e i n g proven w i t h
of
different
e x p e r i m e n t a l techniques ( r e f . 4 ) . A c i d i c z e o l i t e s g e n e r a l l y c o n t a i n Bronsted and Lewis a c i d s i t e s , which b o t h a r e i n v o l v e d i n t h e f o r m a t i o n o f c a r b o c a t i o n s f r o m hydrocarbons. B r o n s t e d a c i d c e n t e r s may be r e s p o n s i b l e f o r t h e g e n e r a t i o n o f carbonium i o n s f r o m p a r a f f i n s (ref. ions
5),
a l k y l carbenium i o n s f r o m o l e f i n s ( r e f .
from d i e n s ( r e f .
7),
6 ) and a l k e n y l
carbenium
w h i l e on Lewis a c i d s i t e s t h e f o r m a t i o n o f
carbenium i o n s from p a r a f f i n s ( r e f .
alkyl
8 ) and a l k e n y l i o n s f r o m o l e f i n s ( r e f . 9 )
may o c c u r . Although is
t h i s generalization o f the formation o f carbocations i n
w i d e l y accepted,
some i m p o r t a n t d e t a i l s c o n c e r n i n g t h e i n f l u e n c e
zeolites of
r a t i o o f t h e Bronsted t o Lewis a c i d i t y (B/L) and/or t h e c o n c e n t r a t i o n o f
the acid
698
sites
on t h e g e n e r a t i o n o f carbenium i o n s have n o t been r e v e a l e d y e t .
For
b e t t e r u n d e r s t a n d i n g o f t h e r o l e o f B r o n s t e d and Lewis a c i d i t y i n t h e case
a of
t r a n s f o r m a t i o n o f u n s a t u r a t e d hydrocarbons o v e r z e o l i t e s a combined I R and UVV I S s p e c t r o s c o p i c s t u d y was c a r r i e d o u t , t h e r e s u l t s o f w h i c h a r e p r e s e n t e d i n this
paper.
Attention
w i l l be focussed on t h e
generation
of
alkenyl-type
carbenium i o n s o r i g i n a t e d f r o m propene and a l l e n e . EXPERIMENTAL HNaY-FAU and HNaZSM-5 z e o l i t e s were used i n t h e e x p e r i m e n t s .
Their
parent
m a t e r i a l s , NH4NaY-FAU and NHqNaZSM-5, were p r e p a r e d by i o n exchange f r o m t h e i r sodium
forms,
c h a r a c t e r i z e d by XRD,
well-crystalline.
K B r - m a t r i x I R and TG,
and f o u n d t o
be
The u n i t c e l l c o m p o s i t i o n s determined by AAS a r e l i s t e d
in
Table 1. F o r t h e s p e c t r o s c o p i c measurements s e l f - s u p p o r t i n g w a f e r s were p r e s s e d f r o m the
ammonium forms
different
of
temperatures
equipped w i t h
t h e samples, (673,
773,
deammonated 873 K ) i n t h e
and
outgassed
respective
e i t h e r q u a r t z o r NaCl windows i n o r d e r t o
at
three
optical
obtain
cell
samples
of
d i f f e r e n t amounts o f B r o n s t e d and Lewis a c i d s i t e s . The a c i d i t y o f t h e samples was
determined
by
I R spectroscopy u s i n g p y r i d i n e
as
probe
molecule
(for
d e t a i l s see r e f . l o ) , and l i s t e d i n Table 1. The I R s p e c t r a were r e c o r d e d on a P e r k i n - E l m e r 225 s p e c t r o m e t e r , experiments were performed on a Cary 17 s p e c t r o m e t e r
UV-VIS
w h i l e the
operating
under
computer c o n t r o l . Quantum chemical c a l c u l a t i o n s were c a r r i e d o u t u s i n g system
developed f o r microcomputers ( r e f .
CNDO/S
(ref.
13)
methods
o p t i m i z e d w i t h i n t h e MNDO a p p r o x i m a t i o n .
PcMol
program
(ref.
were a p p l i e d f o r t h e c a l c u l a t i o n
v e r t i c a l s i n g l e t e x c i t a t i o n e n e r g i e s (LVSEE). were
the
The MNDO-HE
11).
of
12)
the
The g e o m e t r i e s o f
and
lowest
carbocations
For calculation o f the
UV-VIS
s p e c t r a i n t h e case o f a l l y 1 c a t i o n 30 and i n t h e case o f a l l o t h e r
carbenium
ions
the
90
monoexcited c o n f i g u r a t i o n s were taken i n t o
account
in
con-
figuration interaction. RESULTS Fig.
1 shows t h e U V - V I S s p e c t r a o f propene adsorbed on sample No. 6. A f t e r
i n t r o d u c t i o n o f 1.33 kPa propene i n t o t h e c e l l a band was observed a t 318 nm. At
room
temperature t h e p o s i t i o n o f t h i s band was s h i f t e d
to
higher
wave-
l e n g t h s w i t h i n c r e a s i n g c o n t a c t t i m e . Heat t r e a t m e n t a t 323 K f o r 1 h r e s u l t e d in
the
development
o f a new band a t 370 nm
and
simultaneously
frequency
(HF) band s h i f t e d t o 323 nm (see spectrum 4 ) .
298,
and
373
473
K t h e i n t e n s i t i e s o f t h e HF
bands
Upon
the
high-
evacuation
decreased
and
at
their
699
frequencies
were b l u e - s h i f t e d t o 321,
314 and 298
nm,
respectively.
After
e v a c u a t i o n a t 573 K o n l y t h e background remained. Table 1. S p e c i f i c a t i o n s o f t h e z e o l i t e samples used
Sample
U n i t c e l l compositions
Zeolite
No.
Concentrations o f acid s i t e s i n mol /kg Bronsted Lewis
B/L
1
HNaY/673
(NH4)42 Na16
0.441
0.082
5.4
2
HNaY/773
(NH4)42 Na16
0.058
0.305
0.2
0.007
0.156
0.04
HNaZSM-5/673
(NH4)1.2 Na0.06 ZSM-5
0.033
0.008
4.0
5
HNaZSM-5/773
( N H 4 ) i a 2 Na0.06 ZSM-5
0.028
0.013
2.2
6
HNaZSM-5/873
(NH4)1.2 Na0.06 ZSM-5
0.018
0.015
1.2
3
(NH4)42 Na16
HNaY/873
4
The
s p e c t r a a t d i f f e r e n t stages o f a l l e n e a d s o r p t i o n on sample No.
depicted i n Fig.
2.
t h r e e bands appeared a t 295, increased w i t h nounced
6
S h o r t l y a f t e r l o a d i n g t h e sample w i t h 6.65 Pa o f 347 and 450 nm.
The i n t e n s i t i e s o f these
t i m e and t h e i r p o s i t i o n s r e d - s h i f t e d ,
as i n t h e case o f propene.
although n o t
The e f f e c t o f e v a c u a t i o n was
are
allene bands
as
pro-
similar
to
t h a t observed w i t h propene. recorded e i t h e r immediately o r 1 h a f t e r propene exposure
Spectra wafers,
having
been p r e t r e a t e d a t d i f f e r e n t temperatures,
c o n t a i n i n g d i f f e r e n t B/L r a t i o s , the
shift
ratio.
of
From
investigation evident
spectra
similar
of
the
consequently
were s e l e c t e d and drawn i n F i g . 3. O b v i o u s l y
t h e bands o f adsorbed propene i n c r e a s e s the
and
to
a l l e n e adsorbed
on
c o n c l u s i o n s can be drawn (see
with
the
increasing
same
Fig.
4).
from t h e s p e c t r a t h a t t h e l o w e r t h e B r o n s t e d a c i d i t y o f
samples It
is
the
B/L under also
zeolite
t h e more i n t e n s e t h e HF band, c h a r a c t e r i s t i c f o r t h e monoenyl carbenium i o n s . Upon
a d s o r p t i o n o f propene on HNaY as w e l l as on HNaZSM-5 a f a s t
decrease
o f b o t h t h e band a s s i g n a b l e t o t h e C=C s t r e t c h i n g fundamental ( u s u a l l y l o c a t e d at
1612 cm-’) and t h e OH s t r e t c h i n g band o f t h e B r o n s t e d s i t e s
even a t beam temperature,
were
observed
i n d i r e c t l y p r o v i n g t h e enhanced f o r m a t i o n o f
alkyl
carbenium i o n s . R a i s i n g t h e temperature d u r i n g i n t e r a c t i o n a band developed a t 1505-1510
cm-’,
being
characteristic f o r the
a l k e n y l carbenium i o n s ( r e f . the
intensity
14).
CIc-_Cs t r e t c h i n g
mode
of
the
Upon e v a c u a t i o n a t i n c r e a s i n g temperatures
o f t h e bands o r i g i n a t i n g f r o m propene
carbenium i o n s a t 1505-1510 cm-’ decreased,
oligomers
and
alkenyl
w h i l e t h e o r i g i n a l spectrum o f t h e
700
h
300
400
500
x/nm
F i g . 1. U V - V I S s p e c t r a o f propene adsorbed on HNaZSM-5 (sample 6 ) . A f t e r admission o f 1.33 kPa o f propene a t room temperature (RT) ( a ) , a f t e r 1 h ( b ) and 2 h ( c ) a t RT, 1 h a t 323 K ( d ) , a f t e r e v a c u a t i o n a t RT ( e ) , 373 K ( f ) , 473 K ( 9 ) and 573 K ( h ) .
300
400
500
X/nm
F i g . 3. U V - V I S s p e c t r a o f propene adsorbed on samples 1-6 (subsequent t o adsorpt i o n (a), a f t e r 1 h ( b ) ) .
400
300
500
F i g . 2 . U V - V I S spectra o f a l l e n e adsorbed on HNaZSM-5 (sample 6 ) . I m m e d i a t e l y a f t e r l o a d i n g t h e sample t o 6.65 kPa o f a l l e n e a t RT ( a ) , 0.5, 1, 1 . 5 , 2, 2.5, 3, 4 . 5 and 5 h a t RT ( b - h ) , e v a c u a t i o n a t 373 K ( i ) , 473 K ( j ) and 573 K ( k ) .
300
400
500
A/nm
F i g . 4. U V - V I S s p e c t r a o f a l l e n e adsorbed on samples 1 - 6 (subsequent t o adsorpt i o n (a), a f t e r 1 h ( b ) ) .
Nnm
701
z e o l i t e was o n l y p a r t l y r e s t o r e d i n t h e OH s t r e t c h i n g range. A l l e n e t r a n s f o r m s e a s i l y t o propyne o v e r z e o l i t e c a t a l y s t s ( r e f .
15). This
c o n v e r s i o n c o u l d be d e t e c t e d a l s o on a c i d i c z e o l i t e s by t h e appearence o f
the
bands due t o t h e CIC and C-H s t r e t c h i n g v i b r a t i o n s o f adsorbed propyne a t 2116 and 3275 cm-I, r e s p e c t i v e l y . P a r a l l e l t o t h i s i s o m e r i z a t i o n a f a s t o l i g o m e r i z a t i o n v i a a l k e n y l carbenium i o n s takes place, the
i n d i c a t e d by t h e development
band c h a r a c t e r i s t i c f o r these t y p e s o f i o n s a t 1505 cm’ and
o l i g o m e r bands a t 1480 and 1380 c d .
the
of
typical
I t i s w o r t h w i l e t o m e n t i o n t h a t t h e band
o f t h e a l k e n y l carbenium i o n s a t 1505 c f l ’ c o u l d be d e t e c t e d more e a s i l y on t h e zeolites
of
l e s s Bronsted a c i d i t y .
Upon h e a t t r e a t m e n t t h e
changed f r o m w h i t e t o y e l l o w and a t 473
sample
K t o brown.
colour
of
the
I n t h i s case
the
o r i g i n a l spectrum i n t h e OH r e g i o n c o u l d n o t be r e s t o r e d .
DISCUSSION The
bands
observed
i n t h e ranges 280-330,
a t t r i b u t e d t o t h e n-T: t r a n s i t i o n s o f mono-,
360-380 and
430-470
nm
has a l r e a d y been s t a t e d t h a t these wavelengths c h a r a c t e r i s t i c f o r
It
carbenium
ions
formed i n z e o l i t e s agree w e l l w i t h those
different
4c).
Furthermore,
hydrocarbons
the
c a p a b i l i t y f o r a1 k e n y l i o n
and t h e i r d e r i v a t i v e s f o l l o w s t h e
same
by
the
carbenium
i o n s as a f u n c t i o n o f t h e number o f c o n j u g a t e d double bonds i n t h e (ref.
alkenyl
calculated
Sorensen e q u a t i o n d e s c r i b i n g t h e frequency dependence o f p o l y e n y l i c ion
are
d i - and t r i e n y l i c carbenium i o n s .
respective
formation
of
sequence
as
observed i n s u p e r a c i d s o l u t i o n s ( r e f . 7 ) . As
far
as t h e i n f l u e n c e o f t h e a c i d i t y o f z e o l i t e s i s
found
e a r l i e r t h a t t h e f o r m a t i o n o f mono-,
slows
down w i t h decreasing o v e r a l l a c i d i t y ( r e f s .
quantum good
c o r r e l a t i o n t o the
i t was
d i - and t r i e n y l i c carbenium 4c,17).
The
spectroscopic
observations,
of
rendered i .e.
t h e r e l a t i v e p e r m i t t i v i t y o f t h e s o l v e n t (comparable w i t h t h e
z e o l i t e s i n a coarse f i r s t a p p r o x i m a t i o n ) t h e more s t a b l e i s
ions
results
chemical c a l c u l a t i o n s u s i n g t h e MNDO e f f e c t i v e charge model
qualitative
higher of
concerned
the
acidity
the
alkenyl
carbenium i o n ( r e f . 1 6 ) . The common f e a t u r e o f t h e s p e c t r a o f adsorbed propene and a l l e n e (see F i g s . 1-2)
time
i s t h e s h i f t o f t h e U V - V I S band t o h i g h e r wavelengths w i t h c o n t a c t
and i t s r e v e r s i o n upon e v a c u a t i o n . T h i s can be e x p l a i n e d by: t h e band p o s i t i o n of
monoenylic carbenium i o n s (observed g e n e r a l l y i n t h e 275-320 nm
s u p e r a c i d s o l u t i o n ) b e i n g s e n s i t i v e t o t h e s u b s t i t u e n t groups While jump
the by
i n s e r t i o n o f an a d d i t i o n a l double bond g i v e s r i s e t o about
60 nm,
the e x t e n s i o n o f t h e carbon
chain
or
branched o l i g o m e r s r e s u l t s o n l y i n a smooth downscale f r e q u e n c y accomplished by c o n s e c u t i v e r e a c t i o n s ,
range
a
in
18-20).
(refs.
frequency
formation shift,
of
being
as b o t h propene and a l l e n e o l i g o m e r i z e
702
very
fast
at
room temperature o v e r
acidic
zeolites.
Preliminary
chemical
c a l c u l a t i o n s showed t h e l o w e s t v e r t i c a l s i n g l e t e x c i t a t i o n
(LVSEE)
for
methyl
substituted
ally1
carbenium
16).
s i n g w i t h growing methyl s u b s t i t u t i o n ( r e f .
ions
to
quantum energies
be
decrea-
Assuming t h a t t h e f o l l o w i n g
r e a c t i o n s may occur on z e o l i t e s :
t
CH3-CH-CH-CH-CH2-CH3 -----HCH3-CHzCHZ
CH3-CHzCH2
+
CH CH-CH2 - 2:- _ _ -
--5.
y 3 t CH3-CH-CH-CH-CH2
.>
y3+ CH3-C-CH-CH-CH3
-_____
“I
t , CHZ-CH-CH-CH2-CH2-CH3 - - -. - - - -
CH2- C- CH- CH2- CH3 __---
-_--
t h e r e should be an e q u i l i b r i u m m i x t u r e o f v a r i o u s monoenyl carbenium i o n s commonness o f which c o n s i s t s i n a c 6 backbone a b s o r b i n g a t d i f f e r e n t
the
frequen-
c i e s . The s p e c t r o s c o p i c measurements were performed on z e o l i t e s w i t h d i f f e r e n t B/L r a t i o s . I t i s accepted t h a t t h e B r o n s t e d a c i d s i t e s p l a y t h e dominant r o l e i n oligomerization o f olefins, reaction amount
b u t t h e p a r t i c i p a t i o n o f Lewis c e n t e r s i n t h i s
cannot be n e g l e c t e d ( r e f s . of
olefins
Bronsted a c i d s i t e s ,
and
reaction
consequently
the
21,22).
It f o l l o w s t h a t the higher
t h e more pronounced t h e a l k e n y l carbenium
absorb a t h i g h e r wavelengths.
ions
oligomerization formed
during
This i s the experimental
the of this
observation
w i t h b o t h propene and a l l e n e on b o t h t y p e s o f z e o l i t e s (see F i g s . 3 , 4 ) . In range
the
s p e c t r a o f adsorbed a l l e n e t h e p r e v a i l i n g band was
359-380
nm
characteristic f o r dienyl
f o l 1 owing t r a n s f o r m a t i on :
carbenium
found
ions.
in
the
Assuming
the
t
CH2-C-CH-CH-CH2
-_------
t h e g e n e r a t i o n o f i n c r e a s i n g l y u n s a t u r a t e d carbenium i o n s s h o u l d be The
i d e a was t h a t i f t h e number o f B r o n s t e d a c i d s i t e s o f t h e z e o l i t e
was decraesed by d e h y d r o x y l a t i o n , ion
exchange
carbenium creasing Fig,
(ref.
ions
4c),
the
i n s t e a d o f l o w e r i n g t h e degree o f e x i s t e n c e o f more and
s h o u l d be observed by U V - V I S
dehydroxylation the strength o f
increases, in
expected.
more
spectroscopy,
Bronsted acid s i t e s
simple since
sample ammonium monoenyl
with
probably
inalso
t h e r e b y p r o m o t i n g t h e s t a b i l i t y o f a g i v e n carbenium i o n . As shown
4
our
e x p e c t a t i o n came t r u e :
t h e lower the
B/L
ratio
the
more
703
pronounced
290 nm on b o t h z e o l i t e s .
band was observed a t ca.
explanation
detailed
above
f o r the s h i f t s o f U V - V I S
bands
Accepting to
the
the
higher
wavelengths, t h e i n t e r p r e t a t i o n o f t h e r e v e r s e s h i f t o c c u r r i n g upon e v a c u a t i o n o f t h e sample a t i n c r e a s i n g temperatures i s t h e f o l l o w i n g : higher
the
remaining
formed
o l i g o m e r s undergo c r a c k i n g
and/or
A t about 400 K
desorb,
thereby
carbenium i o n s ( i n c l u d i n g t h e a l k e n y l t y p e s ) a r e o f s h o r t e r
or the
carbon
c h a i n type, thus a b s o r b i n g a t h i g h e r f r e q u e n c i e s . The
maximum a b s o r p t i o n o f t h e a l l y 1 c a t i o n b e i n g
carbenium
i o n was
predicted
to
be
a t 293
the
nm as
a
simplest result
monoenyl
of
LCAO-MO
18). The LVSE e n e r g i e s c a l c u l a t e d by CNDO/S and MNDO-HE methods a r e l i s t e d i n Table 2. calculations
(ref.
Table 2. Lowest v e r t i c a l s i n g l e t e x c i t a t i o n e n e r g i e s and s t a n d a r d heats o f f o r m a t i o n o b t a i n e d by s e m i e m p i r i c a l quantum chemical methods Carbocation
LVSE energies/nm CNOO/S//MNDO MNDO-HE//MNDO
Standard h e a t o f f o r m a t i o n bv MNDO k c a l /&l rel. abs.
abs
re1 .
abs.
rel.
220
0
281
0
221.4
is o p r o p y l a1 l y l
237
17
289
8
200.0
-21.4
n-propyla1 l y l
237
17
289
8
195.8
-25.6
l-ethyl-2methyl-ally1
253
33
314
33
195.2
-26.2
l-ethyl-3methyl-ally1
234
14
300
19
187.0
-34.4
241 286
21 0
313 299
32 0
185.4 219.8
-36.0 0
pentadienyl
290
4
306
7
207.3
-12.5
2-methyl pentadienyl
297
11
311
12
214.7
-5.1
305
17
328
29
188.7
-31.1
325
37
330
31
182.8
-37.0
a1 l y l
0
1,1,3-tri methyl -a1 l y l pentadienyl
1 -methyl -
1,1,3,5tetramethylpentadienyl
1,1,5,5-
tetramethyl pentadienyl
It
can be seen t h a t t h e more s u b s t i t u t e d monoenyl and d i e n y l
should ions
absorb a t h i g h e r wavelengts.
Futhermore,
under study a l s o i n c r e a s e d w i t h i n c r e a s i n g
results
carbeniurn
the s t a b i l i t y o f methyl
a r e i n good accordance w i t h t h e e x p e r i m e n t a l
carbenium
substitution.
observations
ions These
discussed
704
above.
According t o t h e r e s u l t s o f quantum chemical c a l c u l a t i o n s , t h e a s s i g n -
ment o f a g i v e n U V - V I S band measured t o a w e l l - d e f i n e d a l k e n y l i o n seems t o be rather d i f f i c u l t .
By t h e e x p e r i m e n t a l l y observed band o n l y a group o f a l k e n y l
i o n s can be c h a r a c t e r i z e d . The
generation
of
unsaturated
carbenium i o n s
was
proven
also
by
IR
spectroscopy. The band g e n e r a l l y observed a t 1505-1510 c d was regarded as t h e t y p i c a l band o f these types o f carbenium i o n s . CONCLUSIONS The lower
the
B/L a c i d i t y r a t i o o f t h e z e o l i t e ,
the higher w i l l
be
the
frequency o f t h e a l k e n y l carbenium i o n s formed f r o m propene and a l l e n e on b o t h types
of
zeolites.
T h i s c h a r a c t e r i s t i c can be a t t r i b u t e d
to
the
enhanced
f o r m a t i o n o f o l i g o m e r s on t h e z e o l i t e s o f h i g h e r B r o n s t e d a c i d i t y . The h i g h e r t h e B/L r a t i o o f t h e sample,
the l e s s intenses the formation o f
monoenyl carbenium i o n s a b s o r b i n g a t 280-310 nm. by
assuming
an
increase o f the strength o f the
T h i s f a c t m i g h t be e x p l a i n e d Bronsted
decreasing c o n c e n t r a t i o n s (as a r e s u l t o f dehydroxyl a t i o n ) ,
acid
sites
with
thereby promoting
t h e s t a b i l i t y o f t h e monoenyl carbenium i o n s formed f r o m a l l e n e . ACKNOWLEDGEMENT One o f t h e a u t h o r s ( I . K.) thanks t h e Alexander von Humboldt F o u n d a t i o n f o r a r e s e a r c h f e l l o w s h i p . The f i n a n c i a l s u p p o r t o f t h e Hungarian Academy o f Sciences (No. OTKA 217/88) and t h e Deutsche Forschungsgemeinschaft i s g r a t e f u l l y acknowledged REFERENCES 1 H F o r s t e r , R . Seelemann, J. C . S . Faraday Trans. I , 74 (1978) 1435 2 H F o r s t e r , J. Seebode, Z e o l i t e s , 3 (1983) 63 I . K i r i c s i , G. T a s i , P . F e j e s , F. Berger, J . Mol. C a t a l . , 51 1989) 341 3 I . K i r i c s i , G. T a s i , I. Hannus, P . F e j e s , H. F o r s t e r , J. Mol. C a t a l . , i n press 4 ( a ) M. Zardkoohi, J. F . Haw, J. H. Lunsford, J. Am. Chem. SOC , 109 (1987) 5278 ( b ) A . S. Medin, V . Yu. Borovkov, V . B. K a z a n s k i i , K i n e t i k a i K a t a l iz. 30 (1989) 177 ( c ) H . F o r s t e r , 3 . Seebode, P . F e j e s , I . K i r i c s i , J . C . S. Faraday Trans. I , 83 (1987) 1109 5 E. A . Lombardo, R . P i e r a n t o z z i , W. K. H a l l , J . C a t a l ., 110 (1988) 171 6 J . P . van den Berg, J. P . Wolthuizen, A. D. H. Clauqe, G. R . Hays, R . Huis, J . H. C . van H o o f f , J. C a t a l . 80 (1983) 130 7 P. Fejes, H. F o r s t e r , I . K i r i c s i , J. Seebode, i n " S t r u c t u r e and R e a c t i v i t y of M o d i f i e d Z e o l i t e s " , D. A. Jacobs,N. I . Jaeger, P. J i r u , V . B . K a z a n s k i i G. S c h u l z - E k l o f f ( e d s . ) , E l s e v i e r , Amsterdam, 1984, p. 9 1 8 M. L . Poutzma, i n " Z e o l i t e Chemistry and C a t a l y s i s " , Am. Chem. SOC. Washington, Ed. J . A . Rabo (ACS Monograph 171) pp. 437 9 H. F o r s t e r , S. Franke,J. Seebode, J. C . S. Faraday Trans. I , 79 (1983) 373 10 I . K i r i c s i , G. T a s i , H. F o r s t e r , P . F e j e s , A c t a Phys. e t Chem. Szeged, 33 (1987) 69 11 G. Tasi, I . K i r i c s i , P . F e j e s , H. F o r s t e r , S. Lovas, Magy. Kem. F o l y o i r a t , 95 (1989) 520
705
12 13 14 15 16 17 18 19 20 21 22
M. J. S . Dewar, J. A. Hashmall, C . G. V e n i e r , J. Am. Chem. SOC., 90 (1968) 1953 H. H. J a f f e , R . L. E l l i s , i n " S e m i e m p i r i c a l Methods o f E l e c t r o n i c S t r u c t u r e C a l c u l a t i o n s " , G. A. Segal ( e d . ) , Plenum P r e s s , New York, 1977 I . K i r i c s i , H. F o r s t e r , J. C . S. F a r a d a y T r a n s . I , 84 (1988) 491 P. Kos. I . K i r i c s i . K. Varqa. - . P. Fe.ies. - . A c t a Phvs. e t Chem. Szeged, 33 (1987) 109 I . K i r i c s i , G. T a s i , H. F o r s t e r , P. F e j e s , J. M o l . S t r u c t . 218 (1990) 369 I . K i r i c s i , H. F o r s t e r , G. T a s i , P. F e j e s , J. C a t a l . , 115 1989) 597 T. Sorensen, i n "Carbonium I o n s " , e d s . G. A. O l a h and R. P S c h l e y e r ( W i l e y - I n t e r s c i e n c e ) New York, 1970, V o l . 2, p . 807 G. A. Olah, C . U. P i t t m a n n , M. C . R. Symons, i b i d . V o l . 1, p. 153 N. C . Deno, i b i d , V o l . 2, p . 783 L . Kubelkova, J . Novakova, B. W i c h t e r l o v a , P. J i r u , C o l l e c Czech. Chem. Commun., 45 (1980) 2290 J. Datka, J. C . S. F a r a d a y T r a n s . I , 77 (1981) 1309, 2633
.
This Page Intentionally Left Blank
70 7
AUTHOR INDEX
.................
573 Aiello. R Albizane. A 603 Babayev. M.K 623 Barrer. R.M 257 Barthomeuf. D 157 Bauer. F 305 Bauermeister. R 367 Baur. W.H 581 Bee, M 445 Bergk. K.-H 185 Bordiga. S 671 Bohlig. H 511 Borovkov. V.Yu 653 BosaEek. V 337 Bragin. O.V. 357 Brunner. E 529 Biilow. M 445.479. 501 Buskens. Ph 47 Buysch. H.-J 297 Camblor. M.A 613 Caro. J 445 Eejka. J 387 Chen. Q 219 Cichocki. A 681 Crea. F 573 Datka. J 681 Dermietzel. J 305 Derouane. E.G 591 Des Courieres. Th 603 Di Renzo, F 603 Dressler. J 663 Dumont. N 591 Eckelt. R 1.415 Ehrhardt. B 663 Eic. M 233 Einicke. W.-D .............. 75. 529 Emig. G ................... 479 Ernst. H 397. 529 Ernst. S 297. 645 Fahlke. B 315 Fajula. F 603 Fejes. P 697 Feoktistova. N.N 287 Fichtner-Schmittler. H 549 Figueras. F 603 Finger. G 501. 537 Forster. H 697 Fraissard. J .............. 219 Freude. D 89. 397 Fricke. R 631 Fujimoto. K 203 Fiirtig. H 185 Gabelica. 2 ...............591 Ganbarov. D.M 623 Geidel, E 511 Geus. E . R 457 Golemme. G 573 Goncharuk. V 581
............... .............. ............... ............. .................. ........... ................. .................... ............... ................ ................. ............ ................ .............. ................ .................. ................ .............. .............. ................... .................. ................... ............... ................... .................. ............. ............. ......... ............... ............... ................. ................... ............... ....................
.................. .................. ................. ................. .................. .......... .... ............... ................. ................
.................. ................. ...............
................. ............. ................. ................. ................ ..............
708
............... 613
Grobet. P.J Gross. T Hadicke. U Hantel, K Hayward. C.-M.T Hilgert. W Hoffmann, J Hunger. B Hunger, M Huybrechts. D.R.C Ito, T Jacobs. P.A Jaeger. N.1 Jancke, K Jansen, A.E Jansen. J.C Jerschkewitz.H G Jobic, H Jockisch. W Joyner. R.W Karge. H.G Karger. J Kazansky, V.B Kiricsi. I Klaeboe. P Kojima. M Kornatowski. J Kubelkova. L Kustov, L.M aaniecki. M Leofanti. G Lindner. D Lischke. G Loffler. E Lohse. U Lucke. B Martens. J.A Martin. A Meier, B Meier, W.M Minachev. Kh.M Mishin, 1.V Muller, B Nagy. J.B Nastro. A Nicolle, M.-A Novakova. J Nowak, 5 O'connor. C.T ohlmann. G O'Malley, P.J Padovan. M Parlitz, B Parton. R.F Petrini. G Peuker, Ch Pfeifer, H Pilz, W Piwowarska. 2 Pradhan. A.R
.................. 415 ................185 .................367 ...........171 ................ 479
............... 367 ................. 367 ................. 397,537 ..........47 .................... 591 ................ 47. 613 ............... 327 ................. 501 ...............457 ............... 457 .- ........... 1. 415 .................. 445 ............... 305 ...............357 ................133 .................. 89. 445 .............117.425 ................ 697 ................ 663
.................491 ............501.537, 581 ..............337. 405 ...............425 ............... 377 ...............671 ................297 ..................1. 415 .................. 1.425. 537 .................. 425. 549 ..................315 ..............613 .................315 ..................529 ................247 ............357 ............... 563 ................. 521 .................573. 591 ................. 573 ............. 603 ...............337. 405 .................. 315 .............491 .................. 1.415. 425 .............689
................ 671 ..................1,415.631 ................ 47 ................671 ................425. 511 .................89,397. 445 ...................511 .............681
.............. 347
709
Preobrazhensky. A.V Prilipko. A.1 Pritzkow, W Rao. B.S Raurell. G.L Rees. L.V.C Reschetilowski. W Richter, M Richter-Mendau. J Rozwadowski, M Ruthven. D.M Salzer. R Schollner. R Schoonman, J Schreier. E Schroder. K.-P Schulz. I Schulz-Ekloff. G Schwartz. S Schwieger. W Senchenya. 1.N Shiralkar. V . P Shpiro, E.S Siegel. H Sonntag. E Spoto, G Springuel-Huet, M.A Steinberg, K.-H Suboti6. B Svensson. A Sychev. M Szulzewsky. K Tagiyev. D.B Tasi, G Tatsumi. T Telbiz. G.M Timm, D Tkachenko. 0.P Tominaga. H Tuleuova. G.J Ullmann. G Van Bekkum, H Vasina. T.V Vaughan, D.E.W Von Ballmoos. R Voogd. P Vorbeck. G Vtjurina. L.M Wehner. K Weitkamp. J Wichterlova. B Wieker. W xu. z Zecchina. A Zhdanov. S.P Zholobenko. V.L Zibrowius, B Zmierczak. W
.......357
.............563 ............... 549 ..................347
.............. 387
................61 .........529 ................631 .........501. 549
............501 .............. 233 .................663 ...............75 ..............457 ................. 1. 631 ............ 435 .................415 ..........327 ............... 491 .............. 185 ............653 ............347 ............... 357 ................. 367 ................ 537 .................. 671 .......219 ...........367. 663 ................ 573 ............... 327 .................581 ............. 549. 631 .............. 623 ...................697 ................203 ...............563 ...................415 ............357 ............... 203 .............357 ................367 ............. 457,467 ...............357 ............275 ........... 171 .................. 467 ................ 631 .............287
................. 415 ................21.297. 645 ............387 .................315 ..................... 233 ...............671 .............. 287 ...........425
................ 1,537,549, 631 ..............377
This Page Intentionally Left Blank
711
SUBJECT INDEX
............... 337 .........405
Acetone adsorption Acetone. conversion from Acid-base pair Acidity Acid site characterization Activation Active sites Adsorption Adsorption of CO Alcohol-water mixtures Al in zeolites Alkali metal exchanged ZSM-5 Alkene oxidation Alkylation of toluene Allene Alloying particles Al-0 bond strength
...................1171.89.563.631.681. 697 ............................ .......133 ....................... 663 ..................... 425 ....................... 689 .................133.157.435.479. 671 ............75 ................... 387 ..... 297 .................. 47 ............387 ........................... 697 ...............357 ...............653 549 AlPO-14 .......................... Alumination...................... 467 Alumino-galloferrierite..........613 Approximate assignment...........511 Aromatic hydrocarbons............233 ........................ 203. 689 ............................ 247 .................... 21 ......................... 1571563 ...............203 ........................ 681 .................... 681 ...................697 .........337 ..................327 .................367 ........................ 157 ........315 ...............397 ..........645 ...........521 ................. 5731645 .................. 415 ........ 337 .................. 405 ........ 501 ......................... 117. 467 .............171 ................... 2871603 .................... 529 ..................2871591 .........573 ..................... ..247 491
Aromatics Atlas Base catalysis Basicity Bimetallic cluster B03 units Boralites MFI Carbenium ions Carbocations on zeolites Carbon deposits Carbon filaments Catalysis Catalyst characterization Catalytic activity Catalytic test reaction Cellular configuration Characterization Coke deposition Coke deposits on zeolites Coke in H-ZSM-5 Controlled crystal growth Cracking Cracking selectivity Crystal growth Crystallinity Crystallization Crystallization kinetics Crystallographic pore diameters Crystal size Curvature of configuration DABCO (triethylenediamine) Deactivation Dealumination Density of acid sites para-Diethylbenzene Diffuse of reflectance
....... 521 ....... 297 ..................... 347. 397 ...................... 1.185.415. 425 ............133 ..............479
...........663
712
........................ 2331305145714671479 ...................... 257 ....................... 327 ................ 157 .............653 ......................... 663 ............ 357 ..........6 3 1 EXAFS ............................ 357 Extraction of aluminium..........415 Extrudates....................... 491 Faujasite........................ 27Sl 327 FCC catalysts.................... 171 Ferrisilicates...................6 3 1 Fe-ZSM-5......................... 631 Framework density distribution...247 Heteroatom incorporation.........5 8 1 Heterogeneous nucleation.........603 n-Hexane cracking................425 High pressure chromatography.....479 H-mordenite...................... 347 Host/guest chemistry ..............2 1 Hybrid catalyst..................203 Hydrocarbon pyrolysis ............315 Hydrothermal crystallization.....623 Hydroxyls......................... ag1563168i H-ZSM-5 zeolite.................. 3 3 7 14 2 5 H-ZSM-12 ....................... ..347 Influence of geometry ............653 In-situ study .................... 663 Interface........................ 257 Intergrowth of structure.........219
Diffusion Diffusivity Dispersion Electronegativity Electronic structure Erionite Ethane hydrogenolysis Ethylbenzene conversion
....1 1 7 ..................4 2 5 1 5 3 7 1 6 6 3 1 6 8 1 .................... 3 0 5 ....5 8 1 .........537 .................... 203 ........347 ......................... 4 4 ......................... 2a 7 1 3 0 5 ......347 ............5 8 1 ........................... ...................35 40 7l I 5 a i ..............347 .................... i1 7 1 6 5 3 .......................... 275 ........... 7 5 MAS NMR .......................... 397 MAS N M R ( 2 7 Al) .................. 529 Matrix activity..................1 7 1 Matrix vs zeolite................1 7 1 Mechanism .......................... il i a 5 Membranes ........................ 2 5 7 1457 Methane aromatization............357 Methane splitting................367 Methanol ......................... 563 Methanol conversion ..............3 1Sl653 Methanol conversion to olefins...4 1 5 IR diffuse reflectance spectr IR spectroscopy Isomerization Isomorphous substituted ZSM-5 Isomorphous substitution Iso-paraffins Isopropylation of benzene Isotherms Kinetics Kinetics of cumene sorption KVS-5 molecular sieve La-H-Y Large crystals Large pore zeolites Lewis acidity Linde L Liquid phase adsorption
713
................1 ...................... 305 .................... 445 ........................ 257,689 ............ 435 ....................... 1. 1171327 ..................501 ...................... 623 MO-Y ............................. 377 Multinuclear N M R .................591 Nay ............................... 4 7 1 61 Ni-Mo-Y .......................... 377 Nitrogen bases ...................297 Ni-ZSM-5 catalyst................367 N M R in alumino-galloferrierite...613 NMR spectroscopy ............337 Methanol-to-olefins Methylation Microdynamics Modelling Modelling by computer Modification Molecular sieve Mountainite
... ....... ....................... (13c)
NMR spectroscopy. multinuclear 537. 549 Normal coordinate analysis 511 Nucleation 287. 591 Nucleation kinetics 573 Oligomerization 491 Organic intermediates 297 Para selectivity 387 Path of oxidation 405 Pelletization 75 Pentasil-type 185 Pentasil zeolites 75 Permeation 457 Phase transitions in AlPOs 549 Piperazine 297 Porosity 247 Propene 697 Properties 623 Proton-MAS-NMR 89 Pt-Cr pentasils 357 Pulsed field gradient N M R 89. 445 Quasi-elastic neutron scattering.445 Realumination 529 Regeneration 315 Restoration of active sites 405 Restoration of void volumes 405
.............. .................. ............ ................. ................ ..................... .................... ................. ....................... ....... ....................... ......................... .......................... ....................... .................... .................. ......... .................... ..................... ...... ...... SAPO-5 ........................... 5011537 SAPO-37 .......................... 591 SAPO-40 ........................ ..591 Secondary synthesis..............529 a gI445 Self-diffusion .................... Separation........................ 61 Shape selective catalysis .........21 Shape selective photochemistry ....21 Shape-selectivity ................387 Ship-in-the-bottle catalysts......21 ....................... ...................... ............ ................... .......................... ..................... ...........
Silicalite 233.457. 4 6 7 1 6 7 1 Silicalite-1 61.435 Silicon incorporation 537 Single crystal 457 Sorption 611257 Spectroscopy 697 Strength of acid sites 133 Structure defect .................219 Structure effect 157
.................
714
.................. 247 ..........511 ...................... 377 ................327 ........................ 1851275157312 .................... 203 ........................ .................... 501 185 ................. 549 ........................ 491 ............... 603 ................ 75 ................ 367 .................... 467 ..................... 471671 .......... 521 ........... 521 . .. ...................521 415 .......... 233 ................... 377 ........ 581 ..........511 ................ 367 ....................... 219 ............................. 47 .................. 377 ........................ 219 .............................. 549 ...613 ...................... 435 ........631 ..................... 479 ........................ 233 ................. 603 171 ........................ .................. .................. 275 ...........397 623 ...................... 511 ......................... 20714451689 ................ ................275 645 ........................ 233 ........................ 563 ................... 645 ............................ 3 0 ~ 1 3 ~ ~ 1 3 a 7 1 4 3 5 1 ~ ~ ~ l ~ ~ ~ 1 ZSM-12 ........................... 603 ZSM-35 ........................... 645 ZSM-57 ........................... 645
Structure types Sub-unit cluster models Sulfidation Syngas conversion Synthesis Synthesis gas Tailoring Template-free Template removal Templates Tetraethylammonium Thermal desorption Thermal reduction Thiele theory Ti-silicalite Topological constraints Topological parameters Topological structure functions TPD of ammonia Transition state theory USY as support Vanadium silicalite KVS-5 Vibrational frequencies Vibration reactor Void space VPI-5 Water-gas shift Xenon-= XRD XRD in alumino-galloferrierite para-Xylene meta-Xylene isomerization Zeolite 13 x Zeolite A Zeolite activity Zeolite 8 Zeolite genesis Zeolite H-ZSM-5 Zeolite-like silicates Zeolite NaX Zeolites Zeolite sub-units Zeolite synthesis Zeolite X Zeolite Y Zeolite ZSM-57 ZSM-~
1
715
STUDIES IN SURFACE SCIENCE AND CATALYSIS Advisory Editors: 6. Delmon, Universite Catholique de Louvain, Louvain-la-Neuve,Belgium J.T. Yates, University of Pittsburgh, Pittsburgh, PA, U S A .
Volume 1
Volume. 2
Volume 3
Volume 4
Volume 5
Volume 6
Volume 7
Volume 8 Volume 9
Volume 10
Volume 11
Volume 12
Volume 13 Volume 14
Preparation of Catalysts 1. Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the First International Symposium, Brussels, October 1417,1975 edited by B. Delrnon, P.A. Jacobs and 0. Poncelet The Control of the Reactivity of Solids. A Critical Survey of the Factors that Influence the Reactivity of Solids, with Special Emphasis on the Control of the Chemical Processes in Relation to Practical Applications by V.V. Boldyrev, M. Bulens and B. Delrnon Preparation of Catalysts It. Scientific Bases for the Preparation of Heterogeneous Catalysts. Proceedings of the Second International Symposium, Louvain-la-Neuve, September 4-7,1978 edited by B. Delrnon, P. Grange, P. Jacobs and G. Poncelet Growth and Properties of Metal Clusters. Applications to Catalysis and the Photographic Process. Proceedings of the 32nd International Meeting of the Societe de Chimie Physique, Villeurbanne, September 24-28, 1979 edited by J. Bourdon Catalysis by Zeolites. Proceedings of an InternationalSymposium, Ecully (Lyon), September 9- 1 1,1980 edited by B. Irnelik, C. Naccache, Y. Ben Taarit, J.C. Vedrine, G. Coudurierand H. Praliaud Catalyst Deactivation. Proceedings of an InternationalSymposium, Antwerp, October 13-15,1980 edited by B. Delmon and G.F. Frornent New Horizons in Catalysis. Proceedings of the 7th InternationalCongress on Catalysis, Tokyo, June 30-July 4, 1980. Parts A and B edited by T. Seiyama and K. Tanabe Catalysis by Supported Complexes by Yu.1. Yerrnakov, B.N. Kuznetsov and V.A. Zakharov Physics of Solid Surfaces. Proceedings of a Symposium, Bechyfie, September 29-October 3, 1980 edited by M. UzniEka Adsorption at the Gas-Solid and Liquid-Solid Interface. Proceedings of an International Symposium, Aix-en-Provence, September 2 1-23, 198 1 edited by J. Rouquerol and K.S.W. Sing Metal-Support and Metal-Additive Effects in Catalysis. Proceedings of an International Symposium, Ecully (Lyon), September 14-16, 1982 edited by B. Irnelik, C. Naccache, G. Coudurier, H. Praliaud, P. Meriaudeau, P. Gallezot, G.A. Martin and J.C. Vedrine Metal Microstructures in Zeolites. Preparation - Properties - Applications. Proceedings of a Workshop, Brernen, September 22-24, 1982 edited by P.A. Jacobs, N.I. Jaeger, P. Jird and G. Schulz-Ekloff Adsorption on Metal Surfaces. An Integrated Approach edited by J. Benard Vibrations at Surfaces. Proceedings of the Third InternationalConference, Asilomar, CA. September 1-4, 1982 edited by C.R. Brundle and H. Morawitz
7 16 Volume 15 Volume 16
Volume 17
Volume 18
Volume 19
Volume 20
Volume 2 1
Volume 22 Volume 23 Volume 24
Volume 25
Volume 26
Volume 27 Volume 28
Volume 29 Volume 3 0
Volume 3 1
Volume 32 Volume 33 Volume 3 4
Heterogeneous Catalytic Reactions Involving Molecular Oxygen by G. I.Golodets Preparation of Catalysts Ill. Scientific Bases for the Preparationof Heterogeneous Catalysts. Proceedings of the Third International Symposium, Louvain-la-Neuve, September 6-9, 1982 edited by G. Poncelet, P. Grange and P.A. Jacobs Spillover of Adsorbed Species. Proceedings of an International Symposium, LyonVilleurbanne, September 12-1 6, 1983 edited by G.M. Pajonk, S.J. Teichner and J.E. Germain Structure and Reactivity of Modified Zeolites. Proceedings of an International Conference, Prague, July 9-1 3, 1984 edited by P.A. Jacobs, N.I. Jaeger, P. Jirir, V.B. Kazansky and G. Schulz-Ekloff Catalysis on the Energy Scene. Proceedings of the 9th Canadian Symposium on Catalysis, Quebec, P.Q., September 30-October 3, 1984 edited by S.Kaliaguine and A. Mahay Catalysis by Acids and Bases. Proceedingsof an International Symposium, Villeurbanne (Lyon), September 25-27, 1984 edited by B. Imelik, C. Naccache, G. Coudurier, Y. Ben Taarit and J.C. Vedrine Adsorption and Catalysis on Oxide Surfaces. Proceedings of a Symposium, Uxbridge, June 28-29,1984 edited by M. Che and G.C. Bond Unsteady Processes in Catalytic Reactors by Yu.Sh. Matros Physics of Solid Surfaces 1984 edited by J. Koukal Zeolites: Synthesis, Structure, Technology and Application. Proceedings of an InternationalSymposium, Portoroi-Portorose, September 3-8, 1984 edited by B. Driaj, S.HoEevar and S.Pejovnik Catalytic Polymerization of Olefins. Proceedings of the International Symposium on Future Aspects of Olefin Polymerization, Tokyo, July 4-6, 1985 edited by T. Keii and K. Soga Vibrations at Surfaces 1985. Proceedings of the Fourth International Conference, Bowness-on-Windermere, September 15-1 9, 1985 edited by D.A. King, N.V. Richardson and S.Holloway Catalytic Hydrogenation edited by L. Cervenl New Developments in Zeolite Science and Technology. Proceedings of the 7th InternationalZeolite Conference, Tokyo, August 17-22, 1986 edited by Y. Murakami, A. lijima and J.W. Ward Metal Clusters in Catalysis edited by B.C. Gates, L. Guczi and H. Knozinger Catalysis and Automotive Pollution Control. Proceedings of the First InternationalSymposium, Brussels, September 8-1 1, 1986 edited by A. Crucq and A. Frennet Preparation of Catalysts IV. Scientific Bases for the Preparationof Heterogeneous Catalysts. Proceedings of the Fourth International Symposium, Louvain-la-Neuve, September 1-4, 1986 edited by B. Delmon, P. Grange, P.A. Jacobs and G. Poncelet Thin Metal Films and Gas Chemisorption edited by P. Wissmann Synthesis of High-silica Aluminosilicate Zeolites by P.A. Jacobs and J.A. Martens Catalyst Deactivation 1987. Proceedings of the 4th International Symposium, Antwerp, September 29-October 1, 1987 edited by B. Delmon and G.F. Froment
717 Volume 35 Volume 36
Volume 37
Volume 38
Volume 39
Volume 40
Volume 41
Volume 42 Volume 43 Volume 44
Volume 45 Volume 46
Volume 47 Volume 48
Volume 49
Volume 50
Volume 5 1 Volume 52
Volume 53
Keynotes in Energy-Related Catalysis edited by S. Kaliaguine Methane Conversion. Proceedings of a Symposium on the Production of Fuels and Chemicals from Natural Gas, Auckland, April 27-30, 1987 edited by D.M. Bibby, C.D. Chang, R.F. Howe and S. Yurchak Innovation in Zeolite Materials Science. Proceedings of an International Symposium, Nieuwpoort, September 13-1 7,1987 edited by P.J. Grobet, W.J. Mortier, E.F. Vansant and G. Schulz-Ekloff Catalysis 1987. Proceedings of the 10th North American Meeting of the Catalysis Society, San Diego, CA, May 17-22, 1987 edited by J.W. Ward Characterization of Porous Solids. Proceedings of the IUPAC Symposium (COPS I),Bad Soden a. Ts., April 26-29, 1987 edited by K.K. Unger, J. Rouquerol, K.S.W. Sing and H. Kral Physics of Solid Surfaces 1987. Proceedings of the Fourth Symposium on Surface Physics, Bechyne Castle, September 7-1 1, 1987 edited by J. Koukal Heterogeneous Catalysis and Fine Chemicals. Proceedings of an International Symposium, Poitiers, March 15-1 7, 1988 edited by M. Guisnet, J. Barrault, C. Bouchoule, D. Duprez, C. Montassier and G. Perot Laboratory Studies of Heterogeneous Catalytic Processes by E.G. Christoffel, revised and edited by 2. Pa61 Catalytic Processes under Unsteady-State Conditions by Yu. Sh. Matros Successful Design of Catalysts. Future Requirements and Development. Proceedings of the Worldwide Catalysis Seminars, July, 1988, on the Occasion of the 30th Anniversary of the Catalysis Society of Japan edited by T. lnui Transition Metal Oxides. Surface Chemistry and Catalysis by H.H. Kung Zeolites as Catalysts, Sorbents and Detergent Builders. Applications and Innovations. Proceedings of an International Symposium, Wurzburg, September 48,1988 edited by H.G. Karge and J. Weitkamp Photochemistry on Solid Surfaces edited by M. Anpo and T. Matsuura Structure and Reactivity of Surfaces. Proceedings of a European Conference, Trieste, September 13-1 6, 1988 edited by C. Morterra, A. Zecchina and G. Costa Zeolites: Facts, Figures, Future. Proceedings of the 8th International Zeolite Conference, Amsterdam, July 10-14, 1989. Parts A and B edited by P.A. Jacobs and R.A. van Santen Hydrotreating Catalysts. Preparation, Characterizationand Performance. Proceedings of the Annual International AlChE Meeting, Washington, DC, November 27-December 2, 1988 edited by M.L. Occelli and R.G. Anthony New Solid Acids and Bases. Their Catalytic Properties by K. Tanabe. M. Misono, Y. Ono and H. Hattori Recent Advances in Zeolite Science. Proceedings of the 1989 Meeting of the British Zeolite Association, Cambridge, April 17-1 9, 1989 edited by J. Klinowski and P.J. Barrie Catalyst in Petroleum Refining 1989. Proceedings of the First International Conference on Catalysts in Petroleum Refining, Kuwait, March 5-8, 1989 edited by D.L. Trimm, S. Akashah, M. Absi-Halabi and A. Bishara
7 18 Volume 54
Future Opportunities in Catalytic and Separation Technology edited by M. Misono, Y. Moro-oka and S. Kimura Volume 55 New Developments in Selective Oxidation. Proceedings of an International Symposium, Rimini, Italy, September 18-22, 1989 edited by G. Centi and F. Trifiro Volume 56 Olefin Polymerization Catalysts. Proceedings of the International Symposium on Recent Developments in Olefin PolymerizationCatalysts, Tokyo, October 23-25, 1989 edited by T. Keii and K. Soga Volume 57A Spectroscopic Analysis of Heterogeneous Catalysts. Part A: Methods of Surface Analysis edited by J.L.G. Fierro Volume 578 Spectroscopic Analysis of Heterogeneous Catalysts. Part 6: Chemisorption of Probe Molecules edited by J.L.G. Fierro Volume 58 Introduction t o Zeolite Science and Practice edited by H. van Bekkum, E.M. Flanigen and J.C. Jansen Volume 59 Heterogeneous Catalysis and Fine Chemicals II. Proceedings of the 2nd International Symposium, Poitiers, October 2-6, 1990 edited by M. Guisnet, J. Barrault, C. Bouchoule, D. Duprez, G. Perot, R. Maurel and C. Montassier Volume 6 0 Chemistry of Microporous Crystals. Proceedings of the International Symposium on Chemistry of Microporous Crystals, Tokyo, June 26-29, 1990 edited by T. Inui, S. Namba and T. Tatsumi Volume 6 1 Natural Gas Conversion. Proccfedings of the Symposium on Natural Gas Conversion, Oslo, August 12-1 7, 1990 edited by A. Holmen, K.-J. Jens and S. Kolboe Volume 62 Characterization of Porous Solids II. Proceedings of the IUPAC Symposium (COPS II),Alicante, May 6-9, 1990 edited by F. Rodriguez-Reinoso, J. Rouquerol, K.S.W. Sing and K.K. Unger Volume 63 Preparation of Catalysts V. Proceedings of the Fifth International Symposium on thescientific Bases for the Preparation of HeterogeneousCatalysts, Louvain-laNeuve, September 3-6, 1990 edited by 0. Poncelet, P.A. Jacobs, P. Grange and 6. Delmon Volume 6 4 New Trends in CO Activation edited by L. Guczi Volume 65 Catalysis and Adsorption by Zeolites. Proceedings of ZEOCAT 90, Leipzig, August 20-23, 1990 edited by G. Ohlmann, H. Pfeifer and R. Fricke Volume 66 Dioxygen Activation and Homogeneous Catalytic Oxidation. Proceedingsof the fourth International Symposium on Dioxygen Activation and Homogeneous Catalytic Oxidation, Balatonfured, September 10- 14, 1990 edited by L.I. Simandi Volume 67 Structure-Activity and Selectivity Relationships in Heterogeneous Catalysis. Proceedings of the ACS Symposium on Structure-Activity Relationships in Heterogeneous Catalysis, Boston, MA, April 22-27, 1990 edited by R.K. Grasselli and A.W. Sleight Volume 68 Catalyst Deactivation 1991. Proceedings of the fifth International Symposium, Evanston, IL, June 24-26, 1991 edited by C.H. Bartholomew and J.B. Butt
