ADVANCES IN CATALYSIS VOLUME 11, Volume 11

ADVANCES IN CATALYSIS AND RELATED SUBJECTS

VOLUME XI

This Page Intentionally Left Blank

ADVANCES IN CATALYSIS AND RELATED SUBJECTS VOLUME XI

EDITED BY

D. D. ELEY Noltingham, England

P. W. SELWOOD

PAULB. WEISZ

Euanalon, Illinois

Paulaboro, N . J .

ADVISORY BOARD

PETERJ. DEBYE Zthaca, N . Y .

W. JOST

P. H. EMMETT

W. E. GARNER

Baltimore, Md.

Bristol, England

E. K. RIDEAL

Cfdtlingen, Germany

Londm, England

H. S. TAYLOR Princeton, N . J .

1959

ACADEMIC PRESS INC. NEW YORK AND LONDON

COPYRIW~T @ 1969, BY ACADEMIC PRESSINC. ALL RIGHTR RESERVED NO PART OF THIS BOOK MAY B E REPRODUCED I N ANY FORM B Y PHOTOSTAT, MICROFILM, OR A N Y OTHER MEANS, WITHOUT WRITTEN PERMISSION FROM THE PUBLISHERS.

ACADEMIC PRESS INC. 111 FIFTHAVENUE NEWYORE3, N. Y.

llnited Kingdom Edition

Published by

ACADEMIC PRESS INC. (LONDON)LTD. 40 PALLMALL,LONDONS.W.1

Library of Congress Catalog Card Number 40-7766

PRINTED I N T H E UNITED STATES OF AMERICA

CONTRIBUTORS TO VOLUME XI

L. G. AUSTIN,Fuel Technology Department, The Pennsylvania State University, University Park, Pennsylvania J. J. CHESSICK, Surface Chemistry Laboratory, Lehigh University, Bethlehem, Pennsylvania

R. V. CULVER, Department of Metallurgical and Chemical Engineering, University of Adelaide, Adelaide, South Australia J. HALPERN, Department of Chemistry, University of British Columbia, Vancouver, B.C., Canada G. KEMBALL, Department of Chemistry, The Queen’s University of Belfast, Belfast, North Ireland G. NATTA, Istituto d i Chimica Industriale del Politecnico, Milan, Italy

I. PASQUON, Istituto d i Chimica Industriale del Politecnico, Milan, Italy

FRANK RUSINKO, JR., Fuel Technology Department, The Pennsylvania State University, University Park, Pennsylvania F. C . TOMPKINS, Chemistry Department, Imperial College of Science and Technology, London

P. L. WALKER, JR., Fuel Technology Department, The Pennsylvania State University, University Park, Pennsylvania A. C. ZETTLEMOYER,Surface Chemistry Laboratory, Lehigh University, Bethlehem, Pennsylvania

This Page Intentionally Left Blank

Catalysis remains a fascinating meeting ground of knowledge and experience, of theories and of experimentation belonging to various disciplines of science. Catalysis is not a science, it is a phenomenon. It may arise in connection with a vital biochemical process, an industrially important chemical reaction, an interesting intramolecular rearrangement, a mesoninfluenced nuclear change, a combustion process, an atomic spin transmutation, or innumerable other rate processes. The phenomenon of catalysis arises in connection with many scientific endeavors, and it involves many and diverse scientific principles. In this volume of the Advances in Catalysis and Related Subjects, we have continued our attempt to collect progress and integrated knowledge toward a better scientific understanding of catalyzed rate processes. A major area of catalytic experience has evolved in the discoveries and observations of man-made stereospecijic chemical reactions (The Kinetics of the Stereospecific Polymerization of a-Olefins), a field of human endeavor which approaches closely on nature’s ability to synthesize macromolecules with great specificity. On the other end of the molecular weight spectrum, the activation of hydrogen dominates as a basic phenomenon a variety of inorganic, organic and bio-organic rate processes, and recent advances make an integrated review appropriate (The Catalytic Activities of Hydrogen in Homogeneous, Heterogeneous, and Biological Systems). Exchange reacticms using (isotopic) hydrogen, have long been important tools in the study of catalytic molecular interchanges, and a review of such exchange phenomena with hydrocarbons is made in the article on Catalytic Exchange of Hydrocarbon with Deuterium. The electronic properties of solid catalytic surfaces have been the subject of much attention in recent years. The importance of the electronic phenomena in the interaction of reactants with catalyst make it imperative to follow, from year to year, at least some of the important developments in this field (Surface Potentials and Adsorption Process on Metals). Furthermore, there is a constant need and a search by the catalytic researcher for tools to examine the locus and sites of heterogeneous catalysis, viz. the surface, and its physical-chemical nature; an examination of new methods of examination and of the type of information it may produce, appears to us to be an important function of this publication (Immersional Heats and the Nature of Solid Surfaces). One elemental solid which has been important in relation to rate processes occurring on its surface is carbon. It has played roles as catalyst, as catalyst vii

viii

PREFACE

support, and as a reactant. Even as a reactant, it may play a catalytic role in relation to individual steps of its self-reaction and many basic phenomena are common to gas-solid rate processes and the solid catalyzed gas conversion process (Gas Reaction of Carbon). The Editors always face the imposing boundary condition of a practical and finite size for each volume, which always makes the ratio of material that could be covered to that which should be covered smaller than unity. They face therefore a constant but inevitable obligation to express regret concerning the accomplishments and areas of interest which could not be included in the present volume.

P. B. W. September, 1969

CONTENTS CONTRIBUTORS TO VOLUMEXI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PREFACE ....................................................................

v vii

The Kinetics of the Stereospeciflc Polymerization of a-Oleflns

BY G . NATTAA N D I . PASOUON

I . Introduction to Anionic Coordinated Polymerization of a-Olefins . . . . . . . . I1. Over-All Kinetics of Polymerization Process ............................ I11. Chain Transfer, Termination Processes, and Molecular Weight of the Polymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV . Steric Composition of Polymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V . Determination of the Number of Active Centers . . . . . . . . . . . . . . . . . . . . . . . . VI . Mean Lifetime of the Growing Polymeric Chains . . . . . . . . . . . . . . . . . . . . . . . VII . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 10 23 46 50 60 64 65

Surface Potentials and Adsorption Process on Metals

BY R . V . CULVERA N D F . C . TOMPKINS I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I1. The Surface Properties of Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I11. The Modification of Surface Properties by Adsorbates . . . . . . . . . . . . . . . . . . IV . The Preparation of Clean Metal Surface V . The Measurement of Work-Function Changes . . . . . . . . . . . . . . . . . . . . . . . . . . V I . Adsorption and Work-Function Studies . VII . Surface-Potential Data ............................................ VIII . Electron Transfer and Bond Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I X . The Kinetics of Desorption and Surface Migration . . . . . . . . . . . . . . . . . . . . . . . X . Surface Reaction Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X I . Heats of Adsorption .................... X I 1. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

68 73 77 82 101 106 111 118 127

Gas Reactions of Carbon

BY P . L . WALKER.JR., FRANK RUBINKO.JR., A N D L . G . AUSTIN I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I1. Thermodynamics of Gas-Carbon Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I11. Review of General Mechanisms for the Gas-Carbon Reactions . . . . . . . . . . IV . Review of Kinetics for the Gas-Carbon Reactions . . . . . . . . . . . . . . . . . . . . . . V . Role of Mass Transport in Gas-Carbon Reactions . . . . . . . . . . . . . . . . . . . . . . VI . Use of Density and Area Profiles on Reacted Carbon Rods for Better Understanding of Gas-Carbon Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII . Some Factors, Other than Mass Transport, Which Affect the Rate of GasCarbon Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

134 135 138 153 164 178 201

X

CONTENTS

Appendix.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

212 217

The Catalytic Exchange of Hydrocarbons with Deuterium

BY C. KEMBALL

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. General Aspects of Exchange Reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. The Exchange of Molecules Possessing a High Degree of Symmetry.. . . . IV. The Exchange of Molecules Possessing 8 Low Degree of Symmetry.. . . . . V. Some Results with Unsaturated Molecules.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Concluding Remarks.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

223 228 239 250 257

259 261

lmmersional Heats and the Nature of Solid Surfaces

BY J. J. CHESSICKA N D A. C. ZETTLEMOYER

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

263

. . . . . . . . . . 268

VII. Site Energy Distribution.. . . . .............. VIII. Solution Adsorption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . .................

291

The Catalytic Activation of Hydrogen in Homogeneous, Heterogeneous, and Biological Systems

BY J. HALPERN I. Introduction.. . ............................. 11. Homogeneous .......................... 111. Heterogeneous IV. Biological Systems. . . . . . . . . . . . . . . ..................... V. Concluding Remarks.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......... References. . . . . . . . . . . . . . . . . . . . ....................... AUTHORI N D E ,X , , ,. . SUBJECT INDEX, ............................. ........................

358 363 365 381

The Kinetics of the Stereospecific Polymerization of a-Olefins G. NATTA

AND

I. PASQUON

Ialiluto di Chimica Induelriale del Politecnico, Milan, Italy Pw 2

I. Introduction to Anionic Coordinated Polymerization of a-Olefins . . . . . . . . A. Generalities.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Stereospecificity of Catalyst .......................... C. Mechanism of Stereospecific merization. . . . . . . . . . . . . . . . . . . . . . . . . . D. Influence of the Crystalline Substrate on the Stereospecific Polymerization. . . . . . . . . ................................................... 11. Over-All Kinetics of Polymerization Process ..... ... A. Catalytic Systems Used. ...................................... B. Influence of the Sizes of 8 Crystals on the Polymerization Rate. Adjustment Period.. . . . . ..................................... C. Catalytic Behavior of aStereospecific Propylene Polymeriza.................................... itions on the Polymerization Steady..................................... State Rate . . . . . . . . . . . . . . 111. Chain Transfer, Termination Processes, and Molecular Weight of the Polymers ......................... A. Catalysts Used and Their B. Independence of the Molecular Wei Polymer of Reaction Time for, Long Reaction Times.. . . . . . . . . . . . . . . . . C. Chain Transfer Process Depending on Alkylaluminum Concentra-

..........................................

6

9 10 10 11 16

17

26

26

D. Chain Transfer Process Depending on the Amount of Titanium Com-

................................................ 32 E. Chain Transfer Process Depending on the Propylene Partial Pressure.. 34 F. Influence of the Temperature on the Single-Chain Transfer and Termina....................................... tion Processes. . . . . . G. Comparison betwee trinsic Viscosity and Specific Radioacti the Polymer Obtained in the Presence of I4C-Labeled Trialkylalumi...... ........... num H. Relative Importance of the Different Chain Transfer Processes. . . . . . . I. Relation between Intrinsic Viscosity and Number Average Polymerization Degree for Polypropylene.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. Remarks on the Catalytic Nature of the Coordinated Anionic Catalysis ............ ............... IV. Steric ................................. V. Determination of t ............. 1

41 43 44 46

46 50

2

G. NATTA AND I. PASQUON

Pwe

A. Adsorption of W-Labeled Alkylaluminums on a a-Titanium Trichlo................................................ ride Number of Active Centers by a Kinetic Method.. . B. n e t VI. Mean Lifetime of the Growing Polymeric Chains. . . . . . . . . . . . . . . . . . . . . . . . A. Determination of the Mean Lifetime from the Number of Active Centers ...................... ....... B. Variation of the Molecular Weight during the Polymerization. . . . . . . . . . C. Block Copolymers (Hetero ). .......................... VII. Conclusions. . . . . . . . . . . . . . . . . . ............................ References... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

50 56

60 60 61 64 64 65

1. Introduction to Anionic Coordinated Polymerization of a-Olefins A. GENERALITIES Considerable interest has been shown in the new processes of stereospecific polymerization, not only so far as they concern the production of new classes of polymers, having unusual characteristics and improved properties, but also because they are representative of a peculiar new type of heterogeneous catalysis, of great interest from the practical and the theoretical points of view (1-5). The discussion of kinetic work will be here preceded by a summarized description of the chemical nature of the polymerization, to which we have attributed a mechanism of anionic coordinated type. Such a definition of the reaction mechanism depends upon the fact that the catalyst is a complex in which, generally, a transition metal acts as a coordinating agent and that a carbon atom, which belongs t o the extremity of a growing polymeric chain, is coordinated to such a complex and, in the activated state, it possesses a negative charge. The stereospecific polymerization of a-olefins takes place only in the presence of heterogeneous catalytic systems, including a crystalline substrate (formed by halides of transition metals, such as T i c & , TiCb , VCh , CrCL , CoClz , etc.) and a suitable metallorganic compound ( 5 ) . Such metallorganic compound or coordination complex contains an electropositive metal for which the carbon-metal bond may be considered a t least partly polarized, so that the carbon atom has a partially ionic character and behaves as a carbanion. The above-mentioned metallorganic compounds must have the property of forming complexes with the halides of transitions metals. It is required, in order to get catalytic complexes, that the metal of metallorganic compounds be able to create a strong localized electric field; therefore, metals having a very small ionic diameter (below 1 A.) jointly with a very electropositive character are to be used. For such reasons, metals such as Cu,

KINETICS OF POLYMERIZATION OF LU-OLEFINS

3

Ba, Sr, K, Rb, Cs, although they have high electropositivity, cannot be employed, their ionic radius being too large, whereas other metals such as B, with a small ionic radius, are not so suitable, because they show an insufficient electropositivity (6, 6). The most suitable metallorganic compounds are those of A1 or Be, since these metals are characterized by small ionic radius [for instance, Al(CzHa)3, A1C1(CzH& , Al(iCdH&, Be(CzH&]. Less efficiency is shown by certain metallorganic complexes containing Zn, Li, etc. (6, 6.) The formation of prevailingly electron-deficient complexes between transition metals of low valency and metallorganic compounds of metals having small ionic radius has been clearly shown. The electron-deficient metallorganic compounds I

R

R

R

\ / \ / \ / \ / Be

Be

Be

Be

I1

where R is an nlkyl group, polymerize ethylene to low-molecular-weight polymers (’7). Crystallizable complexes containing transition metals of the following general formula have been isolated : I11

where R1 is a halide or an alkyl group and Rz is an alkyl group. They polymerize ethylene to high-molecular-weight polymers (8).

B. STEREOSPECIFICITY OF CATALYSTS The metallorganic compounds (I, 11) employed in presence of a heterogeneous phase containing an amorphous compound of a low-valency, strongly electropositive transition metal, generally polymerize a-olefins to amorphous polymers. In a similar fashion, the soluble reaction products of such metallorganic compounds with compounds of transition metals, chemisorbed on amorphous substrates, polymerize a-olefins to amorphous polymers (6, 9). The same compounds (I, 11) employed in presence of solid crystalline halogenated compounds of some transition metals behave as stereospecific

4

0. NATTA AND I. PASQUON

TABLE I 8lereospeci$city of Cala~!ylicSystems: a-Tic1&(CZ&.),, t = 75", PC,H, 2.4 atm.)

-

Metal of the metal alkyl compound

Ionic radii of the metal, A.

Polypropylene not extractable in boiling n-heptane yo

Be A1 Mg Zn

0.35 0.61 0.60 0.74

94-90

80-90 78-86 30-40

catalysts and polymerize the a-olefins to crystalline polymers (6, 9). A greater stereospecificity is shown by catalysts containing metallorganic compounds of metals with very small ionic radius (see Table I) (6, 6 ) . Complexes of type I11 and also traces of soluble halides of strongly electropositive transition metals, being able to form complexes with metallorganic compounds of the type I, 11, increase the activity of the stereospecific catalysts formed by the action of metallorganic compounds on crystalline substrates (10, 11). They can also polymerize in a stereospecific way in the presence of crystalline substrates of transition metals (for instance, CoC12) which are not by themselves sufficiently electropositive, (when used in the presence of metallorganic compounds) to polymerize the a-olefins (10, 11). The stereospecific catalysts polymerize a-olefins, giving linear polymers, by head-to-tail addition containing long sequences of monomeric units, whose carbon atoms show the same relative steric configuration. The nonstereospecific catalysts, on the contrary, give chains whose monomeric units follow each other in a random or not ordered way as far as it concerns the relative steric configuration. Each molecule of a-olefin, at the moment of polymerization, may give rise to two types of monomeric enantiomorphous units, which differ only for the steric configuration, one being the mirror image of the other one (Fig. 1). Only the heterogeneous catalysts, and in particular those acting on a crystalline substrate, contain active centers, each of which makes an asymmetric synthesis, as it converts the monomer molecules which do not yet contain atoms of asymmetric carbon, at the moment of the polymerization, into monomeric units having all the same steric configuration (19,6). An asymmetric structure has been ascribed to such active centers, that justifies their behavior as catalysts of asymmetric synthesis (6). In a heterogeneous not optically active catalyst there is the same probability that an active center shows a given steric structure or the enantiomorphous one; it follows that one half of the present active centers will cause a given configuration of monomeric units (for instance, right-handed) and the other half will cause the opposite configuration (left-handed).

KINETICS OF POLYMERIZATION OF a-OLEFINS

5

monomer having planar srruc ture

k monomeric enantiomorphous units

FIQ.1.

There will be in a raw polymer the same number of chains, whose monomeric units show, with regard to a certain terminal group, a given structure (for instance, right-handed) and the eame number with the opposite steric structure (left-handed). The raw polymers of this type have been called isotactic and may in general crystallize (1).They differ for this reason from atactic polymers, in which the monomeric units follow in the same chain with random steric configuration and are unable to crystallize. In an isotactic chain having a very great length (considered as being of infinite length), there will be no more asymmetric carbon atoms because the asymmetry of the tertiary carbon atoms, which was due to the different structure or length or configuration of the two parts of the chain linked to it, disappears (6). Such carbon atoms show, however, the same steric configuration of the tertiary carbon atoms which follow or precede them, and this makes them different from atactic polymers (6, 13, 14). Before the discovery of stereospecific catalysis, all the known polymers of a-olehs were unable to crystallize, because their structure was chemically irregular (for instance, not rigorously linear or not rigorously head-to-tail) and because it was sterically irregular (16). A difference of structure between isotactic and atactic polymers exists independently of their physical state. If we could, in fact, stretch on a horizontal plane a sterically regular segment of the main chain of an isotactic polymer of the type (CHR-CHs), the R groups linked to the tertiary carbon atoms would range themselves all over or under such plane, whereas in an atactic polymer they would be arranged in a random way, partly over and partly under it (Fig. 2). The chains of isotactic polymers (in which the dimensions of R are much greater than those of the hydrogen atom) have the tendency to assume a helicoidal configuration with a pitch which depends on the dimensions of the R group. There is evidence that a helicoidal structure has the tendency to exist also (at least partially) in the amorphous state. It is detectable,

6

G. NATTA AND I. PASQUON

H

R

H

Fro. 2. Chains of stereoisomeric poly-a-olefins supposing the main chain stretched on a plane. I. Isotactic. 11. Syndiotactic. 111. Atactic.

however, in the crystalline state. In the crystals we may find macromolecules of isotactic polymers with a helix-shaped arrangement, which is ternary (in the case of polypropylene, polystyrene, modification I of polybutene, etc.), quaternary (in the case of poly-3 methyl-butene-I), or heptenary (in the case of poly-4-methylpentene-1, etc. (Fig. 3) (18, 1 7 ) . The helices may show a right- or left-handed winding (independent of the steric configuration of the tertiary carbon atoms. In the crystal lattice of many isotactic polymers, there may be found a chain packing, characterized by the fact that each right-handed chain is surrounded by lefthanded chains and vice versa (Fig. 4) (18, 19). To the high regularity of structure must be ascribed the exceptional properties of isotactic polymers (high melting point, high mechanical characteristics, possibility to form films or fibers made of oriented crystals having high tensile strengths). To them is attributed the great interest arisen in the fields of plastics and synthetic fibers (6).

C. MECHANISM OF STEREOSPECIFIC POLYMERIZATION The process may be ascribed to the coordinated anionic type. Such a process which leads to the addition of a molecule of monomer in a polymeric chain, may be considered as divided into several consecutive steps.

KINETICS O F POLYMERIZATION OF (U-OLEFINS

0

R

I R = - c H ~ . - ~ H.-CH-CH* , -CH*-CH~-CH-(CH& -o-cH,: ~-c~-cH-(cH,),

n

7

UI

R.c~-cH-(cH,)-c,H,

-cH~-cH-wI,)~

R* -CH-bb)*.

-0

FIG.3. Chains of isotactic polymers.

The electron-deficient catalytic complex, containing a transition metal, has the tendency to attract the olefin molecule, whose ?r-electrons tend to compensate the deficiency of electrons of the complex. The catalytic complexes may possess a type of structure as follows:

8

a.

NAWA AND I. PASQUON

Rr

R8

\ / \

Ra/M’\

/

Rs

/”.\ Re where M t is the transition metal and Mp is the strongly electropositive metal to which the alkyl groups are bound. Only when such a complex is chemisorbed or lies on the surface of a crystalline lattice made of a compound of a transition metal does the catalyst act in a stereospecific way in the polymerization of a-olefins. In the first reaction step, the olefin is strongly polarized by the catalyst, as follows: RHC=CH,

R4

00 -, RH&--CH~

At the same time, a dissociation of ionic type of the bridge bond takes place. The bridge bond M-R-M is, in fact, weaker, as demonstrated by

FIG.4. Projection of the crystalline polypropylene lattice, on a plane perpendicular to the axis of the polymeric chains.

KINETICS OF POLYMERIZATION OF (r-OLEFINS

P

R

9

P

e

-

I+!..’

CH2 {.yR

(-I 1+1 CH2-CHR ’ I-(\

RinM!.

,CH2P

__c

Rkfl{.;

,*

.‘/c$?.

‘?l2 R i

.

.\R----M*R 2 n

.

a

I

‘R’

FIQ.6. Hypothesis of addition of a monomer molecule on the bond between the catalytic complex and the growing chain, in the anionic coordinated polymerization.

the greater length of this bond between metal and R group compared with the bonds between side groups R and the metal. This was observed by X-ray examination of different complexes, containing bridge bonds (20, 21). The introduction of a monomeric unit occurs between the electronegative -CH2 at the chain end and the electropositive metal. A new -CH2 group deriving from the new monomeric unit, comes and substitutes in the complex the -CH2 group of the previous monomeric unit (see Fig 5). The reaction of polyaddition is represented by the following equation: (+I (-) [Cat]--CH2P

+ CH,=CHR

(+)

(-1

+ [Cat]--CHd3HRC&P

The termination of the growing polymeric chain may occur through several different processes, mostly by chain transfer. Either the process of chain transfer with the monomer, or the reaction of dissociation to hydride, leads to the formation of terminal vinylidenic groups, whose presence was noticed in the olefin polymers, obtained with the previously described catalysts (22)*

The chain termination processes will be described in detail in the following sections, dealing with the kinetic study of polymerization process.

CRYSTALLINE SUBSTRATE ON THE STEREOSPECIFIC POLYMERIZATION The stereospecificity depends not only upon the electropositivity and the ionic radius of the metal which belongs to the metallorganic compound, used for the preparation of the catalyst, but also upon the lattice structure of the crystalline substrate made of the transition metal compound (6). Besides the chemical composition, the crystalline structure of the substrate exerts a great influence on the stereospecificity of the catalyst. For instance, the halides of the type MX,,which crystallize with layer

D. INFLUENCE OF

THE

10

0. NATTA AND I. PASQUON

FIQ.6. a-TiCls and TiC12crystalline lattices.

lattice (a-TiC13, T i c & , vc13, etc.) (Fig. 6) (23, 24) are all suitable for the preparation of stereospecific catalysts (9,5,25,26'). In the case of TiC13 the less crystalline forms obtained a t low temperature by precipitation from solution of Tic14 and alkylaluminum are less stereospecific than the well-crystallized forms obtained a t high temperature ( 5 ) .TiCl3 crystallizes in three forms at least (27), and the greatest stereospecificity is given by the a-form. For this reason the catalytic systems, which we have mostly employed for studies of stereospecific polymerization of a-olefins, are those made using a-TiCls. II. Over-All Kinetics of Polymerization Process

A. CATALYTIC SYSTEMS USED As already stated in the first chapter, several catalytic systems show n certain stereospecificity in the a - o l e h polymerization.

KINETICS O F POLYMERIZATION O F a-OLEFINS

11

These systems may be differentiated either by the nature of the compound of the transition metal or by the type of the metallorganic compound used for their preparation. The behavior of the different catalytic systems (containing transition metal crystalline compounds) in the a-olefin polymerization, except for the different degree of stereospecificity,may be connected with a definite kinetic scheme. This was shown by experimental work performed a t the Institute of Industrial Chemistry of the Milan Polytechnic. When the compound of the transition metal is changed (e.g., a-TiC13, &Ticla), generally, the molecular weight of the resulting polymer changes. Also the nature of the alkyl group of the metallorganic compound influences the stereospecificity and the molecular weight of the polymer obtained (28). The nature of the olefm exerts a certain influence on the rate constant of the over-all polymerization. This is connected with factors of steric character and to the more-or-less enhanced electron-releasing character of the alkyl group bound to the vinyl group which may influence several steps of the over-all polymerization process. Although the catalysts containing beryllium alkyl are more stereospecific than those with alkylaluminum (6, S ) , nevertheless the greater part of our kinetic measurements were performed using alkylaluminum compounds, since they represent a special practical interest due to the higher availability, and lower toxicity compared with the corresponding beryllium compounds. The results summarized in this paper have been obtained using the catalytic systems: Al(CnHs)t-a-TiCla-n-heptane or

A1(CzH&C1-a-TiCls-n-heptane.

The kinetic measurements reported in the following sections are concerned with the polymerization of propylene; the results obtained with this monomer can, however, be extended to other olefins (e.g., normal: butene-1, pentene-1, or branched). For this reason, although we limit ourselves to recording measurements made with one monomer only and with two types of catalytic system, we have given the most general title to this paper.

B. INFLUENCE OF THE SIZESOF a-Tic13CRYSTALS ON THE POLYMERIZATION RATE.ADJUSTMENT PERIOD The a-TiClr (violet modification), prepared by reduction of Tic14 with flowing hydrogen at high temperature (29), generally shows hexagonal lamellae whose sizes, depending on the method of preparation, lie in the range from 1 p to several hundred microns (see for instance the sample of Fig. 7). Sometimes the a-TiC13lamellae do not show any defined geometric shape, and their dimensions may reach a millimeter (see for instance the sample of Fig. 8).

12

Q. NATTA AND I. PASQUON

.-t t

0, 0

5

i0

15

20 25 h 30 polymerization time

FIG.9. Propylene polymerization rate at constant pressure and temperature as function of polymerization time ( p o , ~ = , 1,460 mm. Hg, 1 = 70"). 1

2 ~~

a-TiCli (sample A ) , g , / l . [A1(C2H1)8]mol./l.

0.80 4.46

x

1.00 10-3

2.94 X 10-1

In Fig. 9 the characteristic behavior of propylene polymerization rate is plotted us. polymerization time. The data were obtained by operating at constant pressure with a catalytic system containing a-TiCls crystalshaving initial sizes between 1 and 10 p (a-TiCls sample A). It may be noticed that during the initial polymerization period (adjust-

KINETICS OF POLYMERIZATION OF O-OLEFINS

13

51 0 Fro. 10. Propylene polymerization rate at constant pressure and temperature 1,450 mm. Hg, t = 70°C) obtained with two samples of unground a-TiClr whose crystals have different sizes (seeFigs. 7 and8). (a-TiC1, : 1.64 g./l., [AI(CIH~)~]: 2.94 X 10-1 mol./l.). ( ~ c , R= ~

ment period) the activity of the catalyst increases until it has reached a value which remains, afterwards, practically constant in the time. This adjustment period has been explained on the assumption that crystals and aggregates of a-TiC1, are smashed and cleaved under the mechanical action of growing polymeric chains, so that we have a consequent increase up to a constant value, in the number of active centers which directly participate in the polymerization. This assumption has been proved by the following experimental data: 1. The polymerization rate, under steady-state conditions, appeared to be almost independent of the initial size of the a-TiC13crystals (Fig. 10). 2. By operating with ground a-TiC13(sizes 5 2 p ) the adjustment period was definitely affected. The initial period, characterized by an increasing rate which might otherwise last for 7-8 hrs., was greatly shortened and modified (see Fig. 11) (SO, 31). 3. The rate that can be reached under steady-state conditions (for instance, by operating at 70") seems to be practically unaffected by a moderate amount of grinding (see Fig. 11). The effect of a moderate degree of grinding on the extent of a-TiCla active surface leads to a final result similar to that resulting from the mechanical disaggregation caused by the action of the growing polymeric chains. It may be assumed that in any case the final size of the cu-TiC13particles reaches approximately the same limiting value. On the other hand, it is most likely that the exceptionally small and active a-TiCla particles, obtained during the grinding, are unstable and lose their activity on ageing. In fact, at the beginning of the reaction, when using ground a-TiC13, the

14

Q. NATTA AND I. PASQUON

6

m 2

C

k

li

0 3

0 4

polymerization time

FIG.11. Effect of previous physical treatments on a sample of a-TiClo on the propylene polymerization rate, at constant pressure and temperature ( t = 70" pc,~,= 1,450 mm. Hg). 1 and 2: ground cuTiCls (sample A ) (sizes 5 2 p ) . 3 and 4 : unground aTiCla (sample A) (sizes within 1 to 10 p ) .

polymerization rate very quickly reaches a maximum and decreases nfterwards, more or less slowly, until attaining the steady-state condition. The presence of such a maximum may be ascribed to very small a-TiC13 particles which lose their activity during the polymerization, either by recrystallization, by reaction with Al(CzH&, or by occlusion in the solid polymeric product. In particular, it may be observed that the maximum disappears when operating with Al(CZH&Cl instead of Al(C*H& (32). 4. By operating with unground a-Tic13 the time ( t 3 1 4 ) which is necessary of the value of polymerization rate in steady-state condito reach the tions varies inversely with the polymerization rate measured under steadystate conditions (SO, 31). In fact, by comparing the polymerization carried out at different temperatures and pressures, referred to the same amount of a-TiC13, we observe different values of t314 mainly depending upon the overall polymerization rate, and consequently upon the value reached by the rate under steady-state conditions (Fig. 12) (31). 5 . The use, in the propylene polymerization, of unground a-TiC13 that had been previously maintained, for many hours, in the presence of solutions of A1(C2H6)3 a t temperatures lower than 80°, does not substantially modify the observed reaction rate and its variation during the adjustment period (31). 6. It has been observed by microscopic examinations that the a-TiC18 lamellae are very thin and brittle. 7. In some tests it has been observed that the polymerization rate at

15

KINETICS OF POLYMERIZATION OF a-OLEFINS

3 1

0.3

0.1

FIQ.12. Dependency of the t a l , index of the adjustment period, on the reciprocal of the propylene polymerization rate in Ateady state conditions. Tests performed with unground a-TiCla (sample A ) .

t , "C PCaHa

,mm. Hg

1

2

3

4

5

6

7

32 1,680

43 1,640

32 2,680

56 1,570

70 750

70 1,450

2,450

70

zero time is not zero; it keeps the initial value for several minutes before starting to increase (see, for instance, the lowest curve of Fig. 11). 8. It has also been found that the polymer formed from the beginning of the reaction is already prevailingly isotactic. This means that, from the start of the reaction, there exist a certain number of active centers on the solid a-Tic13 surface which immediately yield isotactic polymer; consequently, it can be excluded that, a t least for the active centers present on the initial free surface of a-titanium trichloride, there is an initial activation process, whose rate is slow enough to be observed even when operating a t low temperature (30"). The above statements are in good agreement with the fact that, after the reaction has been carried out in steady-state conditions and has been stopped by taking off the monomer, thereafter, when the initial value of monomer concentration has been re-established (Fig. 13), the reaction starts

16

G . NA'M'A

0

20-

1

1 I

I

AND I. PASQUON I,

mohtaincd for 14 t~a t 70% and pc$64

I

16

17h

j

again a t once and at the same steady-state rate. The lowest curve of Fig. 13 shows that the steady-state rate at a given temperature (below SOo) is the same also if the steady-state conditions had been previously reached a t lower temperature (30,31). C. CATALYTIC BEHAVIOR OF a-TiCl3 IN STEREOSPECIFIC PROPYLENE POLYMERIZATION

From the curve reported in Fig. 9, we may observe that the polymerization rate, obtained by operating at constant pressure, after the initial udjustment period, remains practically constant for many hours. This occurs only when operating with pure reagents and solvents, with not too finely ground cr-TiC13, and under such conditions as to get limited polymerization rates per unit volume solvent (some g. of C3Halhr. per liter of solvent) (SO, 33). This time-constant rate is proportional to the a-TiCla amount which proves that, at least formally, the over-all polymerization process is really a catalytic one, with regard to the a-TiCla. The catalytic behavior of a-TiCls is, in any case, connected with the existence on its surface of metallorganic complexes which act in the polymerization only if a-TiC13 is present. This makes stereospecific polymerization processes (of coordinated anionic nature) very different from the better known polymerization processes, initiated with free radicals. In the latter process, the initiator is not a true catalyst, since it decomposes during the reaction, forming radicals which are bound to the dead polymer; on the contrary, in the case of stereospecific polymerization, each molecule of polymer, at the end of its growing period, can be removed from the active center on the solid surface of the catalyst which maintains its initial activity. There is evidence that each active center which initiates a polymeric chain (coordination complex between the titanium salt and a metal-dkyl

KINETICS O F POLYMERIZATION OF (U-OLEFINS

17

compound) retains unaltered its ability to form macromolecules, independent of the number of polymer molecules produced. Many homogeneous catalytic processes, in particular of anionic nature, are known, in which the polymerization takes place by stepwise addition (polymerization of ethylene oxide (34) of ethylene at low pressure and temfor which perature with AIRa (7, 36),of styrene by Szwarc catalysts (M), the growth of the macromolecule can last for a very long time). This led some researchers to talk of a life of macromolecules and of living molecules (37). This attribute is justified by the fact that the growth of the macromolecules does not show any termination; it stops when the monomer is removed, but is resumed immediately at the same rate when the monomer concentration is restored to its initial value. In some cases (e.g., the case of “living polymers” of Szwarc, obtained with anionic catalysts), it is exactly the same macromolecule which continues to grow, yielding polymers whose molecular weight increases with the polymerization time. In the case under examination (heterogeneous catalysis in the presence of coordinated polymetallic complexes) the molecular weight of the polymer is generally almost independent of the polymerization time, whenever the polymerization lasts for more than about 10 min. The macromolecules bound to the catalytic complex can be detached from the active center, but their detachment leaves unchanged the activity of the catalytic center which can initiate the formation of another macromolecule.

D. INFLUENCE OF

THE OPERATINGCONDITIONS ON THE TION STEADY-STATE RATE.

POLYMERIZA-

1. Experimental Apparatus and Operating Conditions. The polymerization of propylene in the presence of a heterogeneous catalyst and a solvent occurred at a relatively low partial olefin pressure and was carried out in an apparatus continuously fed during the reaction with the olefin in the gaseous state at constant pressure (Fig. 14). The amount of olefin consumed was determined by the decrease of pressure with time, measured on the feed vessel, kept at constant temperature by water circulation, where the olefin was maintained in the gaseous state, It has been said that the polymerization rate observed under steady-state conditions, with a given sample of a-TiC1, is practically independent of the initial sizes of the crystals. It is, moreover, convenient to point out that not all samples of o-TiCls prepared by the different methods we have examined, lead in all cases to rates equal to each other. The most active samples of a-TiCla have an activity that does not exceed three times the value given by less active samples which we have here examined. As the initial sizes of a-TiCla crystals seem to have very little influence

18

Q. NATTA AND I. PASQUON

FIG.14. Apparatus used for kinetic measurements of propylene polymerization (reaction vessel rocking 45 times per min. through a 45' angle). PI = pressure gage, PC = pressure control, FI = flow indicator, TC = temperature control.

on the steady-state rate, such differences could be ascribed to the degree of purity of a-titanium trichloride. The most frequent impurities of commercial a-titanium trichloride are generally other chlorides (Tic14 ,TiC12),metallic titanium, titanium nitride, and the products resulting from oxydation or hydrolysis of the titanium chlorides, the latter being unstable at air and moisture. Some of these impurities have opposite effects on the catalytic activity and stereospecificity,depending on their concentration. As we shall show below, the stereospecificity of the catalytic system can be influenced by impurities contained in the a-titanium trichloride. The larger part of the results reported in this paper, have been obtained with an old sample of a-titanium trichloride called a-TiCl,--sample A. This sample is neither one of the most active nor one of the most stereospecific products we have studied. The analytical tests carried out on the product, have given the following results: Methanol insoluble residue: 1%

Ti :C1 ratio

= 1:2.96 in g. atoms

Kinetic data have been obtained with unground a-titanium trichloride, operating at constant temperature and pressure of olefin, during the whole polymerization.

KINETICS OF POLYMERIZATION OF a-OLEFINS

19

In some preliminary tests an examination was made of the influence of some physical factors on the reaction rate in the apparatus employed, such as mass and heat transfer depending on the degree of filling and stirring of the reaction vessel (33). It has been found that, for a given temperature, with the equipment used (see Fig. 14), there is a limiting rate, depending on the volume of the solvent, at which the mass and heat transfer phenomena become determining; operating, for instance, at 70°, in 250 cc. of solvent, the limiting rate is almost equal to 20 g. of polymerized CSHe per hour (33). All kinetic tests have been carried out at polymerization rates lower than this value. The order in which the components of the catalytic system (a-titanium trichloride and trialkylaluminum), the solvent (n-heptane), and the olefin are brought together has no practical influence on the polymerization rate. The rate values are independent of the temperature at which the catalyst is prepared by the action of alkylaluminum solution on a-titanium trichloride, provided that this temperature is not higher than 70" and the concentration of the alkylaluminum in solution is not too low (above 0.5 X mol/l. n-heptane) (30, 33). Most of the kinetic results reported in this study refer to concentrations of trialkylaluminum in solutions higher than 1.4 X mol/l. of solvent. On account of the sensitivity of the catalysts to traces of moisture or oxygen, it is generally not suitable to operate with lower concentrations of alkylaluminum, because the latter acts also as a protector of the solid catalyst. However, by operating with very pure solvents and reagents, the concentration of AlR3 can be reduced to lower values mol/l.). Triethylaluminum/a-TitaniumTrichloride Ratio. Many tests have been carried out with different trialkylaluminum/titanium trichloride molar ratios (from 1 to 8.5) without any considerable difference in the kinetic results obtained with the considered a-Tic13 sample (Fig. 15). For ratios lower than 0.4, the data obtained are of uncertain interpretation, owing to the degree of purity of the solvents and reagents which have been used (33).For such low values of the ratio, the reaction rate against time initially increases, goes through a maximum, and then decreases rapidly without attaining a stationary value. The decrease of the catalytic activity is due to a consumption of A1R3 and, in fact, the activity can be restored with the addition of small amount of A1R3. Overlooking the anomalies due to the lack of absolute purity of the reagents, one must assign a zero reaction order with respect to aluminumalkyl concentration, in the range of the above reported conditions. The result is due to the fact that the alkylaluminum, in the concentrations considered above, is always in excess with respect to the number of active centers existing on the surface of the solid catalyst. [TriethylaZuminum]/[CaH6] Ratio. Using a triethylaluminum concentra-

20

G . NATTA AND I. PASQUON

1

2

3

4

5

1.64 u-TiCls (sample 3.80 4.36 3.80 0.80 A ) , g./l(Al(C2Hs)3],mol./l.2.95 X 10-'8.65 X 10-z11.80 X 10-z4.45 X 10-a5.90 X 10P 5.50 Al/Ti, mol. 1.18 3.10 4.80 8.50 [A1(CzHa)$1 0.048 0.143 0.190 0.072 0.095 [CsHsl

tion higher than about 1.4 X lo-' mol/l. and a triethylaluminum/a-titanium trichloride ratio higher than about 1, the value of this second ratio does not influence the kinetics of the over-all polymerization process. Therefore, in the range of variables tested: [triethylaluminum]/[CaHs] = 0.015 to 0.4,the formation of possible soluble alkylaluminum olefin complexes is not kinetically detectable (SO, 33). This is confirmed by the independence of the olefin solubility in solutions of alkylaluminum in hydrocarbon on the alkylaluminum concentration (58). Amount of a-TiC13.In Fig. 15 the polymerization rate, obtained a t constant pressure of olefin with different amount of cr-TiCL , is plotted US. time. The steady-state rate is found to be proportional to the amount of a-TiCla present in the reaction system (Fig. IS), in agreement with the heterogeneous nature of the catalysis (30, 33). Propylene Partial Pressure. The polymerization rate, under steady-state conditions (Figs. 17 and 18) is proportional to the partial pressure of propylene (SO, $3). Polymerization Temperature. Apparent Activation Energy Based on the Steady-State Rate.* The rates observed at different temperatures, referred

* I n our kinetic calculations, we refer t o the directly observed partial pressure of propylene, rather than to its fugacity, because over the temperature and pressure range examined, we can assume that partial pressures and fugacities are pructicully proportional. I n fact, from the literature data, the variation in propylene fugacity coefficient, in the range of our kinetic tests, is small (about 0.97 a t 30" and 2700 mm. Hg; about 0.99 a t 70"and 450 mm. Hg of propylene partial pressure).

KINETICS OF POLYMERIZATION OF a-OLEFINS

21

FIG.16. Dependency of propylene polymerization rate in steady-state conditions on the amount of a-TiCla (sample A ) in the catalytic system.

FIG.17. Propylene polymerization rate at constant temperature (70") and at different pressures as function of polymerization time (a-TiC13: sample A : 3.60 g J . ; [AI(CZH~)~]: 5.88 X mol./l.).

to a given amount of a-TiCla and to a given partial pressure of propylene, are reported versus polymerization time in Fig. 19. The diagram shown in Fig. 20, which gives the log of the polymerization rate under steady-state conditions, plotted us. the reciprocal of the absolute temperature, was drawn from the above data. The activation energy calculated from the data reported in Fig. 20 is about 10,000 cal./mol. The activation energy referred to the concentration of the olefin in the liquid phase can be deduced from the one referred to the pressure in the gaseous phase, by adding the solution heat of the propylene in n-heptane,

1500

1000

500

2000

2!

m m Hg

FIQ.18. Dependency of propylene polymerization rate in steady-state conditions on the propylene partial pressure (a-TiClr:sample A ) .

FIQ.19. Propylene polymerization rate at constant pressure ( p c , ~ = , 1,500 mm. Hg) at different temperature, as function of polymerization time. a-TiClr (Sample A ) ,

IAl(CaHs)iI, mol. 11.

g./l.

Al/Ti, mol. ~~

1 2

3 4 5 6 7 8 9

3.80 0.80 7.60 1.60 7.60 10.80

1.18 8.50 1.50

5.90

10-9

5.60

10-' 10-* 10-' 10-'

3.60 1.05

x

17.70 X 7.36 X 7.36 X 17.70 X 7.36 X

20.20

12.16 12.10 22

~

2.96 X 10-a 4.45 x lo-' 7.36 X 10-*

lea

0.52 2.20 0.94

KINETICS OF POLYMERIZATION OF &OLEFINS

23

FIG.20. Log of propylene polymerization rate in steady-state conditions as function of 1/T ( p c , ~ = , 1,500mm. Hg, a-TiCl8 : sample A ) .

equal to about 4000 cal./mol. Consequently, the activation energy measured from the steady-state rate and referred to the olefin concentration in the liquid phase corresponds to 14,000 cal./mol. (SO, 31). If we consider the results reported so far, the polymerization rate of propylene under steady-state conditions, catalyzed by the catalytic system Al( CoH6)3-a-TiC13-n-heptane, shows the following relation: =

~~-10,0001RT

GT~PC~H~

For the sample of a-TiC13to which the above data are referred = 2 x 107 e--lO,OOO/RT GTipca H 6

(1)

(2)

where r = polymerization rate under steady-state conditions (g. CaHa/h) pCaHe= partial pressure of propylene (atm.) GTi = g. a-TiC13in the catalytic system. Equation (1) is applicable to other samples of a-TiC13 by varying the value of A . For instance, the sample ( B ) shown in Fig. 7 shows a value A = 3.4 X lo'.

111. Chain Transfer, Termination Processes, and Molecular Weight of the Polymers

The chain transfer and termination processes have been studied by the following methods: Intrinsic viscosity measurements on the resulting polymer. Analysis of end groups of polymeric chains by chemical, radiochemical and physical methods ( I R examination). By operating particularly at 70" it has been observed that every transfer and termination process of polymeric chains involves in the same way

24

Q. NATTA AND I. PASQUON

(from a qualitative point of view) the growing macromolecules, independently of their steric structure. We must, however, notice that the molecular weight of the atactic polymers which are always present in small amount in the crude polymer, is generally much lower than that of the isotactic polymer. In fact, while the intrinsic viscosity of the isotactic polymer generally ranges from 2 t o 5, we found correspondingly 0.5-1 for the atuctic polymer. We shall examine in detail the influence of thedifferent factors controlling the intrinsic viscosity of the isotactic polymers during the polymerization. A. CATALYSTS USEDAND THEIRSTEREOSPECIFICITY The steric composition of the polypropylenes depends on the degree of purity of cu-TiCl3 used in the polymerization. It has been observed, for instance, that the so-called isotacticity index of polypropylene (polymer residue after extraction in boiling n-heptane ) can attain values ranging from 75 to YO%, depending on the catalytic properties of the samples of a-TiC13. Also the average molecular weight depends on the purity of the a-TiCl3 samples used. For instance, the same sample of a type of a-TiC13 which, in the raw state, during a 2-hr. polymerization test of propylene, gives polymers having an intrinsic viscosity equal to 1.5 after a series of washings with anhydrous hydrocarbons (before the polymerization tests) leads to polymers having intrinsic viscosities which increase with the number of washings, until they reach an asymptotic value of about 3.3. For that reason, the study of chain transfer and termination processes in propylene polymerization has been performed by using a standard type of a-TiC13 (sample A ) which is the same as that used in the previously performed kinetic tests, but treated as follows (31): Grinding in a stainless steel bottle, containing spheres of stainless steel (the dimensions of a-TiC13after grinding are 1 2 M ) . Washing with anhydrous n-heptane several times. The a-TiCls treated in this way gives reproducible results for the kinetic behavior, the molecular weights, and the steric composition of the polymer. The atactic amorphous portion (9-16% of the total) contained in the obtained polypropylene has been separated by treating the raw polymer with n-heptane a t room temperature. When operating in such a way, we have not separated the stereoblock fraction (extractable in boiling n-heptane) from the isotactic (not extractable in boiling n-heptane) fraction of polymer. The results reported in this paper are generally referred t o the crystalline fraction, named non-atactic, which contains also some stereoblock polymers (at the considered polymerization temperatures, the latter generally correspond only to 5-7 % of the total) ( 9 )

25

KINETICS O F POLYMERIZATION OF (r-OLEFINS

B. INDEPENDENCE OF THE MOLECULAR WEIGHTAND STERICCOMPOSITION OF THE POLYMER OF REACTION TIME,FOR LONGREACTION TIMES Many polymerization tests have been carried out under different conditions (temperature from 30 to 70" and propylene partial pressure from 450 to 1,450 mm. Hg). When operating under such conditions, we never observed any effect of polymerization time, on the molecular weight and steric composition of the polymer, either after a few minutes of polymerization (e.g., in the interval in which the reaction rate with ground a-Tic13 shows a maximum) or after many hours (when a small decrease in the over-all polymerization rate occurs) (Table 11). This means that, under the tested conditions, the growing time of each polymeric chain is not slow enough to be measured, on the basis of the above reported kinetic data. For this reason, by operating particularly a t 70" and with a propylene partial pressure of about 1 atm. and assuming that all active centers which are present on the surface of cr-TiC13 have the same activity, the average TABLE I1 Polymerization of Propylene to Zsotactic Polymer. Zndependenee of the Molecular Weight and of the Steric Composition from the Polymerization Time

'9

OC

PCaHe s mm. €16

IAI(C*Ft) rl, mol./l.

o-Tic1 I, (sample

U/Ti, niol.

Ahdl.

Polymerization time, hr.

x

1.50

3

1450

2.94 X

3.00

1.5

31

1450

2.94 X

1.50

3

51

1110

1.47 X 10-2

0.30

7.5

70

450

2.36 X

1.20

3

70

450

2.94 X 10-2

1.50

3

70

450

7.36 X

11.30

1

!4

70

950

2.94 X 10-2

1.50

3

>6

70

1450

1.77 X 10-2

700

31

%"

Intrinsic viscosity of le non-atactic polymer: ?I, 100 cm.:,JK.

89 88.5 85 86 86 84 87 87 90 90 91 91 91 90 88.5 88.5 88 88 87

4.40 4.34 4.66 4.65 4.70 4.90 4.50 4.45 3.56 3.47 3.28 3.14 2.18 2.13 3.84 3.78 3.84 4.16 4.20

~-

_______

31

Nonatactic polymer,

2.94

14 31 1 4 8 17 8 24 10 15 4 7 1

0.91

3

2 6 4 7

The data am related to the polymers insoluble in n-beptane at room temperature and include the stereoblock polymers soluble in boiling n-heptane ( b 7 % of the whole polymer) and the isotactic polymer.

26

Q. N A W A AND I. PASQUON

growing time of each macromolecule must not be longer than a few minutes (89, 40).

C. CHAINTRANSFER PROCESS DEPENDING ON ALKYLALUMINUM CONCENTRATION We have separately studied the nature of end groups and the dependence of the molecular weight of the polymeric chains, on the alkylaluminum concentration in the catalytic system. 1. Intrinsic Viscosities The measurements of intrinsic viscosity [q] have been carried out at 135' in tetralin. Under these conditions, the relationship between [q] and viscosimetric molecular weight M , for isotactic polypropylene is (41): [q] =

KMO."

In fig. 21, the values of l/[q]1'0.74= l/[q]1.36(this factor can be assumed as being proportional to the reciprocal of the degree of polymerization 2,) of the considered polypropylene fraction, are plotted vs. the square root of alkylaluminum concentration. The data plotted in Fig. 21, corresponding to constant quantities of a-titanium trichloride, can be assumed to give straight lines and the straight lines obtained for several values of CTi can be assumed to be parallel. The limiting line for CTi = 0 has been calculated from the data plotted in Fig. 22. The dependence of the intrinsic viscosity of the polymer (and consequently of its molecular weight) on the alkylaluminum concentration

*

,

I

0

02 0.3 (mots At (CpH&/I bheptme)"p

01

FIQ.21. Dependency of the reciprocal of the polymeriration degree (proportional to l/[q]lJb) of the non-atactic polypropylene fraction, on the square root of the aluminum alkyl concentration ( t = 70°, %,a, 9W mm. Hg, ground a-TiCla: sample A ) .

-

KINETICS OF POLYMERIZATION OF (r-OLEFINS

27

FIQ.22. Specific radioactivity (and corresponding values of -C2Hs mol. per mol. of polymerized CsHa) of the non-atactic polypropylene fraction, aa function of the square root of the alkylaluminum concentration. (Tests performed with "C-labeled A l ( C ~ H at ~ )1~= 70°, pcla, = 460 mm. Hg, ground cr-TiCls: sample A ) .

shows that this compound takes part (directly or indirectly) in a transfer or termination process of the growing polymeric chains. 2. Radiochemical Determination of End Groups

An attempt was made to determine whether the variations of the molecular weight with the alkylaluminum concentration are due to a chain transfer, with participation of the alkyl groups of alkylaluminum. Polymerization tests were performed, using 14C-labeled ethylaluminum and detecting the radioactivity of the polymer. Careful preliminary tests were necessary to demonstrate the suitability of these methods for the present study. In particular, it was necessary to verify that the radioactivity detectable in the polymer may not be caused by contamination or other processes different from the ones taken into consideration. It was then found that it is possible to remove throughly from the polymer the last traces of unreacted ethylaluminum or of its soluble complexes with titanium compounds by washing with anhydrous hydrocarbon. No alkylation of the preformed polymer caused by triethylaluminum or its derivatives has been observed (4.2). There is evidence that sometimes a permanent radioactive contamination of the polymer appears when one adds to a suspension of the polymer in n-heptane 14C-labeled triethylaluminum and certain samples of a-titanium trichloride (particularly when the a-TiClr contains Tic14 or other compounds of Ti( IV) , as impurities).

28

0. NATTA AND I. PASQUON

V'

-II, E

lXlU4

200 0:1 0:2 0.3. [mols A I [C~H~)J/I n- h ept ane] 'I2

Fxa. 23. Specific radioactivity (and corresponding values of -C1H6 mol. per mol. of polymerized C3Hd of the atactic polypropylene fraction, as function of the square root of the alkyl-aluminum concentration. (Tests performed with W-labeled Al(CnH6)a at t = 70", P C ~ H=~ 460 mm. Hg, ground a-Ticla: sample A ) .

This contamination remains after purification of the polymer from alkylaluminum by washing. Other samples (such as that shown in Fig. 7 ) are not contaminated. The amount of this contamination is in any case limited. This point will be discussed in one of the next paragraphs. All the reported data relating to radioactivity measurements have been corrected for radioactive contamination. The use of "C-labeled triethylaluminum allowed us to demonstrate that the quantity of -C2H6 groups [deriving from A1(C2H6)3]which is bound to the non-atactic polymer a t the end of the polymerization, when operating with a constant amount of titanium trichloride, is a linear function of the square root of the alkylaluminum concentration (Fig. 22), in the considered range of expeiimental conditions (42). Similar results have been obtained by analyzing the fraction of amorphous polymer (Fig. 23) (38). 3. Polymeric Isotactic Chains Bound lo the Aluminum

Determinations of aluminum have been carried out on fractions of polymeric product containing isotactic chains deriving from the polymerization. These measurements were performed in an attempt to establish whether the chain transfer process, depending on the alkylaluminum concentration, leads to the formation of macromolecules which remain bound to the aluminum. The polymer has been therefore purified by physical methods from the unreacted ethylaluminum and from the heterogeneous catalyst. The above determinations were carried out by taking into consideration

KINETICS OF POLYMERIZATION OF (Y-OLEFINS

29

a fraction of the polymeric product containing isotactic chains that may be easily separated from the catalyst in the following way (@) : The unreacted alkylaluminum and the atactic polymer were removed from the vessel in which the polymerization was carried out, by means of decantation and repeated washings with anhydrous n-heptane a t 50'; the separation of most of the polymer from the a-TiC13 was then made by washing a t 100" with anhydrous xylene. The polymer contained in the xylene solution was further purified by repeated precipitations and washings of the polymer precipitated with anhydrous xylene at - 70'. The aluminum was determined spectrophotometrically using 8-hydroxyquinoline in a known quantity of polymer obtained from the xylene solution and previously purified. The absence of titanium from the polymeric product, isolated in this fashion, was checked by conventional analytical methods. The results obtained are plotted in Table 111. I n the same table, for purposes of comparison, are recorded the amounts of ethyl groups derived from the Al(C?H5)3, as found in certain fractions of the polymer by the radiochemical methods. The amount of aluminum bound in the polymer is higher in the test performed with higher triethylaluminum concentrations. A comparison with the tests performed with the same alkylaluminum concentrations, but with different amount of a-TiC13, shows, on the contrary, that the amount of aluminum bound to the polymer decreases remarkably TABLE I11 End Groups i n the Isotactic Polypropylene, Deriving from the Chain Transfer Processes Depending on the Catalyst Concentration (Catalytic System: a-TiCl $-A1(C2H6)-n-Heptane) I

a-TiCIa (sample

4, g.

0.11 0.15 0.50 0.50 0.50

I

AI(CzHd:, mol.

250 cm.8

I , "C.

--Gbend groups/CaHo mol A1 atoms(CaH6 in the mol. In polymer fraction polymer fraction, insoluble in soluble in xylene n-heptane at at loOo room temperature*

x x x x x

Heptane Heptane Heptane Heptane Xylene

70 70 70 70 100

0 . 5 x 10-3 0 . 6 x 10-3 1.35 X 2.7 x 10-3 2 . 2 x 10-3

Polymerization conditions

9.9 9.9 9.9 39.6 9.9

10-3 10-3 10-3 10-3 10-3

Solvent,

2 . 3 x 10-3 2.3 x 10-3 2 . 7 X lo-' 4 . 9 x 10-3 -

-Cab end ~~OUPS/C:HS mol in the isotactic polymer fraction not extractable in boiling n-heptanec

1.4 X 10-3 1 . 4 X 10-3 1 . 7 x 10-3 3.0 X -

Average of the values (which lay in the range 1.3-1.4) obtained in four polymerization tests. Thia fraction contains the isotactic and stereoblock polymers. These data were calculated from the preceding ones, taking into account the percentage and the moleeular weight of the stereoblock polymers contained in the fraction extractable in boiling n-heptane. (I

30

0. NATTA AND I. PABQUON

decreasing the amount of a-TiC13. Probably some soluble titanium compounds act catalytically in the transfer of alkyls from the aluminum to the polymeric chains.

4. Mechanism of the Chain Transfer Process Depending on the Triethylaluminum Concentration The results reported in the preceding paragraphs indicate that the triethylaluminum molecules present in solution take part in a chemical process which affects the molecular weight of the polymers. In polymerization tests carried out in a relatively short time, the amount of reacted ethylaluminum is very small; this is proved also by the fact that the results obtained in these tests are not dependent on the reaction time. Furthermore, it has already been found that in the considered range, the over-all reaction rates do not seem to be affected by the concentration of ethylaluminum. It may be pointed out, therefore, that the ethylaluminum takes part in a chain transfer process. An almost linear dependence of l/[g]'.*' and of the number of -CzHs groups found in the polymer was found on the square root of the ethylaluminum concentration. The number of aluminum atoms chemically bound to the unpurified polymer decreases with the ethylaluminum concentration in the catalytic system. These results make us believe that the rate of the chain transfer process under examination is given by the relation: r3

= k&*C!I

(3)

where C A I = molecular concentration of triethylaluminum C* = number of active centers This dependence on the square root of the ethylaluminum concentration may be interpreted by assuming that the triethylaluminum acts in dissociated form in the chain transfer process. It is in agreement with the wellknown dimeric structure of triethylaluminum which is dissociated as follows : AIz(CnHa)a

2Al(C*Hs)a

(4 )

It has been also related by Bonitz ( 4 )that besides the homopolar dissociation, A12(CnHK)amay be partially dissociated in a heteropolar way: AI?(CzHs)s

+

[AI(C*Ha)al(+) [AI(C*Ha)r](-)

(5)

To explain the results obtained in our researches, we could take into consideration a chain transfer mechanism, such as the following one (at

31

KINETICS OF POLYMERIZATION OF (r-OLEFINS

least in initial stage) : [Cat](+)(-)CHICH(CHZCH),R

I

CH3 [Cat](+)

+ [AI(CZHK)Z](+)

.--)

AH8

(6)

+ (C2HK)zAlCHzCH(CH&H)nR AH3

+ [A1(C2HK)r](-)

[Cat](+)

4

AH1

+

[Cat](+)(-)CH2CH3 Al(CzH&

(71

It is possible that in the later stages of polymerization there may occur transfer processes involving more than one ethyl group per aluminum atom. I n this way, the catalyst will be regenerated while the monomer alkylaluminum takes part in the equilibrium of association with other alkyls in solution. Another interpretation that may be taken into consideration, kinetically equivalent t o the former one, is a total substitution of the polymeric alkylaluminum compound which is bound to a catalytically active complex containing the transition metal, as follows: TiCl.AZYzP

+ A1(CZH&

4

TiCI,A1(C?H6)3

+ AlYzP

(8)

where P = polymeric chain

Y = alkyl group (e.g., ethyl or polymeric chain-one can be substituted by a chlorine atom).

of the two Y

It has been observed that other metallorganic compounds (e.g., Zn(C2H&, which is not associated as A1(C2Ha)3)can be involved in chain transfer processes (46).In this case presumably alkyl groups are exchanged as follows: [CatIP

+ Zn(CzHK)2

-+

[Cat] C2HK

+ CzHIZnP

(9)

The rate of this chain transfer process is of first-order with regard to the Zn(C2H& concentration ($8). The kinetics of the above reported chain transfer reactions seem to be also catalytically affected by the titanium compound present in the reaction system. In fact we have observed (Table 111) that both the numbers of ethyl groups and aluminum atoms bound to the polymeric chains decrease with decreasing amount of titanium compound in the catalytic system. The above-recorded chain transfer processes and the processes of exchange between alkylaluminum in solution and alkylaluminum bound to the catalytic complex could also be effected by soluble polymeric alkylaluminum of the type (C2H6)A1P2or (C2Ha)2A1P(where P = polymeric chain). This

32

G. NAT'FA

AND I . PASQUON

will involve more than one ethyl group per atom of aluminum brought into the polymeric chains. In any case, such chain transfer processes lead to a continuous increase of the molecular weight of the alkylaluminum compounds during the polymerization. It is, however, possible that alkylaluminum molecules having a low molecular weight are regenerated by a mechanism similar to those reported in the study of the kinetic behavior of ethylene polymerization, in the presence of trialkylaluminum ( 3 6 ) or through a dissociation to a hydride: (CzHa)2AlCHzCH(CHzCH)nR ~ H s

-+

(C2Ho)zAIH

+ CHz=C (CH*-CH)nR bHa

AH3

&Hs

(10)

or through a transfer reaction with the monomer: (CzHo)zAlCHzCH(CHzCH).R &HI (CpHs)zAlC& CHZCHs

+ CHz=CHCHs

+

AHs

+ CHz=C (CHZCH)nR AHs

(11)

AH3

In the case of ethylene polymerization with AIRl alone, however, the transfer reaction with the monomer which is thermodynamically displaced towards the right, depends on the temperature (36) and, practically, does not occur any more a t low temperature ( <80") if specific catalysts are not involved. Also in the propylene polymerization the corresponding reaction is strongly dependent on the temperature.

D. CHAINTRANSFER PROCESS DEPENDING ON OF TITANIUM COMPOUND

THE

AMOUNT

As described in the previous paragraph, we investigated the influence of increasing amounts of titanium compounds on the molecular weight, and other characteristics of the polymer. 1. Intrinsic Viscosities

In Fig. 24 the values of l/[q]1.36 for the non-atactic polypropylene obtained at 70" and 950 mm. Hg of partial pressure of propylene are plotted us. the square root of the amount of titanium trichloride introduced in the unit volume of the catalytic system for different values of the trialkylaluminum concentration in solution (39, 40).

2. Radiochernical Determination of End Groups The radioactivity (and consequently the amount of - CaH6groups) observed in the considered polypropylene fraction obtained in polymerization tests carried out with constant concentrations of labeled triethylaluminum

33

KINETICS O F POLYMERIZATION OF (r-OLEFINS

@

0

0

mols Al (CZHs) 3 I n-heptane 8x10-2 6x10-) 4x10-2 2x104

c 1/2

-

Ti 0.1 0.2 (mols Tic!,/ I n-heptane)'l2 FIG.24. Dependency of the reciprocal of the polymerization degree (proportional to l/[q]1.*9 of the non-atactic polypropylene fraction on the square root of the amount of a-TiC1, in the catalytic system ( t = 70", p c , ~= ~950 mm. Hg, ground a-TiCla: sample A).

0

-

u)

E"

C; *

0

0.1

0.2 (mols Ti CI3 Iyn-heptane)'I2

FIG.25. Specific radioactivity (and corresponding values of -CzHs mol. per mol. of polymerized CsHa) of the non-atactic polypropylene fraction, as function of the square root of the a-TiCls amount in the catalytic system. (Tests performed with W-labeled A1(CzHs)s at t = 70°, ~ c , H=~ 450 mm. Hg, ground a-TiC13: sample A).

and with Merent amounts of titanium trichloride is approximately a linear function of the square root of the amount of titanium trichloride introduced in unit volume of the catalytic system (Fig. 25) ( 4 2 ) . Similar results have been obtained by analyzing the amorphous fractions of polypropylene (Fig. 26) (38). 3. Dependence of the Considered Chain Transfer Process on the Pressure

The use of radioactive triethylaluminum enabled us to prove that the rate of the considered chain transfer process depends also on the propylene

34

Q.

NATTA AND I. PABQUON

I

2 4Xl0"-- 200

2

C$

-

FIQ.26. Specific radioactivity (and corresponding values of -CzHs mol. per mol. of polymerized CaHs) of the atactic polypropylene fraction, as function of the square root of thea-TiCla amount in the catalytic system. (Tests performed with W-labeled Al(C2Hs)a a t t = 70°, P C J J ~ = 450 mm. Hg, ground a-TiCls: sample A . )

FIQ.27. Specific radioactivity (and corresponding values of -CzHs mol. per mol of polymerized C8H6) of the non-atactic polypropylene fraction, plotted us. l / p c , ~ ( t = 70", ground a-TiCls: sample A : 1.6 X 10-* mol./l., 1C-labeled Al(C2Hs)s: 3 X lo-* mol./l.).

partial pressure. I'rom the diagram8 plotted in Figs. 22, 25, 27, and 28 it appears, in fact, that the number of ethyl groups contained in the polymer (referred to the polymerized molecules of propylene) is given by the relationship :

+

Et = k f C i i / p c 8 x 6 k"CL

(12)

This equation may be applied only in the range of experimental condi-

KINETICS O F POLYMERIZATION OF (U-OLEFINS

35

Fro. 28. Total number of -GH, mol. found in the non-atactic polypropylene fraction, as function of PC,H@ ( t = 70°C, ground a-TiCls: sample A : 1.6 X 1Wa mol., 1'Clabeled AI(C2Hb)a: 3 X 10-8 mol., solvent: 100 cm.8n-heptane, polymerization time: 136 hr.).

tions considered. In particular, it is not applicable t o the limit conditions

CAI= 0,

or

C,i

=

0

The above relation may be transformed as follows:

If we assume a priori, as will be proved in a further paragraph, that each simple chain transfer process whose rate depends on the catalyst concentration is followed by the insertion in the polymeric chains of -C2Ha groups (deriving from triethylaluminum), the rate of the chain transfer process, depending on the amount of Ti, then would be ~4

=

kc&ipcSH,c*

It is also in agreement with the data obtained from the determination of intrinsic viscosities of the polymers (46). In Fig. 29 the reciprocals of the intrinsic viscosities (raised to 1.35) of the polymers obtained by operating a t M e r e n t pressures and with the same Al/Ti ratio, are plotted us. the square root of the alkylaluminum concentration in the catalytic system. For a given Al/Ti ratio, the lines which correspond to different pressures are almost parallel between themselves; according to the fact that l/[q]'.a6 can be represented by the relationship where Z r r

=

sum of the rates of chain transfer processes, depending on the catalyst concentration

36

0 . NATTA AND I. PASQUON

k,pcsH,C*

= over-all polymerization rate [ q ] 0 = limit value of the polymer intrinsic

viscosity obtained extrapolating the data up to the catalyst concentration equal to zero. It follows that the function @ is not strongly influenced by the pressure; i.e., Zrt should also be a function of the partial pressure of propylene.

4. Interpretation of the Chain Transfer Process Depending on the Amount of Titanium Compound Introduced into the Catalytic System It follows from the above reported results that an agent whose concentration depends upon the amount of titanium compound introduced into the reagent system takes part directly or indirectly in a chain transfer process. It must be remembered that several compounds or soluble complexes of titanium which, alone, and also in the presence of AlRa , do not polymerize the propylene, can effect the polymerization rate of the catalytic systems containing cu-TiCl3 and the molecular weight of the polymers obtained (10, 11). The analogy between this last process and the one whose rate depends on the triethylaluminum concentration (like, for instance, its influence on the

0.3

%I

7 0.2

9 E u

0 0

.5

0.1

(mols Al (C2H&/I n-heptane)'12 FIG.29. Dependency of the reciprocal of the polymerization degree (proportional . ~ non-atactic ~) polypropylene fraction on the square root of [Al(C2H,)a]. to l / [ ~ ] of~ the Comparison between tests performed for few pressure values ( t = 70°, ground a-TiClo: sample A ) .

KINETICS O F POLYMERIZATION OF a-OLEFINS

37

number of aluminum atoms and of -C2Hs groups bound to the polymeric chains) enables us to state that in both cases an exchange occurs between alkyl groups deriving from alkylaluminum compounds and growing polymeric chains or between metallorganic compounds containing, respectively, alkyl groups and polymeric chains. It may be that these compounds act catalytically on the chain transfer processes which depend on the alkylaluminum concentration. We can also consider that non-catalytic complexes containing alkylaluminum, exchange their metallorganic component with the catalytic complexes bound to the active centers of the crystalline substrate.

+

[Ti,CIm]AIY~P TiX,,AlRs

+ AlRs

TiX,,AIYzP

+

(15)

+ AlYzP

(10)

-+

[Ti,,Clm]AIR~ TiX,AlYJ'

-+

TiX,,AlRs

where [Ti,,Cl,] means the active center (of the crystalline substrate) to which a metallorganic compound of the aluminum is bound, and TiX,AlRa means a generic complex of titanium which acts catalytically as an intermediate compound in the exchange of organoaluminum groups among complexes in solution and solid catalyst. The dependence of the rate of the considered chain transfer process on the olefin pressure may be justified if we assume that such a process requires an activation state of the solid catalyst. Such an activation state would correspond t o the activated intermediate complex which is formed during the polymerization process a t the particular stage in which a monomeric unit bounds itself to the catalytic complex. As a result, the rate of the transfer process would depend on the rate of the chain-growing process, since the two processes (growing and chain transfer) would be considered as parallel and deriving from the same activated complex. The chain transfer process whose rate depends only upon the alkylaluminum concentration (or upon the zinc-diethyl concentration) is, on the contrary, independent of the partial pressure of the olefin and may occur even if the polymerization stops for lack of monomer.

E. CHAINTRANSFER PROCESS DEPENDING ON THE PROPYLENE PARTIAL PRESSURE A study of the intrinsic viscosities of the polymers and of the chain-end groups enabled us to establish that there are processes affecting the molecular weight which are independent of the catalyst concentration. They are now discussed on the basis of experimental data obtained by different tests. 1. Intrinsic Viscosities

Taking into consideration t,he results so far observed (see also Fig. 30), it may be assumed that the reciprocal of the degree of polymerization of

38

Q. NA'TTA

AND I. PASQUON

A

FIG.30. Dependency of the reciprocal of the polymerization degree (proportional to 1 / [ ~ ] ~of. 9the non-atactic polypropylene fraction on the square root of [A1(C2HJI]. (Tests performed for few values of Al/Ti ratio, t = 70°, = 950 mm. Hg, ground cr-TiCls: sample A ) .

propylene is given by the expression

For the term l/znoit may be assumed:

where Zrt, = sum of the rates of chain transfer and termination processes independent of the catalyst concentration. Tests performed at different pressures (see Fig. 29) have proved that l/[q]o , and consequently l / Z n a , in the considered range of pressure, is only slightly affected by the partial pressure of olefin. This means that the term Zrr, is a function of the pressure. We assumed that Z r r , consists of two terms, one being of first order with regard to the pressure and the other one independent.

-1 --

h C * + kzpc,~$* (19) kppc,~aC* kppc,~,C* The term klpCsHsC*represents the rate of a process of first order with regard to the partial pressure of the monomer, and klC* the rate of a process xno

39

KINETICS OF POLYMERIZATION OF a-OLEFINS

independent of the olefin pressure. The first of these two processes can be equivalent t o a chain transfer with the monomer. Under the polymerization conditions previously described, (temperature below 80") the term kl seems to be very small with respect to kzpcaH6; in fact, a t low pressure (450 mm. Hg) the value of 1/[7]:," may appear somewhat higher than the corresponding value obtained a t 950 mm. Hg or higher pressures (see Fig.29). The values of 1/[7]:." obtained, respectively, a t 950 mm. Hg and 1,450 mm. Hg of propylene partial pressure are almost coincident (40,46). The numerical value of l / x n 0 ,in the considered field, is then mainly determined by the term k 2 / k , , e.g., depending on the chain transfer with the monomer. 2. Infrared Spectra Analysis The infrared spectra of the obtained polymers have shown that they contain vinylidenic end groups (22). Quantitative data for the amorphous fraction of polypropylene are reported in Table IV (38). 3. Mechanism of the Chain Transfer Process with the Monomer and of the Spontaneous Termination Process From the results reported in the preceding paragraphs, the mechanism of the chain transfer process with the monomer may be as follows: [Cat]CH&H(CH,CH).R I I &HI 6Hs [CatICHzCHZCHs

+

+ CHI=CHCH3

--t

CHz=C (CHzCH),R I

bH8

I

(!!HI

It may be that another process (whose rate is independent of the monomer and catalyst concentrations and can be detected at the considered TABLE IV End Groups in the Small Atactic Fraction (Eztractable in Boiling Ether) Obtained b y Polymerization of Propylene with the Catalytic System: a-TiCla-Al(CzHs)a-n-Heptane Polymerization conditions

_

1 . 5 X 10-2 3 X 1 . 5 X 10-2 3 X lo-* 1 . 8 x lo-' 3 X

-(CHI) C-CHa end groups/CaHa mol. in atactic polym-*

_

450 950 1,450

_

70 70 70

~

2.5 X 2 . 3 X lo-' 2 . 6 X lo-'

-C& end groups/C~H!

mol. in atactlc polymer

Ratio -crHa

-(CHI) C-CHX

~

10.3 X lo-' 4.7 X 5.5 X

4.15 2.05 2.10

40

G. NATTA AND I. PASQUON

temperature ‘70” only at low partial pressure of propylene) affects the molecular weight of the polymeric chains. This process could be a process of spontaneous dissociation which leads to the formation of an active hydride : [ CatICHz CH(CHzCH),R AH3

+ [ Cat]H

+ CHz=C (CHZCH),R

AH3

AH3 AH3

(21)

The complex [CatIH could initiate a new polymeric chain as follows: [CatIH

+ CHz=CHCHa

-+

[Cat]CHzCHzCH3

(22)

and therefore the spontaneous dissociation of the polymeric chain should not be considered a priori a real termination of the reaction chain. At high temperature there exists an equilibrium between Alz(CnHz,+l)s Ft Alz(CnH2n+i)aH

+ CnH2,

nevertheless, it is most likely that the addition reaction of the monomer to the hydride, requires a greater energy of activation than the one corresponding to the addition of a monomeric unit to a growing chain, so that we may admit that complexes containing hydride cannot be alkylated immediately by the olefin. Hence, we may consider the reaction (22) as the first stage not only of a new polymeric chain but also of a new chain of reactions. I n this case, the spontaneous dissociation could be considered as a real termination process of the chain reaction. Indeed, reaction (20) can be considered as a real chain transfer with the monomer. In fact, by making a comparison between the reactions (20) and (6) and (7), it can be observed that, after these reactions, the catalytic complexes [Cat]CH&H&Ha and [Cat]CH&Ha are respectively formed and both these complexes can add monomeric units in the polymerization process. The reactions (6) and (7) are equivalent to a real chain transfer bccause the over-all polymerization rate appears to be independent of the triethylaluminum concentration. Considering the chemical analogics of the catalytic complexes, resulting from the reactions (20) and (7), it may also bc assumed that the transfer reaction whose rate is of first order with regard to the monomer may be considered a real chain transfer (from a kinetic point of view).

F. INFLUENCE OF THE TEMPERATURE ON THE SINGLE-CHAIN TRANSFER AND TERMINATION PROCESSES As the temperature decreases from 70 to 30”, either for limited or high concentrations of catalyst, it may be observed that the molecular weight of the obtained polymer, somewhat increases (Fig. 31). The fact that the I value of the slope of the lines obtained by plotting 1/[~$,.~’vs. C ~slightly

41

KINETICS OF POLYMERIZATION OF (U-OLEFINS

x

0 0 z

4

0.1-

0

4

4

4

/

0

I

fi

112

“A1

03 [mols AI(C2Hs)J/I 0:2 n-heptane] 0:3 112 Fro.31. Effect of temperature upon the polymerization degree of the non-atactic polypropylene fraction (ground a-TiCla: sample A ) .

1 2 3 4 5

31 31 51 70 70

1,450 700 1,110 1,450 950

1.18 0.57 0.63 0.62 0.41

3 3 3 3 3

varies with the temperature shows that the activation energy of the chain transfer processes, depending on the catalyst concentrations, are little different from the activation energy of the chain propagation process. In order t o define the value of the activation energy of the chain transfer process with the monomer, it would be necessary to know quite exactly the values of l/[~]h’~’ at different temperatures. From the data plotted in Fig. 31, it may be observed, therefore, that also the activation energy of this process is just slightly greater than the activation energy of the chain propagation process.

G. COMPARISON BETWEEN INTRINSIC VISCOSITY AND SPECIFIC RADIOACTIVITY OF THE POLYMER OBTAINEDIN THE PRESENCE OF “C-LABELED TRIALKYLALUMINUM In this paragraph we intend to confirm the hypothesis that each process of chain transfer, depending on the concentration of Al(CzHJa and

42

Q. NAlTA AND I. PASQUON

on the amount of a-titanium trichloride present in the catalytic system, is followed by the introduction into the polymeric chain, of group, -C2Ha deriving from the ethylaluminum. According to the results acquired through the study of the intrinsic viscosities of the polymer, we have assumed that the reciprocal of the number average degree of polymerization is represented by the relationship :

1 - klC*

+ k2pca,,C*

Xn

-I- ksC!IC*

+

k4C'!'ipCaH,c*

(23)

k~pC8H6C*

The terms of the numerator indicate the rates of chain transfer and termination processes depending, respectively, on the concentration of the growing chains, on the partial pressure of olefin, on the concentration of alkylaluminum, and on the amount of titanium compounds. Let us consider now the equation

which gives the amount of ethyl groups found in the polymer after polymerization. The terms on the numerator show the rate of the processes which lead to the introduction of ethyl groups in the polymer. The relation (24) can be substituted in Equation (23) if ICs and lcr

$21""

b

-

Et x 103 rnds-G&/mol $+la polymrrizrd

FIQ.32. Relation between the reciprocal of the intrinsic viscosity raised to 1.36 (proportional to the reciprocal of polymerization degree) of the non-atactic polypropylene fraction and the number of mol. of -GHs groups found in the polymer. (Tests performed with W-labeled Al(C2HI)a at 70' and 960 mm. Hg ~ G ~ ,Hground , aTiC1,: sample A .)

KINETICS O F POLYMERIZATION OF a-OLEFINS

43

have, respectively, the same numerical value as 6 and & . In this case, we shall find

+ Et

1/2, = l/Zno

(25)

The reciprocal of the intrinsic viscosities of the polymer would be a unique and linear function of the amount of ethyl groups (deriving from ethylaluminum found in the polymer. This appears to be confirmed (except for some small irregularities, for high concentrations of ethylaluminum) by the data plotted in Fig. 32, where the values of l/[q]1.36 for polypropylene, obtained by operating with several titanium trichloride and trialkylaluminum concentrations are plotted v5. the amount of ethyl groups found in the polymer.

H. RELATIVEIMPORTANCE OF THE DIFFERENT PROCESSES CHAINTRANSFER In an attempt to calculate approximately the relative importance of the different chain transfer processes, we can put (41) [q] =

K' X0n' 7 4

(26)

and taking into account Equations (26) and (23), we obtain

K

- 1 - k~

[t111'36

+

k@CsHp,

+ ~C!I +

xn

k4ckipC3H6

(27)

k~pc3H 6

where K = K".35and

Let us consider the ratio

- l/[~Ii'~' - r3 + ~4

1/[q11'36

l/[Tli.35

Tl

+

rz

-

k3C!1

ki

+ +~ P P c ~ H ~ k4Ckip~aH6

(29)

The numerator includes the rates of the chain transfer processes which depend on the catalyst concentration; the denominator includes the rates of the processes which do not depend on the catalyst concentration. Operating at low temperature ( <SO"), kl can be neglected with respect to kzpcsH0.Therefore, the relationship (29) is equal to the ratio between the number of polymeric chains interrupted by the chain transfer processes, depending on the catalyst concentration, and the number of polymeric chains interrupted by the chain transfer with monomer. The values of

44

G. NATTA AND I. PASQUON

FIQ.33. Values of the ratio between the rate of the chain-transfer proceslies depending on the catalyst concentration and the rate of the chain-transfer process wit>h the monomer (plus the rate of the spontaneous termination process) (1 = 70°, pcraa= 950 mm. Hg, ground a-TiCls: sample A ) . The values were calculated assuming for I/ the isotactic polymeric fraction zn = K , [ v ] ~ . *=* KZCAL".

have been computed from the diagrams of Fig. 30, concerning polymerizntion experiments carried out at 70" and 950 mm. Hg partial pressure of propylene. The data so obtained are summarized in Fig. 33. We observe that the calculated values are strongly dependent on the value of l/[7]:'". On the other hand, the latter value cannot be exactly determined with the reported data (see, for instance, Fig. 30). Therefore, the data plotted in Fig. 33 are largely approximate. The importance of the chain transfer process depending on the concentration of alkylaluminum, when compared with the one depending on the amount of a-titanium trichloride present in the catalytic system, may be easily deduced from the diagrams plotted in figures 22 and 25. From such diagrams it follows for the catalytic system considered: ks/k4 = 1.27

(30)

I. RELATION BETWEEN INTRINSIC VISCOSITY AND NUMBER AVERAGE POLYMERIZATION DEGREE FOR POLYPROPYLENE If the amount of polymeric chains containing an end group -GH6 and the fraction of such chains compared with the total number of them are

KINETICS O F POLYMERIZATION OF (Y-OLEFINS

45

known, it should be possible to calculate the number average degree xn of polymerization and compare this latter with the intrinsic viscosity. If the value of znis calculated in such a way it will necessarily differ from the viscosimetric molecular weight (since the polypropylene is a polydispersed polymer) as may be calculated by already known relations (41). Combining Equations (23), (24), and (26), we obtain

Drawing a straight line through the experimental points plotted in Fig. 32, we approximately obtain Zn =

95[qy6

(32)

where [q],measured at 135' in tetralin, is expressed in 100 ~ m . ~ / g . This relation gives likely approximate values for the limited range of polymers obtained under the conditions previously examined, namely, t = 70', pCsHs = 950 mm. Hg; CTi and CA1 = 1 to 15.10-' mol./l. n-heptane. The number average degree of polymerization of non-atactic polypropylene, as evaluated from this relation, is much lower than the viscosimetric molecular weight resulting from the suggested relationship because the above-considered fraction of polypropylene contains isotactic and stereoblock macromolecules which have low molecular weight. On the other hand, that fraction is strongly dispersed (47) and, as it is usual in these cases, the number average molecular weight is lower than the viscosimetric one. Moreover, we confirmed the results previously ascertained through radiochemical measurements. In fact, the comparison between such data and those obtained by I R measurements on the atactic polymer fraction (Table IV) shows that the ratio between the number of polymeric chains with a -CzH6 end group (corresponding to a chain transfer process depending on the catalyst concentration) and the number of polymeric chains with a vinylidenic end group of polymeric chains (corresponding to the chain transfer process with the monomer) is closely in accordance with the data reported in Fig. 33.

J. REMARKS ON THE CATALYTIC NATUREOF THE COORDINATED ANIONICCATALYSIS From the above summarized data, it follows that, during the polymerization, in the presence of triethylaluminum, there is a consumption of A1 atoms and ethyl groups bound to the aluminum, because the triethylaluminum is involved in chain transfer processes without being regenerated. Therefore, the polymerization process is now thoroughly catalytic only with

46

0 . NATTA AND I. PASQUON

respect to titanium trichloride. It must be noticed, however, that not all chain transfer processes lead to a consumption of alkylaluminum. If we take into consideration the chain transfer processes with the monomer, we can consider the coordinated anionic catalysis as thoroughly catalytic in the strictest sense. I n practice, an excess of alkylaluminum is necessary if the polymerization is t o be of long duration. I n fact, the catalyst obtained by treating a-titanium trichloride with trialkylaluminum and successively washed so as to eliminate the excess of chemisorbed alkylaluminum, makes only n very small amount of polymer (unless a further amount of alkylaluminum is added). It is most likely that the polymerization stops because of the presence of impurities (traces of O2 , moisture) which act as poisons of the catalyst when alkylaluminum is absent.

IV. Steric Composition of Polymers The amount of amorphous polymer, which is generally produced in small percentage (9-16 %) contemporaneously with the non-atactic polymer, is independent of reaction time (see Table 11). It is on the contrary closely connected with the nature of the catalytic system employed and changes, for instance, when the triethylaluminum is substituted by other metal ulkyls (beryllium alkyls, propylaluminum, isobutylaluminum, etc. ) ( 5 , 2 8 ) . It also depends on the purity of the a-titanium trichloride, in particular increasing in the presence of other crystalline modifications of titanium trichloride [i.e. /3-TiC19 (27)] and of titanium compounds obtained by reduction of titanium tetrachloride a t low temperature with aluminum alkyls. In the Tables V-IX we reported the percentages of the fraction of n-heptane insoluble polymer at room temperature and the respective intrinsic viscosities. The data concern the polymers obtained from polymerization tests carried out with triethylaluminum concentrations varying from 1.18 X lo-* to 14.75 X mol./l., with ratios Al/Ti between 1 and 10, with propylene concentrations included between 0.19 and 0.63 mol./l. at temperatures of 31", 51" and 70°C. In a recent work (@), the influence of the variation of some factors, concerning the polymerization, on the steric composition of the polymer has been studied subjecting the obtained raw polymer to subsequent extractions with the following series of solvents employed a t their boiling point: ether, n-heptane and n-octane. We have examined polymers obtained in polymerization tests carried out a t 70"C, with the following catalytic systems: a-TiC13/A1(CzHa)zC1 and (r-TiCl3/A1(C2H6)9. We have observed that with the decrease of the propylene partial pressure a t which the polymerization is carried out, from 1600 mm. Hg to 250 mm. Hg,the percentage of each fraction of polymer extracted with the

KINETICS OF POLYMERIZATION OF a-OLEFINS

47

Catalytic System. Tests Performed at 70", 950mm. H g p c , ~,, [CaHa] = 0.41 mol./l.

[A1(CzHs)rl, mol./l.

1.47 X 1.47 X 1.47 X 1.62 X 1.77 X 2.36 X 2.22 x 2.65 X 2.80 x 2.94 X 2.94 X 2.94 X 2.94 x 2.94 X 2.94 X 4.42 X 4.42 X 4.42 X 5.88 X 5.88 x 5.88 X 5.88 x 8.82 x 8.82 X 8.82 X 8.82 x 10.30 X 11.75 x 11.75 X 11.75 X 11.75 X 11.75 X 14.75 x 14.75 X 14.75 X

10-* 10-9

10-* 10-* 10-* lo-* lo-* lo-* lo-* lo-* lo-* 10-' lo-* 10-* 10-* 10-* 10-* 10-* 10-* lo-' 10-' lo-' lo-* lo-* lo-' 10-2

lo-' lo-* 10-* 10-2

lo-* 10-0 10-2

a-TiCls, (sample A ) ,

Al

-

g4.

Ti mol

0.75 1.50 1.50 0.83 0.91 1.20 0.34 2.72 0.87 3.00 3.00 1.50 1.50 1.50 0.45 4.52 4.52 1.36 3.02 3.02 1.82 1.82 9.05 9.05 4.52 4.52 1.58 6.05 6.05 6.05 3.62 1.82 2.27 2.27 2.27

3 1.5 1.5 3 3 3 10 1.5 5 1.5 1.5 3 3 3 10 1.5 1.5 5 3 3 5 5 1.5 1.5 3 3 10 3 3 3 5 10 10 10 10

.

Intrinsic PolymeriNon-atactic viscosity of zation polymer," non-atactic time, polymer" : % hr . [TI100 cma./g. 4.52 2 89 3.92 2 89 4.06 2 88.5 4.00 2 88.5 4.30 9 88.5 4.00 2 89 4.16 10 89 3.60 H 89 3.92 7 89 3.43 2 89 3.51 2 89 3.84 H 88.5 3.78 2 88.5 3.84 6 88 4.08 4 89 3.07 2 89 3.12 2 89.5 3.84 2 90 3.12 2 90 3.18 2 90.5 3.47 2 88.5 3.51 2 89 2.38 1 90 2.42 1 89.5 2.76 1 90 2.86 1 90.5 3.28 2 90 2.44 2 90.5 2.50 2 90 2.60 2 90 2.86 2 90.5 3.08 2 90.5 2 90 2.84 3.03 2 90.5 3.18 2 91

The data are related to the polymern insoluble in n-heptune a t room temperature and include a h the

TABLE VI Polymerization of Propylene to Isotactic Polymer with the a-TiCls-A1(C2Ha)s-n-Heptane Catalytic System. Tests Performed at t = 51", 1,110 mm. Hg P C ~ E ~ , [CsHa] = 0.63 mol./l. n-Heptane Polymerization time, hr.

1.47 1.47 1.47 1.47 1.47 2.36 2.94 4.42 11.75

X lo-' X 10-2 x lo-* X lo-* X 10-8 X 10-2 x lo-* x 10-1 X 10-2

1.13 0.75 0.75 0.30 0.30 0.60 1.50 2.26 6.05

2 3 3 7.5 7.5 6 3 3 3

Intrinsic viscosity Jon-atactic polymer, if non-atactic polymer": % ,I, 100 cm.a/g. 0

83 85 86 87 87 85 87 88.5 89

8 9 9 8 24 9 5

1% 2%

4.20 4.38 4.35 4.50 4.45 4.00 4.16 3.60 2.92

a The data are related to the polymers insoluble in n-heptane a t mom temperature and include also the stereoblock polymers soluble in boiling n-heptane ( b 7 % of the whole polymer).

TABLE VII Polymerization ojPropylene to Isotactic Polymers with the a-TiCla-A1(CtHs)s-n-Heptane Catalytic System. Test8 Performed at t = 70", 450 mm. Hg p ~ $ ~ , [CsHs] = 0.19 mol./l. -Tic1 sample A ) ,

A1 Ti mol .

Polymerization time, hr.

0.62 0.62 1.82 3.18 1.20 1.20 3.00 3.00 1.50 1.50 5.00 6.80 4.54 11.30 11.30

3 3 1.5 1 3 3 1.5 1.5 3 3 1 1 1.5 1 1

30 30 11 11 10 15 2% 2% 4 7

(Y

Intrinsic viscosity Von-atac tic polymer,a of non-atactic polymera : % [TI, 100 cm.J/g. ~~

1.18 x 1.18 x 1.77 X 2.06 X 2.36 X 2.36 X 2.94 X 2.94 X 2.94 X 2.94 X 3.24 X 4.42 X 4.42 X 7.36 x 7.36 X

10-2 10-1 lo-* 10-2 lo-* lo-' lo-' lo-' lo-*

lo-' lo-* lo-*

94 5 3

ki

1

89 89 90 90.5 90 90 90 90 91 91 90 91 90.5 91 90

4.10 3.70 3.52 3.22 3.56 3.47 3.14 3.22 3.28 3.14 2.86 2.63 2.74 2.18 2.13

The data are related to the polymers insoluble in n-heptane a t room temperature and include also the stereoblock polymera soluble in bailing n-heptane (b7% of the whole polymer). (I

48

49

K I N E T I C S O F POLYMERIZATION O F &OLEFINS

TABLE VIII Polymerization of Propylene to Zsotactic Polymer wilh the a-TiClr-Al(CzH,) n-Heptane Catalytic System. Tests Performed at t = 70", 1450 mm. H g P B~ , [CaH,] = 0.62 mol./l.

I3.h

Al T i' mol .

0.75 0.91 0.91 2.10 1.20 1.20 3.00 1.50 1.50 4.52 5.45 4.54

3 3 3 1.5 3 3 1.5 3 3 1.5 1.5 3

a-TiClr (sample A )

1.47 1.77 1.77 2.06 2.36 2.36 2.94 2.94 2.94 4.42 5.30 8.84

x x x

lo-* 10-9 10-2 X 10-0 X lo-* X X X X X

x x

10-2 lo-* lo-* 10-2 10-2 10-0

Polymerization time, hr.

8 4 7 2 2 2 1

6 6 1% 2

%

Intrinsic viscosity Ton-atac tic polymep, of non-atactic polymer": [q] % 100 cm.*/g. 87 88 87 87 87 87 86 89 89

90 88 88

4.16 4.16 4.20 3.87 4.04 4.12 3.57 3.83 3.88 3.12 3.03 2.92

" The data are related to the polymers insoluble in n-heptane a t r w m temperature and include also the stereoblock polymers soluble in boiling n-heptane (b7%of the whole polymer). TABLE IX Polymerization of Propylene to Zsotactic Polymer with the a-TiC13-Al(C2Hs) n-Heptane Catalytic System. Tests Performed at t = 31", 700 mm. Hg. P C , H ~ , [CaHe] = 0.57 mol./l. a-TiC13 :sample A ) ,

1.47 2.94 2.94 5.88

X X X X

lo-* 10-8 10-2

10-2

0.75 1.50 1.50 3.00

A1 Ti' mol .

Intrinsic Polymeri- Non-atactic viscosity zation polymer," of non-atactic time, hr. polymero: [ q ] % 100 cm.S/g. 40 14 31 2

89.6 89 88.5 89.5

5.85 4.40 4.34 4.00

" The data are related to the polymers insoluble in n-heptane at rmm temperature and include aLo the stereoblock polymera soluble in boiling n-heptane (67% of the whole polymer). above-mentioned solvents increases. This phenomenon can be easily observed working a t the lowest pressures; it was not observed in previous works in which generally we worked at partial pressures of propylene superior to 1 atm.

50

G . NATTA AND I. PASQUON

We have also observed an increase of the percentage of polymer extractable in n-octane when varying the concentration of the triethylaluminum to 12 X mol./l. from 1.5 X These results made us conclude that the inversions of steric configuration of the monomeric units which happen during the growing of the atactic or stereoblock polymeric chains is determined by several different processes. The rate of one of these is independent of the pressure of the olefin, and therefore, being the growing rate of the polymeric chain of first order with regard to the olefin pressure, the frequency of the inversions along the chain increases when the olefin pressure decreases. The inversion phenomena during the growing of the polymeric chains are also connected with the chain transfer phenomena, whose rates depend on the concentration of the components of the catalytic system.

V. Determination of the Number of Active Centers The coordinated anionic catalysis is one of the few examples of heterogeneous catalysis in which it is possible to estimate the actual number of active centers, present on the catalyst surface, which directly take part in the chemical catalytic process. In the stereospecific polymerization of propylene, such active centers, which are formed by treating a-TiC13 with alkylaluminum, have been determined by using 14C-labeled alkylaluminum. The two systems studied were A1(C2Ha)&TiCIS A1 (CZHs)ZCl/cu-TiC18

In both cases the determination of active centers has been performed by means of absorption tests of “C-labeled alkylaluminum on samples of ground a-titanium trichloride. In the first case, two a-Tic13 samples were used, one of them (sample A ) having low catalytic activity, the other one (sample B ) high activity. In the second case, only the sample A was used, and the determination of active centers was also made by a kinetic method, still using labeled alkylaluminum. ON A. ADSORPTIONOF 14C-LABELEDALKYLALUMINUMS

CY-TITANIUM TRICHLORIDE The following determinations have been performed: Direct determination of “C-labeled ethyl groups bound on the surface of titanium trichloride samples, treated either with triethylaluminum or diethylaluminum monochloride solutions, at different temperatures. Determination of active centers by the number of 14C-labeled-CzHs

KINETICS OF POLYMERIZATION OF WOLEFINS

51

groups found in a polymer, obtained using, as catalyst, a-titanium trichloride pretreated with labeled alkylaluminums. Before carrying out such determinations in an attempt to estimate the number of active centers correctly, it has been necessary to define the magnitude of the eventual radioactive contaminations of the polymers, caused by phenomena extraneous to the polymerization process. 1. Measurements of Casual Radioactive Contaminations

It has been pointed out that the product obtained by treating certain samples of ground a-titanium trichloride (particularly those which contain traces of TiC14 or other Ti(1V) compounds) with radioactive alkylaluminum, shows a certain degree of radioactivity also after submitting it to the action of an acid or an alcohol in an attempt to decompose the metal-carbon bonds. Such radioactivity is due to a contaminant, the nature of which depends on the degree of purity and the amount of crude a-titanium trichloride employed. It generally decreases, eventually attaining very low values if the crude a-titanium trichloride is repeatedly washed with anhydrous benzene before its use. We have observed that the radioactive contamination is practically independent of the temperature (49). We believe that this radioactive contamination is due to the presence of traces of radioactive polyethylene resulting from ethylene polymerization. Ethylene can result, in fact, from the disproportionation of C2H*radicals released by decomposition of ethyl titanium compounds, which derive from the reaction between ethylaluminum and traces of titanium tetrachloride or other tetravalent titanium compounds that are sometimes present as impurities in the a-titanium trichloride. This assumption is confirmed by the fact that adding titanium tetrachloride to a-titanium trichloride catalysts a greater contamination in the reaction products with labeled ethylaluminum is obtained (42). When other samples of a-titanium trichloride are used, for instance, unground a-titanium trichloride (sample B , see Fig. 7), having a high catalytic activity, we do not observe radioactive contamination. Other types of contamination due, for instance, to the incomplete removal of alkylaluminum from the polymer or to secondary reactions of its alkylation, are not present in the reported tests. 2. Adsorption Tests with 14C-LabeledTriethylaluminum and Diethylalurninum Monochloride on &-Titanium Trichloride The following method was applied in an attempt to evaluate the amount of ethyl groups fixed on the a-titanium trichloride surface by treating it with ethylaluminum solutions:

52

a.

NATTA AND I. PASQUON

Treatment of a given amount of titanium trichloride, at a given tempernture, with a solution of radioactive ethylaluminum. Filtering and washing of the solid phase at a given temperature, under nitrogen, with carefully purified anhydrous benzene (or other hydrocarbon solvent) until any radioactivity disappears from the washing solvent. Subsequent addition to the solid phase of an amount of inactive alkylaluminum in solution and decomposition of all alkyl-metal bonds with 10 % HzSO,. Purification of the gas released at high temperature, by washing at -78" in order to separate the solvent and other hydrocarbons having an high molecular weight, which are entrained by the gas. Combustion of the gas released. Absorption of the COa on Ba(0H)Z. Determination of the 14C/12Cratio in the BaC03. The radioactivity measurements of BaC03 allowed us to calculate, according to the law of the isotope dilution, the amount of ethyl groups, fixed on the a-titanium trichloride surface. The results obtained in such tests, in the interval - 18-+ loo", for the considered system a-titanium trichloride-triethylaluminum and a t 70" for the considered system a-titanium trichloride-diethylaluminum monochloride, are tabulated in Table X (49). In all tests, the initial treatment of a-titanium trichloride, with the ethylaluminum solution, was carried out under conditions suitable for the practically complete saturation of the a-titanium trichloride surface at the temperature considered by the metallorganic compound. The above conditions were found by varying, in preliminary tests, the concentration of the alkylaluminum solution and the contact time. From the results obtained, one may conclude that: Metallorganic complexes containing ethyl groups are present on the a-titanium trichloride surface. Within the temperature range 4-20 to - 18"C, after raising the equilibrium conditions, the amount of ethyl groups strongly fixed on the a-titanium trichloride surface, seems to be almost independent of the temperature. For the considered sample of a-TiC13, above 20" approximately, the amount of the fixed ethyl groups decreased with increasing temperature and depends on the temperature a t which, after treatment with alkylaluminum solution, the washings of the a-titanium trichloride are carried out. At the lowest temperatures considered ( - ISo), it has been observed that the adsorption process is relatively slow. The adsorption process is partially reversible. However, we can suppose that almost a portion of the ethyl groups (probably bound to a metallor-

53

KINETICS OF POLYMERIZATION OF WOLEFINS

TABLE X Number of Conventional Active Centers i n a Sample of Ground a-TiCla , Determined by Adsorption Tests of 1%-Labeled Alkylaluminum Compounds, Followed b y Polymerization of Propylene. (a-Ticla-Ground: Sample A : 0.5 g . ; Alkylaluminum Compound: 0.5 cm.3; Solvent: 30 cm.*). -Ct&

Solvent

Temperaturt washin wit} anhy ous solvent, “C

mol. adsorbed per mol. of a-TiCh INumber of alkvl

&

on a-TiCIa surface after washing

ifound plfi”y”iin% ,E,” tge

I

- 18 - 18

(I

36

n-heptane

3

w

20 46 70 70 100 70

M M

70

4i

36 56

w

1I

benzene 11

- 18

- 18 20 46

11

20

11

toluene n-heptane

70 100 20

n-heptane

20

17.0 X 45.0 x 48.2 x 17.7 x 10.5 X 6.2 X 3.0 X 3.0 X 5.0

nd. 10-3 9.3 x 10-3 10-3 10.1 x 10-3 10-3 10.5 x 10-3 10-3 10.8 X 10-3 6.7 X 10-3 3.0 X lo-’ 10-3 3.1 X

x 10-3

<0.2

x

10-3

This teat waa performed with u n w u n d a-TiClr: a m p l e B (See fig. 7).

ganic complex) is strongly fixed to the crystalline substrate (a-titanium trichloride) ; actually repeated washings with anhydrous solvent are not sufficient to give a complete desorption of the alkyl compounds. The amount of ethyl groups bound to the a-titanium trichloride surface, treated with diethylaluminum monochloride, is somewhat smaller than the one observed by the treatment with triethylaluminum (49). It has been observed by many authors that the catalytic activity of several crystals is related to some particular faces of the crystal (60). In the case of a-titanium trichloride, microscopic examination reveals that the widest faces are those (001). Owing to the sandwich structure of the a-titanium trichloride, such faces are constituted by C1 atoms only. There is some evidence that the faces on which titanium atoms may exist, e.g., lateral faces (thus, not the 001 faces supposed free from fault), show a greater catalytic activity. From adsorption measurements of radioactive triethylaluminum on an a-titanium trichloride sample having well developed crystals (sample B , see Fig. 7 ) , one may observe that the total amount of alkylaluminum which can be adsorbed (Table X, last line) is remarkably greater than the one sufficient to form a monomolecular layer on the lateral faces of the crystals (38).It is most likely that the alkyl-

54

Q. NA'ITA AND I. PASQUON

aluminum adsorbed on that a-titanium trichloride sample is mostly adsorbed along the 001 faces of the crystals. In the following paragraph it will be, however, demonstrated that, m for the sample here considered, the whole amount of alkylaluminum adsorbed does not correspond to the active centers (see Table XI last line). 3. Evaluation of th Active Centers

a-Titanium trichloride, on whose surface a given amount of ethyl groups was previously bound, was employed in the propylene (or ethylene) polymerization, generally after a further addition of non-radioactive alkylaluminum. The ethyl groups were bound on a-titanium trichloride surface by treating titanium trichloride with a radioactive alkylaluminum compound, followed or not by repeated washings with anhydrous n-heptane. Further addition of alkylaluminum is not, in any case, indispensable for the formation of the catalyst, given the fact that an olefin polymerization occurs also without it although for a short time. In tests carried out with further addition of non-radioactive alkylaluminum, we obtained 0.5 to 2 g. of isotactic polypropylene per 0.5 g. of a-titanium trichloride, according to the polymerization time. The polymer obtained in these tests was found to be always radioactive. It is most likely that the radioactive carbon found in the polymer comes from the ethyl groups initially contained in the catalytic complexes bound to the active centers of the a-titanium trichloride surface. The results obtained in such measurements are summarized in Table X (49). It has been assumed that all radioactive carbon found in the polymer is contained in it as -CzH6 groups. In the interval 70-100", by using the sample A of a-titanium trichloride, all ethyl groups, initially bound to the a-titanium trichloride surface, were later found in the obtained polymer. Within the range -18 to 50" the amount of ethyl groups found in the polymer is smaller than the total amount, initially bound to the a-titanium trichloride surface. In other tests, carried out with other samples of a-titanium trichloride, the same behavior was also observed when operating a t 70" (see Table XIlast line). If the temperature of washing of the a-TiClo , after being treated with radioactive ethylaluminum, is kept below about 50°, the number of ethyl groups found in the polymer remains practically independent of the temperature a t which the adsorption of the triethylaluminum was made and of the amount and kind of polymer (polypropylene or polyethylene) obtained in each experiment. If the washing-temperature is higher (e.g., 70°), the number of ethyl groups found in the polymer is lower.

55

KINETICS OF POLYMERIZATION OF a-OLEFINS

TABLE XI Ptopylene Polymerization with the a-TiCla-A1(CnHs) 8-n-Heptane Catalytic System. Comparison between Two Samples of a-TiCls Sample A : Ti/Cl = 2.96 (Initial Sizes of the Crystals 5 2 p . Sample B : Ti/Cl = 3.00 (Initial Sizes of the Crystals: within 1 to 200 p-See fig. 7). a-TiCls sample

A

B

B / A ratio

Catalytic activity at 70°C expressed as: g. CaHspolymerized in steady-state conditions/hr. g. T i c l a . atm. PC,H, Conventional active centers (number of -C9Ha groups initially fixed on the active centers) mol./mol. of ground a-TiCIJ

8

13

1.63

1.7 X 10-1

1.7

1

x

10-2

The number of ethyl groups, and, therefore, the number of the corresponding active centers found in the catalytic system diethylaluminum monochloride-a-titanium trichloride, is smaller than the one found in the system triethylsluminum-a-titaniumtrichloride. It is interesting to notice that for the two samples of a-TiC13 ( A and B ) , the activity ratio in the propylene polymerization is almost equal to the ratio (determined from ethyl groups) between the number of active centers (Table XI) (38).

4. Discussion of the Results The results summarized in the previous sections show that, for the considered sample of a-TiC1, , there are two types of adsorption of the triethylaluminum on a-titanium trichloride, one of them is connected with active centers which are directly active in the stereospecific polymerization of propylene. Moreover, it has been possible to confirm that the formation of isotactic macromolecules occurs through a process of polymerization of the monomeric units on a metal-carbon bond. The actual structure of the complex containing ethyl groups bound to the a-titanium trichloride surface is not completely known. For this reason, it is not possible to say whether the number of ethyl groups found in the polymer may be considered equal to the number of active centers, which are expected to exist on the a-titanium trichloride surface, where the growing process of macromolecules occurs. The number of ethyl groups found in the polymer may be a multiple of the molecules of the chemisorbed complex, and therefore a multiple of the actual active centers, considered as sites of the a-titanium trichloride surface, on which that complex is chemisorbed. This factor will be the larger the larger

56

G . NATTA AND I. PASQUON

the number of ethyl groups present in the complex adsorbed on a-titanium trichloride. For the above reasons, we have named the active centers so determined "conventional active centers." It was observed that, during the treatment with radioactive alkylaluminum of some samples of a-titanium trichloride containing small amounts of Ti(1V) compounds, traces of ethylene may be released which may polymerize to give traces of polymer that could be considered as the cause of radioactive contamination. The polymerization of ethylene would occur on active centers that, a t least partly, may be also active for propylene polymerization. For that reason, one can suppose that, after having prepared the catalyst, more or less long polyethylenic chains, instead of the -CzH6 groups only, are fixed on some active centers. However, we think that, such an occurrence does not substantially alter the value determined for the -CzHb groups, which correspond to the active centers. In fact, the radioactive contamination, expressed in terms of numbers of carbon atoms, is only a small fraction of the carbon atoms corresponding to the active centers. In some tests, moreover, it has been possible to observe that the product which may be considered the cause of the radioactive contamination is not soluble in boiling ether and is, only partly soluble in heptane; it contains, therefore, long chains of ethylene groups. This assumption was confirmed by some adsorption tests of triethylaluminum carried out a t 70" on a-titanium trichloride (sample A ) (Table X ) . In these tests the number of ethyl groups fixed on a-titanium trichloride, measured from gas evolution, by decomposing the catalyst,, was found to be practically equal to the corresponding number of ethyl groups found in the polymer after polymerization, corrected for the radioactive contamination. One must consider also that the determination of the radioactivity of the evolved gas was carried out after cooling the gas at -78", so as to separate the hydrocarbons higher than ethylene (49).

B. DETERMINATION OF THE NUMBEROF ACTIVECENTERS BY A

KINETICMETHOD

In the above paragraph the number of conventional active centers, initially present on the surface of the catalyst, have been evaluated. It has already been noticed that the number of active centers, which the monomer can directly reach, on the surface of the catalyst, may vary after the start of the polymerization, until it twsurnes to a constant value. The previously described method for the determination of the number of active centers gives, therefore, a value that may not, correspond to the number of active centers found in steady-state conditions.

KINETICS OF POLYMERIZATION OF (Y-OLEFINS

57

I n the present section, there will be summarized the measurements employed for the determination of the whole number of centers which participate in the polymerization until it reaches the steady-state conditions. 1. Introduction

The method employed to calculate the total number of active centers relies upon the determination of the variation occurring in the ratio between ethyl groups (deriving from the alkylaluminum) which are present in the polymer and the whole amount of polymerized propylene, on increasing the time of polymerization. Such a ratio can be directly determined on the polymei free of catalyst, provided that the polymerization of propylene is carried out in the presence of "C-labeled alkylaluminum. The processes leading to the entry of ethyl groups into the polymer are: Formation of the catalyst, followed by the start of the polymerization on carbon-metal bonds, present in the catalyst, C h i n transfer between the ethyl groups contained in the metallorganic compounds (ethylaluminum in particular) and the growing polymeric chains. Since the catalyst is formed by the reaction of alkylaluminum with a-titanium trichloride, each active center is of the type [Cat]C2H5,and the first polymeric chain originated by each active center has a terminal -C2H5 group. During the polymerization, many different chain transfer processes are involved, some of them not involving a transfer of ethyl groups. Therefore, only a portion of the chains ends with a -C2H5 group. We shall indicate by (ra r k ) the sum of the entry rates of ethyl groups into the polymeric chains, during the polymerization, and by C* the amount of conventional active centers (in mols) , involved in the polymerization within the interval of time 0, t. Since a t the end of the reaction the catalyst is decomposed and then the polymeric chains, which were growing a t the moment when the polymerization was interrupted, were separated from the catalyst, the whole amount of ethyl groups that may be found in the polymer is as a result (we suppose that a t the beginning of the reaction, one -CzH6 groups corresponds to each active center) :

+

rt

and, with respect to the amount of polymer obtained a t the time t : rt

58

Q. NATTA AND

I. PASQUON

where El = -C2Ha mol./mol. of polymerized propylene a t the time t Qt = mols of propylene polymerized at the time t By assuming r3 r4 independent of the time and

+

Qt

=

jotrpdt = r,t,

then (34) becomes

Such a relationship should allow the determination of C*, provided that a regular variation of ethyl is detectable experimentally, on varying the r,) is small enough polymerization time. This will occur only when (ra to make possible the evaluation of C*.(rg 7-4) can be deduced from experimental data. I n fact, for large enough polymerization times, we get

+

+

E~ = E ~ = , 4*

=

constant

TP

where El, = the asymptotic value to which El tends with the increasing of polymerization time. Thus, it follows

C*

=

(Et

- E1,)Qd

(37)

It is necessary to notice that Equation (37) can give reliable results only if (rg n ) / r , remains practically invariable during the whole polymerization, so that its value can be calculated after large polymerization times. Furthermore, this method enabled us to evaluate separately the number of the conventional active centers from which, respectively, each of two particular fractions of polymer, having different steric composition, is originated: an amorphous one, ether soluble, and another one (ether extraction residue) containing stereoblock and isotactic chains.

+

2. Total Number of Actice Centers In Fig. 34 has been plotted the specific radioactivity (corrected for the radioactive contamination) and the corresponding number of -CPHI groups found in the polymer against, the polymerized propylene mols. The evaluated number of active centers C* docs not depend in practice on the polymerization time. In fact, the curve of Fig. 34, drawn by assuming C* = 0.6 X lo-' mol. of -C2H6 per mol. of a-titanium trichloride and

KINETICS OF POLYMERIZATION OF a-OLEFINS

E' '0

a

'

59

Ltr/min.

-300

0

Q

I I

I

I I

0

I

I 1

I I

1/2h l b h 0 1 2 3

3h 4 5

6

?

c. T I

I d m e r i z o t i o n time? 5,h 7,112h llh 8 4 i o 11 12x10-" molsC3Ha polymerized

FIG.34. Specific radioactivity of the polymer (and corresponding values of -C2Hs mol. per mol. of polymerized CsHa) plotted us. the amount of polymer obtained at different polymerization times (t = 70", PC,A, = 450 mm. Hg, 14C-labeledAI(C2H&Cl: 4.88 X 1Czmol./l., ground a-TiCls: sample A : 1.95 X 10-* mol./l,).

CzH6 mol. per mol. of polymerized -CzH6 , agrees with Era = 0.5 X the values deduced from the experimental tests quite well (32). Therefore, with the catalyst under consideration, ground a-titanium trichloride-A1C1(C2H6)z,the totality of active centers is involved in the catalytic process, once the polymerization starts. We may deduce from the time constancy of the polymerization rate from the very beginning of the reaction that, in this case, the number of active centers is invariable and that the calculated value corresponds to the number of conventional active centers involved in the polymerization, in steady-state conditions. 3. Active Centers Generating Non-Atactic Polymers

The eame methods employed with raw polymers were used for fractions of isotactic polypropylene (ether extraction residue), containing also some stereoblock polymers (32). The results obtained are plotted in Fig. 35. In this case, the value 0.4 X lo-* mol. of CaH6 mol. of a-titanium trichloride was found for C*. Such active centers, although they represent only 35 of the whole amount of centers present in the polymerization, generate nearly 93% of the whole amount of polymer. Such a result may be interpreted by the assumption that the growth rate of the amorphous polymeric chains (having a molecular weight which is lower than that of of the growth rate of isotactic chains. the isotactic chains) is about These data must be considered as being largely approximate.

60

Q. NATTA AND

I. PASQUON

FIG.35. Specific radioactivity of the non-atactic polymeric fraction (and corresponding values of -C2Hb mol. per mol. of polymerized C3Hs) plotted us. the amount of polymers obtained at different polymcrieation times. (Tests performed in t,he conditions reported in Fig. 34.)

4. Discussion of the Results

By comparing the above reported data with the ones of Table X, it becomes evident that the amount of --CzHK groups corresponding to the active centers, which were determined by the kinetic method for the system a-titanium trichloride-diethyl monochloride aluminum, are of the same order of magnitude as those determined by adsorption, although the former results are somewhat higher (0.6 instead of 0.3% CzHK mol./TiCla mol.). Such a discrepancy of values may be ascribed to many reasons: 1. The data obtained through adsorption measurements may be defective because, by this method, only the ethyl groups, which are present on the surface of a-titanium trichloride since the start of the polymerization, have been determined. 2. The reported data were corrected for radioactive contamination; such contamination was measured before polymerization, in conditions that could not be strictly related to the ones occurring during the polymerization.

VI. Mean Lifetime of the Growing Polymeric Chains A. DETERMINATION OF THE MEANLIFETIME FROM THE NUMBER OF ACTIVECENTERS The data summarized in the previous paragraphs (namely, over-all polymerization rate, average polymerization degree for isotactic polymer, and number of active centers) enabled us to evaluate the mean lifetime of the

KINETICS OF POLYMERIZATION OF (Y-OLEFINS

f

12

61

avertge life of the growing chains

FIG.36. Approximate values of the mean lifetime of the chains of polypropylene growing on the active centers of a catalytic system: ~~-TiCl~-Al(CzH~)~-n-heptane at 70" and 950 mm. Hg p c , ~ (The ~ . calculations were performed assuming the number of conventional active centers: C = 1 mol. per 100 mol. of ground a-TiC13: sample A and zn = 95[q]1,56.)

growing macromolecules of polypropylene, in the a-titanium trichloridetriethylaluminum catalytic system. The assumption was made that the growth life is related to the time during which the polymeric chain remains bound to the active center. The determination of the mean lifetime of the growing macromolecules was made under the following conditions: t = 70°, pcaHs= 950 mm. Hg. We have considered that polymeric fraction for which the relationship between intrinsic viscosity and average numerical polymerization degree is known (see Sec. 111, I). The number of isotactic chains growing at the same time, was assumed as being equal to the number of conventional active centers: 1 X lo-' mol./ mol. of a-TiC1,. The results obtained are plotted in Fig. 36. The value so calculated is approximate and conventional and could not be otherwise because of the data assumed for its determination. On the other hand, the life of a single macromolecule may differ remarkably from the average value, since the catalytic system under consideration gives rise to largely polydispersed polymers. Furthermore, the growth rate of a single macromolecule may be far away from the average value; actually, we believe that the more crystalline fractions which generally correspond to higher molecular weight, have a higher growth rate.

B. VARIATION OF

MOLECULAR WEIGHTDURING POLYMERIZATION

THE

THE

In a polymerization process the number average degree of polymerization a given instant, is given by the relationship

(2,) a t

62

Q. NATTA AND I. PASQUON

xn =

monomer polymerized at time t macromolecules present in the system at time t

Hence, xn =

where

6’

rpdt/(Cf

+

4‘

Z r r dt)

(38)

(39)

T , = polymerization rate Zrt = sum of the rates of chains transfer and termination processes c,*= number of growing chains a t time t

If r, , Zrl , and C1* are independent of the time, l/x, = C*/r,t

+ ZrJr,

By substituting the intrinsic viscosity and considering that

+

l / ( [ e I / ~ ) ” “= ~ * / r , t Zrt/rp

(40) [T] =

Kxa, (41 1

A variation of [ q ] against polymerization time is observed when the ratio C*/Zrt is not too small. In Figs. 37 and 38 are plotted the values of the intrinsic viscosity and its reciprocal raised to 1.35 for the isotactic polymer , fraction, obtained with two catalytic systems: a-TiC13/Al(C2HL)3 These data enable us to determine the mean lifetime ( L ) of the growing macromolecules (38),with the result that

L

c*xn/rp It follows from the relationship (41) assuming p = l/a, =

(42)

and consequently

From Figs. 37 or 38 we obtain for the asymptotic value of [TI, respectively, LA = 16 min. and LD = 10 min. for the two considered catalytic systems. We have assumed B = 1.35. These values are smaller than the ones which may be found from the data tabulated in the above section, but as previously noticed, this discrepancy might be justified by the fact that the value assumed for the conventional active centers, which give rise to isotactic chains, is undoubtedly higher than the real ones. We have also observed that the polymerization rate obtained in the considered conditions is higher than the one calculated with the relationship (2). It really occurs a t the beginning of the

KINETICS OF POLYMERIZATION OF a-OLEFIN8

63

v)

I

0

i

2

3

4

5 6 7 8 h Polymeritat ion time

5

FIQ.37. Intrinsic viscosity raised to 1.35 (proportional t o polymerization degree) of the non-atactic polypropylene fraction, plotted vs. the polymerization time. (The tests were performed with 4 g,/l, of two samples (A and D)of grounda-TiC18 ,2.35X 10-1 mol. Al(CzHs)s/l., t = 15", PC,H~ = 200 mm. Hg.)

FIQ.38. Reciprocal of the intrinsic viscosity (raised to 1.35) plotted us. the reciprocal of the polymerization time. (The experimental points are those reported in Fig. 37.)

polymerization process by employing ground a-TiClr (see, for instance, Fig. 11). On the other hand, the slope of the lines plotted in Fig. 38, from the value of which the mean life of the chains is calculated, may be greatly effected even by small errors in the determination of the intrinsic viscosity of polymers obtained by shorter time tests. The resulting data are mostly concerned by eventual errors. Finally, we believe that the growth rate of the longest and sterically purest polymeric chains is higher than that for the shortest chains.

64

G. NATTA AND I. PASQUON

C. BLOCKCOPOLYMERS (HETEROBLOCK POLYMERS) * From the data summarized in the above sections, it follows that The mean lifetime of polypropylene chains growing on each active center can reach several minutes. The catalytic complexes can “be alive” for a long time in the absence of monomer. The active centers of the catalytic system a-TiCls/Al( C2Ha)3may polymerize either propylene or ethylene. Such findings allow us to foresee the possibility of synthesizing block copolymers formed alternately by sequences of monomeric units having different natures. To attempt this polymerization, experiments have been performed by feeding a catalytic system, kept at 15’, alternately during a few minutes: with propylene at low pressure ( < 1 atm.), with nitrogen in order to stl ip the propylene, with ethylene at low pressure ( < I atm.), with nitrogen, and so on. The presence of block copolymers has been established by submitting the polymer obtained to extractions with various solvents, by examination of each fraction by X-rays, by measurements of its melting point,, and hy resilience measurements (52, 10).

VII. Conclusions The data here related on the kinetics of the propylene polymerization and of the transfer processes and the studies of the catalysts carried out with 14C-labclledalkylaluminums, derive from a series of resesrches mostly carried out some time ago, when the knowledge of the mechanism of the considered catalytic processes was still rather limited. Nevertheless, it helped remarkably to know these new processes of anionic coordinated polymerization: their true catalytic nature (which regard to a-TiC13) differentiates them from the more usual polymerization processes (radicalic) which, actually, are not catalytic. They substantially contributed to demonstrate that the anionic coordinated polymerization is a step-wise addition process in which each monomeric unit inserts itself into a metal carbon bond of the catalytic complex. We think that the reported studies, in spite of their preliminary character, can contribute not only to open new fields of investigation in the branch of macromolecular researches and to know the nature of a particular kind of catalytic complexes, but also t o deepen the knowledges in the field of the heterogeneous catalysis. We wish to thank Miss Mirella Bersani for translating this work.

* The denomination “heteroblock polymers” was suggested in order t o differentiate these polymers from the stereoblock ones whichcontain only one type of monomer (61).

KINETICS OF POLYMERIZATION OF a-OLEFINS

65

REFERENCES 1. Natta, G., Atti accad. nazl. Lincei, Mem. Classe sci.fis. mat. e nat. [8] 4,61 (1955); J . Polymer Sci. 16, 143 (1955). 2. Natta, G., Pino, P., Corradini, P., Danusso, F., Mantica, E., Mazzanti, G., and Moraglio, G., J . Am. Chem. SOC. 77,1708 (1955). S. Natta, G., Chim. e ind. (Milan) 37, 888 (1955). 4. Natta, G., Chim. e ind. ( M i l a n ) 38, 751 (1956). 6. Natta, G., Opening lecture, 16th Intern. Congr. Pure Appl. Chem., Ezperientia Supplementum 7, 21 (1957). 6. Natta, G., Paper presented at Intern. Conf. on Chem. of Coordination Compounds, Rome (1957) suppl. to Ricerca sci. 28, 1 (1958). 7 . Ziegler, K., Brennstoff-Chem. 33, 193 (1952). 8. Natta, G., Pino, P., Mazzanti, G., and Giannini, U . , J . Am. Chem. SOC.79, 2975 (1957). 9. Natta, G., Pino, P., Mazzanti, G., and Longi, P., Gazz. chim. ilal. 87, 570 (1957). 10. Natta, G., Paper presented at Intern. High Polymer Conf., Nottingham (1958), J . Polymer Sci., 34, 21 (1959). 1 1 . Natta, G., Pasquon, I., and Giachetti, E., Italian Patent 580,507 (1957). 12. Natta, G., Angew. Chem. 68,393 (1956). I S . Natta, G., and Corradini, P., Atti accad. nazl. Lincei, Rend. Classe sci.fis. mat. e nat. [8] 4, 73 (1955). 14. Natta, G., and Corradini, P., Makromol. Chem. 16.77 (1955). 16. Flory, P. J . , “Principles of Polymer Chemistry,” p. 237. Cornell University Press, Ithaca, N. Y., 1953. 16. Natta, G., Corradini, P., and Bassi, I. W., A t t i accad. nazl. Lincei, Rend. Classe sci. fis. mat. e nat. [8] 19, 404 (1955). 17. Corradini, P., and Pasquon, I., Alti accad. nazl. Lincei, Rend. Classe sci. Jis. mat. e nat. [8] 19, 453 (1955). 18. Natta, G., Corradini, P., and Cesari, M., Atti. Accad. nazl. Lincei, Rend. Classe sci. j s . mat. e nut. [8]21, 365 (1956). 19. Natta, G., Corradini, P., and Bassi, I. W . , Makromol. Chem. 21,240 (1956). 20. Natta, G., Corradini, P., and Bassi, I . W . , J . Am. Chem. SOC.80, 755 (1958). 21. Cartmell, E., and Fowles, G. W., “Valency and Molecular Structure.” Butterworths, London, 1956. 22. Natta, G., Pino, P., Mantica, E., Danusso, F., Mazzanti, G., and Peraldo, M . ,

Chim. e i n d . ( M i l a n ) 38, 124 (1956). 2.9. Baenziger, N., and Rundle, R. E., Acta Cryst. 1. 274 (1948). 24. Klemm, W . , and Krose, E., 2.anorg. Chem. 263, 218 (1947).

26. Natta, G., Pino, P., and Mazzanti, G., Italian Patent 545,332 (1954). 26. Natta, G., Danusso, F., and Pasquon, I., Czechoslov. Chem. Communs. 22, 191 (1957). 27. Natta, G., Corradini, P., Bassi, I. W . , and Porri, L., Atti accad. nazl. Lincei, Rend. Classe sci. fis. mat. e nat. [8] 24, 121 (1958). 28. Natta, G., Pino, P., Mazzanti, G., and Longi, P., Gazz. chim. ital. 88,219 (1958). 29. Stiihler, A., and Bachran, F., Ber. 44, 2907 (1911). SO. Natta, G . , Pasquon, I., and Giachetti, E., Angew. Chern. 69.213 (1957). $1. Natta, G., Pasquon, I., and Giachetti, E., Chim. e ind. (Milan) 39, 1002 (1957). 3%.Natta, G., Pasquon, I., Pajaro, G., and Giachetti, E., Chim. e ind. ( M i l a n ) 40. 556 (1958). SS. Natta, G., Pasquon, I., and Giachetti, E., Chim. e ind. (Milan) 39, 993 (1957). 34. Natta, G., Rend. isl. lombardo sci. Part 1,78. 307 (1945).

66

G. NATTA AND I. PASQUON

Natta, G., and Simonetta, M., Rend. ist. lombardo sci. Part 1, 78, 336 (1945). Natta, G., and Mantica, E., J. Am. Chem. SOC. 74,3152 (1952). 36. Natta, G., Pino, P., and Farina, M., Atti Simposio Intern. Chim. Macromol., Torino (1955), Suppl. t o Ricerca Sci. 26, 120 (1965). 36. Szwarc, M., Levy, M., and Milkovich, R., J . Am. Chem. SOC. 78, 2666 (1956). 37. Szwarc, M., Nature 178, 1168 (1956). 38. Unpublished data, obtained in the Inst. of Ind. Chem. of the Polytechnic of Milan. 39. Natta, G., Pasquon, I., and Giachetti, E., Chim. e Znd. (Milan) 40, 97 (1958). 40. Natta, G., Pasquon, I., and Giachetti, E., Makromol. Chem. 93, 258 (1957). 41, Moraglio, G., Chim. e Ind. (Milan),in press. d.a. Natta, G., Paequon, I., Giachetti, E., and Pajaro, G., Chim. e ind. (Milan) 40, 267 (1958). @. Natta, G., GiuffrB, L., and Pasquon, I., Atti accad. nazl. Lincei, Rend. Claese sci. fi.mat. e nal. [8] 26, 417 (1958). 4.4. Bonitz, E., Angew. Chem. 67, 525 (1955). 46. Natta, G., Giachetti, E., and Pasquon, I., Italian Patent 587,506 (1957).

46. Natta, G., Pasquon, I., Giachetti, E., and Scalari, F., Chim. e ind. ( M i l a n ) 40, 103 (1958).

47. Natta, G., and Pegoraro, M., Chim. e id. (Milan)in press. 48. Natta, G., and Pasquon, I., Atti accad. nazl. Lincei, Rend. Claese sci. $8. mat. e nat., (81 20,617 (1869). 49. Natta, G., Pajaro, G., Pasquon, I., and Stellacci, V., Atti accad. nazl. Lincei: Rend. Classe sci. fis. mat. e. nat. [8] 24, 479 (1958). 60. Gwathmey, A. T., and Cunningham, R. E., Advances in Catalysis 10.57 (1958). 61. Natta, G., and Danusso, F., Chim. e ind. (Milan)40, 743 (1958). 68. Natta, G., Giachetti, E., and Pasquon, I., Italian Patent 594,018 (1958).

Surface Potentials and Adsorption Process on Metals R. V. CULVER Dept. of Metallurgical and Chemical Engineering, University of Adelaide, Adelaide, South Australia AND

F. C. TOMPKINS Chemistry Department, Imperial College of Science and Technology, London, England Page

I. Introduction. . . . . . . . . . . . . . . . . . ............................... 11. The Surface Properties of A. The Work Function.. ... ..................................... B. The Contact Potential Difference.. . . . . . . . . . ........... 111. The Modification of Surface Properties by Adso ........... A. The Change in Work Function. .................... ............... B. The Change in the Contact Pot ............................. C. The Dipole Moment.. . . . . . . IV. The Preparation of Clean V. The Meaaurement of Work-Function Changes.. . . . . . . . . . . . . . . . . . . . . . . . . . A. Thermionic Method.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Field-Emission Microscope

....................

68 74 76 77 77 78 82

92

D . Magnetron Measurements.. . . .

. . . . . . . . . 106 .......................

A. Desorption Processes.. B. The Free Energy of De

110

. . . . . . . . . 113

.....................

67

118

G8

R. V. CULVER AND F. C. TOMPKINS Page

XII. Conclusions . . . . _ . .. . 127 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

I. Introduction When a gas is adsorbed a t a mctal surface, the observed change in work function is brought about by electronic interaction between the metal and the adsorbate. Most chemisorptions involve electron transfer,* the nature of which is related to the electronic structure and thc surface properties of the metal. At the outset, therefore, it is desirable to consider the adsorption process and the formation of chemical bonds at metal surfaces in general terms. The interaction of a molecule such as Hz with a metal surface may he illustrated by means of the potential energy diagram ( 1 ) shown in Fig. l , where curves ( 1) and (2) refer to the potential energy of a molecule as a function of itdsdistance from the surface of the metal. In (1) the molecule is physically adsorbed, whereas in ( 2 ) it is dissociated into H atoms prior to chemisorpt,ion. The behavior of a molecule approaching the surface depends upon the relative positions of the two curves. Near the surface the molecule is tittractcd hy long-range forces and its potential energy is described hy curve (1). If, howcver, the molccule can acquire an activation energy equd t,o, or greatcr than, Ea , then at point S it may transfer t,o curve ( 2 ) : i d become adsorbed in the atomic state. As shown in Fig. I , physical adsorption is normally characterized by a low heat of adsorption ( - A H ) , . Thc heat of chemisorption ( - A H ) is largcr and may be expressed in t,erms of the atomic heat of adsorption (-AH), and the heat of dissociation ( - A H ) d , viz., (-AH) = 2( -AH)a - (-AH)d . A large ( - A H ) value is indicative of a strong chemisorption bond, and the magni, CO suggests that the bonds tude of ( - A H ) for gases such as Hf , 0 2 and are chcmical in nature and thereby involve some transfcr and sharing of elect.rons. The adsorption of an alkali metal atom may be considered in a similar manner wit,h reference to Fig. 2. The minimum pot,entinl cnergy lcvcls in curves ( 1 ) and (2) correspond to the adsorption of Na atoms and NLLions, respectively. Since level A is lower than level B, atoms will adsorb in the ionic form, and hence a Na atom approaching t,hc metal surface will transfer from curve (1 ) to curve (2) at point S. On desorptiori the ion cxtracte an electron from the mctal and is removed in the atomic state. The heat of adsorption for the adsorption of Nu atoms as Na ions is given by the difference in energy levels A and C . *The term “electron transfer” has been used throughout to signify either partial or complete transfer.

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

69

-AH

Distance from surface

FIG. 1 . Potential energy curve for the adsorption of HI on a metal.

D i s t a n c e from surface

*

FIG.2. Potential energy curve for the adsorption of Na on tungsten.

The participation of the solid in electronic interaction with adsorbates is conveniently discussed in terms of the band theory. Here the energy levels are grouped into allowed bands and the distribution of the energy of the electrons within the band is regarded as being continuous. Figure 3 shows the potential energy of an electron in a metal as a function of distance along

70

R. V. CULVER AND F. C. TOMPKINS

FIG.3. Potential energy of an electron at a metal surfuce.

a line of atom centers. The position of the surface cannot be exactly defined, but the potential energy curve approaches a horizontal asymptote which represents the potential energy of an electron outside the metal. Within the metal, the allowed energy bands extend upwards indefinitely. At any temperature, however, most of the states with an energy exceeding EF (the Fermi energy) are empty, and the energy required to remove an electron, initially possessing the energy E p , from the metal is given by e+, where t$ defined in this way is the work function expressed in volts and is an important parameter in various electron-emission phenomena such as thermionic, photoelectric, and field emission and changes of surface potentials. A more exact definition will be given in Sec. 11. The presence of adsorbates modifies the potential barrier at the metal surface and this may lead to the adsorption of positive or negative ions. In the former case, as seen in Fig. 4a, the highest occupied level of the adsorbate of ionization potential I lies above the Fermi level of the metal so that an electron may transfer from the adsorbate to the metal. When, for example, a Th atom is located near a W surface (Fig. 4b), the neutral atom is unstable as the ionization potential of T h (4.0 v.) is less than the work function of W (4.5 v.). Consequently, the outermost electron of the T h atom transfers to the metal, and a positively charged layer of T h ions is

Metal

Adsorbate

(0)

Metal

Thorium

(b)

FIQ. 4. Chemisorption of positive ions.

SURFACE POTENTZALS AND ADSORPTION PROCESS ON METALS

71

FIQ.5. Chemisorption of negative ions. EF

Metal

Adsorbate

deposited on the surface. The dipole layer with its positive side outwards lowers the metal work function by increasing the electrostatic potential energy and thereby decreasing the potential energy of an electron just outside the layer. Alternatively, if, as seen in Fig. 5 , there is a vacant energy level in the adsorbate below the Fermi level of the metal, an electron may transfer from the metal to the adsorbate. This lowers the Fermi level, and as a result the work function is increased. Such is the case when O2 is chemisorbed on a metal surface in the form of negative ions. The electron affinity of the 0 atom is greater than the work function of the metal, and during chemisorption electrons are accepted into the vacant energy levels of the 0 atom. In covalent bonding, the chemisorbed phase is stabilized by electron exchange rather than by electrostatic forces. This may involve the interaction of an electron from a filled band of the metal and an electron from the adsorbate to form a bonding orbital. Chemisorption bonds of this type, however, always display a degree of ionicity as revealed by the presence of a dipole layer on the metal surface. The factors which favor positive-ion, negative-ion, and covalent bonding on metal Eurfaces have been considered in detail by Dowden ( 2 ) . He concludes that covalent bonding will be favored by ( 1 ) large values of the work function, ( 2 ) a positive and large change in the density of energy levels at the Fermi surface, and ( 3 ) the presence of unfilled atomic orbitals. For a semiconductor like Ge, the pattern of electronic interaction between the surface and an adsorbate is more complex than that for a metal. Semiconductors possess a forbidden gap between the filled band (valence band) and the conduction band. Fig. 6a shows the energy levels for a semiconductor where E v represents the energy of the top of the valence band, Ec the bottom of the conduction band, and E p is the Fermi energy level. The clean Ge surface is characterized by the presence of unfilled orbitals which trap electrons from the bulk, and the “free bonds” give rise to a spacecharge layer Z and hence a substantial dipole moment. Furthermore, an appreciable field is produced inside the semiconductor, as distinct from a metal, and positive charges may be distributed over several hundred A.

r

R. V. CULVER AND F. C. TOMPKINS

72

r

4

FIQ.6. Energy levels at a germanium surface (a) before and (b) after exposure to

02.

beneath the surface. The effect of bringing 0 atoms t o the surface is seen in Fig. 6b. Electrons are accepted from the semiconductor, and the space charge is reduced. The total change in the surface dipole therefore comprises the change in the work function due to the adsorption of oxygen and a contribution from the potential barrier. From an experimental viewpoint, early attempts to measure the changes in work function which occurred during adsorption were generally frustrated by inadequate high-vacuum techniques, while the surfaces employed were frequently ill-defined and seldom reproducible. Richardson and Iiobcrtson (3), for example, in 1922 were unable to interpret the complicated results obtained when HP was adsorbed on 1% and W filaments. Modern developments largely stem from the work of Langmuir and Kingdon (/t) (1923), who investigated the adsorption of alkali metals on W. Later :I series of papers by Bosworth ( 6 ) (1937-1945) served to draw attention to the importance of work-function measurements in the study of adsorption problems. Considerable progress has resulted from the development of modern high-vacuum techniques and the use of extended surfaces provided by metal films. Consequently, reliable data relating to various metal alkali metal and metal gas systems are steadily accumulating. The essential features of this review are concerned with ( 1 ) the metal surface and the influence of adsorbates on the surface properties, (2) the experimental determination of the work function, ( 3 ) the interpretation,

+

+

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

73

and (4)the application of the work function data. Thus, we first define the work function for a clean metal surface and consider the effect of a surface dipole layer arising from the presence of an adsorbate. This is followed by the experimental sections which deal with the preparation of clean metal surfaces, the measurement of work-function changes, and the nature of the results obtained. The work-function change is then discussed in terms of current adsorption theory with particular reference to bond formation in van der Wads' and chemisorption phenomena. Finally, we draw attention to the application of work function measurements in the determination of activation energies of desorption and surface migration, and to the use of work-function data in the computation of heats of adsorption.

II. The Surface Properties of Metals There is a double charge layer a t the surface of a metal which has its origin in the unsymmetric distribution of the electronic charge ( 6 ) .On quantummechanical grounds, the electronic charge distribution spreads beyond the limits normally imposed by the presence of adjacent cells in the interior of the metal in order to lower the kinetic energy of the electrons. The consequence is seen in Fig. 7. In Fig. 7a the charge distribution about the sur-

(C)

FIG.7. Charge distribution at a metal surface (a) as for atoms in bulk metal, (b) actual distribution, and (c) charge density (b minus a).

74

R. V. CULVER AND F. C. TOMPKINS

face atoms corresponds to that of the atoms in the interior. The actual distribution is shown in Fig. 7b, the mean charge density being indicated by the density of stippling. The net result (a minus b ) in Fig. 7c is a double layer with negative charges on the outside and positive charges on the inside. The spread of electronic charge is accompanied by (and limited by) the increase in the potential energy, so that in Fig. 7b the electrons are in positions of higher electrostatic potential than in Fig. 7a. This partly accounts for the increase in potential energy; exchange and correlation effects, however, make an additional contribution. The difference in electrostatic potential which exists between the inside and the outside of the metal is termed the surface potential. The related properties-the work function and the contact potential differencerespectively measure free energy changes when electrons are moved from one conductor to a vacuum and from one conductor to another. The thermodynamic basis of these properties has been reviewed by Herring and Nichols (6),and Chalmers (7) has considered the theory of contact potentials.

A. THE WORK FUNCTION The state of a gaseous or solid phase containing electrons may be described thermodynamically by the electrochemical potential of the electrons. Thus, in an isolated body of volume V containing n electrons, the electrochemical potential p is defined as

P

=

(aF/an)T,V

where F is the Helmholtz free energy of the system at temperature T and F = U - TS, U being the total internal energy of the system and S its entropy. This is equivalent to defining p as the work done in bringing an electron from infinity (zero energy) and adding it isothermally to the system. For the more practical case, i.e., a composite system of several phases, the electrochemical potential in the ith phase p i is given by pi =

(aF’/andT,v.nkZi

where ni is the number of electrons in the ith phase and F’ = U - Z T k S k . Now any change A% in the electrostatic potential Q, of a body to which electrons are added will alter ii by -eA@, and since the electrostatic potential inside the conductor can be varied by moving external charges, etc., p depends upon surface and external conditions as well as on the internal state. It is possible, however, to isolate the effect of charge distribution on surface conditions by defining the chemical potential p as p =

P

+

dinner

so that it depends only upon the temperature and the internal conditions

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

75

of the body. Here we assume (8) that the surface energy is small compared with the total energy of the system and that there are no large electrical or thermal gradients. The significance of the electrochemical potential is apparent when related t o the concepts of the usual statistical model of free electrons in a body where there are a large number of quantum states c populated by noninteracting electrons. If the electronic energy is measured from zero for electrons at rest a t infinity, the Fermi-Dirac distribution determines the probability P( 6 ) that an electron occupies a state of energy c given by

P ( c ) = l/{exp[(e

-

ii)/kTl

+ 1)

Thus, the electrochemical potential ii occurs as a parameter in the energy distribution of the electrons. The work function #J of a uniform surface is defined in terms of the difference between the electrochemical potential of an electron inside the surface and its electrostatic potential energy -eeiPouter at a point just outside the surface. This is a distance of about lo-’ cm. from the surface where the electron is regarded as being removed from the range of significant surface interactions (9). Thus,

9

= -+outer

-

’E

= +inner

-@outer

- c1-E

(1)

where @inner is the electrostatic potential inside the conductor. The work function 6 comprises the “inner work function” p/e and the surface poten- @ i n n e r ) , the double charge layer being produced by tial ( l o ) ,x = the unsymmetrical arrangement of electrons at the metal surface. The value of the surface potential is given by x = 4irMt, where M t is the total dipole moment per unit area. The surface potential is a structure-sensitive property, so that for a metal crystal there is a different value of x for each crystal face; p, on the other hand, is a bulk property which is independent of the surface structure. This leads t o a change in the work function from face to face-a well-established experimental fact ( 6 ) . Work function differences also occur on a finer scale. The chemical potential p is sensitive to surface geometry, particularly when small radii are involved, since it is obtained by integrating the image force ( -e’/a)/2(x2 2az) of an electron at. a distance x from a spherical surface of radius a. The inner work function may (If) account for more than 80% of the total value of 9; consequently, #J depends upon the regularity of the surface on an atomic scale. For projections on a surface of radius lop7cm. Lewis (fa)calculates that the work function may be diminished by as much as 50 %. These variations in work function due to polycrystallinity and irregularities in surface geometry result in patch surfaces and

+

76

R. V. CULVER AND F. C. TOMPKINS

give rise to patch fields ( 1 3 ) in electron emission. In such cases, the surface properties are more correctly described in terms of an average work funcion 4.

B. THECONTACTPOTENTIAL DIFFERENCE When two isolated and electrically neutral conductors A and B with electrochemical potentials j i A and j i B , respectively, are placed in electrical contact, j i A and j i B change in value so that a t equilibrium j i A = p B . This effect is due t o the transfer of electrons from the conductor with the higher is equal to the initial differelectrochemical potential until ( @ A - aB)inner ence p A - p B , p, the chemical potential, being insensitive t o variations in electron density. Thus from Equation ( 1 ) it follows that an electrical potential diffcrence V A B , equal to (iPA - @ B ) o u t c r , and hence to the difference in work function + B - + A , appears between points just outside the surface of the two conductors, i.e., V A B

=

(@A

- @B)Outor

=

+B

- +A

(2)

where V A Bis the volta potential or contact potential difference (C.P.D.) between the two conductors. The equililxium conditions at the junction of the two metals A iuid B are illustrtited in Fig. 8. Here it is seen that when the two metals are in contact the Ferini levels, E l p must coincide. In this case 4 B > + A ; hence, metal A is charged positively with respect to metal B , arid there is a doriblc charge layer at the junction between the two metals due to electron transfer from A to B. As the work function specifically refers t o a homogeneous crystal face, the C.P.D., as defined by Equation (2), will, in general, be nonzero when the subscripts A and B refer to different crystal faces of the same conductor or to different grains in the surface of a polycrystalline conductor. The effect of these patch surfaces on the C.P.D. may be treated as follows (6): Consider first an electronic conductor with a patch surface, each patch having its own double charge layer and consequently its own particular work function. Then, in the absence of an externally applied electric field, the electrostatic potential outside the ith type of patch at a distance which

1 -

c

A

0

A

FIG.8. The C.P.D. at a metal junction.

$I

E

SURFACE POTENTIALS AND ADSORPTION PROCESS O N METALS

77

is very small compared with the patch dimensions is given by +Pi,outer

= -ii/e

-

6i

(3)

where 6 ; is the work function of the ith type of patch and ji is the electrochemical potential inside the conductor. Equation ( 3 ) shows that the value of varies from patch to patch according t o the difference in work function. However, a t a distance from the surface sufficiently large compared with the dimensions of the patches, the electrostatic potential becomes constant. This constant potential is given by &outer

=

Zafi+i,outer

=

-p/e -

6

where f i is the fraction of the surface occupied by the ith type of patch and = &fa$, . Now consider two isolated and electrically neutral conductors A and B , both with patchy surfaces and at the same temperature, connected electrically. Then, if the distance of separation is large compared with the size of the patches, so that the patch fields between the two conductors are negligibly small, the equilibrium state corresponds to that in which the electrochemical potentials in the two conductors are equal. Thus,

6

- +B)outer

( 9 . 4

=

6 B

- 6.4

and a difference in potential exists between the outer neighborhood of the two patch surfaces which is equal to the difference of the average work functions.

111. The Modification of Surface Properties by Adsorbates Adsorbates modify the inherent dipole structure of a metal surface and hence change the work function and other surface properties. This effect may be evaluated in terms of the dipole moment produced by the adsorbed layer on the metal surface. A. THECHANGE IN WORKFUNCTION When u dipoles per unit area are aligned perpendicular to the surface of a conductor, the resulting system approximates to a double charge layer, one layer positive and the other negative, of density ue and separated by a distance d. If the dipole moment of the individual adatom complexes is independent of coverage, the system may be regarded as a capacitor, or dipole sheet, whose potential is proportional to the number of individual dipoles per unit area. The product of the electronic charge and the distance d is the dipole moment M . There is no external field with a perfect double layer, and, according t o classical electrostatics, the electrostatic potential changes sharply from (@inner - +outer) to (@inner - @outer) f 4 ~ u Min moving from a point inside the metal across the double layer to a point outside

78

R. V. CULVER AND F. C. TOMPKINS

the surface. The value of p / e , the inner work function, is unaffected by adsorption, since it is a bulk property. Hence, when dipoles are deposited on the surface of a metal, the work function, as defined by (@inner

- *outer - p / e > ,

changes by 4mM, increasing when the double layer is negatively charged outwards and decreasing in the opposite case. When the distribution of charge is not continuous ( l d ) , the region over which this potential change occurs is somewhat smoothed out (9). Thus, if A and A' refer to the state of a metal surface before and after the deposition of a dipole layer on the surface, +A#

=

+A

zt 4 r ~ M= + A f 4rumBM

where u = Bumdipoles per cm.2, 0 being the fraction of the total number of sites per cm.', urn,available to the adsorbate. The change in the value of the work function corresponding to a coverage 0 is, therefore, A+ = 4ru,BM

B. THECHANGEIN

THE

CONTACT POTENTIAL DIFFERENCE

Since it has been shown that the deposition of dipoles on the surface of a conductor changes the work function, it is apparent that the C.P.D. between two metals A and B of surface potentials x A and x B , respectively, will be modified by the presence of adsorbate on one of the metal surfaces. Thus, if surface A' (representing surface A plus adsorbate) is substituted for surface A in Equation (2), the new C.P.D. is given by C$B - C$Al . It i,s possible, therefore, t o determine the value of by difference. Then if xAtand x A are the corresponding surface potentials related to the work function by +,, = - x A - p / e , etc., we have

VA,, = + A - + A t = x A ~ - X A IrA,, now represents the C.P.D. between a clean and covered metal surface A . It is equal to the difference in surface potential A x which is a function of the change of the double-layer strength a t the metal surface, viz., A x = 4*A(uM). Some ambiguity has arisen in the adsorption field regarding the use of x defined in Sec. I1 as x = (@outer - @'innor). Clearly, the surface potential consequent upon adsorption, or more correctly the change in S.P., Ax, and the C.P.D., are equal in magnitude and opposite in sign to the change in the work function A+, C. THEDIPOLEMOMENT The evaluation of the dipole moment from the observed change in work function on adsorption demands a knowledge of the nature of the adsorption

SURFACE POTENTIAL8 AND ADSORPTION PROCESS ON METALS

79

(4 FIG.9. Dipoles at a metal surface (a) van der Waals’, (b) ionic, and (c) covalent adsorption.

process. There are three cases to be considered ( a ) van der Waals’ (physical) adsorption, ( b ) ionic adsorption, and ( c ) covalent chemisorption. 1. V a n der Waals’ Adsorption. Referring t o Fig. 9a, it is seen that the adatoms or adsorbed molecules are polarized by the surface field. Alternatively, a small amount of electron transfer may take place. In either case the resulting dipole moment is small and

M

= Atp/4ramB

where M = ed, , d, being the distance between the positive and negative charges. 2. Ionic Adsorption. For a completely ionized adsorbate on a metal surface, e.g., Na ions on W, Langmuir (16) considered that the double charge layer was formed by the adions and their electrical images in the metal. Thus, electrons emitted from the metal surface crossed only half the double layer and the measured work function change was related t o the dipole moment by the expression A 4 = 2sumBM’, where M’ = ed, d being the distance between the charge and its image. Recently, however, the reality of these image charges has been questioned (16), and it has been shown from the theory of image forces that the electrical image has no real existence; it is, in fact, concentrated almost entirely on the surface opposite the adatoms ( 17 ) . Consequently for ionic adsorption on a conducting surface, if the effective dipole length is the distance do between the positive charge and the surface (see Fig. 9b), then M = eda may be written as M = A4/4samB 3. Coualent Adsorption (with Ionic Character). I f it is assumed that metallic conduction starts at the metal surface, the effective dipole length is dun/2, where d M Ais equal to the distance between the adsorbate and metal atoms as shown in Fig. 9c. Thus, for a dipole moment M = edMA, M = Atp/2sumB

IV. The Preparation of Clean Metal Surfaces In the study of work-function changes of metals caused by the presence of adsorbates, the most troublesome aspect is that of obtaining a sufficiently well-defined surface. There is considerable experimental evidence based on

80

R. V. CULVER AND F. C. TOMPKINS

thermionic emission, field emission, and C.P.D. measurements, for the nonuniformity of metal surfaces. Thus, for W the values of the work function derived from thermionic emission experiments (6), Crystal face. . . . . . .

111

Workfunction,+V.. . 4.39

112

116

001

110

011

4.69

4.39

4.56

4.68

4.53

lead to a variation from face t o face of at least 0.3 v. Such differences promote preferential adsorption; e.g., Cs appears to be adsorbed on the face with the highest work function ( I t ? ) , while 0 atoms are attracted t o faces of low work function (19),as might be expected from a consideration of the energy levels in Figs. 4 and 5. Furthermore, there is an inherent lack of definition on an atomic scale owing to the heterogeneous nature of the surface which may arise from lattice defects, dislocations, edges, corners, etc. However, the effect of impurities in the solid or presorbed gas is more serious; in fact, contamination may so radically change the surface of a metal that normal chemisorption processes are inhibited ( 2 0 ) .Various techniques have been evolved to prepare surfaces which are free from presorbed gas. These include the use of flashed metal filaments, metal films deposited from the vapor phase, and positive-ion bombardment of solid surfaces. Although the technique is limited t o cases where the oxides of the m e h l are unstable, clean surfaces of highly refractory filaments such as W, or Mo, may be readily produced by flashing at high temperature in vucuo. The area Hence, on cooling, the surface is of the surface is only of the order of 1 highly susceptible to contamination by readsorption of residual gas in the system. Nevertheless, if the pressure of the residual gas is reduced with the aid of getters t o a sufficiently low value, the rate of contamination may be so diminished that the surface remains tolerably clean during the course of an investigation. Since a monolayer of N2 , for example, is adsorbed on a W filament in 100 sec. at a pressure of mm. Hg, it appears that the maximum residual gas pressure in the system should not exceed mm. Hg (21). A pressure of this magnitude can be easily measured with the inverted ionization gauge ( 2 2 ) . The necessity for adopting modern high-vacuum techniques in preparing the filament has been emphasized by Thomas and Schofield ( 2 3 ) , who showed that even the W filaments used by Roberts (24) in his classical experiments on the accommodation coefficient were not properly cleaned. This method of obtaining a clean surface by flashing is basic to field-emission microscopy and is of fundamental importance in being probably the only method to yield unambiguously clean surfaces; other methods require the test of agreement with it. The surface produced may be, to some extent, polycrystalline, but the advantage of the metal filament lies in the

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

81

simplicity with which a fresh surface may be prepared for successive measuremen ts . Metal films are produced by the thermal evaporation of an outgassed metal, which may be in the form of a filament or a bead supported in a refractory crucible, and subsequent condensation on a cooled substrate ( 2 5 ) ;according to the mode of preparation, a preferred orientation of one crystallographic plane is sometimes obtained ( 2 6 ) .Evaporated metal films deposited a t low temperatures have a high (surface arealfilm weight) ratio and also a high surface energy, and hence are unstable, the microcrystallites comprising the film tending to change spontaneously t o more stable forms. The area of a metal film may be as high as lo6 cm.*/g., as for nickel deposited on n glass substrate, and, moreover, the area is approximately proportional t o the weight of the film (27). In the unsintered state, metal films are, in general, unsuitable for measurements of the work function, since their electrical behavior, as determined by the microcrystallites, does not correspond to that of the bulk metal (28). Many films exhibit an expanded lattice structure of 1 t o 2 % (29), and it is only after sintering that the normal metallic properties are observed. These structural changes have been detected by work-function measurements which disclose that the work function of the metal film increases during sintering ( 3 0 ) .An evaporated metal film weighing 50 mg. may be composed of 10,000 atomic layers with an internal surface of about 5000 cm.*, and since 10'' molecules are required to cover this area-a figure considerably in excess of the reservoir of gas adsorbed on the walls of a well-outgassed vessel-the film may be expected to remain uncontaminated for some time ( 3 1 ) .Many arguments have been adduced for the cleanliness of the metal surface obtained by evaporation ( 3 1 ) . In particular, it is found that ( a ) Robert's heat of adsorption data for W filaments agree with those of Beeck ( 3 2 ) for evaporated metal films so that, apparently, similar surfaces were produced in each case and it is reasonable to assume that both were clean; ( b ) surface areas calculated from physical adsorption data agree with those measured by chemisorption-an unlikely situation if some shes were contaminated, since the areas determined by chemisorption would be less than those obtained by physical adsorption which is nonspecific; and ( c ) chemisorption is found to be a function of film weight, whereas if the films were seriously contaminated, the effect should be more marked for a 5-mg. film than, say, for a 50-mg. film. The results of Hickmott (33), however, have caused these criteria for the cleanliness of evaporated metal films to be accepted with some reserve. In this instance a W film was prepared by evaporation from a W filament, commencing a t a pressure of lo-'' mm. Hg, in an all-glass system which had been subjected to a rigorous outgassing in accordance with ultra high-

82

R. V. CULVER AND F. C. TOMPKINS

vacuum procedure. Hickmott observed that after an initial drop, the pressure rose and did not return to its original value even after evaporation ceased. In view of the rapid chemisorption of gases by W, it was concluded that the film so produced was contaminated probably by gas desorbed from the glass system. Later, when the evaporation was carried out in short bursts, the initial low pressure fell steadily, showing that the W was acting as a getter; thus, it was possible to prepare a clean film by a slight modification of the conventional technique. Ion bombardment is equally applicable to single crystals and polycrystalline surfaces. Farnsworth et al. (34) have shown that this treatment is effective for various metals and for the semiconductor Ge, which has been used in the form of a single crystal for work-function measurements ( 3 5 ) . Cathode sputtering (positive argon ion bombardment) removes the contamination and up to several hundred surface atomic layers, depending upon the time and intensity of the treatment, but positive ions still remain trapped in the surface. The surface also contains a number of defects due to the removal of metal atoms from their equilibrium lattice position. It is necessary, therefore, to anneal a t an elevated temperature to remove lattice defects and occluded argon. There is considerable risk of contamination during the annealing period, and the pressure in the system must be kept below lo-'' Hg ( 3 6 ) .The actual state of the surface is determined by the low-energy electron diffraction method. However, it has been correctly pointed out (37) that the reproducible diffraction maxima obtained after sputtering and annealing are not necessarily proof of a clean surface, although they correspond to the diffraction maxima of the surface metal lattice; a regularly contaminated surface might also give reproducible patterns which could be interpreted as arising from an oriented surface.

V. The Measurement of Work-Function Changes The change in work function which accompanies adsorption on a metal surface may be determined directly by thermionic, field-emission, and photoelectric methods. Indirect methods rely on the measurement of a C.P.D. between a reference electrode and the original and covered surfaces, respectively.

A. THERMIONIC METHOD For thermionic emission from a uniform surface, the maximum electron flux j across the uniform surface at a temperature T in the absence of an applied field is given by j

=

A ( 1 - P > Texp(-e+//cT) ~

where 7 is the mean reflection coefficient of electrons a t the surface, assum-

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

83

ing a Maxwellian dist.ribution of velocities, and A is a constant. An earlier relation j = AT2 exp( -e+*/kT) (4) where +* is temperature-independent was shown by Bridgman (38) to involve certain approximations which were not strictly valid. Nevertheless, if j is measured as a function of T and (log j ) / T 2is plotted against 1/T, a straight line is usually obtained. The slope of this “Richardson plot” is -e+*/kT, +* being the apparent work function. A similar relation applies for a covered surface, and the work function change A+* refers t o differences in $* values for covered and bare metal surfaces. The emission equation is valid for each of the patches on a nonuniform surface, but the total measured current from the various patches depends on the relative magnitudes of the collecting field and patch fields, as well as on the work function. Hence, the effect of some patches may be out of proportion to their area and the average work function of a polycrystalline surface measured thermionically may differ somewhat from the true average work function 6. In practice, unless p is small, measurable thermionic currents are obtained onlyat high temperatures. Thus, when the work function is increased by an adsorption process, the electronegative film may be partly evaporated before the requisite temperature for thermionic emission has been reached. An exceptional case is the adsorption of 02 on W. However, the thermionic method has proved very useful for studying the electropositive films produced by alkali metals (39). Cs, for example, reduces the work function t o such an extent that thermionic measurements may be made a t temperatures aslow as 150”. The electrode assembly described by Becker (81) is shown in Fig. 10.

FIG.10. Electrode assembly for measuring thermionic emission.

81

R. V. CULVER AND F. C. TOMPKINS

It comprises an axially mounted W filament and three cylindrical collectors C1, C2 , and Ca. All collectors are maintained at the same positive potential with respect t o the W filament, but only the emission current collected from C2 is recorded. In the region of CZthe temperature and the surface concentration of adsorbate on the emitter are assumed to be uniform.

B. THEFIELD-EMISSION MICROSCOPE The presence of an external field F v./cm. at the surface of a metal will modify the burrier t o electron emission, comprising the work function 4

and the Ipermi energy E p, so that there is a finite probability that electrons approaching the surface will tunnel through the barrier and be emitted. The electrons concerned are those with energies near E , , and the tunneling probability is very sensitive to the tunneling distance, which, in turn, is inversely proportional to the applied field. The theoretical expression describing the electron emission is the Fowler-Nordheim equation ( 4 0 ) ,

j/V’

=

abexp[-6.84

7

3/2

X 10 4 / k V ]

(5)

where j is the total current, V t8heapplied voltage, a the emitting arm, and k ( = F / V ) is a constant relating the applied potential V to the field F a t 4). the surface. The constant b is given by 6.2 X 106k2(Ep/+)”2/(E, Now, assuming that 4 is independent of F , the work function may be readily deduced from the slope of the plot of log (j/V’>against 1/V; alternatively, 4 may be calculated from values of the current density by means of Equation ( 5 ) . This leads to an average work function increment & when measurements are made for a clean and then for a covered metal surface. In its uncorrected form, however, the observed value of due t o an adsorbed layer is somewhat less than that predicted by Equation ( 5 ) , since the potential of a discrete layer reaches $ only at some distance from the surface. Thus, the contribution of the adsorbed layer to the energy barrier is reduced-an effect which is most marked at low coverage (41 ) .

+

Anode

-Fluorescent

screen

i

Pump

FIG. 11. Field-emission microscope (diagrammatic).

SURFACE POTENTIALS AND ADSORPTION

PROCESS ON METALS

85

Experimentally, work-function measurements which rely on the cold emission of electrons are carried out in the field-emission microscope (F.E.M.) (21).The apparatus, as shown in Fig. 11, consists of a W tip P sharpened by electrolytic polishing so that the radius of curvature is cm., and an anode in the form of a film of Aquadag. A variable potential of 3 to 15 kv. is applied to the anode, and the electrons, pulled out from the point, travel in approximately straight lines to the fluorescent screen. The linear magnification obtained is of the order of lo6to lo6.The secondary electrons from the screen are collected by the anode, and the field-emission current is measured by a sensitive microammeter.

C. PHOTOELECTRIC METHOD The energy of photoelectrons liberated from a metal surface is h( v - YO), where vo is the threshold frequency, and, if V , is the voltage required to stop the electrons when light of frequency v falls on the surface, vo may be calculated from the relation eV, = h( v - YO). In turn, the threshold frequency is related t o the work function a t 0" K by the Einstein equation hvo = ~I#JO , which retains its validity at room temperature. When the photocurrent for monochromatic light of frequency v is plotted against the potential, as shown in Fig. 12a, the retarding potential V Ris given by the intercept on the voltage axis, although the curve may be distorted because the collector itself exhibits photoelectric activity. If the emitter and collector are of identical composition, there is no C.P.D. between them and V R = V , . Alternatively, the photoelectric yield, or the photocurrent per unit light intensity, is measured for different frequencies of incident light. The point

-

wL- - R e t o r d i n g potential 4

(a)

V

-

~~

h

\

La

Wavelength

c

(b)

FIQ. 12. The variation in the photoelectric yield with (a) retarding potential and (b) wavelength.

86

R. V. CULVER AND F. C. TOMPKINS

at which the spectral distribution curve cuts the wavelength axis is the threshold wavelength A0 , As shown in Fig. 12b, the distribution of electron velocities gives a pronounced “foot” to the curve, although an approximate value of XO is usually obtained by inspection. The change in the work function is calculated from the relation

4.4 - 4.4t = hc/(X/

- XOA’)

where XoA and XO-” are the threshold values obtained for the clean and covered surfaces A and A’, respectively. Fowler’s analysis of spectral distribution curves involves the equation (@), log ( I / T 2 )= B

+ F(x)

where I is the photocurrent, B a constant independent of v,

x = h(v - vo)/kT, and F ( z ) is a universal function of z whose form is the same for all metals and all temperatures. Accordingly, when log (ZIT’) is plotted against hv/kT, the curve obtained will fit the theoretical curve of log ( I / T 2 )= F ( z ) if it is moved horizontally by an amount hvo/kT and vertically by an amount B. The horizontal shift directly determines the true threshold frequency vo and hence the work function. In the photoelectric method, the measured average work function is always less than the true 4, since patches of high work function tend to be excluded from the emission process. Thus, the nonuniform distribution of adsorbate on a patch surface may cause a slight discrepancy in the evaluation of G. Experimentally, the photoelectric method has various limitations. Photocurrents of the order of lo-“ A. must be measured accurately in the region of y o , and for films of work function greater than 5 v., the threshold frequency lies in the far ultraviolet-a practical disadvantage. Furthermore, the method is inapplicable at pressures in excess of mm. Hg because of ionization of the gas by collision. Figure 13 shows a photocell used recently for work-function measurements ( 4 3 ) . A metal coating inside the cell serves as the anode, and B is the cathode, which may be a metal foil or a film previously evaporated from a filament at C.Also provided at C are a W emitter for bombarding the cathode surface and a Pt electrode for thermally dissociating gas molecules. The monochromatic light which passes through the quartz window Q strikes the cathode B, and its energy can be measured by a calibrated photocell.

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

87

Pump

1

To adsorbate rrsrrvoir

FIG.13. Photocell for measuring the photoelectric work function.

D. CONTACT-POTENTIAL-DIFFERENCE METHODS Here either the C.P.D. between a surface A and a reference surface R is measured directly, or else some property dependent on the C.P.D. is measured. In each case, when the surface A is changed to A’ by the process of adsorption, n varying potential is applied to A’ until it behaves as A ; alternatively, the new C.P.D. is measured directly. This change in potential is the required C.P.D. between A and A’. Thus, if V A R and V A r R are the C.P. differencesbetween the reference electrode R and the clean and covered surfaces A and A’, respectively, V A t A

=

V A ’ R

-VAR

=

+A

-

where and + A are the work functions of surfaces A’ and A . The C.P.D. is obtained directly by the magnetron and the capacitor (or Kelvin) methods. Other methods which have proved successful rely on the variation of anode potential in a diode with constant cathode conditions, In this case, since adsorption changes the effective anode potential, the applied potential necessary to restore the anode current to its original value is equal to the C.P.D. between the two surfaces. As considered in Sec. 11, a true average work function is measured in the C.P.D. method when the two conductors are separated by a distance which is much greater than the size of the patches on the surface. These conditions are invariably fulfilled in the capacitor and the space-charge-limited diode methods. 1. Magnetron. This method depends upon the magnetron action of an

88

R. V. CULVER AND F. C. TOMPKINS

axial magnetic field which reduces the saturation emission current i n a cylindrical diode. C.P. differences are measured between the clean and covered anode surfaces and a central W cathode which acts as a reference electrode. The total potential V between the anode and cathode is given by (44): =

V,

+ VAR + Vt

where V , is the applied potential, VAR is the C.P.D. between the anode and reference surfaces, and V t is a temperature-dependent term related to the kinetic energy of the electrons emitted from the cathode. I t has been shown (45) that if Ilo is the field strength required t o reduce the anode current to half its original value, then V = H,"R2e/8m,R being the radius of the anode and elm the specific charge of the electron. Since E l 0 is proportional to the current 1 0 producing it,

Va

+ VAR + Vc = kl,"

where k is a constant characteristic of the geometry of the system. During an experiment, values of 1 0 are measured a t several applied potentials V , for a given anode surface and cathode temperature. When I," is plotted against V , , the intercept on the voltage axis is - (VAR V t ) ,and VAR is readily calculated by using the appropriate values of V 1, which are usually about 0.2 ev. The arrangement of the electrodes in the magnetron method ( 4 6 ) is siniilar to that shown in Pig. 10. The anode-consisting of the central collector and two guard cylinders, all made of Ta-may be moved independently of the cathode filament. After removal from the vicinity of the cathode, the anode may be cleaned by electron bombardment, and then, if required, a metal film deposited on the inside of the central collector. The accuracy of the measurements largely depends on the symmetry of the system, so that the whole electrode assembly must be sufficiently rigid to withstand rigorous outgassing without deformation. 2. Capacitor. The potential difference between two conductors A and B at the same temperature connected by a circuit containing no source of e.m.f. is given by the difference in work function $ B - $ A according to between the Equation (2). The application of a potential V O= $ B two conductors brings them t o the same potential, but for a value V # Vo there will be a net charge on one of the conductors of ( V - V,)/C, where C is the capacitance. If V is adjusted so that V = Vo , no current flows when ~ . This is the basis of the capacithe capacitance is varied and V = c $ tor method, where the surface under consideration and a reference electrode form the plates of a capacitor. The reference electrode must remain inert at normal temperatures when exposed to gas or vapor during the measure-

+

SURFACE POTENTIALS A N D ADSORPTION PROCESS ON METALS

89

ment of the C.P.D., and oxidized Ni and Pt surfaces apparently fulfil this requirement ( 4 7 ) .A continuous reading can be obtained when one capacitor plate is vibrated by means of a flexible diaphragm ( 4 8 ) .I n this case, except at the null point, an ax. signal is produced in the external circuit, where it may be conveniently detected, after amplification, by a cathode-ray oscillograph. A recently developed cell in the form of a hollow glass tuning fork is shown in Fig. 14 ( 4 7 ) . The upper part is held rigid by cementing it into a metal block, while the two glass limbs are vibrated by the electrical excitation of a piece of iron attached to B. The vibrating capacitor plate is joined to the inside wall of limb A , and the adjacent nonvibrating plate is fixed to the top of the cell. The metal to be examined is deposited on the vibrating plate from a filament (not shown), the nonvibrating plate being shielded during evaporation. Alternatively, the film is evaporated on to the inside of limb A , which then acts as one plate of the capacitor. 3. Space-Charge-Limited Diode. This method is equally applicable t o electropositive and electronegative adsorbed layers. In a diode the maximum current which can be drawn from the cathode is given by the emission formula in Sec. V,A. Electrons emitted from the cathode a t low current densities build up an electron atmosphere or space charge a t the surface, and this space charge presents an electrostatic barrier t o electrons flowing to

t I @

Iron block

-

,~

Vibrating electrode

Pump

_Electromognet

Non-vibrating electrode

FIG.14. Cell for C.P.D. measurement by the capacitor method.

90

R. V. CULVER AND F.

C. TOMPKINS

the anode. Since the electrostatic barrier is higher than the surface barrierthe Fermi inner potential energy plus the metal work function-the anode current is controlled by the space charge, which is a function of temperature, and not by the cathode work function. The theory of the diode has been considered by Gysae and Wagener (49).They showed, on the basis of a Maxwellian distribution of emitted electron velocities, that the current flowing between the anode and cathode depended almost entirely on the applied potential and the mean anode work function. Thus, the shift of currentvoltage characteristics due to a change at the anode surface represents the change of anode work function. In the diode the cathode is usually a W filament which can be flashed and maintained as a reference electrode at a temperature above which adsorption occurs. The anode may be of similar construction or take the form of a metal film evaporated from an adjacent filament. Experimentally, current polarization curves are obtained, first for the clean anode surface A and then for the covered anode surface A'. Alternatively, resistance-voltage characteristics are measured (SO). The potential difference comprises the applied polarization and the C.P.D. between the emitter and collector. For a given anode current j , va

+

V A R

=

vat

+

VA'R

where V , and Vat are the applied potentials, and

va -

vat

=

VA'R

-

V A R

=

VA'A

Thus, the displacement of the characteristic current voltage curves along the voltage axis represents the required C.P.D., VAfn (Fig. 15).

"a Potrntiol

FIQ. 16. Current-voltage curves obtained in the space-charge-limited diode.

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

91

VI. Adsorption and Work-Function Studies A. THERMIONIC MEASUREMENTS The effect of alkali metals on the emission from refractory metal filaments was examined by Langmuir and Kingdon ( 4 ) in 1923. The adsorbed film reduced the work function, so that it was concluded that the adsorbate was ionized to form a dipole layer on the metal surface. Furthermore, they suggested that the formation of positive ions and their subsequent evaporation at high temperature required that the ionization potential of the alkali atoms should be less than the work function of the metal filament. Thus, for Cs with an ionization potential of 3.9 v., positive Cs ions were obtained with W and Ta filaments whose work functions are about 4.5 v., but not with a Th-coated W filament having a work function of 3.0 v. During the adsorption of Cs on a metal surface, the formation of positive ions ceased when the value of the work function of the metal plus adsorbate fell below that of the ionization potential of Cs. With a W surface, the work function was found to vary with coverage, and Taylor and Langmuir (60) attributed this effect to ionic adsorption with mutual depolarization of the resulting dipoles. Continuing these investigations, Becker (21,61),showed that (1 ) Cs was adsorbed as ions as well as atoms, (2) as the concentration of adsorbed Cs increased, the ratio of adions to adatoms decreased, and (3) the temperature for migration of adatoms and adions over the surface was lower than that required for evaporation. Becker also examined the adsorption of Ba (62) and Th ( 6 3 ) on W filaments. Recently, Moore and Allison ( 6 4 ) inBa and W Sr. The magnitude of the apvestigated the systems W parent electric moment of the adatom complex was found to be of the same order as that expected for ionic adsorption. Nevertheless, Moore and Allison prefer to explain the thermionic activity in terms of polarized atoms instead of making the usual assumption of ionization, for which they stress there is no direct experimental evidence. On the W surface, a similar decrease in work function was observed for all crystal faces; and it was argued that if adions were present, they should be concentrated on faces such as the (110), where the highest metal work function obtains. It is difficult, however, to distinguish between polarization and partial ionization, particularly on a heterogeneous surface at low coverage, and it is possible that both polarization and ionization effects contribute t o the total dipole moment. The surface potentials experimentally determined for the adsorption of alkali metals on W are always positive, and the maximum values recorded for Cs are +2.9 v. (66) and +3.0 v. (60); for Ba +2.4 v. (62) and +1.9 v. ( 6 4 ) ;and for Sr +2.2 v. ( 6 4 ) ,so that the dipole layer is oriented with its positive pole directed away from the metal surface.

+

+

92

R. V. CULVER AND F. C. TOMPKINS

The adsorption of O2 on W has been investigated by Kingdon ( 5 6 ) . In this case, the temperature required to produce a satisfactory emission current was also sufficient to evaporate 0 2 from the filament at a measurable rate, and additional Oz was continuously supplied to the filament to compensate for the loss. The results were difficult to interpret, but Reimann (57) used the data to show that the adsorption of 0 2 on W increased the work function of W by 1.75 v. at 1500" K. B. THEFIELD-EMISSION MICROSCOPE Work-function measurements have been made recently in the F.E.M. for t$e systems W O2 , W H2 , and Ni HZ. As considered in Sec. V,B, an average work function may be calculated from the Fowler-Nordheim equation, although several assumptions are inherent in this procedure, uiz., the constancy of the emission area with adsorption and the use of a single average work function 6 to describe a patch surface. Using a clean W surface (+ = 4.5G v.) as a reference, Klein (68) found that the addition of a monolayer of 0 to the metal surface raised the work function by 1.7 v. Gomer and Hulm (69) observed a similar increase of approx. 1.5 v. when a W tip was exposed to O2 at 30" K, but after heating t o 300" K and returning to the original temperature, A$ was 1.9 v. This observation suggests that activated adsorption occurs on heating, the oxygen atoms moving t o less accessible sites, thereby vacating high-energy sites for further lowtemperature adsorption. Becker and Brarides ( 6 0 ) have followed the desorption of 0 2 from the various crystallographic planes of a W tip in the F.E.M. The fraction of the total current which came from a particular plane was deduced from the optical density of the photographs, and the current densities computed on this basis were used in the Fowler-Nordheim equation. For the ( 1 1 1 ) plane, a considerable amount of 0 2 was desorbed a t 600" to 700" K, and 4 decreased from 7 to 6 v. No further change in the work function was observed until 1400"K; then further 0 2 was desorbed and $ fell gradually with increasing temperature t o 4.4 v. a t 1600" K. This value of 4.4 v. is the work function of the original ( 11 1 ) W plane. For the adsorption of H2 on W at 4.2"K, Gomer, Wortman, and Lundy (61) found that the work function increased by 0.5 to 0.6 v. at a surface coverage of 0.75 to 0.90. At 300" K, Muller (62) obtained a value of A+ = 0.43 v. The adsorption of H Pon Ni increased the work function by about 0.5 v. at 20" K ( 6 3 ); t$ appeared to decrease at higher coverages, and this positive effect was attributed to the adsorption of HZmolecules.

+

+

+

C. PHOTOELECTRIC MEASUREMENTS One of the earliest determinations of the change of photoelectric work function of a metal surface due to adsorption was made by Ives and Olphin

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

93

(64), who obtained results for films of Na, K, Rb, and Cs on W. Later the sensitivity of the photoelectric work function to electropositive films was utilized by Bosworth (66) to study the mobility of Na and K on a W surface, and for the system W K the S.P. was found t o be about +2.7 v. (66). Other measurements of the change in work function which occurred during the deposition of K and Cs on W filaments were made by Mayer ( 6 7 ) , and the results confirmed those of Taylor and Langmuir (50) obtained by the thermionic method. For the adsorption of K, the work function decreased from a value of 4.53 v. for pure W to a minimum value of 1.76 v. at a coverage somewhat less than a monolayer. Further adsorption of K increased +, so that after five monolayers had been deposited on the surface, the measured work function was that of the adsorbate itself. The increase in 0 in the region of monolayer coverage was attributed to mutual interaction of adsorbed dipoles, but it has also been related t o differences of electronic interaction on the various faces of the W crystals (68). The maximum S.P. values for the adsorption of K and Cs on W surfaces were +2.77 and +2.83 v., respectively, and for a Pt surface ( 6 7 ) the values were +3.68 v. for K and +3.76 v. for Cs. Initial measurements of the photoelectric emission for gas films on metal surfaces were carried out by Suhrmann and Csech (69). They evaporated films of Ag, Al, and T1 on to a Pt plate and determined the work function before and after saturating with Hz , The spectral distribution curves were analyzed by Fowler's method, and the maximum S.P. values, probably reHz , +0.81 v.; A1 Hz , ferring to contaminated surfaces, were Ag -0.81 v.; and Pt Hz , +2.2 v. Suhrmann and Sachtler ( 4 3 ) investigated the adsorption of various gases on a Pt foil and found that the adsorption of Hzmolecules increased the work function of the Pt, whereas atomic H decreased it. The effect of electron bombardment on Pt surfaces already partly covered with hydrogen was also noted. In one case, where the work function decreased initially, but later increased, it appeared that bombardment had first dissociated the adsorbed Hz molecules into atoms and then completely removed them from the surface. With N2, Suhrmann observed no change in the work function of the Pt foil until the molecules had been dissociated by a glow discharge and then the photocurrent fell t o zero (68). When CeHa was adsorbed on a Pt surface, the maximum photocurrent occurred in the region of monolayer coverage, and it was concluded that ?r electrons were transferred from the adsorbate to the metal (70). Suhrmann (71, 72) has also confirmed some of the results obtained by C.P.D. methods (73) for the adsorption of H P and Xe on evaporated Ni films. As seen in Fig. 16, when Hz was added to an unsintered Ni film at 90" K, the work function initially increased, but a t higher coverages each dose of gas produced a positive drift in potential. The maximum S.P. was -0.39 v., falling to -0.21 v. at a pressure of mm. Hg. To account for

+

+

+

+

94

R. V. CULVER AND F. C. TOMPKINS

HOW8

Fia. 16. Change in photoelectric yield on the adsorption of hydrogen on an sintered nickel surface.

tin-

this behavior, Suhrmann suggested that with progressive adsorption H2 molecules migrate to sites with a higher work function, where they are dissociated and adsorbed as a negative layer. Finally, at high coverage, the mobility of the adsorbate is considerably reduced, and the undissocinted Ha molecules are polarized to give Ha+ ions which form a positive layer. With Xe, the S.P. was large and positive, varying from +0.7 to +0.9 v. for a fully-covered Ni surface. The adsorption of N2 on an evaporated W film a t 20" was found to increase the work function from 4.66 to 4.88 v. (72),but it fell to 4.60 v. when the temperature was reduced t o -183". Again the effect was attributed to the presence of a polarized molecular layer, in this instance containing N2' ions. Sachtler and Dorgelo (74) measured the change in photoelectric work function when Nz and HZwere adsorbed on evaporated films of Ni and Ta. These films were deposited under a vacuum of lo-* mm. Hg, and the maxiHZ, mum surface potentials observed were 0.1 v. for the system Ni -0.44 v. for Ta HZ, and -0.38 v. for Ta Nz , On the other hand, the adsorption of H2 on a Ni surface prepared under less satisfactory experimental conditions decreased the work function (76).As a result of this work, some doubt arises as to whether the positive S.P. values reported for the adsorption of Hz on various metals, particularly Pt ( 6 8 ) ,refer to clean surfaces. A photoelectric method was also used by Baker and Rideal (76) for studying the adsorption of Hz , CO, and C2HI on evaporated metal films of Ta, Fe, Ni, and Co. Photocurrents near the threshold were too small to be measured accurately, so the threshold frequencies were obtained by in-

+

+

+

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

95

spection. The S.P. was determined as a function of surface coverage, and in some cases the relation between the S.P. and the coverage 0 was approximately linear up to 8 = 0.6; i.e., the effective dipole moment of the sdsorption complex remained constant until 60 % of the surface had been covered. This important conclusion has not been confirmed (77). With the exception C2H4 system, all the S.P. values obtained by Baker and Rideal of the Ni refer to negative layers. The maximum surface potentials were

+

Hz (V) CO (V) C& (V)

Ta

Fe

co

Ni

-0.43

-0.19 -1.15

-0.06 -0.26

-0.12 -0.46

-0.67

+0.89

+

In the Ni H2 system at room temperature, the S.P. passed through a maximum value of -0.12 v. and then fell t o -0.10 v., but prolonged pumping restored it to -0.12 v. Eisinger has correlated the observed change in work function with the number of CO molecules (78) and N2 atoms (79) adsorbed on a monocrystalline W ribbon whose surface was normal t o the [113] direction. Values of t$ for clean and covered W were obtained from Fowler plots or by comparing the photocurrents due t o the incidence of monochromatic light (1900 to 2800 A.) on the W ribbon. The number of molecules adsorbed per cm.*, N , , was determined by flashing the ribbon and noting the increase in pressure in the system. Figure 17 shows the work function plotted against N , for the adsorption of CO on W. Here it is seen that the maximum value of t$(At$ = 0.86 v.) corresponds to a coverage of 3.7 X 1014mol./cm.*, and this figure is identical with the number of atoms, Nw , in the (1 13) plane

0

2 4 N,~IO -14 m o l . / c m ?

6

FIQ.17. Change in work function with coverage for the adsorption of CO on tungsten.

96

R. V. CULVER AND F. C. TOMPKINS

of W. There is a slight decrease in 4 for coverages in excess of Nw and the maximum number of CO molecules adsorbed is approximately 2Nw . Nitrogen was rapidly chemisorbed on the W ribbon until N , was 3.7 X 10" atoms/cm.2, and 4 changed linearly with coverage up to a value of N, = 2Nw , when the work function had increased by 0.4 v.

D. MAGNETRON MEASUREMENTS The original C.P.D. determinations of Oatley (44) were made with H2 and 0 2 on a Pt anode. The gases were ionized, and the work function of the anode increased by 1.2 v. with O2 and decreased by 1.15 v. with Hz . Similarly, Weissler and Wilson ( 4 6 ) obtained C.P.D. values of -1.2 v. and up to -1.4 v. for the adsorption of O2on W and Ag films which were evaporated on to the inside of a cylindrical T a anode. Adsorption was promoted by means of a glow discharge. Less reliable positive S.P. values were obtained for the adsorption of H2 and N2 . The results were not always reproducible, and it is possible that the anode was a source of contamination.

E. CAPACITOR MEASUREMENTS Using the simple vibrating electrode arrangement of Zisman ( 4 8 ) ,Duhii (80) obtained values for the S.P. of metals such as Cu, Ni, Ag, Ta, aiid Pt at H2pressures varying from loe4to 10 mm. Hg. For Ni the S.P. increased from +0.3 v. at lo-' mm. to +0.7 v. at 10 mm. Hg and for Cu and Ag the S.P. was about -0.1 v. at lop4mm. Hg. The metal surfaccs were probably contaminated, since the design of the apparatus imposed a severe limitation on the outgassing procedures which could be employed. More reliable results have been obtained by Ogawa, Doke, and Nakada (81) for the adsorption of H2 and O2on evaporated films of Ni, Ag, Zn, and Cd. The maximum surface potentials recorded for the various metal-gas systems were Ni H2 , -0.40 v.; Ni 0 2 ,-0.55 v.; Ag Hz , -0.49 v.; and Ag O2, -0.60 v. Complicated effects were noted with Zn and Cd, no doubt because of the difficulty of outgassing these metals prior to evaporation. The effect of 0 2 on surfaces prepared by scraping the bulk metal in uacuo was investigated by Giner and Lange (8.2),who determined the C.P.D. between a Ag reference electrode and Pt, Pd, and Cd surfaces before and after oxygenation. Unfortunately, the results obtained do not apply to clean metal surfaces. Hackerman and Lee (83)have also used the simple vibrating electrode to examine the effect of O2 and other gases on evaporated metal films of Fe, Ni, Cr, Al, and Pb. Employing a Pt plate as a reference electrode, they found that the work function increased when the transition metals were exposed to O2 ; later it decreased as the chemisorbed 0 2 was converted to an oxide layer. With A1 and Pb, however, the work function steadily decreased

+

+

+

+

97

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

in the presence of Oz . Again, it is unlikely that the techniques employed were sufficiently rigorous to insure clean metal surfaces; hence, the results are of limited value. A series of papers by Mignolet (SO, 73, 84, 86, 86) describes the most comprehensive investigations carried out to date. With the apparatus shown in Fig. 14, the maximum S.P. values obtained for the adsorption of various gases on evaporated films of Ni and W were as follows: Hz Ni (V)

-0.35

(V)

-0.48

w

Nz

0 2

- 1.6 -1.90

+0.21 -0.50

Xe

CzH,

+0.85 +1.14

+0.83

GHK +0.77

Two types of adsorption were distinguished, viz., atomic adsorption giving rise to negative surface potentials, and van der Waals' adsorption, where the sign of the S.P. was positive. At - 196" the addition of H:! to a Ni surface already covered with a layer of H atoms decreased the work function by about 0.1 v., and this effect was ascribed t o physical adsorption. A similar effect was observed with Cr and W films ( 7 3 ) .The van der Waals' adsorption of Ar, Xe, NZ, C2H4 , and CZH6 on bare Ni and a Hz-covered Ni surface was also investigated. The gases adsorbed on bare Ni were always stable in vucuo, but reversible positive surface potentials were obtained on Ni-Hz surfaces. With Nz the dipole moment was about 10 times as great on a Ni-Hz surface as on clean Ni. Apart from Ni and W, the adsorption of Xe produced a substantial decrease in the work function with many other metals such as Fe, Cr, Ti, Zn, and Hg. The polarization of the Xe atoms was accounted for in terms of donor-acceptor interaction, the complex M-Xe resulting from the adsorption of the inert atom on the metal surface M being described as essentially "no-bond" ( 8 7 ) (see Sec. VIII). Bloyaert, D'Or, and Mignolet (88) have repeated these investigations with evaporated Cu films at -196" C.; HZ was not adsorbed, but stable van der Wnals' layers of Xe, CZH4, and CZH6 were obtained as with Ni. The maximum values of the S.P. were:

cu

(V)

HZ

02

Nz

Xe

CzHi

-

-0.68

+0.45

+0.66

+1.23

CZHK +0.69

CO +0.28

Bloyaert et ul. also found that the adsorption of CO on evaporated films of Fe and Ni increased the work function by 1.33 and 1.35 v. (max.), respectively. Mignolet (89) has also examined the adsorption of Hz on Pt and Pt-H: surfaces. The S.P. for the system Pt Hzat -196" was -0.14 v., in con-

+

98

R. V. CULVER AND F. C. TOMPKINS

trast t o the positive effect observed by the Suhrmann school (68, 4 3 ) . On the other hand, positive values of up t o +0.23 v. at a pressure of 4.5 X lo-‘ mm. Hg resulted from the adsorption of H2 on a Pt surface covered with atomic hydrogen, and the behavior of these positive layers was analogous to that found with Ni and W. Dillon and Farnsworth ( 3 6 ) determined the work function of a Au-plated reference electrode photoelectrically and then measured the C.P.D. between the reference surface and the clean and covered surface of a (nearly intrinsic) Ge crystal by the capacitor method. On exposure to O2a t a pressure of lo-‘ mm. Hg, the work function increased by 0.17 v., but a t a pressure of mm. Hg, Ad was reduced t o 0.12 v. There was no further change a t lo-’ mm. Hg, and after pumping the work function slowly returned t o its maximum value of 0.17 v. Hence, this loosely bound 0 2 was associated with the formation of a second adsorbed layer. The adsorption of CO also increased the work function ( A 4 = 0.11 v.), but Nf and Hz decreased it by 0.04 and 0.3 v., respectively. The observed changes in work function for the adsorption of 0 2 are in accord with recent evidence for the large density of surface states localized at the Ge surface ( 9 0 ) .These surface states are the unfilled orbitals, which act as acceptor states by trapping bulk electrons and forming a p-type conducting layer, Presumably, adsorbed gases like O2bond covalently using the unfilled orbitals. Following the original investigations of Brattain and Bardeen (91 ), I’ratt and Kolm ( 9 2 ) have examined “long-time” work function changes induced by electrostatic fields for Au, Ge, and Si. In these experiments the application of an electrostatic field of 30,000 v./cm. between two electrodes forming the plates of a capacitor changed the work function of both of them. The effect built up slowly over the period of an hour and then decayed on removal of the field. When the induced change was measured as a function of time after removal of the applied field, a logarit,hmic decay curve was observed. Illumination of the electrodes produced a similar effect. Pratt and Kolm attribute the slow surface reaction to chemisorbed oxygen on the surface of the electrode acting as an electron trap, and in arriving at this conclusion, a model based on a polar-covalent adsorption bond was found t o be in good agreement with the experimentally determined adsorption kinetics.

F. DIODE MEASUREMENTS Langmuir and Kingdon ( 9 3 ) used the diode method in conjunction with thermionic experiments to study the adsorption of alkali metals and O2on a W anode. The arrangement of the electrodes in the diode is shown in Fig. 18a. Their results were not very accurate; for the adsorption of Cs and Th, the maximum surface potentials were +2.8 and +1.4 v., respectively, but the value for O2was only -0.8 v. The technique was greatly improved by

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

99

G;

Fro. 18. Electrode assemblies used in the determination of the C.P.D.by the diode method.

Reimann (57), who mounted two U-shaped hairpin filaments of different radii, one inside the other, with their mid-points about 1 mm. apart (see Fig. 18b). Under these conditions, the emission produced by a small accelerating field was restricted to the tip of the filament, thus eliminating the need for the internal shielding used by Langmuir and Kingdon (93).In this apparatus the S.P. for the adsorption of 0 2 on W was -1.70 v., in agreement with the value recalculated from Kingdon’s thermionic data (57). Later Reimann (94) obtained a C.P.D. of +1.7 v. between clean and thoriated W filaments. A system of crossed filaments, as shown in Fig. 18c, was used by Bosworth and Rideal (95), who investigated the condensation and evaporation of alkali metals on W and determined the C.P.D. as a function of the surface coverage. The amount of adsorbed alkali metal could not be measured directly, owing to the small area of the filament; instead, the coverage was taken as proportional to the intensity of a beam of alkali metal vapor directed on to the filament and to the time the filament was exposed. Maximum values of the S.P.-W Na, f2.78 v. and W K, +2.90 v.-were reached in the vicinity of monolayer coverage. With 0 2 and N2, the S.P. values recorded by Bosworth and Rideal (96) were W O2 , - 1.74 v. Nz , -1.38 v. Again the surface coverage was determined inand W directly; in this instance S.P.-timerelationships were transformed into S.P,coverage plots from a knowledge of the gas pressure and the use of several parameters assumed t o be independent of coverage. On this somewhat arbitrary basis, an almost linear relation was observed between the S.P. and the coverage. The same C.P.D. method gave surface potentials of - 1.04 v. (97) and -1.26 v. (95) for the adsorpt)ion of H:! on W. In order to investi0 2 , Bosworth (98) evaporated Ni from a surroundgate the system Ni ing cylinder on to one of the W filaments, and the adsorption of oxygen on

+

+

+

+

+

100

R. V. CULVER AND F. C. TOMPKINS

this Ni surface increased the work function by 1.4 v. Crossed W filaments were also used by Copley and Spence ( 9 9 ) , who found that the adsorption of 0 2 increased the work function by 1.71 v. Eley and Rideal (100) obtained a similar value of 1.60 v. for the system W 0 2 , but for H2the S.P. varied from -0.1 t o -1.0 v.; this variation was attributed t o radiation from the cathode, which on occasions raised the temperature of the W collector to 800". The same electrode arrangement was used by Burshtein (101) and his collaborators to study the adsorption of O2on metals. With an Fe filament the adsorption of 0 2 at - 120"increased the work function as expected. At temperatures of loo", 150°, and 270°, however, the work function decreased initially, apparently because of the migration of O2 atoms into the bulk metal, and a plot of the C.P.D. against the number of molecules adsorbed passed through a maximum. This corresponded to the formation of two t o five protective oxide layers, depending upon the temperature of the experiment. In a modified form of the diode used recently, the anode consisted of an evaporated metal film deposited on the cylindrical cell wall (see Fig. 18d). With this type of electrode system, Mignolet (SO) checked the results obttained by the capacitor method, and the surface potentials for the adsorpN 2 , and Xe on W, as well as Xe on Hg, showed excellent tion of H2 , 02, agreement; e.g., for the W Xe system the values were +1.11 v. (diode) and 4-1.14 v. (capacitor), and for the adsorption of H2 on W, -0.50 v. (diode) and -0.48 v. (capacitor). At -196" the S.P. for the system mm. Hg owing W H2fell with increasing pressure t o -0.41 v. a t 2 X to the positive S.P. effect of the adsorbed molecular H2 . When H2 was adsorbed, the potentials measured in the diode were more negative and the adsorption somewhat greater than that found in the capacitor method. This was attributed to the presence of H atoms produced by the incandescent W cathode. Mignolet (89) also confirmed the S.P. data given by the Capacitor method for the adsorption of H2 on evaporated Pt films, the values being -0.15 v. (diode) and -0.14 v. (capacitor). By the same diode method, Bloyaert, D'Or, and Mignolet (88) obtained surface potentials of +0.30 and -0.33 v. for the adsorption of CO and atomic H on evaporated Cu films, and the value for CO agreed with that measured by the capacitor method (88). Further diode measurements have been made by Culver, Pritchard, and Tompkins (77) for the adsorption of H and CO on evaporated metal films of Cu, Ag, Au and Fe, Co, and Ni. The S.P. data obtained a t - 183" were :

+

+

+

H (V)

co (V)

CU

Ag

AU

Fe

-0.36 +0.30

-0.34 +0.31

-0.18 +0.92

-0.43 -1.64

co

Ni

-0.33 -1.48

-0.35 -1.35

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

101

- 0.35 S.P. V

I

I

10 H, atoms x

2

16'~

co

mol. x

16"

FIG. 19. The change in the surface potential with coverage for the adsorption of HZ and CO on nickel.

The data for Cu, Fe, and Ni are in reasonable agreement with those of Bloyaert et al. (88). Hydrogen was not adsorbed by Cu, Ag, or Au until dissociated into the atomic form: in most cases a plot of the S.P. against coverage was a smooth curve of steadily decreasing slope rising to a maximum value before the surface was fully covered, as shown in Fig. 19 for the adsorption of Hz and CO on Ni films.

VII. Surface-Potential Data Tables I-VI summarize the more reliable results for the adsorption of alkali metals and gases on polycrystalline metal surfaces. Many of the older TABLE I Surface Potentials for Alkalies Adsorbed on Tungsten System

S.P., volts

Method

W-Na W-K w-cs

+2.8 +2.9 +3.0 +2.8 +2.8 +2.4 +1.9 +2.2 +I .7

Diode Diode Thermionic Diode Photoelectric Thermionic Thermionic Thermionic Diode

W-Ba W-Sr W-Th

Reference Bosworth and Itideal" Bosworth and Itidealo Taylor and Langmuirb Langmuir and Kingdonc Mayer' Beckera Moore and Allison' Moore and Allison' Reimannu

Bosworth, R. C. L., and Rideal, E. K., Proc. Roy. Soc. (London) A162, 1 (1937). Taylor, J. B., and Langmuir, I., Phys. Rev. 40, 463 (1932); 44, 423 (1933). c Langmuir, I., and Kingdon, K . H., Phys. Rev. 34, 129 (1929). d Mayer, H., Ann. Physik 33,419 (1938). * Becker, J . A., Trans. Faraday SOC.28, 148 (1932). f Moore, G. E., and Allison, H. W., J . Chem. Phys. as, 1609 (1955). 0 Reimann, A. L., Proc. Roy. SOC.(London) A163,499 (1937). 0

b

R. V. CULVER AND F. C. TOMPKINS

102

TABLE I1 Surjace Potentials for HI Chemisorbed on Metals System

S.P., volts

Fe-HZ CO-HZ Ni-HZ

N

CU-Hz Ag-Hz Au-Hz W-Hz

-

Ta-Hz Pt-Hz

-

-0.19 -0.43 -0.47 -0.33 -0.1 -0.12 -0.35 -0.35 -0.39 -0.40 - 0.5 -0.33 -0.36 -0.34 -0.49 -0.18 -0.48, -0.50 -0.55 -1.04, -0.44 -0.43 -0.14 -0.15

-0.65

-1.26

Method Photoelectric Diode Capacitor Diode Photoelectric Photoelectric Capacitor Diode Photoelectric Capacitor F.E.M. Diode Diode Diode Capacitor Diode Capacitor Diode F.E.M. Diode Photoelectric Photoelectric Capacitor Diode

Reference Baker and RidealO Culver et a1.h Mignolet" Culver et aLb Sachtler and Dorgeld Baker and RideaP Mignolet" Culver et aZ.b Suhrmannf Ogawa et a1 - 0 Wortman et aLA Bloyaert el al.' Culver et a1.b Culver et al.* Ogawa et aZ.0 Culver el al.b Mignolettsk Mignoleti Comer et al.1 Bosworth and Ridealm Sachtler and Dorgelod Baker and RideaP Mignolet" Mignoletn

Baker, M. McD., and Rideal, E. K . , Nature 174, 1185 (1954). Culver, R. V., Pritchard, J . , and Tompkins, F. C., Proc. 2nd Intern. Conyr. of Surjace Activity, Volume 2, p. 243, Butterworths, London, 1957. e Mignolet, J. C. P., Bull. soc. chim. B e l g . 64, 126 (1955). d Sachtler, W. M. H., and Dorgelo, G . J. H., J. chim. phys. 64, 27 (1957). Mignolet, J. C. P., Discussions Faraday SOC.8, 105 (1950). f Suhrmann,R., 2.Elektrochem. 80,804 (1956). Ogawa, I., Doke, T., and Nakada, I., J . A p p l . Phys. (Japan) 21.223 (1952). * Wortman, R., Comer, R., and Lundy, R., J. Chem. Phys. 27, 1099 (1957). i Bloyaert, F., D'Or, L., and Mignolet, J. C. P., J . chim. phys. 64.53 (1957). i Mignolet, J. C. P., Rec. trav. chim. 74, 686 (1955). k Mignolet, J. C. P., J . chim. phys. 47, 172 (1950); J . Chem. Phys. 20, 341 (1952); b

@

21, 1298 (1953).

Comer, R., Wortman, R., and Lundy, R., J . Chem. Phys. 26, 1147 (1957). Bosworth, R. C. L., Proc. Cambridge Phil. SOC.33, 394 (1937); Bosworth, R. C, L.,and Rideal, E. K . , Proc. Roy. Soc. (London) AM%, 1 (1937). "Mignolet, J . C. P., J . chim. phys. 64, 19 (1957). 1

m

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

103

TABLE I11 Surface Potentials for 02 Chemisorbed on Metals System

S.P., volts

Method

Reference

Ni-02

-0.55 -1.40 -1.60 -0.68 -0.60 -1.60 -1.70 -1.70 -1.71 -1.74 -1.89 -1.90 -1.90 -1.2

Capacitor Diode Capacitor Capacitor Capacitor Diode Diode F.E.M. Diode Diode Diode Capacitor F.E.M. Magnetron

Ogawa et a1.a Bosworthb Mignoletc Bloyaert et al.d Ogawa et a1 .a Eley and Rideale Reimann' Kleino Copley and Spenceh Bosworth and Rideal' Bourionj Mignoletk Gomer and HulmJ Orttleym

cu-0, Ag-Oz w-02

Pt-02

Ogawa, I., Doke, T., and Nakada, I., J. Appl. Phys. (Japan) 21, 223 (1952).

* Bosworth, R. C. L., Trans. Faraday SOC.36,397 (1939).

Mignolet, J. C. P., Discussions Faraday Sac. 8 , 105 (1950). Bloyaert, F., D'Or, L., and Mignolet, J. C. P., J . chim. phys. 64, 53 (1957). Eley, D., and Hideal, E. K., Proc. Roy. Sac. (London) A178, 429 (1941). Reimann, A. L . , Phil. Mag. 171 20, 594 (1935). u Klein, R., J . Chem. Phys. 21, 1177 (1953). Copley, M. J., and Spence, R . W., J . A m . Chem. SOC.61, 3027 (1939). Bosworth, R . C. L., and Rideal, E. K., Proc. Roy. Sac. (London) A162, 1 (1937). j Bourion, R., J. phys. radium 12, 930 (1951). Mignolet, J. C. P., Rec. trau. chim. 74, 685 (1955). Gomer, R., and Hulm, J. R., J . Chem. Phys. 27, 1363 (1957). Oatley, C. W., Proc. Roy. SOC.(London) A166. 218 (1936); Proc. Phys. SOC.(London) 61, 318 (1939). TABLE IV Surface Potentials for N ZChemisorbed on Metals System

S.P., volts

Method

Reference

W-N,

-0.50 -1.38 -0.38

Capacitor Diode Photoelectric

Mignolet" Bosworth and Ridealb Sachtler and Dorgeloc

Ta-Nz

"Mignolet, J. C. P., Rec. trau. chim. 74, 685 (1955).

* Bosworth, R. C. L., and Rideal, E . K., Physica 4, 925 (1937).

Sachtler, W.M. H., and Dorgelo, G. J. H., J . chim. phys. 64, 27 (1957).

R. V. CULVER AND F. C. TOMPKINS

104

TABLE V Surface Potentials For CO Chemisorbed on Metals System

S.P., volts

Method

Reference

Fe-CO

-1.33 -1.64 -1.15 -1.48 -1.35 -1.20, -1.35 +O. 28 +O. 30 +O. 30 +O. 31 +0.92 -0.86 -0.67

Capacitor Diode Photoelectric Diode Diode Capacitor Capacitor Diode Diode Diode Diode Photoelectric Photoelectric

Bloyaert el al.a Culver e t al.b Baker and Ridealc Culver el al.* Culver el al.b Bloyaert el a1.O Bloyaert el al.a Bloyaert et al.a Culver et aLb Culver et al.b Culver et a1.b Eisingeld Baker and RideaP

co-co Ni-CO cu-co Ag-CO Au-CO we-co Ta-CO

Bloyaert, F., D’Or, L., and Mignolet, J. C. P., J . chim. phys. 64, 53 (1957). Culver, It. V . , Pritchard, J . , and Tompkins, F. C., Proc. 2nd Intern. Congr. on Surface Activity, Volume 2, p. 243, Butterworths, London, 1957. c Baker, M. McD., and Rideal, E. K., Nature 174, 1185 (1954). d Eisinger, J . , J . Chem. Phys. 27, 1206 (1957). (113) face. a

b

TABLE VI Surface Potentials for Gases Physically Adsorbed on Metal Surfacesd System Ti-Xe Cr-Xe Fe-Xe Ni-Xe Cu-Xe Zn-Xe Hg-Xe Hg-Oz Hg-CHd Hg-CtHa

S.P., volts +O. 84 +0.95 +0.66 +O. 85 +O. 66 +o. 21 +O. 23 +O. 03b +O. 16b +O. 23

System

S.P., volts +O .27

+o. 21

+O. 77 +O. 83 +1.3 $1.1 +0.45” $0. 14b +O. 69 +1.23

‘Mignolet, J . C. P., in “Chemisorption” (W. E. Garner, ed.), p. 118. Butterworths, London, 1957. b incomplete coverage.

data referred t o in the previous section have been omitted, since S.P. measurements made under unsatisfactory experimentd conditions have lit,tle significance. An inspection of the S.P. values reveals that in many cases, e.g., the sys-

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

105

tem Ni-H2, there is satisfactory agreement between the results obtained by Hz and W 0 2 also, the values various methods. For the systems W given by the F.E.M. operating under the most stringent high-vacuum conditions provide a valuable check both as to the sign and the magnitude of the S.P. measured by other methods. In these cases, therefore, it is reasonable to assume that the data refer to clean metal surfaces. Occasionally, divergent S.P. values may be directly attributed to defects in experimental technique. As pointed out by Bourion ( l o g ) , this is the most likely explanation for the results obtained by Bosworth and Rideal in a diode with crossed filaments for the systems W Hz and W Nz , viz., - 1.26 (95) and - 1.38 v. (96), respectively, compared with values of -0.48 and -0.50 v., respectively, observed by Mignolet (30). I n general, the variation in the S.P. obtained by various workers may be ascribed to the use of metal surfaces contaminated to a greater or lesser degree with foreign gases, although crystal orientation and experimental artifacts may be contributing factors. Most metal surfaces are polycrystalline, but if in some cases there is a preferred orientation of crystals in the surface, then the corresponding work function of the crystal face will predominate in S.P. measurements. Apart from this, as seen in Sec. V, the method of measuring the S.P. may account for small variations in the results obtained; a true average work function 4 is measured by C.P.D. methods, but the photoelectric work function, for example, is always less than 6. It may be deduced from the data in Tables I-VI that ( 1 ) the S.P. for the chemisorption of alkali metals on W is positive and about 2 t o 3 v., ( 2 ) the chemisorption of simple gases gives rise to small negative potentials, (3 ) both negative and positive surface potentials are produced by CO, and ( 4 ) the S.P. for physical adsorption is invariably positive. In chemisorption processes the observed surface potentials and the metal properties are not readily correlated as shown by the data for various metal Hz systems:

+

+

+

+

+

Metal

W

Ta

Fe

Co

Ni

Cu

Ag

Au

Pt

Work function, volts Ionization POtential, volts Surface potential M-Hz, volts

4.50

4.12

4.77

4.18

5.03

4.61

4.33

4.71

4.72

7.89

7.89

7.82

7.87

7.44

7.75

7.54

9.18

9.00

~

-0.48 -0.44 -0.47 -0.33 -0.35 -0.36 -0.36 -0.17 -0.14

The significance of the sign and the magnitude of the S.P. for the chemisorption of alkali metals and gases will be dealt with in Sec. VIII.

R. V. CULVER A N D F. C. TOMPKINS

106

For van der Waals' adsorption, the data of Table VI suggest a rough parallelism between the S.P. and the bond energy of the metal (103). This is exemplified by the metal Xe systems, where W (bond energy 54 kca1.l mole) and Hg (bond energy 5 kcal./mole) give surface potentials of 1.1 and 0.23 v., respectively when Xe is adsorbed. A further point of interest lies in the large S.P. values which are associated with the transition metals of the 4th period. Such behavior is not unexpected, since these metals possess holes in the d band and hence function as electron acceptors. In this respect, Xe system is surprisingly high, particuthe value of 0.67 v. for the Cu Xe falls sharply to 0.21 v. It has been suglarly when the S.P. for Zn gested, therefore, that the d-orbitals at the Cu surface are incompletely filled (108).

+

+

+

VIII. Electron Transfer and Bond Formation Chemisorption of a gas on a metal surface may involve heteropolar bonding represented by structures (1) or (2) : (1) M+-A-

(2) M--A+

(3) M-A

or covalent bonding with a small dipole moment arising from the resonance between the three structures ( 1 , 2 , 3 ) as envisaged by Pauling (104) for diatomic molecules. Electronic interaction between the metal and the adsorbate, on the one hand, or the polarization of the adsorbed molecules, on the other, is readily detected by S.P. measurements. Moreover, such changes in the electron distribution at a metal surface may be confirmed by measuring the change in electrical resistance or magnetization of the metal during the adsorption process, A. CHEMISORPTION ON METALSURFACES The positive S.P. of up to 3 v. which is obtained for the adsorption of alkali metals on a W surface provides evidence for positive ion formation with electrons transferred from the adsorbate to the metal. For Na and Cs there is reasonable agreement between the values of the experimental dipole moment (M = 1018A~/4a300a,B,as considered in Sec. II1,C) measured at low coverage, where the effects of mutual depolarization may be neglected, and the calculated values based on the electronic charge ( e = 4.80 X e.s.u.) and the radius of the positive ion do (106) : d o , A.

W-Na

w-cs

1.83 2.02

M (expt.) 11.3 12.5

M (calc.) 8.75 8.1

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

107

In this instance, therefore, it appears that the bonding is essentially ionic, but for Ba and Sr a more refined calculation shows that the ionic character of the bond is rather less than 100 % (106, 107). When gases such as Oz, Hz , and Nz are adsorbed on a metal surface, the metal usually acts as an electron donor and a double charge layer is produced with the negative side outwards. For 0 2 the magnitude of the S.P., which often exceeds 1 v., suggests that the chemisorption bond is largely ionic. The adsorption of Hz on metals has been the subject of continued investigation. Two models have been advanced to account for the observed negative S.P., uiz., (1) covalent bonding of the H atoms with electrons from the conduction band of the metal (16) ; this depresses the Fermi level and increases the work function, and ( 2 ) essentially metallic binding with resonant covalent bonds, the adsorbed H, carrying its positive charge, being located well below the metal double layer, so that the effective dipole moment is still negative outwards (208). The latter-a type of interstitial adsorption-may well be the first step in the bulk d 8 u sion of hydrogen in W or Ni. The covalent bonding of Hz on metals will be considered in Sec. VII1,B. The second model-characterized by a negative S.P. and an electron transfer to the metal-has received some support from electrical resistance measurements carried out by Suhrmann ( 7 2 ) during the adsorption of H2 on a sintered Ni film. The reverse effect, however, was noted for a freshly prepared unsintered film; i.e., the film resistance increased with the adsorption of Hz , and the difference in behavior exhibited by the two films was attributed t o the fact that the work function of the sintered film was 0.45 v. greater than that of the unsintered film. This interpretation, although plausible, may require modification, since Sachtler ( 7 5 ) has shown by S.P. and electrical resistance measurements that the positive S.P. effect may be ascribed t o contamination of the original metal surface due t o unsatisfactory experimental conditions. It is difficult to predict the sign of the S.P. when radicals are adsorbed on a metal surface. During the chemisorption of N20 or CO, for example, the bond between the adsorbate and the metal contributes t o the S.P., but the adsorbed radical also possesses a dipole moment and may participate in a charge transfer process with the metal surface. Thus, with the chemisorption of CO on Cu where the S.P. is positive, it has been suggested on the basis of spectrographic evidence (109) that coordinate binding occurs via the 0 of the CO molecule (88). Here, apparently, the S.P. is partly due t o the permanent dipole moment of the CO molecule and partly to electronic interaction with the metal. With Ni, where the S.P. is negative, it appears from the infrared spectrum of CO adsorbed on reduced Ni that two Ni atoms are involved in binding the CO molecule t o the surface

108

R. V. CULVER AND F. C. TOMPKINS

through the C atom at lower coverages, (109, 110) viz., 0

II

C

Ni

/ \

Ni

Under these conditions, the two metal atoms provide orbitals required for the covalent bond with C. For simple gases an explanation has been offered ( 1 1 1 ) for the sign of the S.P. based on the difference of the electronegativities of the metal and the adsorbate. In this analysis, the electronegativity of the metal is taken as the work function 4, and the electronegativity of the adsorbate is given by = % ( I a ) ( 1 l a ) , where I is the ionization potential and @ is the electron affinity of the atom. This relationship, (4 - +) has been of importance in the evaluation of dipole moments. Thus, in terms of Mulliken’s electronegativities, we have

+

+

M

=

(1/3.15)(4 - +), D

the figure 3.15 being an empirical conversion factor used in the transform:htion of the l’auling expression M = ( x u - xA), where xu and xn correspond t o 4 and +, respectively, on the l’auling scale of electronegativities. In Table VII, S.P. data for the adsorption of alkali metals and gases on clean W surfaces are compared with values for the difference (4 - +), and it is seen that in all cases the sign of the S.P. agrees with the sign of the electronegativity difference (4 - +). Clearly, on the basis of electronegativity, the S.P. change for the adsorption of gases will be negative, since 4 TABLE VII Electronegativity Diflerences and Surface Potentials

a

System

$, volts

W-Ba w-Cs W-Ha w-02 W-N,

2.84 2.21 7.17 9.95 7.35

$2,

volts 4.50 4.50 4.50 4.50 4.50

(.$ - $), volts

S.P., volts

+1.16 +2.29 -2.67 -5.45 -2.85

+2.4O +3.0* -0.65‘ -1.9od -0.5W

Beckor, J. A., Phys. Rev. SS, 1082 (1929); Trans. A m . Electrochem. Soc. 66. 155

(1929).

bTaylor, J. B . , and Langmuir, I . , Phys. Rev. 40,463 (1932); 44,423 (1933). Mignolet, J. C. P., J. chim. phys. 47. 172 (1950); J . Chem. Phys. 20, 341 (1952); 21, 1298 (1953). d Mignolet, J . C. P., Rec. trav. chim. 74, 685 (1955).

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

109

for most metals is between 3 and 5 v., while the electronegativity of the a), is considerably greater. adsorbates, # = $h(I An empirical relation (113),

+

V

=

0.29 (4 - 3.15xA)

has been proposed by Mignolet t o account for the S.P. in terms of the metal work function 4 and the electronegativity of the adsorbate xA . This relation describes the behavior of the systems W 0 2 ,W IZ, W Hz , Ni Hz , Fe Hz ,and Pt Hz , with deviations of up to 30 % for surface potential values varying by a factor of 9.

+

+

+

+

+

+

B. COVALENT BONDING ON A NICKEL SURFACE Broeder el al. (111) have carried out S.P. and magnetization experiments t o distinguish between ionic and covalent bonding for the adsorption of Hz and O2 on Ni. Nickel contains 9.4 electrons, 0.6 hole, and 0.G unpaired electron spin per atom in the d band, the latter being responsible for the magnetic properties of the metal. The S.P. measurements were made on an evaporated Ni film and the magnetization studies on a nickel-silica catalyst, the properties of which were regarded as strictly comparable with the metal film. The theory underlying the experiments is as follows. In the formation of a heteropolar bond by the transfer of an electron, the number of unpaired d electrons in the metal increases or decreases by one for each atom adsorbed. This means that the adsorption of a positive ion-an electron donor4ecreases the magnetization, while the adsorption of a negative ion-an electron acceptor-increases it. When a covalent bond is formed, however, an electron from the adsorbate and an electron from the d band of the metal, both with unpaired spins, occupy a bonding orbital in which the respective electron spins are paired. Since the bonding electron is withdrawn from the metal d band, as evidenced by the change in magnetization, it follows that the magnetization will decrease during covalent bond formation irrespective of the polarization or direction of the resulting bond. The magnetic method alone does not, however, distinguish between covalent bonded hydrogen and electrostatically bonded protons. A knowledge of the direction of electron transfer derived from work function or resistance measurements is required as well. If, for example, the work function change on adsorption shows that electrons have been transferred to the adsorbate, an increase in the magnetization denotes ionic bonding and a decrease indicates that the bonding is covalent. Now for the adsorption of Hz and 0 2 on Ni the surface potentials are negative, and Broeder el al. (111) observed a decrease in the magnetization in each case. The negative S.P. shows that the metal is an electron

110

R . V. CULVER A N D F. C. TOMPKINS

donor, and this behavior taken together with the decrease in magnetization suggests that H2 and O2 form covalent bonds with Ni. However, there is other evidence t o the contrary. The resistance measurements of Suhrmann (72, 11.4) showed that when Hz was adsorbed on a sintered Ni film, the metal acted as an electron acceptor, and in these circumstances a decrease in magnetization could be interpreted as positive ion formation. With 0 2 it is the change in magnetization and not the direction of electron transfer which is in dispute. As opposed to the findings of Broeder et al. ( l l l ) , Moore and Selwood (116) noted an increase in the magnetization of L: nickel-silica catalyst during the adsorption of 02-an effect which cannot be reconciled with covalent adsorption. These data together with those for the adsorption of CO and NtO are shown below: H*

Gas adsorbed

S.P.change

- (76) - (75) Film resistance (76) - (114) (114) change Electron transfer M+ Mt M+ Magnetizationchange ( I f f ) - (116) - ( I f f )

+

-

CO

0 2

-

+

NzO

(88)

+ (116) + ( 1 1 4 ) +

M-t

M+ (116)

+

(116)

+ (116)

Clearly, further work along these lines is required before any definite conclusions could be drawn regarding the nature of the chemisorption bond.

ADSORPTION ON METALSURFACES C. PHYSICAL The positive S.P. observed when gases are adsorbed on a metal surface has been atrributed t o ( a ) polarization of the adsorbate by the electron field of the metal double layer (73) and ( b ) charge-transfer effects (103). The importance of charge-transfer forces has been stressed by Mulliken (87) in his general theory of donor-acceptor interaction. If, as suggested, these charge-transfer forces contribute t o the van der Waals' attraction, then they probably take part in the physical adsorption process. The complex M - X resulting from the adsorption of an inert gas on a metal surface M has been described as essentially no-bond with a small contribution from the structure M - - X + . As seen in Table VI, the S.P., and hence the calculated dipole moment, may be of some magnitude. I n the adsorption of Xe on Nil for example, the Xe atom has an induced dipole moment of about 0.4 D, assuming the adsorbed layer to be fully occupied. This donor-acceptor interaction contributes to the heat of adsorption, thereby increasing the stability of the complex. No complete explanation has been advanced for the positive sign of the S.P., although some generalization has been made on the basis of the overlap conditions of the surface orbitals ( 8 6 ) .With the exception of 0 2 , these

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

111

appear to favor the formation of positive ions, particularly with Xe. Donoracceptor interaction cannot be interpreted according to differences in electronegativities ( I O S ) , since the molecules CH4 , CzHs , and 0 2 are unlikely to play the role of electron donors. Oxygen, for example, has an electron affinity of some 6 v., while Hg has a work function of 4.5 v., so that on the basis of electronegativities the physically adsorbed 0 2 layer on Hg should be charged negatively outwards. Thus, while the suggestion has been made that the positive effect may be accepted as experimental evidence of donor-acceptor interaction, it appears that at this stage an explanation in terms of polarization of the adsorbate along classical lines may also be acceptable (117').

IX. The Kinetics of Desorption and Surface Migration Once the change in the work function of a metal has been determined as a function of coverage, the kinetics of a desorption or migration process may be readily followed by noting the variation in the S.P. with time. Further, the application of the Arrhenius equation to the relative rates obtained leads to activation energies of desorption or mobility. The advent of the F.E.M. has stimulated interest in desorption and mobility studies. Here the spreading of the adsorbate over the metal surface is followed directly by observing the nature of the F.E.M. pattern, while the measurement of the S.P. serves to identify the coverage for the conditions obtained . A. DESORPTION PROCESSES The activation energy for desorption comprises the heat of adsorption and the activation energy of adsorption, E,, (see Fig. l ) , but, as the adsorption of alkali metals and most gases on clean metal surfaces is nonactivated, the activation energy of desorption is, in fact, equal to that of adsorption. Two classes of measurements have been made: ( 1 ) those in which desorption occurred without subsequent readsorption, and (2) those where equilibrium conditions were approached during the desorption process. A true desorption velocity is observed in the first case only. Taylor and Langmuir (60) followed the desorption of Cs ions and atoms from a W surface by a thermionic method. At low coverage the rate of desorption of positive ions was measured by the positive ion current. At higher coverages, the rate of desorption of atoms was determined by allowing a calculated fraction of the total number of atoms to impinge on an adjacent incandescent filament where they became ionized; then the rate of evaporation of atoms from the original surface was calculated from the increase in the ion current. The electron emission was measured simultaneously so that the rate of evaporation of ions and atoms could be ex-

112

R. V. CULVER AND F. C. TOMPKINS

pressed in terms of the work function. In turn, the work function was related t o the surface coverage, 8, in an independent series of experiments. For the desorption of Cs atoms from a W surface, E, , computed from the evaporation rates expressed as a function of coverage and temperature, was found to vary from 65 kcal./g. atom at 0 = 0 to 40 kcal./g. atom at e = 1. The desorption of Ba and Sr from a W ribbon was recently investigated by Moore and Allison ( 5 4 ) . The emission current was correlated with the surface coverage by means of data obtained in a series of calibration tests with radioactive Sr, in which the emission was determined as a function of the amount of radioactive Sr deposited on a W filament. The thermionic activity was measured by determining the temperature required for a fixed emission density. Thus, if A in Equation (4), Sec. V, is constant throughout the experiments, 90 kT d=-++-lnTO e

T TO

where cp and T refer to the coated W filament at a fixed emission density and 90 and To apply to the clean W surface. Desorption experiments carried out between 950" and 1150" K. enabled E, to be computed from Arrhenius plots; for Ba, E, was 80 to 86 kcal./mole, and for Sr, E, was 75 to 82 kcal./mole. Johnson and Vick (118) measured the rate of change of the work function during the desorption of 0 2 from a W cathode by passing the saturation emission current through a resistor and recording the voltage drop on a cathode-ray oscillograph. It was assumed that the coverage 0 varied linearly with the work function-a reasonable approximation at low coverage-so that in terms of the Richardson equation for an emission current j , 8 is given by

e

= (lnj,

- lnjO)/(lnjl - l n j O )

where the subscripts refer to coverages of 0, 8, and 1. The activation energy a t low coverage, as determined from measurements at various tempcratures, was about 150 kcal./mole. A C.P.D. method was adopted by Bosworth and Rideal (95,119) to investigate the evaporation of Na from a W filament. Desorption was accompanied by a negative drift in the S.P. when the coated filament was held at a temperature in the range 610' t o 795' K., and the resulting S.P.time curves were converted into coverage-time curves by the use of calibration data previously obtained. The results represent the mutual effect of adsorption and desorption processes on the W filament. Hence, the heat, of evaporation Eevap may be calculated from the temperature coefficient of

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

113

the rate of evaporation by means of the Clapeyron equation

E,,,,

=

Rd In P / d ( l / T )

the pressure P being related t o the surface coverage 0 by the Herz-Knudsen equation,

r = du/dt

=

umdO/dt = crP/(21rmkT)"~

where r is the rate at which atoms strike the metal surface in atoms/cm.2 sec., 6 = u/um as defined in Sec. 111, a is the condensation coefficient which is unity for Na, and m is the atomic weight. Thus, d6/dt is a measure of the pressure in equilibrium with an adsorbed layer of coverage 6 a t a temperature T , and EevBpis given by

The heat of evaporation of Na from W was measured for coverages of 0 to 10 monolayers of Na, and EevRp varied from 32 kcal./mole a t zero coverage to a minimum value of 17 kcal./mole a t 0 = 0.75. Bosworth and Rideal (96) also measured the rate of change of the C.P.D. of an oxygenated W filament which was maintained at temperatures varying from 1270" t o 1930" K. Again the experimental conditions permitted the heat of evaporation to be calculated by the Clapeyron equation and the value recorded a t low coverage-about 150 kcal./mole-is in satisfactory agreement with the measured calorimetric heat of adsorption of 0 2 on W (24).

B. THEFREEENERGY OF DESORPTION The activation free energy of desorption may be computed from the rate of desorption as determined experimentally from the change in the surface potential with time. The theory of absolute rates has been applied to desorption by Eley (120) and Higuchi et al. (107) to obtain energies and entropies of activation as a function of coverage. The rate of desorption is given by,

dt d6

- ( 1 - e)pk', exp [-a%],

(6 )

where 6(AF') is the decrease in the activation free energy for desorption i.e., 6(AF') = AFo' - AF', AFO' being the activation free energy for desorption a t zero coverage, and 6 ( A F , # ) is the decrease in the activation free energy for adsorption: k' and k', are the rate constants a t zero coverage for desorption and adsorption, respectively, expressed as k' = K ( k T / h )

114

R. V. CULVER AND F. C. TOMPKINS

exp (-AFo'/RT), etc. Now in a true desorption process the value of k', is very small, so that Equation (6) may be written as

if it is assumed that 6(AF') = 6(AFlu)B, 6(AFln) being a constant and equal to the decrease in the activation free energy at a coverage of e = 1. When Equation (7)is compared with the well-known Elovitch ( 1 2 1 ) relation, -&/at

= a0

exp ( b e )

used to describe desorption phenomena on metal surfaces, it is seen that a = k' =

( k T / h ) exp ( - A F o u / R T )

(8)

and b = G(AF')/BRT = S(AF,')/RT

We may express Equation (7) in the form log ( - d In 8 / d t ) = log k'

+ [6(AFln)/RT]B

so that when values of ( - d In B/dt) are plotted against 8, the slope of the linear plot is given by 6( AFl'/RT) and the quantity k' is obtained from the intercept at 0 = 0. Furthermore, if AFoMis evaluated from Equation (8) at Merent temperatures and the thermodynamic relationship 6(AF') = 6(AH') - T ( A S ' )

is used, which for nonactivated adsorption where AHo' = AH0 and AH' = AH, becomes 6 ( A F n ) = 6 ( A H ) - T ( A S ' ) , the desorption heat, AH0 , and AS,' , the activation entropy for desorption at zero coverage, may be readily determined. This analysis has been applied by Higuchi, Reel and Eyring (106, 107) to data for the desorption of Na (Bosworth, 119) and Cs (Langmuir, 122) from a W surface. For Na, AFou varied from 63.2 kcal./mole at 920" K to 68.9 kcal./mole at 1070" K,and for Cs from 64.5 kcal./mole at 600" K to 69.7 kcal./mole at 900" K. The thermodynamic quantities AFo' , 6( AFl'), AHo' , and A&' calculated from the more extensive data of Moore and Allison (64) for the desorption of Ba and Sr from W are given in Table VIII. Similarly, the results of Bosworth and Rideal (96) for the system W 02 yielded the thermodynamic quantities shown in Table IX. In his original paper, Eley (117)derived values for W Na, W O2and W Nz from Bosworth's data, and where comparison is possible, the present figures agree reasonably well with his values,

+

+

+

+

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

115

TABLE VIII Thermodynamic Quantities for the Systems W-Ba and W-Sr AF,’

System

AF,’

(lOOOoK)l

kcal./mole

~ A F ~ $

kcal./mole

(llOO”K)’

kcal./mole

AH$ AS,’ kcal./mole cal./mole deg.

TABLE IX Thermodynamic Quantities for the System W-02

0 2

~ A F ~ $

AF,$

T(”K)

kcal./mole

kcal./mole

1375 1575 1770 1820

124 118 121 120

41.8 37.8 44.6 45.7

s( A F ~ T V~R s (AF~ 15.2 12.1 12.6 12.6

~

~

15.0 13.1 11.6 11.3

As calculated from b = 6(AFlru)/RT= e V l / R T , where VI for the system W is 1.78 v . at 0 = 1 (6).

+

An examination of the data in Table IX reveals that AFo’ is practically independent of the temperature, i.e., ASO’ 2 0, and that s(AF1’) is approximately equal to the value of 41.0 kcal. as calculated from the expression (to be considered in Sec. XI) 6 ( A H I ) = S(AH1’) = eV1, where V1 = 1.78 v. Thus, it appears that for the desorption of O2from W

6(m1) = AH,')

= ~(AF,’) =

eVl

and, as a consequence of ALSO’ == 0, AFo’

=

AHo’

=

AH0

=

E (the bond energy)

This is borne out by the value of AFO’ , about 122 kcal./mole, which is approximately equal to the calculated bond energy of the system W 0, viz., E = 116 kcal./mole.

+

C. THEACTIVATION ENERGY FOR MIGRATION Thermionic and photoelectric methods have been successfully employed in the measurement of the mobility of alkali metals on W surfaces. The migration of adsorbed gases like H z , 0 2 , and CO over a metal surface may be followed in the F.E.M.,and data concerning the mobility of these adsorbates are of particular interest, since they are frequently involved in surface reactions and other chemisorption phenomena.

~

.

116

R. V. CULVER AND F. C. TOMPKINS

1. Thermionic and Photoelectric Investigations. Brattain and Becker (53) deposited a Th film on the front face of a W filament and determined the rate of migration to the back face at 1535" and 1655' K by comparing the electron emission from the two faces. The emission from the front face fell, while the film spread and the emission from the back rose steadily until the whole of the filament was covered uniformly. Assuming that the work function varied linearly with the surface concentration, the rate of migration was equated to the rate of change in work function and the activation energy for migration, Em, as determined from the temperature coefficient of the rate of migration, was 110 kcal./g. atom, or approximately half the heat of adsorption. The mobility of Cs on W was studied by Taylor and Langmuir ( 5 0 ) , who measured the thermionic emission from a coated W filament in a cell equipped with three cylindrical collectors, as shown in Fig. 10. At low coverage, Cs was desorbed essentially as positive ions; hence, when the Cs-covered filament was heated with the outer collectors at a positive potential and the inner one at a negative potential, a sharp boundary was obtained between the outer section of the filament-still covered with n dilute layer of Cs-and the central part, which was bare. Spreading of the Cs film occurred when the filament was heated with all the collectors at a positive potential to prevent desorption. The total amount of Cs which migrated to the center of the filament was measured by flashing with all three collectors at a negative potential and noting the positive ion current to the central collector. From measurements made within the temperature range 652' to 812" K the activation energy of migration, Em,was found to be 16 kcal./g. atom. Bosworth (66) examined the mobility of Na and K on a W ribbon photoelectrically. I n these experiments the ribbon was flashed at 2200" C. The deposit of Na or K was obtained by directing a beam of ions from a Kunsman source on to the center of the W ribbon, the amount of deposit being determined from the positive ion current between the source and the ribbon. Spreading from the adsorbed patch of Na or K was followed by focusing a fine beam of ultraviolet light on to various points along the length of the ribbon, and the photoemission was measured as a function of the distance along the ribbon, the temperature, and the time for migration. Sodium tended to migrate into the interior of the metal, so that diffusion measurements were restricted to relatively thick layers; with K, however, the effect was less marked, and measurements were made over quite a wide range of concentrations. For Na, migration measurements were made up to 800" K, and a series of curves were obtained for the concentration of the adsorbate in the vicinity of the patch, as shown in Fig. 20. It was found that a t any one temperature the total emission, and hence the con-

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

I

117

0 min

Distance

m

FIG.20. The migration of Na on tungsten at 20"

centmtion, was related t o the time of epreading t by the equat,ion

l/c2=

K(t

+ to)

K and to being constants. The reaction constant K was shown to be proportional to the surface diffusion coefficient, so the activation energy for migration was computed from the temperature coefficient of K , i.e.,

Em

=

Rd In K / d ( l / T )

At room temperature the value of E m for dilute Na films varied from 5.8 t o 6.9 kcal./g. atom. For K films, E mfell with increasing surface concentration from 16 kcal./g. atom at a very low coverage t o 6.7 kcal./g. atom for a surface concentration of 4.8 x 10" atoms/cm.2, and the fall in E min this case was attributed to the existence of a spreading force due t o the mutual repulsion of adions. 2. F.E.M. Investigations. Here the work function and hence the surface coverage, assuming A4 is proportional to 0, is determined from the FowlerNordheim plot (see Sec. V,B) in the usual way and Em , the activation energy for migration, is obtained from the temperature dependence of the diffusion rates observed for equal doses of gas at different temperatures. Recent F.E.M. surface-diffusion studies have included the systems W O2 (59),W Ha ( G I ) , W CO (I23),and Ni Hz (63). It was found that the nature of the spreading observed often depended upon the initial surface coverage, as well as the temperature, and with 0 2 , Hz , and CO, on W,

+

+

+

+

118

R. V. CULVER AND F. C. TOMPKINS

three types of surface migration were distinguishable. At low temperatures (-20" K ) and high coverage, the sharp diffusion boundary which accompanied spreading was attributed to the migration of physically adsorbed gas over a chemisorbed layer followed by "precipitation" on the bare metal surface. At higher temperatures, diffusion occurred within the chemisorbed layer, and the diffusion boundaries moved outwards radially from the closepacked W faces, following the precipitation of adsorbate atoms in trap sites on the rougher part of the surface. The presence of such traps will predominate on all but the close packed faces because of the open structure of the b.c.c. W lattice, and saturation of these traps must precede further migration. At low coverage, boundary-free diffusion occurred with migration of adsorbate atoms from unsaturated trap sites. For O2 adsorbed on W, the value of E m was 0.9 kcal./mole at high coverage, -25 kcal./mole for diffusion in the chemisorbed layer at 400" to 500" K, and -30 kcal./ mole for migration from trap sites a t low coverage. H2 a similar mechanism was suggested for the lowIn the system Ni temperature spreading observed at high coverage (63). In the chemisorbed layer, however, at 250" to 280°K diffusion took place without a sharp diffusion boundary, the activation energy being 7 kcal./mole. This behavior is in marked contrast to that observed for 0 2 , Hz , and CO on W and suggests that the f.c.c. structure of Ni is inherently homogeneous.

+

X. Surface Reaction Mechanisms A knowledge of the change in work function which occurs during physical adsorption or chemisorption processes on a metal surface may help to elucidate the mechanism of a surface reaction. The adsorption of nitrous oxide on a Pt surface at low temperature (83" K ) has been examined photoelectrically by Suhrmann and Sachtler ( 4 3 ) . They found that the work function increased during adsorption, a similar effect being noted with oxygen atoms but not with 0 2 or N2 molecules (68). This increase in the work function suggests an electron shift from the metal to the 0 of the N2O molecule, and supporting evidence is provided by the increased electrical resistance observed when N20 is chemisorbed on a Ni surface (114). It is assumed, therefore, that O2molecules are not concerned in the reaction mechanism. Accordingly, the catalytic decomposition of N20 is presumed to commence with the binding of N 2 0 to the surface through the 0 atom with a consequent loosening of the bond between the 0 atom and the nitrogen. The nitrogen is thermally dissociated from the bound 0 atom at higher temperatures and neighboring 0 atoms, providing they can gain the necessary activation energy, migrate, combine, and desorb as 0 2 molecules. Thus, it appears that the decomposition of

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

119

NzO on a metal surface may take place in accordance with the reactions, N20

+ 2e

( NzOs; ;1

=

(NzO)ZZ

=

Nz

0% =

+OL

>5oz+ 2e

with the previously bound electrons returning to the metal. A further insight may be obtained into the mechanism of the Hz Dz exchange reaction as a result of recent S.P. investigations which have revealed the presence of adsorbed atomic and molecular H a on a W suiface (124). Interrupting the flow of Hz t o a W film at -190' gave rise to a positive drift in potential characteristic of molecular H Zadsorption. Further, it was found that (1) adsorption was competitive (and activated), in that H atoms were replaced by Hz molecules before the surface was covered with H atoms, and ( 2 ) an increase in pressure favored the adsorption of molecular H2 . Similar results were obtained by Wortman et a2. ( 6 3 ) ,who noted a field-dependent transformation in the adsorbed film between 2" and 4" K. during the low-temperature spreading of large doses of Hz on a Ni surface in the F.E.M. This was accounted for by a shift in the equilibrium between adsorbed Hz molecules and the terminal fraction of the adsorbed H atoms. These experimental results lead to two significant deductions. I n the first place, the presence of a stable adsorbed layer of molecular HZat low temperature on a W surface lends support t o the Rideal mechanism (166)for the Hz Dz exchange reaction, aiz.,

+

+

since the exchange may proceed via H and H2 adsorbed at temperatures too low to permit mobility of the H atoms within the monolayer proper. Secondly, since the exchange process proceeds through substitutional adsorption and the rate of catalytic H, exchange with D2has been shown to be pressure-dependent ( l o o ) , the available evidence does not favor the original Bonhoeffer-Farkas mechanism (166), H

k-4

D

+ HD

+

2w.

XI. Heats of Adsorption In an adsorption process involving ionic or covalent bonding, the adsorption heats of principal interest are -AH,,, the heat of adsorption a t zero coverage and 6( - A H ) , the decrease in the heat of adsorption with coverage. It is in connection with the latter that the role of the work func-

120

R. V. CULVER AND F. C. TOMPKINS

tion is most significant, but it is desirable to consider the evaluatioii of - AHo before dealing with the variation in - A H with coverage.

HEATOF ADSORPTION A. THEINITIAL 1, - AH0 for the Adsorption of Alkali Metals. If an alkali metal atom is located at an infinite distance from a metal surface at zero potential, then the heat of adsorption comprises the work done in ( 1 ) transferring an electron from the atom to the metal, and (2) bringing the positiveion to its equilibrium distance from the metal surface (la?‘).In the first step, the energy change is (e4 - e I ) , where I$is the work function of the metal and I is the ionization potential of the alkali metal atom. In the second, the force of attraction on the positive ion at a distance d from the metal surface, i.e., the electrostatic image force, is e2/4d;hence, the heat liberated is e2/4do, where do is the equilibrium distance of the adsorbed ion from the metal surface. This distance is often assumed t o be equal to the ionic radius, which is 1.83 A. for the Na ion. The initial heat of adsorption, therefore, is -AH0 = @ - e l

+ e2/4do

(9)

The significance of the terms in Equation (9) is readily appreciated by referring to Fig. 2, which shows the potential energy curves for the adsorption of Na ions on W metal. Here the difference between the work function of the metal and the ionization potential of the Nu atom is given by the difference between the potential energy levels C and D, while the difference between the minimum A in the potential energy curve and the energy level C corresponds t o e2/4do, the interaction energy between the ion and its image. Some initial heats of adsorption are shown in Table X. In calculating 3te2/4do (3t = Avogadro’s number), it has been assumed that the distance of the adsorbed ion from the surface is equal to the ionic radius, and since TABLE X Initial Heals of Adsorption, kcal./g. atom“ System

9

I

W-Na

104 104

118 89.4

w-cs

xe’/4do 44.5 31.1

-AHo (calc.) -AH0 (exptl.) 30.5 45.7

32* 64c

a Trapnell, B . M. W., in “Chemieorption”, Chapter 6. Butterworths, London, 1955. * Boaworth, It. C. L., Proc. Roy. Sac. (London) A162,32 (1937); Eley, D. D., Trans. Faraday Sac. 49, 643 (1953). c Taylor, J . B., and Langmuir, I . , I’hlls. Rev. 40. 463 (1932); 44, 423 (1933).

SURFACE POTENTIALS A N D ADSORPTION PROCESS O N METALS

121

the value of e2/4do may be modified by polarization forces, van der Waals' forces, and repulsion forces, the agreement between the calculated and experimental data is regarded as satisfactory. 2 . Calculation of -AH0 from Bond Energies and Dipole Moments. Several attempts have been made to calculate the initial heat of adsorption from bond energies and dipole moments. Recently Higuchi et al. (106, 107) have shown that by approximating the complex formed between a metal atom and an adsorbate atom to a diatomic molecule, the bond energy E of the surface complex M-A may be calculated from a knowledge of the energy of pure covalent and ionic bonds and the dipole moment of the complex, as derived from work-function measurements. For a monatomic gas the evaluation of - AH0 for adsorption on a nearly bare surface follows immediately, as under these conditions - AH0 = E. The bond energy for the complex M-A is related to the energies of the ideal ionic and covalent bonds, H i and H , , respectively, and the fraction Ci' of the ionic bond in the adsorption bond, by the expression (128)

E - Hi - I - = 1 ~E - Hc To a good approximation,

and

H,

=

[E'(M-M)

+ E'(A-A)]/2

where a is the electron affinity of the metal M , and I is the ionization potential of the adsorbate A when an electron is transferred from A t o M . If the direction of electron transfer is reversed, then a is the electron affinity of A , and I is the ionization potential of M . E ( M - M ) and E ( A - A ) are the bond energies of the single bonds ( M - M ) and (A-A ) ,respectively ; E ( M - M ) is calculated from the heat of vaporization of the metal, A, and has the value 2A/12 for f.c.c. metals and 2A/8 for b.c.c. metals. The fraction Ci" of the ionic bond in the adsorption bond is expressed in terms of the dipole moment by the relation

C? = M/edyA where d y A is the sum of the atomic metal radius and the covalent bond radius of the adsorbed atom except for alkali metal atoms when d y A is replaced by the monovalent ionic radius. The dipole moment M is evaluated from the work function change observed during adsorption, viz., M = A1$/2?r300u for covalent adsorption, u being the number of dipoles per

122

R. V. CULVER A N D F. C. TOMPKINB

cm? of surface. A value of M for an almost bare surface is required, but as the experimentally determined work functions usually refer to a fully covered surface, it is necessary to disregard the possible effects of mutual depolarization a t high coverage and assume that M8-o = Me-1 . Finally for a homonuclear gas which chemisorbs dissociatively, -AH0 = 2E - E ( A - A )

Initial heats of adsorption have been calculated in this way for various systems including (1) Ba and Sr on W, (2) H2 on Ni, Fe, Ta, Rh, Cr, Cu, Co, and Pt, ( 3 ) O2 on W, Ni, and Fe, (4) N2 on W, Ta, and E'e and ( 5 ) CO on Fe and Ni. With the exception of the system Ni 0 2 , where oxidation may have occurred, there is fair agreement between calculated and experimental values of - AHo , as shown in Table XI. . The initial heat of adsorption of H 2 on a number of metals can be predicted (Eley, 129) with a fair degree of accuracy from the equation

+

-AH0 = H c

+ 23.06M2

(10)

which was based on Pauling's empirical relation (104) -AH0

=

Hc

+ 2 3 . 0 6 ( 2 ~-

~

4

)

~

Here z M and x4 are the respective electronegativities of thc metal and t)hc adsorbate atoms and the difference ( x M - z A )may bc approximated to the dipole moment M at 0 = 0. Trapnell (127)has extended these calculations to other adsorbates such as 0 2 , N2 , CO, and C2H4, but the agreement between experimental and calculated values of - AH0 is less satisfactory for a molecule such as C2Ha than for 02,owing to the uncertainty of the nature of the surface bonds. If, however, Equation (10) is applied t o the adsorption of alkali metals, it is found that the calculated values of -AH0 TABLE XI Initial Heats of Adaor lion Calculated from Bond Energies and Dipole Moments"

I

I System M-A

Dipole Dipole moment Length M(D) d ~ , A d.

mole

mole

Initinl heats

= E

mole

W-Be W-H w-0 Ni-CO 4

~ _ _ _ _ _ ~ _ ---~ 87.0 104.2 87.0 aa.7 iao.1 18.4 a.eo 64.4 0.212 0.778

0.610

1.67 1.m a.oi

64.4 64.4

103.4 ma/a

16.1

68.6

183.9 183.9 176.1

16.4 60.7 31.6

80.9

68.4

117.2 61.1

118.2 61.1

Higuchi I.. Rae, T., and Eyring, H., J . Am. Chsm. Soc. T9, 1330 (1967); ibid. ? 4969 I , (19666). Cham. Phus. za, 1808 (1966). Roberta, J. K.,Prw. Roy. Soc. (London) A l l , 446, 464 (1836). Beeak, O.,Aduancss in Calalums 2, 161 (1960)

* Moore, 0.E., and Alliaon, H.W., J

mole

_

83O 4.5c IW 3Kd

_

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

123

are too high, e.g., for the adsorption of Ba on W, -AH0 (calc.) is 4,280 kcal./mole, as opposed to a value of 83 kcal./mole for - AH0 (exptl.) (106, 107).Thus, the method appears to be limited to cases where the bonding is essentially covalent and the resulting dipole moment comparatively small. The procedure of Eyring, which has a quantum-mechanical basis is, however, equally applicable to both ionic and covalent adsorption.

B. THEVARIATION IN

THE

HEATOF ADSORPTION WITH COVERAGE

The change in the work function which accompanies adsorption on a metal surface has an effect on the heat of adsorption. In the formation of a dipole layer which is negatively charged outwards, the work done in transferring an electron from a point inside the metal to the region of the adsorbed atom is given by (eA+) e. v., as considered in Sec. 111. Furthermore, as the value of A+ increases with adsorption, the work of transferring electrons will increase correspondingly. Alternatively, if the double layer is positive outwards, as in cationic adsorption, eA+ is the work done in transferring an electron from a point outside the metal to a surface of diminishing electron affinity. Thus, the heat of adsorption-comprising the heat of binding, i.e., the initial heat - AH0 ,and the change in the work functiondecreases with coverage irrespective of the sign of the double charge layer. Accordingly, -AH

=

- A H o - eA+

and if - AHo is independent of coverage, 6(-AH)

=

-AHo+

=

2r~,,,OMe

AH = eA+

(for covalent adsorption)

(11)

where 6( - A H ) is the decrease in the differential heat of adsorption. De Boer (130)first drew attention to the contribution of the “workfunction effect” to the heat of chemisorption of alkali atoms on a metal surface. With Cs on W, for example, the heat of adsorption is described by the equation, - A H = -AH0 - eA+

=

-AHi - e l

+ e+ - eA+

which may be written in terms of the dipole moment by introducing the relation eA+ = 4rumOMe (for ion adsorption) as -AH

=

- AHi - e I

+ a$ - 4 ~ ~ , e M e

so that if - A H i , the heat of adsorption of the Cs ion, ( -e2/4d0), is constant and the dipole moment, M , does not vary during the adsorption process, the heat of adsorption, - A H , should decrease linearly with cover-

124

R. V. CULVER A N D F. C. TOMPKINS

age. Such a relationship is not observed experimentally, however, and the discrepancy is attributed to the discontinuous nature of the charge distribution of the surface dipoles and their mutual depolarixation. With regard to the former, de Boer has shown that if the dipole layer formed by the chemisorbed Cs ions consists of a double layer of discrete charges, then the potential gradient extends to the region outside the double layer (14, 180). As a result, Cs ions are adsorbed more strongly by a partly covered W surface than by a bare one and the heat of adsorption of the Cs ion increases with coverage. The mutual depolarixation of the dipoles becomes important at high coverage and leads to a decrease in the dipole moment. For several metal H2 systems, it has been found that the experimental heat of adsorption is a linear function of the surface coverage (24,131). This implies that for these systems the distribution of dipoles approximates to a continuous layer of charge and that Equation (11) correctly describes the change in the heat of adsorption when an electron is moved through the double charge layer. Originally Boudart (108) used the relation

+

6( - A H )

=

>4eA$

to account semiquantitatively for the observed decrease in the d8erenti:il heat of adsorption for the systems W Cs, W H2 ,W 0 2 , W N2, and Ni H2 in terms of the change in work function. This relation was derivcd for a double layer with a homogeneous charge distribution in which the binding electrons spent most of their time in the region midway between the bound atoms-a situation which has been criticized (16) on the grounds that the electrons must form part of the double layer. Further, the assumption of a continuous charge distribution for adsorbates other than H2 may not be justified. If the charge distribution is discrete, rather than continuous, the potential energy curves evaluated by Gomer (41) show that at low coverage the potential at the midpoint in the double layer is very much less than >4A+. In considering the magnitude of this work function effect Mignolet (16) has pointed out that in moving an electron work is done against both electrostatic and quantum mechanical forces and not against the electrostatic field alone as frequently assumed. For this reason, he prefers t o derive a relation between the decrease in the differential heat of adsorption 6( - A H ) and the change in the work function eat$ in accordance with a model for the formation of a covalent bond involving the excitation of an electron from the conduction band of the metal into a vacant adsorbate orbital. The electron shift reduces the height of the highest occupied level and gives rise to a double charge layer with a negative sign. Accordingly, the increase in the work function is given by

+

+

4 = AE/e

+

+

+

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

125

where A& is the decrease in energy attending an electron transfer, and it follows that there will be an increase in the excitation energy for the transfer of succeeding electrons. Then assuming that the decrease in the heat of adsorption 6( - A H ) may be entirely attributed to the reduction in the height of the highest occupied level, i.e., 6( - A H ) = A&, he obtains the expression 6( - A H )

=

eAq5

which is identical with Equation ( 11) derived from electrostatic considerations. An examination ( 1 6 ) of a few metal-gas systems for which reliable data are available, viz., W H2, Ni Hz , and Fe H2 has shown that there is reasonable agreement between 6( - A H ) referring to the difference in the differential heat of adsorption a t 8 = 0 and 8 = 1 and A4 corresponding to the change in work function, as disclosed by S.P. measurements, over the same range of coverage. It would appear, however, more satisfactory to restrict such investigations to a limited range of coverage, such as 0 = 0.2 t o e = 0.8, and so avoid the effect of surface heterogeneity at 8 2 0 and lateral interactions between adsorbed atoms at 8 2 1 on the heat of adsorption. The analysis of S.P. and heat of adsorption data has been extended by Higuchi et al. (106,107) to the systems Ni Hz , Ni 0 2 , W H2, and Hz . They used Equation (11) t o express the difference between the Fe initial heat of adsorption, -AH0 , and - A H , the heat of adsorption a t a coverage 8, in terms of the change in work function over the range of coverage 0 t o 8. Aq5 was calculated for various values of 0 on the assumption that a linear relation existed between the S.P. and the coverage, and while it has been shown that this is only approximately true (77) , the divergence from linearity is normally not serious. A comparison between the calculated and experimental plots of 6( - A H ) against 8 showed that the relation of 6( - A H ) to 8 was correctly described, although there was considerable deviation from the experimental 6( - A H ) data in the systems W H2 and W O2for coverages above 8 = 0.5. Figure 21a and b shows 6 ( - A H ) against 8 curves for the chemisorption of H2 on Ni and Fe, respectively. In Fig. 21a, 6 ( - A H ) values were calculated from the data of Eeeck ( 3 2 ) and Schuit and de Boer (131 ), and the value of Aq5 at 0 = 1 was taken as 0.33 v. ( 1 6 ) ; in Fig. 21b, the respective values of - A H and - AHo were selected from the data of Bagg and Tompkins (is,%'), - AH0 = 31 kcal./mole being obtained by extrapolating the adsorption heats from 0 = 0.2 to 8 = 0, and A+ a t 8 = 1 was taken as 0.43 v. (77). An inspection of Fig. 21 reveals that the agreement between the experimental and calculated data for 6 ( - A H ) is satisfactory in both cases; it is significant,

+

+

+

+

+

+

+

+

+

126

R. V. CULVER AND F. C. TOMPKINS

0

1.0

e

0

8

1.0

(b)

(a)

FIG.21. The decrease in the differential heat of adsorption with coverage (a) for Ni-H1 and (b) for Fe-H2

.

however, that the experimental values of 6( - A H ) are less than those calculated from the change in work function. In the system Ni Hz , de Boer (133)has recently offered an explanation for the difference between the calculated and the experimental heats of adsorption in terms of a model in which H atoms are chemisorbed on top of the Ni atoms-an arrangement which may account both for the negative S.P. (75) and the decrease in the electrical resistance of the metal film (114), as observed experimentally. The bonding is covalent, and the sharing of electrons between the metal and the adsorbate leads t o an array of n atoms per unit area containing m additional electrons ( m < n ) moving freely among the chemisorbed atoms; consequently the charge per atom is (m/n)e.An atom adsorbed on the surface has the probability m / n that it will take up an electron from the bulk metal, so that when an electron passes through the double layer (assumed homogeneous) at the metal surface, the decrease in the heat of adsorption is modified by the factor m/n; hence,

+

6( - A H ) = (m/n)eA+ Such a picture for the adsorption process leads to a linear decrease in the heat of adsorption with coverage and a value of 6( - A H ) essentially less than that computed from the relation 6( - A H ) = eA+.

C. ADSORPTION IN TERMS OF

SURFACE ELECTRON GAS While the decrease in the heat of chemisorption has been attributed t o a change in the electron distribution a t the metal surface, it is doubtful A

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

127

whether the transfer of electrons with consequent changes in the Fermi level would give rise t o the observed values of 6( - A H ) in covalent adsorption. An alternative mechanism has been proposed by Temkin (134). He suggests the presence of a two-dimensional surface electron gas which takes part in the chemisorption process, and derives an equation for changes in the kinetic energy of the electrons, viz.,

- AH

=

h2~,e/4~m

(12)

whcre h is Planck’s constant, m is the mass of the electron, and a,e is the number of adsorbed atoms per cm.’ of surface. Usually the calculated values of - A H are too large, so that it is necessary to replace m by an “effective mass.” There are few quantitative data available for testing the equation. Apart from the work of Federova and Frumkin ( 1 3 5 ) , which showed that the heat of solution of hydrogen in the p phase of the Pd H system (Pd containing GO at. % H ) was a function of the concentration of the dissolved hydrogen, other data in the literature point t o the nondependence of - AH with the concentration. The heat of solution of hydrogen in @ Ti (136), for example, does not fall with an increasing concentration of dissolved hydrogen, and it may be argued in this case that as the heat of solution of hydrogen in the metal is almost constant, it is unlikely that electronic interaction between chemisorbed atoms and surface electrons would produce a substantial change in - A H . Hence, a t the present time it is difficult t o accept the validity of Equation (12) ; should it be so, then work-function changes would be related to changes in the energy levels of the electron gas, and it follows that the concepts of dipole moments and surface potentials would need to be revised (133).

+

XII. Conclusions The tabulated S.P. values show that there is a need for more precise experimental measurements, employing modern high-vacuum techniques and the use of more refined methods of preparing clean metal surfaces, e.g., ion bombardment with subsequent examination of the surface by electron diffraction. However, the results obtained by a variety of measuring techniques, including the F.E.M., are generally consistent with regard t o the sign of the S.P. and the approximate magnitude. An advance from an experimental viewpoint has been made by Eisinger ( 7 8 ) , who used the “flash filament” method of Becker (137) to determine the coverage (molecules per cm.2) of a monocrystalline W ribbon with its surface normal in the 11131 direction and measured the work function change photoelectrically. Thus, he was able t o directly relate the change in S.P. t o the density of metal atoms in a particular crystal plane. The S.P. values cannot be generally interpreted in terms of a simple

128

R. V. CULVER AND F. C. TOMPKINS

model for the adsorption bond. For ionic adsorption the work fiiiictioii change is usually attributed t o the modification of the surface double layer by the deposition of dipoles. For covalent adsorption, however, this mechanism may not apply and consequently the change in work function may not be correctly expressed by the relation A4 = 2ru,OM. Clearly, the complete interpretation of the work function change for simple gases and radicals must await the results of a more critical analysis of the bonding process at a metal surface. Turning now t o the applications considered in Secs. IX, X, and XI, we note that the activation energies of desorption and migration are conveniently measured by S.P. methods, while the results obtained for the adsorption of NZO on Pt show that there is considerable scope for the application of work-function measurements in investigating catalytic renctions at metal surfaces.

REFERENCES 1 . Lennard Jones, J. E., Trans. Faraday SOC.a8, 333 (1932). 2 . Dowden, D. A., J . Chem. SOC.p. 242 (1960).

3. Richardson, 0. W., and Robertson, F. S., Phil. Mag. [7] 43, 162 (1922).

4. Langmuir, I., and Kingdon, K. H., Phys. Rev. 21, 380 (1923).

6. Bosworth, R . C. L., J. Proc. Roy. SOC.N . S. Wales 79, 63, 166, 190 (1945). 6 . Herring, C., and Nichols. M. H., Revs. Modern Phys. 21, 185 (1949). 7. Chalmers, J . A., Phil. Mag. [7) 99, 399, 416, 496, 506, 599, 608 (1942). 8. Hulburt, H. M., in “Catalysis” (1’. H. Emmett, ed.), Vol. 11, p. 200. Reinhold, New York, 1961. 9. Parsons, R., i n “Modern Aspects of Electrochemistry’’ (J. O’M HockriN, cd.), Chapter 111. Butterworths, London, 1954. 10. Lange, E., and Miscenko, K. P., 2.physik. Chem. (Leipzig) A149, 1 (1!)30). 1 1 . Bardeen, J., Phys. Rev. 49, 663 (1936); Oldekop, W., arid Sunter, P.,%. Z’hysik 136. 534 (1964). 1%. Lewis, T. J., Proc. Phys. SOC.(London) B67, 187 (1954). 13. Becker, J. A,, Revs. Modern Phys. 7 , 95 (1935). 1.6. de Boer, J. H., and Veenemans, C. F., Physica 1, 960 (1934). 16. Langmuir, I., Chem. Revs. 18, 147 (1933). 16. Mignolet, J. C. l’., Bull. soc. chim. Belg. 64, 126 (1955). 17. Coulson, C. A., “Electricity.” Oliver, London, 1951. 18. Johnson, R . P., and Shockley, W., Phys. Rev. 49, 436 (1936). f9. Muller, E. W., 2.Physik 106, 541 (1937). 20. Porter, A. S., and Tompkins, F. C., PTOC. Roy. SOC.(London) A217, 544 (1953). 21. Becker, J. A., Advances i n Catalysis 8 , 135 (1957). 2%.Alpert, D., J. Appl. Phys. 24, 860 (1953). 23. Thomas, L. B., and Schofield, E. B., J. Chem. Phys. 2S, 861 (1955). 24. Roberts, J . K., Proc. Roy. SOC.(London) Al62, 446, 464 (1935). 26. Allen, J. A., Revs. Pure and Appl. Chem. (Australia) 4, 133 (1954). $6. de Boer, J. H., Koninkl. Ned. Akad. Wetenschap. Proc. 49, 1103 (1946). BY. Beeck, O . , Smith, A. E., and Wheeler, A,, Proc. Roy. SOC.(London) A177, 62 (1940).

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

129

28. de Boer, J. H., and Kraak, H. H., Rec. trav. chim. 66, 941 (1936). 29. Shishakov, N. A,, J . Exptl. Theoret. Phys. U.S.S.R. 22, 241 (1952). 30. Mignolet, J. C. P., Rec. trav. chim. 74, 685 (1955). 91. Wheeler, A., i n “Structure and Properties of Solid Surfaces,” (R. Gomer and C. S. Smith, eds.), p. 439. Univ. of Chicago Press, Chicago, 1952. 32. Beeck, O., Advances i n Catalysis 2 , 151 (1950). 33. Hickmott, T. W., Advances i n Catalysis 9, 167 (1957). 34. Farnsworth, H. E., Schlier, R. E., George, T. H., and Burger, R. M., J . Appl. Phys. 26, 252 (1955). 36. Dillon, J. A , , and Farnsworth, H. E., Phys. Rev. 99, 1643 (1955). 36. Schlier, R. E., and Farnsworth, H. E., i n “Semiconductor Surface Physics” (R. H. Kingston, ed.), p. 5. Univ. of Pennsylvania Press, Philadelphia, 1957. 37. Autler, S. H., and McWhorter, A. L., i n “Semiconductor Surface Physics” (R. H. Kingston, ed.), p. 47. Univ. of Pennsylvania Press, Philadelphia, 1957. 38. Bridgman, P. W., “The Thermodynamics of Electrical Phenomena in Metals.” Macmillan, New York, 1934. 39. Reimann, A. L., “Thermionic Emission,” Chapter 111. Chapman & Hall, London, 1934. 40. Fowler, R. H., and Nordheim, L. W., Proc. Roy. SOC.(London) A119, 173 (1928). 41. Gomer, R., J . Chem. Phys. 21, 1869 (1953). 42. Fowler, R . H., Phys. Rev. 38, 45 (1931). 43. Suhrmann, R . , and Sachtler, W. M. H., Proc. Intern. Symposium on Reactivity of Solids, Gothenburg, 1962, p. 601 (1954). 44. Oatley, C. W., Proc. Roy. SOC.(London) A166, 218 (1936); Proc. Phys. SOC. (London) 61, 318 (1939). 46. Hull, A. W., P h y . Rev. 18, 31 (1921). 46. Weissler, G. L., and Wilson, T. N . , J . Appl. Phys. 24, 472 (1953). 47. Mignolet, J. C. P., Discussions Faraday SOC.8 , 326 (1950). 48. Zisman, W. A., Rev. Sci. Znstr. 3, 367 (1932). 49. Gysae, B., and Wagener, S., Z. tech. Physik 19, 264 (1938); 2. Physik 116, 296 (1940). 60. Taylor, J. B., and Langmuir, I., Phys. Rev. 40, 463 (1932); 44, 423 (1933). 61. Becker, J. A., Phys. Rev. 28, 341 (1926). 62. Becker, J. A , , Phys. Rev. 33, 1082 (1929); Trans. A m . Electrochem. SOC.66, 155 (1929). 63. Brattain, W. H., and Becker, J. A., Phys. Rev. 43, 428 (1933). 64. Moore, G. E., and Allison, H. W., J . Chem. Phys. 23, 1609 (1955). 66. Becker, J. A., Trans. Faraday SOC.28, 148 (1932). 66. Kingdon, K . H., Phys. Rev. 24, 510 (1924). 67. Reimann, A. L., Phil. Mag. (71 20, 594 (1935). 68. Klein, R., J . Chem. Phys. 21, 1177 (1953). 69. Gomer, R., and Hulm, J. R., J . Chem. Phys. 27, 1363 (1957). 60. Becker, J. A., and Brandes, R. G., J . Chem. Phys. 23, 1323 (1955). 6 f . Gomer, R . , Wortman, R., and Lundy, R., J . Chem. Phys. 26, 1147 (1957). 62. Muller, E. W., Ergeb. exakt. Naturwiss. 27, 290 (1953). 69. Wortman, R., Gomer, R., and Lundy, R., J . Chem. Phys. 27, 1099 (1957). 64. Ives, H. E., and Olphin, A. R., Phys. Rev. 34, 117 (1929). 66. Bosworth, R. C . L., Proc. Roy. SOC.(London) A160, 58 (1935); A164, 112 (1936). 66. Bosworth, R . C. L., Trans. Faraday SOC.32, 1369 (1936). 67. Mayer, H., Ann. Physik 33, 419 (1938). 68. Suhrmann, R., Advances in Catalysis 7, 303 (1955).

R. V. CULYER AND F. C. TOMPKINS

130

69. Suhrmann, R., and Csech, H., Z. physik. Chem. (Leiptig) B28, 215 (1935). 70. Suhrmann, R., and Sachtler, W. M. H., Arbeitstagung Festkorperphysik,

Dresden, 1952. 71. Suhrmann, R., J. chim. phys. 64, 15 (1957). 78. Suhrmann, R., Z. Elektrochem. 60, 804 (1956).

73. Mignolet, J. C. P., Discussions Faraday SOC.8 , 105 (1950). 74. Sachtler, W.M. H., and Dorgelo, G. J. H., J . chim. phys. 64, 27 (1957). 76. Sachtler, W.M. H., J. Chem. Phys. 26, 751 (1956). 76. Baker, M. McD., and Rideal, E. K., Nature 174, 1185 (1954). 77. Culver, R. V., Pritchard, J., and Tompkins, F. C., i n “Surface Activity” (J. H. Schulman, ed.), Vol. 2,p. 243. Academic Press, New York, 1958. 78. Eisinger, J., J. Chem. Phys. 27, 1206 (1957). 79. Eisinger, J., J . Chem. Phys. 28, 165 (1957). 80. von Duhn, J. H., Ann. Physik 43, 37 (1943). 81. Ogawa, I., Doke, T., and Nakada, I., J. Appl. Phys. (Japan) 21, 223 (1952). 88. Giner, J., and Lange, E., Nalurwiss. 40, 506 (1953). 83. Hackerman, N., and [Lee,] E. H., J . Phys. Chem. 69, 900 (1955). 84. Mignolet, J. C . P., J. chim. phys. 47, 172 (1950). 86. Mignolet, J. C. P., J . Chem. Phys. 20, 341 (1952). 86. Mignolet, J. C. P., J. Chem. Phys. 21, 1298 (1953). 87. Mulliken, R.S . , J . A m . Chem. SOC.71, 600 (1960);74, 811 (1952). 88. Bloyaert, F.,D’Or, L., and Mignolet, J. C. P., J. chim. phys. 64, 53 (1957). 89. Mignolet, J. C. P., J . chim. phys. 64, 19 (1957). 90. Handler, P., i n “Semiconductor Surface Physics” (R. H. Kingston, ed.), p. 23. Univ. of Pennsylvania Press, Philadelphia, 1957. 91. Brattain, W. H., and Bardeen, J., Bell System Tech. J . 32, 1 (1953). 98. Pratt, G.W., and Kolm, H. H., in “Semiconductor Surface Physics” (R. 1%. Kingston, ed.), p. 297. Univ. of Pennsylvania Press, Philadelphia, 1957. 93. Langmuir, I., and Kingdon, K. H., Phys. Rev. 34, 129 (1929). 94. Reimann, A. L., Proc. Roy. Soc. (London) A163, 499 (1937). 96. Bosworth, R.C.‘L., and Rideal, E. K., PTOC. Roy. SOC.(London) A162, 1 (1937). 96. Bosworth, R. C. L., and Rideal, E. K., Physica 4, 925 (1937). 97. Bosworth, R. C . L., Proc. Cambridge Phil. SOC.59, 394 (1937). 98. Bosworth, R. C. L., Trans. Faraday SOC.36, 397 (1939). 99. Copley, M. J., and Spence, R. W., J . A m . Chem. SOC.61, 3027 (1939). 100. Eley, D. D., and Rideal, E. K., Proc. Roy. SOC.(London) A178, 429 (1941). 101. Burshtein, R., J . chim. phys. 64, 106 (1957). 108. Bourion, R.,J. phys. radium 12, 930 (1951). 103. Mignolet, J. C. P., i n “Chemisorption” (W. E. Garner, ed.), p. 118.Butterworths London, 1957. 104. Pauling, L., “The Nature of the Chemical Bond.” Oxford Univ. Press, London and New York, 1948. 106. Trapnell, B. M. W., “Chemisorption,” Chapter VII. Butterworths, London, 1955. 106. Higuchi, I., Ree, T., and Eyring, H., J. A m . Chem. SOC.79, 1330 (1957). 107. Higuchi, I., Ree, T., and Eyring, H., J. Am. Chem. Soc. 7 7 , 4969 (1955). 108. Boudart, M., J . A m . Chem. SOC.74, 3556 (1952). 109. Eischens, R . P., Pliskin, W.A., and Francis, 8. A., J . Chem. Phys. 22,1786 (1954). 110. Eischens, R . P., Francis, 8. A., and Pliskin, W. A., J. Phys. Chem. 60,194 (1956). 111. Broeder, J. J., van Reijen, L. L., Sachtler, W. M. H., and Schuit, G. C. A., Z. Elektrochem. 60, 838 (1956). 118. Mulliken, R. S., J . Chem. Phys. 2 , 782 (1934).

SURFACE POTENTIALS AND ADSORPTION PROCESS ON METALS

131

113. Mignolet, J. C. P., Bull. SOC. chim. Belg. 66, 837 (1956). If.&. Suhrmann, R.,and Schulr, K. Z., 2. physik. Chem. (Frankfurt) [N.F.] 1, 69 (1954. 116. Moore, L. E., and Selwood, P. W., J . Am. Chem. SOC.78, 697 (1956). 116. Suhrmann, R.,i n “Chemisorption” (W. E. Garner ed.), p. 108. Butterworths, London, 1957. 117. de Boer, J. H., i n “Chemisorption” (W. E. Garner, ed.), p. 171.Butterworths, London, 1957. 118. Johnson, M., and Vick, F. A., Proc. Roy. SOC.(London) A161, 308 (1935). 119. Bosworth, R. C. L., Proc. Roy. SOC.(London) A162, 32 (1937). 180. Eley, D.D., Trans. Faraday SOC.49, 643 (1953). 181. Elovich, S.Y., and Zhabrova, G. M., J. Phys. Chem. (U.S.S.R.) 13, 1761, 1775 (1939). 188. Langmuir, I., J. Am. Chem. Soc. 64, 2798 (1932). 183. Gomer, R.,J. Chem. Phys. 28, 168 (1958). 1.84. Mignolet, J. C. P., Rec. Irao. chim. 74, 701 (1955). f86. Rideal, E. K., Proc. Cambridge Phil. Soc. 36, 130 (1939). 186. Bonhoeffer, K.F., and Farkas, A., Z. physik. Chem. (Leipzig)Bl2, 231 (1931). f87. Trapnell, B. M. W., i n “Chemisorption” (W. E. Garner, ed.), Chapter VI. Butterworths, London, 1955. 188. Ree, T., and Muroyama, N., PTOC. Imp. Acad. Japan 20.93 (1944);Chem. Abstr. 43, 5240 (1949). 189. Eley, D . D., Discussions Faraday SOC.8, 34 (1950). 130. de Boer, J . H., “Electron Emission and Absorption Phenomena.” Cambridge Univ. Press, London and New York, 1935. 131. Schuit, G. C. A., and de Boer, N. H., Rec. trau. chim. 72, 909 (1953). 138. Bagg, J., and Tompkins, F. C., Trans. Faraday SOC.61, 1071 (1955). 133. de Boer, J. H., i n “Chemisorption” (W. E. Garner, ed.), p. 27. Butterworths, London, 1957. 134. Temkin, M. I., “Symposium on Problems of Chemical Kinetics, Catalysis, and Reactivity.” Akad. Nauk, S.S.S.R., 1955. 136. Federova, A. J., and Frumkin, A. N., J. Phys. Chem. (U.S.S.R.) 27, 247 (1953). 136. McQiiillan, A. D., Proc. Roy. SOC.(London)A204, 309 (1950). 137. Becker, J. A., and Hartman, C. D., J . Phys. Chem. 67, 157 (1953).

This Page Intentionally Left Blank

Gas Reactions of Carbon P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN Fuel Technology Department, The Pennsylvania State University, University Park, Pennsylvania Page

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 11. Thermodynamics of Gas-Carbon Reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 A. Heats of Reaction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 B. Equilibrium Constants and Product-Reactant Ratios. . . . . . . . . . . . . . . . 136 111. Review of General Mechanisms for the Gas-Carbon Reactions. . . . . . . . . . . 138 A. General Remarks.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 B. Mechanisms.. . . . . . . . . . . . . . . . . ............. IV. Review of Kinetics for the Gas-C A. Orders of Reactions. . . . . . . . . . 156 B. Activation Energies of Reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Relative Rates of Gas-Carbon Reactions.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 V . Role of M:tss Transport in Chs-CtLrbon Reactions, . . . . . . . . . . . . . . . . . . . . . . 164 A. General Remarks.. . . . . . . . . . . . . . ................................. 161 B. Three Temperature Zones in (;a rliori Reactions. . . . . . . . . . . . . . . . . . 165 C. General Discussion of Zone I1 for the Gas-Carbon Rc?actions... . . . . . . . 167 D. Comprehensive Rate l
134

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

1. Introduction A substantial portion of the world’s energy requirements is met directly or indirectly through the utilization of the gas reactions of carbon and carbonaceous materials. To be particularly considered are the reactions of oxygen, steam, carbon dioxide, and hydrogen with carbon. The exothermic reaction of carbon with oxygen has been, and still is, the major source of energy in the world. The endothermic reaction of carbon with steam produces carbon monoxide and hydrogen, which are used either directly as gaseous fuels or as synthesis gas can be converted catalytically to a series of hydrocarbon fuels or organic chemicals, Since carbon dioxide is a direct product of the carbon-oxygen reaction and an indirect product of the carbon-steam reaction through the water-gas shift reaction, the secondary reaction of carbon dioxide with carbon in fuel beds is closely tied to the former gas-carbon reactions. The reaction of hydrogen with carbon to produce methane is not of great industrial significance at the moment but appears to have an important future. The gas-carbon reactions have other major contributions besides those related directly to fulfilling our energy requirements. Active carbons are produced almost entirely through the activation of carbonaceous materials with steam and/or air. The regeneration of coked or spent catalysts by burning the coke with air is an essential part of the process involving the catalytic cracking of petroleum. The production of carbon monoxide and hydrogen, which serve as reducing agents for the direct processing of ores, shows considerable promise. Paradoxically, there is a necessity in many operations for retarding the gas-carbon reactions. When carbon is used as an electrode material, it is desired that the carbon not react with either the carbon dioxide produced by reduction of the ore or with the ambient atmosphere, Carbon has excellent high-temperature strength properties which suggest its use for nozzles in rockets and nose cones in missiles, but again good oxidation resistance is a necessity. Graphite is being used to a considerable extent as a moderating material in atomic reactors; and when carbon dioxide is used as the coolant, its reaction with the graphite can be a problem. Even though the gas-carbon reactions have been an integral part of our industrial economy for many years, a fundamental understanding of their reaction mechanisms and kinetics has lagged far behind their practical use. This primarily has been caused by the lack of experimental techniques to define the properties of one of the reactants-the carbon. With the relatively recent ability to determine quantitatively such properties of solids as surface areas, pore distributions, crystallographic parameters, average crystallite sizes, chemisorption of gases, trace impurities, rates of internal gas transport, and electronic properties, the possibility of clearly understanding the gas-carbon reactions is closer at hand. Certainly, workers in

GAS REACTIONS OF CARBON

135

the field of catalysis, who have employed many of the above experimental techniques to describe their catalysts, have, in recent years, made rapid strides at understanding reactions occurring on solid surfaces. Consequently, many of the concepts developed by workers in the catalysis field have been used extensively by the authors and other workers studying the gas-carbon reactions. Even more extensive use of these connecting and related concepts is to be encouraged. With this main purpose in mind, this article has been written. In this article, the authors have attempted to supply a reference to the majority of pertinent papers on gas-carbon reactions. Reasons for the large amount of apparently conflicting data on orders and activation energies for the reactions are advanced. A detailed quantitative discussion of the role which inherent chemical reactivity of the carbon and mass transport of the reactants and products can play in affecting the kinetics of gas-carbon reactions is presented. The possibilities of using bulk-density and surface-area profile data on reacted carbons for better understanding of reaction mechanisms is discussed. Finally, some factors, other than mass transport, affecting gas-carbon reactions are reviewed.

It. Thermodynamics of Gas-Carbon Reactions A. HEATSOF REACTION Heats of reaction at 18" and 1 atm. for the important gas-carbon reactions are presented. When secondary and/or concurrent reactions are possible, data on these reactions are included. In the equations, the carbon is taken to be in the form of &graphite. On the basis of @-graphitehaving a zero heat of formation, various types of amorphous carbons are reported ( 1 ) t o have positive heats of formation (+AH)ranging from 1.7 to 2.6 kcal./mole :

136

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

TABLE I Equilibrium Constants for Gas-Carbon and Associated Reactions Temperature, O K .

1

2

3

4

5

6

7

-20.81 -13.28 -8.74 -5.72 -3.58 -1.97 -0.71 $0.28 $1.08 +1.76 +2.32 +2.80

-15.86 -10.11 -6.63 -4.29 -2.62 -1.36 -0.37 +0.42 +1.06 S1.60 +2.06 +2.44

+4.95 +3.17 $2.11 +1.43 +0.96 +O. 61 +0.34 +0.14 -0.02 -0.16 -0.26 -0.36 -

+8.82 +5.49 +3.43 +2.00 +0.95 +0.15 -0.49 -1.01 -1.43 -1.70 -2.10 -2.36

~~

300 400

+68.67 +51.54 +41.26 +34.40 +29.50 -+25.83 $22.97 +20.68 +18.80 +17.24 +15.92 +14.78 +5.14

500 600 700 800 900 lo00 1100 1200 1300 1400 4000

+23.93 +19.13 +l6.26 +14.34 +12.96 +11.93 + l l . 13 +10.48 +9.94 +9.50

+9.12 +8.79 +5.84

+44.74 +32.41 $25.00 S20.06 +16.54 +13.89 +11.84 +10.20 +8.86 +7.74 +6.80 $5.99 -0.70

-

-

-

B. EQUILIBRIUM CONSTANTS AND PRODUCT-REACTANT RATIOS Equilibrium constants for the gas-carbon arid associated reactions ( 1) to ( 7 ) , listed in the previous section, are presented in Table I. The individual concentrations of the species in the equilibrium constants are expressed as partial pressures in atmospheres. From the data (see ref. 2 ) , it is evident that the oxidation of carbon to carbon monoxide and carbon dioxide is not restricted significantly by equilibrium Considerations at temperatures even up to 4000'K. Figures 1 t o 3 present calculated equilibrium molar ratios of products to reactants as a function of temperature and total pressure of 1 and 100 atm. for the gas-carbon reactions (4),( 7 ) , and ( 5 ) , ( G ) , (4),( 7 ) , respectively. Up to 100 atm. over the temperature range involved, the fugacity coefficients of the gases are close to 1; therefore, pressures can be calculated directly from the equilibrium constant. From Fig. 1, it is seen that at temperatures above 1200°K. and at atmospheric pressure, the conversion of COn F? 2CO carbon dioxide to carbon monoxide by the reaction C essent ially is unrestricted by equilibrium considerations. At elevated pressures, the possible conversion markedly decreases; hence, high pressure has little utility for this reaction, since increased reaction rate can easily be obtained by increasing reaction temperature. On the other hand, for the 2H2 S CHI , the production of mrthnne is srriously liniitcd reaction C a t one atniosphcre pressure and practical opcrating temperatures, as see11 in Fig. 2. Obviously, this reaction must be conducted a t elevated pressures to realize a satisfactory yield of methane. For the carbon-steam reaction,

+

+

GAS REACTIONS OF CARBON

137

TOTAL PRESSURE 0

0

-

I ATM. 100

FIG.1. Equilibrium carbon monoxide-carbon dioxide ratio as a function of temperature and pressure for the reaction C COz F? 2CO. Perfect gas law assumed.

+

it is seen in Fig. 3 that the nniounts of carbon monoxide and hydrogen which can be produced above 1100" K. up to 100 atm. pressure are essentially equal, even when tJhe possible side reactions are considered. However, as in the carbon-carbon dioxide reaction, the possible extent of con-

138

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

and

version of the steam to carbon monoxide and hydrogen decreases with increasing total pressure.

111. Review of General Mechanisms for the Gas-Carbon Reactions A. GENERALREMARKS A large amount of evidence has been accumulated which shows that, one of the steps involved in a gas-carbon reaction is the chemisorption of the

139

GAS REACTIONS OF CARBON

TOTAL I

/

PRESSURE ATM.

100

1

/

0.11

'

I 1000

900

I

1100 T, K.

I

1200

I

I

1400

1300

I

FIG.3. Equilibrium product-steam ratios as a function of temperature and pressure for the reactions C H20 & CO H Z, CO HzO C o t Hz , C COz F? 2C0, and C 2Hz F? CH4 . Perfect gas law assumed.

+

+

+

+

+

+

140

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L.

o. AUSTIN

gas (in whole or in part) on the carbon surface. Further, it is known that

some of the products of the gas-carbon reactions chemisorb on the carbon surface under certain conditions. Therefore, an understanding of the chemisorption of gases on carbon is essential to the understanding of the gascarbon reactions. The modern concepts of chemisorption of gases on solids, including carbon, are reviewed by Trapnell ( 3 ) . Briefly, workers are in agreement that the chemisorption of gases on carbon occurs on a relatively small fraction of the total surface. On four amorphous carbons, Loebenstein and Deitz ( 4 ) find oxygen to chemisorb on less than G % of the total surface a t 200". Savage (6) reports that up to 4 % of a freshly formed graphite "wear-dust" surface chemisorbs hydrogen and water vapor. Methane is chemisorbed on ca. 2 % of the surface. Even nitrogen is cheniisorbed on ca. 0.4% of the surface. No chemisorption of helium or argon is found. Gadsby and co-workers ( 6 ) find that only ca. 0.5 % of a charcoal surface chemisorbs carbon monoxide a t 850". Keier and Man'ko ('7) find that the rate of chemisorption of oxygen (at 182') and hydrogen (at 485") by carbons is markedly affected by the type and amount of mineral impurity present. Zelinski (8) finds that oxygen chemisorbed on artificial graphite imparts eithcr oxidizing or reducing power to the surface, depending upon the adsorption temperature and oxygen pressure. Many workers find that the exposure of carbons to oxygen or carbon dioxide at different temperatures and pressures drastically changes the acid and base-adsorbing power of the surface. Studebaker and co-workers ( 9 ) , Garten and Wciss ( l o ) , and Graham (11) have looked a t the nature of carbon-oxygen complexes on carbon Rurfaces in considerable detail. Probably the major conclusion to he drawn from the numerous findings is that there is more than one type of carbon-oxygen complex which can form on a carbon surface. Workers find that the chemisorbed oxygen species never can be removed from the carbon surface as such. When the surface is degassed, the oxygen is removed as oxides of carbon. Upon outgassing carbons at elevated temperatures, Anderson and Emmett ( l a ) , Carter and Greening ( I S ) , and Norton and Marshall (14) find that considerably more carbon monoxide than carbon dioxide is released. The majority of the carbon dioxide is released at temperatures below GOO", and the majority of the carbon monoxide is released at higher temperatures. The workers do find that hydrogen can be desorbed from carbon as such, with the majority of it being released at temperatures above 900". In chemisorption, it is known that the surface atoms must have free valence electrons in order to form strong chemical bonds with gas molecules or atoms. Much recent work (16-18) using electron paramagnetic resonance absorption techniques has confirmed the presence of unpaired

GAS REACTIONS OF CARBON

141

electrons in various t,ypes of carbons. Observing the marked effect of exposure of carbon to oxygen on the nature of the resonance absorption, Ingram and Austen ( 1 5 ) conclude that these unpaired electrons are located primarily at, or close to, the carbon surface. Ingram and Austen (16) and Winslow and co-workers (16) find that the number of unpaired electrons is a complex function of carbon heat-treatment temperature, apparently being affected primarily by the number and nature of imperfections in the carbon structure. It is not surprising that chemisorption experiment,s have shown the carbon surface to be heterogeneous. In addition to the normal sources of heterogeneity (holes and dislocations in the lattice), carbon is a multicrystalline material, which means that its surface, in most instances, will be composed of different crystallographic planes. The crystallites in carbon are also of widely varying size, ranging from 10 A. in some amorphous materials up to thousands of angstroms in natural graphite. The degree of heterogeneity in carbon surfaces will vary depending upon the percentage of different crystallographic planes composing the surface and their size. More will be said about the possible relation between the surface heterogeneity of carbon and its gas reactions later, but it is well t o keep this heterogeneity in mind while discussing mechanisms and kinetics.

B. MECHANISMS 1. Carbon-Oxygen Reaction. The major question concerning the mechanism of the carbon-oxygen reaction has been whether carbon dioxide is a primary product of the reaction of carbon with oxygen or a secondary product resulting from the gas-phase oxidation of carbon monoxide. The obvious experimental approach to answering this question has been to retard the gas-phase oxidation of carbon monoxide. Mainly, this has been done by three methods: (1) use of low pressures (19-23), (2) use of high gas velocities (24-27)) and (3) addition of substances known to inhibit the oxidation of carbon monoxide (24, 28-31). Other workers (32, 33) have conducted the carbon-oxygen reaction at sufficiently low temperatures t o be able t o assume that the rate of carbon monoxide oxidation is negligible. Almost total agreement has been reached that both carbon monoxide and carbon dioxide are primary products of the oxidation of carbon, or that (1)

(2)

+ Odd CI + OZ(6)

2c/

-+

2 C ( O ) + 2CO(g)

+

C(02)

+

COdd

where C, represents a carbon-free site capable of reaction and C ( 0 ) and C(0Z) represent a chemisorbed oxygen atom and molecule. Likewise, it is agreed by most workers that the primary CO-C02 ratio increases with

142

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

increasing reaction temperature. Using a flow method and POC1, to inhibit the gas-phase oxidation of carbon monoxide, Arthur (SO) determined the CO-CO2 ratio for two carbons of widely different reactivities at atmospheric pressure in the temperature range 460 to 900". For these two carbons, he finds that the CO-COaratio can be expressed as CO/COz= 103.4~e-12-m/RoT (1) over the entire temperature range. Rossberg (Sf?), using two different carbons and thoroughly drying his oxygen t o prevent secondary oxidation of carbon monoxide, finds that the temperature dependence of the CO-COz ratio over the temperature range 520 t o 1420" is quite similar t o that found COZ + 2CO by Arthur. Arthur and Rossberg feel that the reaction C is not a factor in affecting the CO-COa ratio even at the highest temperatures which they use. Arthur also states that CO-COa ratios predicted from Equation (1) over the temperature range 900 to 2000" are consistent with the relative rates of formation of these two species observed by other workers (20,22, 23) in low-pressure experiments a t these temperatures. Lewis and co-workers ($3) investigated the oxidation of carbon at a total pressure of 1.1 atm. in a fluidized bed. They confirm that carbon dioxide is a primary product of carbon oxidation, but find that the CO-COz ratio is essentially constant below 520" and is equal to ca. 0.3. According to Equation ( l ) , the CO-COz ratio a t 520" should be ca. 0.9. In agreement with Arthur's findings, Lewis and co-workers report that the CO-COz ratio is relatively independent of the carbon types which they used-hardwood charcoal, metallurgical coke, and natural graphite. Day (24),studying the carbon-oxygen reaction a t atmospheric pressure and high gas velocities, finds the CO-CO2 ratio to be independent of oxygen concentration in the range of 37 to 99.6mole % a t a total gas velocity of 20,000 ft./min. over the temperature range 1300 to 1900". He also finds the CO-COzratio t o be independent of gas velocities between 10,000 and 60,000 ft./min. over the same temperature range. At somewhat lower gas velocities (5,000 ft./min., for example), the products leaving the carbon surface are not removed rapidly enough, and gas-phase oxidation of the carbon monoxide is in evidence. Day finds that the type of carbon reacted does affect the CO-COaratio under otherwise identical conditions. For example, graphitized lampblack-base electrodes have ca. a sevenfold smaller CO-COz ratio at comparable temperatures than do the corresponding ungraphitized electrodes. Day finds that the CO-COZratio increases exponentially with temperature between 1500 and 1900", but its magnitude is substantially less than that predicted by Equation (1 ). At 2100", the maximum CO-COP ratio found is ca. 28 compared with a predicted ratio of 123. Day concludes that carbon dioxide is a primary product of carbon oxidation. Arthur and co-workers (34) have made a study of a number of inhibitors

+

QAS REACTIONS OF CARBON

143

of the oxidation of carbon monoxide. They find that the best vapor inhibitors are the phosphorus halides and suggest that their main purpose is to remove water molecules from the gas phase. Many workers, including Walker and Wright ( 3 6 ) , have shown that water increases the oxidation rate of carbon monoxide. Arthur and Bowring (36)also find that inorganic halides (particularly copper chloride) deposited on carbon markedly increase the CO-C02 ratio. In summary, it is found that 1. Carbon dioxide, as well as carbon monoxide, is a primary product of carbon oxidation. 2. The ratio of the primary products, CO/CO,, generally is found to increase sharply with increasing temperature. 3. It is not well established that the magnitude of the ratio of the primary products is solely a function of temperature and independent of the carbon reacted. Lack of agreement between workers could be caused either by the inability to prevent completely secondary reactions which will change the CO-C02 ratio or by actual variations in the primary CO-COa ratio coming from different carbon surfaces. 2. Carbon<arbon Dioxide Reaction. There is general agreement (6, 3742) that experimental data on the rate of gasification of carbon by carbon dioxide fit an equation of the form

where pcoq and pco are the partial pressures of carbon dioxide and carbon monoxide and the constants kl ,k 2 , and k) are functions of one or more rate constants. There are several hypothetical mechanisms which give the required form of rate Equation (2). A completely general expression is difficult to formulate, but the following steps may be postulated: (1)

(2) (3) (4)

(6) (6)

c, + C O d d C(C02) c, + C(C02) Ft C ( 0 ) + C(C0)A c, + C ( 0 ) F? C ( C 0 ) B C(C0)A Ft C O ( d + c, C ( c 0 ) B F! cob) + c, C O ( d + c, * C(C0)C

This is analogous to the general scheme used to represent a catalytic reaction : (1)

(2)

(3)

S+RaSR SR F! SP SPF!S+P

144

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

where S represents the catalytic surface, R represents the reactant( s), and P represents the product (8). To reduce to the simple rate Equation (2), it is necessary that some of the steps in the general scheme occur at a negligible or an extremely fast rate. It has been observed (37) that carbon monoxide gas is an immediate product of the chemisorption of carbon dioxide on carbon and that the adsorption of carbon dioxide is not reversible to give immediate desorption of carbon dioxide. Therefore, it may be assumed that the lives of C(C02) and C( CO), are short. Consequently, the general expressions can be simplified t o

+ COdQ) c/ + c(0)

CI

(1)

(2)

+ CO(d

@

C(0)

@

C(C0)B

C(C0)B ’# c o ( g ) f Cf

(3)

CO(I7)

(4)

+ c,

’# C(C0)C

+

There is the further possibility that the transition Cj C(0) 4 C(CO)B is either slow (Case 1) or fast (Case 2) in comparison with C( CO) CO(g) Cf The rate expression to be derived is the same in either case, but the interpretation of the individual rate constant, j , , in Equation (5) will be different. When Case 1 holds, j 8 represents the rate constant for the surface rearrangement; when Case 2 holds, j 8 represents the rate constant for the desorption of (CO)s. It is not possible, on the basis of present experimental evidence, to decide which case is operative. It is conceivable that each case will be operative but in different temperature ranges. Assuming for the moment that Case 1 holds, the general expressions given above can be simplified to ---f

+ .

(1)

Cf

+ CO,(S) Ft C(0) + C0(!.7)

Equation (2) can now be shown to be consistent with at least two mechanisms where carbon monoxide retards the gasification reaction. Mechanism A applies where the rates of the back reactions of reaction ( 1 ) and (2) are negligible. Mechanism A : (1) (2)

(3)

c, + CO&) -% C ( 0 ) + CO(I7) C(0)

3CO(g)

145

GAS REACTIONS OF CARBON

in which il , j 3 , iz , and j z are the rate constants for these reactions. At steady state, the rates of formation and removal of the surface complexes are equal. If el and e2 are the fractions of the active surface covered by oxygen atoms and by carbon monoxide molecules, respectively, then the relative number of active carbon free sites (C,) can be expressed as ( 1 el - 02). Therefore,

which gives ilPC0,

Rate = j3el = 1

+

i2

372- pc,

which is identical to Equation ( 2 ) , where

kl

+

(5)

il 3T 3 pco,

=

i~ , k2

=

iz/j2,and

k3

=

idj3.

Mechanism R applies where the rate of the back reaction of reaction (2) is negligible and where reaction (3) is not import:int,. Mechanism I?:

C(0)

(2)

LCO(g)

Equating the rates of formation and removal of C ( 0 ) and again letting 8, be the fraction of active surface covered by oxygen atoms, Rate = j361

ilPC0,

=

1

+

5

pco

7

33

+ 33 pco,

which is again identical to Equation (2), where

k,

=

il

7-

kl

= il

, kz

= j 1 / j 3 and ,

'il/j3.

Mechanisms A and B both state that carbon monoxide retards the gasification of carbon by carbon dioxide by decreasing the fraction of the surface which is covered by oxygen atoms under steady state conditions. In mechanism A , el is decreased by the chemisorption of carbon monoxide by a fraction of the active sites. In mechanism B, is decreased by the reaction of a portion of the chemisorbed oxygen with gaseous carbon monoxide to produce gaseous carbon dioxide. Reif (37) shows that only one of these reactions can control retardation at one time. Gadsby and co-workers (6) support mechanism A for a t least three reasons. First, experiments were performed in which mixtures of carbon

146

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

dioxide and carbon monoxide in varying proportions were introduced to charcoal at 750" for a period of 20 min., after which the quantity of oxygen adsorbed on the surface was determined. They report that a wide variation in the final pressure of carbon dioxide was not accompanied by a corresponding change in the amount of oxygen complex adsorbed but followed more closely the smaller variation in the final pressure of carbon monoxide. This led to the conclusion that a large part of the oxygen on the surface at the end of this time interval was probably due to the adsorption of carbon monoxide. Second, assuming mechanism B to' be correct, Gadsby and co-workers find an activation energy of - 16.8 kcal. for reaction ( 1 ) reverse. This they conclude cannot be correct. Third, two of the above authors ( 4 3 ) discuss at length their findings that both carbon monoxide and hydrogen retard the carbon-carbon dioxide reaction but only hydrogen retards the carbon-steam reaction. They argue that the carbon-steam reaction can take place on edge carbon atoms possessing only one unshared electron and that carbon monoxide, which would be expected to chemisorb only on carbon atoms containing two unshared electrons, would not be expected to poison the carbonsteam reaction. On the other hand, it is suggested that the carbon-carbon dioxide reaction takes place on edge carbon atoms containing two unshared electrons; hence, these reacting sites can be blocked by the chemisorption of either carbon monoxide or hydrogen. If retardation in the carbon-steam reaction were occurring by reduction of the surface-oxygen complex, carbon monoxide, as well as hydrogen, should inhibit the reaction. The conclusion is that retardation in both the carbon-steam and carbon-carbon dioxide reaction is by chemisorption. Reif (37), on the other hand, supports mechanism B . He argues that Gadsby et al. incorrectly interpret their chemisorption experiments (reason one above) and further states that his own chemisorption experiments for carbon monoxide on a coke surface (37, 44) make mechanism A unlikely. Insofar as reason two offered above is concerned, Reif (37) counters with the fact that Wu (40) finds an activation energy of +21.4 kcal. for reaction (1) reverse under mechanism B. Reif does not comment on reason three given by Gadsby and co-workers but acknowledges that there is a possibility that the two retarding reactions may be operative far different types of carbon under different conditions of temperature and pressure. Ergun ( 4 6 ) presents results which very strongly support mechanism B. Experiments were conducted in a fluidized bed using three different types of carbon (Ceylon graphite, activated carbon, and activated graphite). These samples had a considerable range of mineral content (from a trace to 0.5 %) ;and although not reported, it is certain that they also had a wide

147

GAS REACTIONS OF CARBON

10-

-

1400 1300 1200 1100

I

I

I

1000

-

800

900

700

1

I

I

0

- ACTIVATED

CARBON

@

- ACTIVATED

GRAPHITE

0

- CEYLON

GRAPHITE

I1

-

-

-

-

-

-

-

0.1 -

-

ul3 Y

-

-

-

I --

$

C.

-

-

-

-

-

-

-

0.01

I

0.6

I 0.7

I

0.8

I

I

0.9

1.0

1.1

FIG.4 . Equilibrium constant of reaction (1) for mechanism B in the carbon-carbo dioxide reaction as a function of temperature. [After S. Ergun, J . Phys. Chem. 80 480 (19561.1

range of specific surface area. In spite of this, as is shown in Fig. 4,Ergun finds the equilibrium constant for reaction ( 1 ) of mechanism €3 to be independent of the carbon used and the reaction to have an average AH of +23 kcal./mole over the temperature range 800 to 1400". Because of its high temperature coefficient, Ergun feels that the equilibrium has a pronounced effect on the rate of gasification. If, for example, in the gas phase, the CO-CO, ratio equals 1, the fraction of the total active sites which are

148

P. L. WALKER, JR., FRANK RUSINKO, JR., A N D L. G . AUSTIN

occupied, C ( 0 ) in this case, increases from 0.0215 to 0.81 in going from 700 t o 1400".Since the gasification rate is proportional t o the number of occupied sites, the effect of the equilibrium constant on the rate is through its influence on the concentration of occupied sites. Key ( 4 6 ) and Strickland-Constable ( 4 7 ) also support mechanism B for the carbon-carbon dioxide reaction. Strickland-Constable concludes from earlier measurements ( 4 8 ) that the rate of adsorption of carbon monoxide on carbon is too low to account for the retardation. At, 450" and a total pressure of 1.1 atm., Paxton (4.9) finds that for oxygen pressures between 0.21 and 0.5 atm. the reaction rate with carbon monoxide dilution is more than twice that with nitrogen dilution. This finding also appears to support indirectly mechanism B for the carboncarbon dioxide reaction. If chemisorption of carbon monoxide were occurring a t a significant rate in Paxton's work, blockage of additional carbon-free sites would occur, which should retard the carbon-oxygen reaction. Instead, the carbon monoxide is presumably removing relatively stable carbon-oxygen complex, which is produced by the product carbon dioxide through the back reaction, as discussed shortly. Workers have used radioactive carbon, CI4, as a tracer to study oxygen and carbon exchange reactions occurring during the over-all gasification of carbon with carbon dioxide. Bonner and Turkevich ( 5 0 ) find that reaction ( 1 ) of mechanism A and reaction ( 1 ) forward of mechanism B is rapid on charcoal at temperatures of 735 and 840" and initial carbon dioxide pressures of 180 and 330 mm. Hg. On the other hand, under these conditions they find reaction ( 2 ) of both mechanisms t o be slow. They confirm that some carbon from the original carbon dioxide has also transferred t o the charcoal surface. Brown ( 5 1 ) investigated carbon transfer to the surface of graphite and sugar carbon during their reaction with carbon dioxide. He finds carbon transfer for both materials but states that it is much greater for the sugar carbon. He suggests that when carbon dioxide reacts with a small fraction of the active surface (perhaps 2% of the active surface for the sugar carbon), the carbon dioxide deposits its carbon atom on the surface and its oxygen atoms depart with two new carbon atoms. Orning and Sterling ( 5 2 ) find that the rate of oxygen transfer to a carbon surface depends upon the nature of the solid, presence of catalytic agents, and gas composition. Potassium carbonate, which is known to catalyze carbon gasification, also enhances oxygen transfer to a high temperature coke. Orning and Sterling find the specific radioactivity of the product gas equal to that of the entering gas as long as the temperature is low enough for gasification to be negligible. This indicates that chemisorption of carbon monoxide is also negligible under these conditions. 3. Carbon-Steam Reaction. There is general agreement ( 4 ,43, 53-55)

GAS REACTIONS OF CARBON

149

that experimental data on the rate of gasification of carbon by steam fit an equation of the form Rate =

1

+

klPH,O k2pH2

(7)

+

k3pHzO

where pHzOand pH, are the partial pressures of steam and hydrogen and the constants k1 , k 2 , and k3 are functions of one or more rate constants. The form of this equation is identical to that for the carbon-carbon dioxide reaction.* The mechanism suggested by Gadsby et al. ( 5 3 ) and Johnstone et al. ( 5 4 ) is as follows: Mechanism A : (1)

CI

+ HsO(g)

il

C(Hz0) 31

(2)

C(Hn0) I C o b ) 3s

+ Hz(g)

(3)

At steady state, the rates of formation and removal of the surface complexes are equal. If e3 and e4 are the fractions of the active surface covered by water and hydrogen molecules, respectively, then ilPH2Ou ilPH,(i

- e3 - 6,)

+ j3e3

jle3

(8)

- ea - e4> = j2e4

(9)

=

which gives

If the rate of evaporation of water molecules from the surface is negligible ( j , is small), kl = il , kz = i 2 / j 2 , and k 3 = iJj3 . Mechanism A can be written in slightly more detailed form as

* As in the carbon-carbon dioxide reaction, mechanisms A and B can be treated for the cases where either the surface rearrangement or desorption of the carbonoxygen complex is the slow step. This has no effect on the discussion except that the significance of the rate constant j 3 in Equation (10) is altered, as previously discussed.

150

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

(4)

If j , << j , and jl is small, the correct rate equation may be derived. Alternatively, it is found that if j , << j , and js << j , , which implies that the surface reaction is fast compared with the desorption of C ( 0 ) as CO, a rate expression identical to Equation (10) is obtained. Under these conditions the mechanism can be expressed more simply by equations similar to those for mechanism A of the carbon-carbon dioxide reaction as (1)

CI

+ HzO(u) -% C(0) + Hdg) C(0)

(2)

(3)

Hi(g)

2CO(g) in

+ Cf

C(&)

3n

The mechanism of the carbon-steam reaction is discussed in more detail by Long and Sykes ( 43) .They propose that the steam molecule decomposes at the carbon surface into a hydrogen atom and hydroxyl radical both of which chemisorb rapidly on adjacent carbon sites. This is followed by the hydrogen atom on the chemisorbed hydroxyl radical joining the hydrogen atom on the adjacent carbon site and leaving as a hydrogen molecule. Therefore, a further breakdown of the steps in mechanism A may be written as (2)

+ HaO(g) C(H) + c(OH)

(3)

C(0)

(4)

c(HJ

(1)

2Cf

+

C(H)

+

C(Hd

-+

CO(g)

Fr?

CI

+ C(OH) + C(0)

+ Hz(g)

A second mechanism for the carbon-steam reaction, similar to mechanism B of the carbon-carbon dioxide reaction, may be operative. Mechanism B : (1)

CI

+ HP(g)

il

C(0)

+ HAg)

31

(2)

C(0)

5 CO(g)

This mechanism also gives Equation ( 7 ) directly.

GAS REACTIONS OF CARBON

151

Whereas there has been considerable discussion as t o the possibility that the retardation of the carbon-carbon dioxide reaction by carbon monoxide is caused by reduction of the amount of chemisorbed oxygen on the carbon surface, the like possibility for hydrogen retardation in the carbon-steam reaction generally has not been discussed. Reif (37), using Key's suggestion that the carbon-steam reaction follows an analogous reaction mechanism to the carbon-carbon dioxide reaction and noting that the attainment of equilibrium in the water-gas shift reaction is catalyzed by carbon ( 5 6 , 5 7 ) , suggests the following equations as part of the carbon-steam reaction (1) (2)

+ HzO(g) F? C(O) + Hz(g) CO(g) + C ( 0 ) F? CO&) + c, CI

In addition t o these reactions causing the rapid attainment of equilibrium in the water-gas shift reaction, they should retard the rate of carbon gasification by reduction of the concentration of chemisorbed oxygen on the carbon surface. However, Gadsby and co-workers ( 5 3 ) find that the addition of carbon monoxide does not inhibit the gasification of carbon with steam other than its resulting in the production of more hydrogen which does inhibit the reaction. Recently, Ingles (58) concludes that a carbon surface accelerates the water-gas shift reaction by acting as a chain initiator to the following reactions (1) (2)

(3)

+ H(g) H(g) + H2O(g) F? O H b ) + Hdg) OHb) + F! CO&) + H(g) C(H) + C,

which means that the acceleration of the water-gas shift reaction by carbon and the lack of retardation of the gasification reaction by carbon monoxide (and possibly hydrogen) need not contradict each other. Strickland-Constable ( 4 7 ) , observing that hydrogen is not only strongly but very rapidly adsorbed on carbon, supports the view the hydrogen retardation in the carbon-steam reaction is caused by its chemisorption on active sites. 3. Carbon-Hydrogen Reaction. Surprisingly little work has been published on the carbon-hydrogen reaction. Zielke and Gorin (69) investigated the gasification of a low-temperature char in a fluidized bed at temperatures between 810 and 928" and hydrogen pressures of 1 t o 30 atm. They propose the following mechanism for the conversion of carbon and hydrogen to methane:

152

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

H H H H

H H

H

/c=c\ /' ---c/c=c\ c-c /c=c\ \

\\ /I'

c-c

//

\

c=c/c-c\ ,'

\

\&-A/ / \

\

/c=c\

-- C \

p-

/

c-c

,//

'\

c-c /

\

\

\

/

/c-c\ c=c\

/

'\

/

c=c/' C--

\\

'\

H H H H H

H\

ks + Ho -+

/CH3

--.c/c=c\ c-c /c=c\ \ / / \ \ / / c-c ,,' \ /c-c\, c=c /'

H

\

, /

/H

H\

C.--

'\

/'

---c/c=c\ c-c /c=c\ C--. \ \ c-c / c-c /

+ 2CH4

Zielke and Gorin suggest that edge groups -CH=CHare always regenerated by resonance considerations. On the basis of the assumptions that (1) on the average, an equal number of new active sites represented by -CH=CHare regenerated for each one consumed, (2) reaction 3 is rapid compared to reactions 1 and 2, and ( 3 ) a steady-state concentration of the product of the forward reaction 1 is established, they state that the rate of methane production is given as

GAS REACTIONS OF CARBON

153

where A represents the number of active groupings per unit of carbon. However, they do not find that this equation correlates the rate data a t 870". They suggest that this is a result of a conglomeration of different types of carbon reacting at different rates. At 9 2 8 O , they say that they would expect the carbon to become more uniform and they indeed find that an equation similar t o Equation ( 1 1 ) of the form

does correlate the rate data a t 10 and 20 atm. hydrogen pressure a t different percentages of carbon gasified. The authors further confirm that methane does not retard the carbon-hydrogen reaction other than through equilibrium considerations. It is obvious t o the writers that much more research must be done on the carbon-hydrogen reaction before it is well understood.

IV. Review of Kinetics for the Gas-Carbon Reactions A. ORDERSOF REACTIONS When the rate of a gas-carbon reaction is being controlled solely by the inherent chemical reactivity of the solid (and not in part by mass transport of t,he reacting gas to the surface of the solid), a relatively simple qualitative discussion of reaction order is possible. Icor simplicity, the rate of reaction (weight loss of carbon) for the carbon-carbon dioxide, carbonoxygen, and carbon-steam reactions can be assumed to be determined by the rate of surface rearrangement of the carbon-oxygen complex to a rapidly desorbable product. (The discussion would follow in a similar manner if the rate of reaction was determined by the rate of release of the desorbable product.) If the fraction of the surface covered by a carbonoxygen complex is 8, the rate of reaction is proportional to the product of 0 and a rate constant. At a particular temperature, the order of reaction depends upon the relationship between the change in 0 with the change in pressure of the reacting gas. At one extreme, if 0 approaches one throughout the range of pressure change investigated, the reaction will be zero order. At the other extreme, if 0 is small, the change in 0 will be directly proportional t o the change in pressure, and the reaction will be first order. At intermediate values of 0, the order of the reaction will vary from zero to one. The value of e is a function of the magnitude of the individual rate constants for the formation of the surface complex and its conversion to a desorbable product and the pressure of the reacting gas. If the product of the rate constant for the formation of the surface-oxygen complex and the pressure of the reacting gas is large compared with the rate constant for

154

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. 0 . AUSTIN

the conversion of the surface-oxygen complex to a rapidly desorbable product, 6 + 1. On the other hand, if the product of the rate constant for the formation of the surface-oxygen complex and the pressure of the reacting gas is small compared with the rate constant for the conversion of the surface-oxygen complex to a rapidly desorbable product, 0 + 0. It i8 apparent that at a particular temperature, where the values of the rate constants for the formation and conversion of the surface-oxygen complex are fixed, the pressure of the reacting gas can affect 0 and, hence, the order of the reaction. If the pressure is sufficiently low, the product of the rate constant for the formation of the surface-oxygen complex and the pressure will be small compared with the rate constant for the conversion of the surface-oxygen complex and 0 + 0. At sufficiently high pressures, 0 + 1. Therefore, the order of the reaction at a particular temperature can range from zero to one, as the pressure of the reacting gas is decreased over a wide range. Reaction temperature can also affect the order of a reaction. It is generally agreed that the rate constant for the conversion (or desorption) of the surface-oxygen complex has a higher activation energy than the rate constant for the formation of the complex. Therefore, a reaction which is zero order at low temperatures and a given pressure can become first order at the same pressure and a sufficiently high temperature. Unfortunately, insofar as a clear understanding of the true orders of gascarbon reactions is concerned, the problem is made more difficult when the gasification rate is affected by product retardation and by the rate of mass transport of reactants to the surface of the solid. Product retardation can result in the obtaining of orders of reaction which are too low, while mass transport retardation can either raise or lower the apparent order depending upon the true order of reaction and the nature of the mass transport control. In Secs. V and VI, these complicating factors will be discussed in more detail. In the remainder of this Section, pertinent references on the orders of gas-carbon reactions will be given. 1. Carbm-Carbm Dioxide Reaction. It is seen from Equation (2) that the carbon-carbon dioxide reaction will be zero order when k2pco << 1 and k3pcon >> 1. At low temperatures, the production of carbon monoxide is small and the first inequality is satisfied. At high carbon dioxide pressures, the second inequality is satisfied. On the other hand, it is seen from Equation (2) that the carbon-carbon dioxide reaction will be first order when k2pc0 << 1 and krpcoz << 1. These inequalities will be satisfied at low temperatures and low carbon dioxide pressures, Workers have shown ( 6 , 3 9 , 4 0 ) that kZ and k0 decrease sharply with increasing temperature; therefore, at high temperatures and increasing pressures the inequalities kzpco << 1 and kspco << 1 still can be found to hold, resulting in a first-order reaction.

GAS REACTIONS OF CARBON

155

Workers reporting orders of reaction for the carbon-carbon dioxide reaction include Graham (41 ), Strickland-Constable ( 4 8 ) , Vulis and Vitman (SO),Thring and Price (61) , Armington (62), Vastola ( 6 3 ) , Duval ( 6 4 ) , and Karzhavina (66). As expected, they find reaction orders which vary from zero to one depending upon temperature, pressure, type of carbon reacted, purity of carbon, and geometric dimensions of the sample. 2. Carbon-Steam Reaction. The analysis of the order of the carbon-steam reaction as deduced from Equation ( 7 ) is identical to that for the carboncarbon dioxide reaction. Orders ranging from 0 to 1 are expected. As for the carbon-carbon dioxide reaction, it has been shown that k2 and ka in Equation (7) decrease exponentially with temperature (41,63, 6 4 ) , resulting in a first-order reaction a t sufficiently high temperatures. Batchelder, Busche, and Armstrong (66) have taken the data of Johnstone et al. ( 6 4 ) for the variation of k2 and ka with temperature and have shown that at a total pressure of 1 atm., the carbon-steam reaction is expected to be first order above 1370". Workers reporting on orders of reaction for the carbonsteam reaction include Graham (41 ), Strickland-Constable (@), Mayers ( 6 7 ) ,Pilcher, Walker, and Wright (68), Key and Cobb ( 6 9 ) ,James ( 7 0 ) , Goring and co-workers (71), Tuddenham and Hill ( 7 2 ) , and Binford and Eyring ( 7 3 ) .They find reaction orders varying from 0 to 1. 3. Carbon-Oxygen Reaction. Discussion on the reaction orders of the carbon-oxygen reaction was deliberately postponed until after presenting results for the carbon-carbon dioxide and carbon-steam reactions to emphasize the considerable difference in experimental findings between these reactions. The majority of results under varied experimental conditions show the carbon-oxygen reaction to be first order, or close t o first order, only. The findings of Strickland-Constable (22),Day (24), Rossberg ( S 2 ) , Lewis et al. (SS), Armington (62), Mayers ( 7 4 ) , Scott and Jones ( 7 6 ) , Sihvonen ( 7 6 ) ,and Chen et al. (77) substantiate this statement. From the previous discussion, the implication of the first-order reaction is that under all experimental conditions used by the above authors, the fraction of the total active carbon surface occupied by an oxygen complex at any given instant during the reaction approaches zero. Two notable exceptions to the carbon-oxygen reaction being first order are found. Gulbransen and Andrew ( 7 8 ) , working with spectroscopic graphite, find that at reaction temperatures of 450 and 500" the order is nearly zero at pressures below 0.15 cm. Hg. They do state further that at pressures above 10 cm. Hg the reaction is first order. Blyholder and Eyring (79) reacted extremely thin coatings of graphite, which were supported on a ceramic base, with oxygen a t 800" and pressures less than 100 p Hg. From limited data (at least as presented in the paper), they conclude that the reaction is of zero order. These results are of extreme interest, since they

156

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. (3. AUSTIN

appear to contradict the reasoning behind the orders of heterogeneous reactions. As was discussed previously, a reaction should become zero order when 0 --+ 1. This state should be favored at high reacting pressures, not a t the low pressures reported by the above authors. Further, Gulbransen and Andrew's statement that the reaction order increases with increasing pressure is difficult to explain on theoretical grounds. Indeed, their data a t 450" over the entire pressure range could be better expressed as a half-order renction, as pointed out by Blyholder and Eyring (79). The results of the above two groups of workers indicate the necessity of more experimental work being done before the reasons behind orders of reaction for the carbonoxygen reaction are well understood. 4. Carbon-Hydrogen Reaction. Limited data are available on the reaction orders of the carbon-hydrogen reaction. Equation ( 11) suggests that at low pressures a maximum order of two should be obtained; and at high pressures, the order should go to one. Zielke and Gorin (69) find that at 928", a t hydrogen pressures between 10 and 20 atm., and a t 20% carbon gasified, the reaction order is 1.60. Between 20 and 30 atm., the reaction order decreases to 1.27. On some very limited data, Gilliland and Harriott (80) conclude that, a t low hydrogen pressures and 540" the react,ion may be 0.5 order.

B. ACTIVATION ENERGIES OF REACTIONS It has been popular for workers to determine activation energies for the gas-carbon reactions. However, the correct interpretation of the meaning of the activation energies frequently has not been made. Primarily they have failed to recognize the part which mass-transport resistance can play in affecting the activation energy values. In Secs. V and VI the effect of varying degrees of mass transport control on activation energies will be discussed. At, the moment, we are concerned about the values for activat'ion energies of gas-carbon reactions when the rate of reaction is controlled solely by resistance t o chemical reactivity. Wicke and his school (3f,32, 81, 82) have been particularly concerned about the true activation energies for the gas-carbon reactions. Rossberg (32)suggests that the slow step in these reactions is the separation of tin oxygen atom from the reactant species. Therefore, he suggests that the activation energies for the different reactions should be related t o the energy necessary to dissociate the reacting species. Table I1 presents a comparison between the dissociation energies of the reactant gas and the true chemical activation energies of the'corresponding gas-carbon reactions, which, according to Rossberg ( 3 2 ) ,confirms the above hypothesis. It would appear inconsistent, however, in light of the previous discussion on orders of reaction, t o say that the separation of an oxygen atom from the reac-

157

GAS REACTIONS OF CARBON

TABLE I1 Comparison of True Activation Energies i n Reactaons of Carbon with OxygenContaining Gases and the Dissociation Energy of an 0 Atom from the Reactant (after Rossberg") Reaction

c + coz

-+

2co

+ HZ0 CO + Hz C+~OZ-+CO C + NzO CO + Nz C

-+

-+

I

True activation energies, kcal./mole

I

Dissociation reaction and energy, kcal ./mole COn* CO

86 ca. 80 50-58

HzO

+

won -+

40-50

NzO

-+

+ 0,

+ 0, N Z+ 0,

HZ 0,

AH = +I26 = +I16 = +59 = +39

Rossberg, M . , Z. Elektrochem. 60: 952 (1956).

tant species need necessarily be the slow step in the over-all gasification reaction. Indeed, a t least for the carbon-carbon dioxide and carbon-steam reactions a t low temperatures and a t pressures not too far removed from atmospheric, the reaction is found to be of zero order. As discussed, the implication of the zero order reaction is that the over-all gasification rate is being controlled by the rate of removal or rearrangement to a desorbable product of the surface-oxygen complex and not by the rate of its formation. Therefore, the activation energy is that for the breakdown of the surfaceoxygen complex t o release carbon oxides. Even for the carbon-oxygen reaction proceeding under first order conditions, it is doubtful whether Rossberg's concept has any significance, since i t would appear unsound to compare half the dissociation energy of oxygen with the activation energy of the reaction. Clearly, the activation energy will be the same whether the reaction rate is expressed in terms of moles of oxygen or atoms of oxygen reacting per unit time; therefore, the correct dissociation energy for comparison is 118 kcal./mole and not the value of 59 used by Rossberg (unless it is postulated that oxygen chemisorbs in the form of a peroxide structure). Perhaps a more reasonable explanation for the relative activation energies for the reactions between carbon and the oxygen-containing gases (oxygen, carbon dioxide, and steam) is more in line with the following picture, which has been indirectly suggested in a paper by Long and Sykes ( 4 3 ) .For these reactions, the process of going from a reacting gas molecule and a carbon free site to a surface-oxygen complex, C(O), is exothermic. The exothermicity of this process for the carbon-oxygen reaction is estimated t o be nearly twice that for the carbon-steam or carbon-carbon dioxide reactions. The magnitude of this excess energy could determine the lifetime of the carbon-oxygen complex on the surface. For the carbon-oxygen reaction, this duration could be relatively small, the surface coverage

158

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

in turn small, and the over-all activation energy determined by the adsorption step. For the carbon-steam and carbon-carbon dioxide reactions, this duration could be relatively long, the surface coverage in turn large, and the over-all activation energy determined by the desorption step.* Since it is generally agreed that the desorption of the carbon-oxygen complex has a higher activation energy than the initial adsorption of the complex, the carbon-steam and carbon-carbon dioxide reactions would be expected to have an over-all higher activation energy than the carbon-oxygen reaction. The same reasoning can be extended t o the carbon-nitrous oxide reaction if desired. In any event, there appears to be reasonable experimental evidence that the activation energy for the carbon-oxygen reaction is almost always less than that for the carbon-carbon dioxide and carbon-steam reactions. Some particular results on the different gas-carbon reactions can now be considered. 1. Carbon-Carbon Dioxide Reaction. It is important to realize that t,he activation energy determined for the carbon-carbon dioxide reaction need not refer to the same rate-controlling step in every case. In Equation ( 5 ) , it is seen that at low temperatures and pressures the rate of reaction is proportional t o il , the rate constant for the formation of the surface-oxygen complex. At low temperatures and higher pressures, the rate constant is proportional to j , , the rate constant for the removal of the surfaceoxygen complex. Where none of the terms can be dropped from the denominator of Equation ( 5 ) , an understanding of the exact physical meaning of an over-all activation energy is made difficult. Rossberg (32) bases his recommended activation energy of 86 kcal./mole for all carbons undergoing gasification with carbon dioxide in the chemical control region on the experimental findings of Wicke ( 3 1 ) , who finds the same activation energy for a high-purity electrode carbon and a mediumpurity activated charcoal. Since Wicke’s experiments were conducted in a flow sytitem close to atmospheric pressure, the rate-determining step presumably was the desorption of the surface-oxygen complex. Indeed, observing that the frequency factors determined from the reaction rates are not consistent with the concept of activation energies produced by molecules impinging on the Burface, Wicke also concludes that desorption from

* In support of the hypothesis regarding the relative lifetime of the carbon-oxygen complexes on the surface for the different reactions, Paxton (49) finds the carbonoxygen reaction to be accelerated by carbon monoxide. The writers suggest that the addition of carbon monoxide to the incoming oxygen drives the reaction C ( 0 ) CO(g) @ COp(g) Cf in the forward direction and reduces the extent of Burface coveragc by the relatively stable carbon-oxygen complex, which is produced by the product carbon dioxide through the back reaction. Other workers who have shown that a stable surface-oxygen complex can retard the carbon-oxygen reaction are Arthur, Newitt, and Raftery (83) and Lambert (8.4).

+

+

GAS REACTIONS OF CARBON

159

the surface, energetically supported by thermal vibrations of the graphite lattice, probably is the rate-determining step. This is important for this means that Wicke’s activatioii energy probably represents the value belonging t o j 3 in Equation ( 5 ) . Ergun (45) reports the activation energy for the product ( j 3 ) ( C t for ) three different types and purities of carbon (Ceylon graphite, activated carbon, and activated graphite), using a fluidizing bed operating close to atmospheric pressure. He finds the same activation energy in each case (59 kcal./mole), and assuming that C t (the total number of active sites) does not change with temperature, concludes that this is the activation energy for the rate constant j 3 .Within each set of results, both Wicke and Ergun have confirmed that different carbons can have the same activation energy for the carbon-carbon dioxide reaction, but obviously the lack of agreement of the two investigators on the same activation energy still leaves the issue unsettled. Armington (62) estimates the activation energy for the gasification of a series of graphitized carbon blacks and graphite ((wear-dust” at 0.1 atm. carbon dioxide pressure. A zero-order reaction is found for all samples investigated, indicating that the over-all gasification rate is proportional to j 3 . For seven different samples, Armington finds the activation energies to vary from about 73 t o 97 kcal./mole, with the arithmetic average being 88 kcal./mole. These values are in considerably better agreement with Wicke’s value than with Ergun’s value. The range of activation energies found, however, does keep open the question, “In the region of chemical reactivity control, do all carbons have the same over-all activation energy for a given gas-carbon reaction?” Other workers (67, 85-89) have determined over-all activation energies for the carbon-carbon dioxide reaction, but the values have been affected to some extent by mass-transport control. Workers (6, 39, 40, 41) have also determined activation energies for the individual rate constants in Equation (5) but do not agree on their magnitude. The values of activation energy reported for rate constant il vary from 26.5 (41) to 61.5 kcal./ mole (40). 2. Carbon-Steam Reaction. The discussion of activation energies for the carbon-steam reaction using Equation (10) is analogous to the prior discussion for the carbon-carbon dioxide reaction using Equation (5). It is not clear from which source Rossberg ( 3 2 ) obtained his recommended activation energy of ca. 80 kcal./mole for the carbon-steam reaction. According to Hedden (90), however, recent data on the carbon-steam reaction, using the same experimental arrangement and carbons as used by Wicke (31) on the carbon-carbon dioxide reaction, yield an activation energy of 71 kcal./mole for il in Equation (10). Recently, James (70),

160

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

using a flow system and reacting graphite rods with steam close t o atmospheric pressure, determined an over-all activation energy of 69 kcal./ mole in the chemical control region. Since the reaction was also reported to be of zero order, this should be the activation energy for the rate constant j , in Equation (10). Binford and Eyring (‘73),using a flow system and reacting graphite rods at pressures below 100 p Hg, report an activation energy of 60 kcal./mole concurrent with a zero order reaction. Again, this activation energy should be for rate constant j 3 in Equation (10). Other workers (68, 91-93) have determined over-all activation energies for the carbon-steam reaction, which in most cases undoubtedly are low because of some mass-transport control. Again, workers ( 4 2 , 43, 53, 54) have determined activation energies for the individual rate constants in Equation (10) but do not agree on their magnitude. This apparently is not surprising, for, as Johnstone, Chen, and Scott (64) show, the activation energies even vary with per cent burn-off of the carbon. Long and Sykes (94) investigated the effect of removal of impurities from coconut shell charcoal on the individual rate constants. They find that the activation energy for the step in which adsorbed oxygen atoms are converted to gaseous carbon monoxide is increased from 55 f 7 t o 83 f 5 kcal./mole upon purification of the charcoal. 3. Curbon-Oxygen Reaction. Rossberg (32) apparently bases his recommended activation energy of 50 to 58 kcal./mole on two experiments. Wicke (31), working with the same experimental set-up as used for the carbon-carbon dioxide reaction, reports a value of 58 f 4 kcal./mole for crushed electrode carbon. Rossberg (32), using spectrographic carbon tubes of apparently the same source as those used by Wicke, finds an activation energy of 49.5 kcal./mole. Rossberg is certain that his lower value is not caused by partial mass-transport control. Actually, upon looking at Rossberg’s data, it is seen that his activation energy cannot be reported to an accuracy better than f 5 kcal./mole, which means that there is little, if any, significant difference between the two above results. Armington (M), reacting three graphitized carbon blacks and two graphite “wear-dust” samples in 0.1 atm. of oxygen between 550 and 600”, reports activation energies ranging from 46 t o 58 kcal./mole. He finds the reaction to be close t o first order. As in the case of the order of reaction, the results of Gulbransen and Andrew (‘78)and Blyholder and Eyring (79) again are difficult to resolve on the basis of the above data. Both groups have determined activation energies under conditions where mass transport should not affect the results. Gulbransen and Andrew, reacting thin spectroscopic graphite plates between 425 and 575” under 0.1 atm. of oxygen, report an activation energy of 36.7 kcal./mole. They base their value on reaction-rate data at

QAS REACTIONS OF CARBON

161

zero time. Armington (62) finds considerable difficulty in duplicating rate data at extremely low burnoffs ( t 0) but has little difficulty a t burnoffs above 2%, using an apparatus quite similar to that of Gulbransen and Andrew. Using Gulbransen and Andrew's rate data a t 425 and 575" after 1-hr. reaction time, the writers calculate a somewhat higher activation energy, ca. 40 kcal./mole. Unfortunately, rate data are not available a t 575" for longer periods of time. The writers have found in their laboratory that invariably after a certain burnoff (depending upon the reactor, temperature, and sample) , a subsequent extended period of constant reaction rate, expressed in grams of carbon reacting per unit time, is attained. In this burnoff region, there obviously is equilibrium between the rate of formation of the surface-oxygen complex and its removal with a carbon atom. It is felt that this is the reaction rate most characteristic of a given temperature and should be used in kinetic calculations. In principle, Wicke (31) concurs with this reasoning and reports reactivity data only after the sample has attained a total surface area which is virtually constant. Blyholder and Eyring (79), reacting very thin coatings of spectroscopic graphite, report an activation energy of 80 kcal./mole a t pressures below 100 p Hg and temperatures around 800". As discussed before, the reaction is reported to be zero order. While the writers do riot understand why the order of the reaction is zero, the activation energy is in line with such a value. The zero order is indicative of the building up of a more stable surfnce-oxygen complex in a manner similar to the carbon-carbon dioxide and carbon-steam reactions. Therefore, the activation energy for Blyholder and Eyring's experiment should be, and is, comparable to that for the gasification reactions. Other workers reporting activation energies for the carbon-oxygen reaction include Meyer (21) , Lewis et al. (33), Chen et al. (77), Lambert (95), Letort and Magrone (96) , Golovina (97), Klibanova and Frank-Kamenetskii (98),and Earp and Hill (99). Activation energies are found to vary from 17 ( 3 3 ) to 100 f 30 kcal./mole (98). Some of the lower activation energies are influenced by mass transport resistance. Both sets of experimenters who found high activation energies (21,98) worked a t low pressures, as did Blyholder and Eyring (79). 4. Carbon-Hydrogen Reaction. Zielke and Gorin (59) determined the activation energy for the reaction of a low temperature char between 810 and 928" a t a hydrogen pressure of 30 atm. They find that the activation energy increases from 15 to 48 kcal./mole as the per cent burnoff increases from 0 to 60. They attribute this increase to the heterogeneous char structure approaching that of graphite progressively more closely with increasing burnoff. The possibility that the lower activation energies are in part con---f

162

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. 0. AUSTIN

trolled by mass transport resistance is considered but concluded not t o be a factor. Unfortunately, insofar as understanding the activation energy data is concerned, the char had seen a maximum temperature of only 600", prior to the above runs, which means that a material of different properties was reacted at each temperature. Gilliland and Harriott (80) investigated the reactivity with hydrogen at one atm. pressure of carbons deposited from hydrocarbons on porous carriers. The porous carrier usually consisted of 28 to 200 mesh silica gel impregnated with nickel. The carbon deposition and subsequent reactivity studies with hydrogen were both carried out in a batch fluidized reactor. In the temperature range 538 t o 660",the activation energy for the reaction of all carbons with hydrogen is found t o be roughly the same, 36 f 6 kcal./ mole. The authors also conclude that mass transport resistance is not affecting the gasification rates and, hence, the activation energies.

REACTIONS C. RELATIVERATESOF GAS-CARBON Under fixed experimental conditions, the rate of a gas-carbon reaction (rate of removal of carbon atoms from the surface) is dependent upon the reacting gas and the nature of the carbon. A discussion of the effect of the nature of the carbon on particular gas-carbon reactions is postponed until Sec. VII. In this section, existing data on the relative rates of gas-carbon reactions, where an investigator has reacted the same carbon, is presented. Results of primary interest are those where the reaction rates are not affected by mass transport resistance. Obviously, since the gas-carbon reactions have different activation energies and orders of reaction, the relative rates of these reactions will be a function of the temperature and pressure selected for the correlation. Unfortunately, the authors can find no reactivity data for all four of the gascarbon reactions using the same carbon. Furthermore, data for even two of the gas-carbon reactions on the same carbon are limited. The available data will be taken and extrapolated, where necessary, to give at least a qualitative idea of relative rates of the gas-carbon reactions a t 800" and 0.1 atm. gas pressure. Extrapolation of these relative reactivities to other temperatures and pressures by the reader will require the assumption of activation energies and orders of reaction. Gadsby and co-workers (63) report that for a coal charcoal, the rate of the carbon-steam reaction is greater by a factor of about three than the carbon-carbon dioxide reaction ut 800" and a pressure range of 50 to 500 mm. Hg. The rcsults of Pilcher et al. (68) and Walker et al. (M), using the same graphitized carbon rods and apparatus, essentially agree with this finding. At 1100", the former workers report a reaction rat#eof 1.6 g./hr. at a steam partial pressure of 142 mm. Hg, which can be extrapolated t o 4.8

GAS REACTIONS OF CARBON

163

g./hr. a t 1 atm. using their experimental order of reaction of 0.66. The latter authors report a reaction rate of 1.7 g./hr. for the carbon-carbon dioxide reaction a t 1100" and 1 atm. pressure, giving a ratio of reaction rates of 2.8. The important point to be made from these results is that the rates of the carbon-steam and carbon-carbon dioxide reactions are quite similar. On the other hand, available data show the rates of the carbon-oxygen and carbon-carbon dioxide reactions to be markedly different. Gulbransen and Andrew (78), reacting spectrographic graphite plates, find reaction rates of 2.5 x 10-8 g./cm.2/sec. at 575" and 0.1 atm. of oxygen and 1.1 X lo-' g./cm.2/sec. at 900" and 0.1 atm. of carbon dioxide. Using activation energies of 36.7 kcal./mole for the carbon-oxygen reaction (78) and 84 kcal./mole for the carbon-carbon dioxide reaction (6.2),the ratio of the rates of 800" and 0.1 atm. is calculated to be 6 X lo4. Wicke ($1 ), reacting spectroscopic electrode carbon at 0.1 atm. reactant pressure, gives the following equations for the rates of the carbon-oxygen and carbon-carbon dioxide reactions Ratec-o,

=

2.9 X 109e-58j=41R0T

Iiatec-coz = 2.6 X 108e-85*3/R0T

(13)

(14)

where the units are cc. of gas consumed per sq. cm. of surface per sec. If these rates are to be on the basis of weight of carbon consumed and if carbon monoxide is assumed to be the primary product of the carbonoxygen reaction, the ratio, ratec-ol/ratec.co, , should be multiplied by 2. At 800" and 0.1 atm. pressure, the ratio of the rates is calculated to be 6 X lo5. Armington (6.2) reports the reactivity of graphite "wear-dust" to be 4.9 X lo-" g./cm.2/sec. at 600" and 0.1 atm. of oxygen and 6.2 X lo-'' g./cm.2/sec. at 900" and 0.1 atm. of carbon dioxide. Using his activation energies of 46 and 84 kcal./mole for the carbon-oxygen and carbon-carbon dioxide reactions, the ratio of the rates at 800" and 0.1 atm. is calculated to be 4 x lo4. For a graphitized carbon black (P-33), Armington also reports reaction rates of 1.1 X lo-'' g./cm"/sec. at 600' and 0.1 atm. of g./cm.a/sec. a t 900"and 0.1 atm. of carbon dioxide. oxygen and 7.3 X Using his activation energies of 54 and 89 kcal./mole for the carbon-oxygen and carbon-carbon dioxide reactions, the ratio of the rates a t 800" and 0.1 atm. is Calculated to be 2 X 10'. No data have come to the authors' attention on a direct comparison of the reaction rates for the carbon-oxygen and carbon-steam reactions. To the authors' knowledge, the only data available which can relate the relative rate of the carbon-hydrogen reaction t o the rates of the above gas-

164

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

TABLE I11 Approzimate Relalive Rates of the Gas-Carbon Reactions at 800" and 0.1 Atm. Pressure Reaction

Relative rates

c-02 C-HnO

1 x 106 3 1 3 x 10-3

c-co2 C-HI

carbon reactioiis are that of Goring and co-workers ( 7 1 ) . They present data for the reactivity of a low-temperature char with hydrogen and hydrogen-steam mixtures at 870" and total pressures from 1 to 6 atm. after 10 % carbon burnoff. Extrapolation of the data to 0.1 atm. and 870" gives and 1.3 X lo-' g. carbon gasified per g. carbon in reactor rates of 5.5 X per second for steam and hydrogen, respectively. Using activation energies of 69 and 27 kcal./mole for the carbon-steam (70) and carbon-hydrogen reactions ( 6 9 ) ,the ratio of the rates at 800"and 0.1 atm. is estimated t o be 1 x lo3. Table I11 presents relative rates for the gas-carbon reactions at 800" and 0.1 atm. based on the experimental data discussed. It is to be emphasized that these are approximate, relative rates. However, it is seen that there is a wide variation possible in the rates of gas-carbon reactions depending upon the reacting gas.

V. Role of Mass Transport in Gas-Carbon Reactions A. GENERALREMARKS Heterogeneous reaction rates involving a porous solid and a gas may be controlled by one or more of three major steps: 1. Mass transport of reacting gas and product or products across a relatively stagnant gas film between the exterior surface of the solid and the main gas stream. 2. Mass transport of the reacting gas from the exterior surface to an active site beneath the surface and mass transport of the products in the opposite direction. 3. Chemisorption of reactant, wholly or in part; a rearrangement of chemisorbed species on the surface to a desorbable product (s) ; and desorption of product or products from the surface. It is imperative that anyone attempting t o understand the kinetics of the gas-carbon reactions also understand the role which the above steps (separately or in combination) can play in affecting values determined for orders of reaction, activation energies, and reaction rates. In the field of

165

GAS REACTIONS OF CARBON

catalysis, Thiele (loo),Wheeler (101, 102),Weisz and Prater (I&?), and Frank-Kamenetskii (104) have made major contributions to the understanding of the role which steps 2 and 3 jointly play in affecting the kinetics of reactions. In this section, their quantitative concepts will be used and extended in an attempt to clarify the kinetics of gas-carbon reactions.

B. THREETEMPERATURE ZONESIN GAS-CARBON RE.4CTIONS Ideally, the variation of reaction rate with temperature for gas-carbon reactions can be divided into three main zones, as shown in Fig. 5 and as previously discussed by Wicke ( 3 1 ) and Rossberg and Wicke (82). In the low-temperature zone, Zone I, the reaction rate is controlled solely by the chemical reactivity of the solid (step 3). The measured or apparent activa-

T

<< I

Ea- 0

')=I

E,= 112 E,

-I T

E,= E,

FIG.5. Ideally, the three zones representing the change of reaction rate of a porous carbon with temperature.

166

P. L. WALKER,

JR.,

FRANK RUSINKO, JR., AND L.

a.

AUSTIN

tion energy, E’. , is equal to the true activation energy, E t . Furthermore v, which is defined as the ratio of the experimental reaction rate to the reaction rate which would be found if the gas concentration were equal to C , throughout the interior of the sample, virtually equals 1. (Obviously, there must be some concentration gradient of reactant through the sample, even in Zone I; but it is so small that a concentration of C, can be assumed.) In the intermediate-temperature zone, Zone 11, the concentration of the reactant species goes t o zero a t a distance from the exterior surface less than the radius R . The reaction rate is controlled jointly by steps 2 and 3. Wheeler (101) and Weisz and Prater (108) have shown that the apparent activation energy is one-half of the true activation energy in this zone. Further, q is less than one-half. In the high-temperature zone, Zone 111, the concentration of the reactant species goes to a small value at the exterior surface of the solid. (This does not necessarily mean that reaction penetration into the porous carbon is zero.) The reaction rate is controlled by step 1. Increasing temperature affects the reaction rate by determining how much additional reactant can reach the exterior surface per unit time. Since bulk mass transport processes have low activation energies, the apparent activation energies for the gas-carbon reactions in Zone I11 are also low. Obviously, q is << 1. Before discussing in more detail the intermediate and high temperature zones under ideal conditions, it is well to emphasize that in practice there are good reasons why the simplified picture presented in Fig. 5 is not necessarily obeyed : 1. The reactant concentration gradient across the stagnant film thickness, 6 , can deviate significantly from zero before the reactant concentration goes to zero in the solid. This results in the disappearance of Zone I1 and a longer transition region from Zones I t o 111. This situation is most likely t o occur with low gas flow rates past the sample ( 6 becomes larger) and with small particle sized samples, where the external-surface-area-tovolume ratio becomes large and the possibility of the reactant concentration going to zero in the particle becomes less. 2. The rate controlling part of step 3 (the chemical step) can change with increasing temperature. If, for example, this rate-controlling part of step 3 changes from desorption in Zone I to adsorption in Zone 11, the true activation energy for the over-all reaction will have changed. The apparent activation energy in Zone I1 will correspond t o one-half the true activation energy in this zone, which will be different from one-half the true activation energy for Zone I. 3. I n Zone I, the concentration of products within the porous solid is negligible and reaction retardation is likewise negligible. In Zone 11, the concentration of products within the solid becomes comparable with that

167

GAS REACTIONS OF CARBON

of the reactant, and reaction retardation can become significant. The true activation energy for the over-all reaction rate then becomes a complex mixture of activation energies for different rate constants, as discussed in Sec. I11 and IV.

C. GENERALDISCUSSION OF ZONE I1 FOR GAS-CARBON REACTIONS

THE

1. Relation between True Activation Energy and Apparent Activation Energy Found in Zone ZZ. It has been shown (101, 103) that the rate of reaction in the diffusion controlled zone is given by

where dw/dt is the rate of react,ion per unit area of exterior surface, C Ris the reactant gas concentration at the exterior surface of the reacting specimen, k , is the specific rate constant per unit volume, m is the order of reaction, and D,rr is the effective diffusion coefficient through the material. It is of considerable importance to be aware of the assumptions made in this derivation, especially when applied t o porous carbons, which have a complicated internal pore structure. It is assumed that all of the pore surface area at a given penetration corresponding to a gas concentration C, is available for reaction at the concentration C, . Also, the derivation implicitly assumes that the penetration of gas into porous carbons takes place along a series of pores of varying dimensions and shapes (106),each pore joining into other pores, thus providing a tortuous path to the interior. Statistically, it is assumed that the gas concentration at any depth of penetration into the specimen is constant over the specimen; that is, the gas concentration profile is the same in each series of pores reaching the center of the sample. This will be true if the pores are interconnected at relatively short distances. As discussed in Sec. VI, there is some experimental evidence that this assumption is not justified. The general Equation (15) still holds when applied to any part of the reaction occurring in pores of constant effective diffusion coefficient; but if a unit of exterior surface area is composed of elements of area dA, in which is found a range of effective diffusion coefficients, determined by the pore size in element d A , then Equation (15) becomes

Experimental evidence, presented in Sec. VI and elsewhere (31,106),suggests that after a relatively small burnoff (ca. 5 %) the surface area available for reaction and the over-all reaction rate remains virtually constant over a

168

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

considerable burnoff range. Further, as discussed in Sec. VI, the specific surface area a t any point in the carbon which has undergone the initial burnoff does not vary greatly with reaction temperature. These are additional assumptions made in the derivation of Equation (15), in the case of gas-carbon reactions. Assuming the change in effective diffusion coefficient with temperature to be small compared with the change of specific rate constant with temperature, Equation (15) may be re-expressed as

Thus, the increase in the over-all reaction rate is proportional to the square root of the specific or true rate constant. Since the apparent activation energy for the reaction is defined by

and the true activation energy by k = (constant)

e-"lRuT

(19)

and since dw/dt is proportional to
where E, = E , / 2 . The physical meaning of Equation (17) is simply this: if the specific rate constant, goes up, say nine times, because of an increase of temperature, the concentration profile must be steeper in order to diffuse in the extra amount of reactant gas. Consequently, the penetration into the carbon decreases. Obviously, equilibrium is reached when the concentration gradient increases threefold, the penetration distance decreases threefold, and the over-all reaction rate (proportional to k times penetration distance) increases threefold, where 3, in this example, is the square root of the factor of specificrate-constant increase. For these conditions, the over-all reaction rate has increased threefold, but so has the diffusion gradient ; that is, equilibrium has been reached. 2. Criteria for Ihe Prediction of Gas-Carbon Reactions Entering Zone II. Thiele (100) has derived equations to predict under what conditions plane or spherical specimens undergoing reaction will enter Zone 11.Aris (107) has discussed the effect of specimen shape on the Thiele equations. The au-

GAS REACTIONS OF CARBON

169

thors, using an approach similar to Thiele, derive an equation to be used to determine when cylindrical (rod) specimens undergoing reaction will enter Zone 11. Equations pertaining t o all three geometric shapes are reviewed and implications of the equations discussed. Let the reaction be first order, and assume that the specific rate constant and effective diffusion coefficient are constant throughout the rod. It can easily be shown that, for a cylindrical specimen, the differential equation to be solved is

d2C dr2

- + - -r1+dC dr

- 2 r C = O

where C is the reactant gas concentration at a penetration r from the center axis of the specimen. This equation can be solved using Bessel functions (as shown in the Appendix). The over-all rate of reaction per unit area of external surface of the rod, dw/dt, is given by

Therefore, in Zone 11, when 4

> 4,(4 = R d k , , / D e f f )

dw -= dt

C R d m i

and q = 2/4

(24)

where q is the Thiele utilization factor defined as the ratio of the actual rate of reaction t o that which would occur if the reacting gas concentration were uniform throughout the material. The criteria used for the prediction of gas-carbon reactions entering Zone 11, for first order reactions, are presented in Table IV, with the results of Thiele (100) for plane and spherical specimens included. Zone I1 is entered when 4 > where dIIis the value of 4 for the start of Zone I1 and is 2, 4,or 6 for a plane, cylinder, or sphere, respectively. I n all cases, the specimens approach uniform internal reaction, that is chemical control, when 4 is I#Q~. The rate of reaction per unit external surface area in Zone I1 is given by

C R d K D Z or C I t d i i T i , irrespective of the geometric shape of the sample.

170

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

TABLE IV Criteria for the Prediction of Gas-Carbon Reactions Entering Zone Varioue Geometrically Shaped Samples

Geometric shape

4

for

Rate of reaction v for 4 per unit area of > A 4'1 exterior surface Rate of reaction per unit area of exterior surface when 4 > concentration

1

sample

I

Plane: thickness R Cylinder: radius R Sphere: radius R

2. The transition region between Zones I and I1 will occur over a temperature range sufficient to increase 4 by ca. 20. Since 4 = RZ/k,/Derrand

where n = 1 for a plane = 2 for a

cylinder

= 3 for a sphere

Thus, 4'7 a dw/dt, and since over the transition region t#~ increases by ca. 20 while 7 goes from 1 to the over-all reaction rate, d w l d t , must increase by ca. 200 over the transition range. This implies, as discussed by Weisz and Prater (IOS),that the transition region between Zones I and I1 can cover a considerable temperature range. 3. Equation (25) can be used to determine whether a reaction is proceeding in Zone I or 11. Since Zone I is closely approximated when4 < 0.3 and 7 is close t o 1, if 4 ' is ~ 2n2,the reaction is in Zone 11.

s,

* This is a more stringent requirement than that given by Weisz and Prater (103) who give 4% < 1.0 in the chemical control zone. Weisz (108) later gives the safe limit of Zone I as 441 < 0.8 for 8 2 0.95 (taking into account uncertainty in order of reaction).

GAS REACTIONS OF CARBON

171

4. If the order of the reaction is m, then the formulas given in Table IV are modified as follows:

4 = R d(kvCF1)/Deff

(26)

and dw/dt is given by Equation (15).

D. COMPREHENSIVE RATE EQUATIONS COVERING THREE TEMPERATURE ZONES IN GAS-CARBON REACTIONS When a solid is reacting with a gas stream flowing over its surface and the reaction rate is dependent on the partial pressure of the reacting gas, the over-all picture of the process of reaction may be represented as shown in Fig. 6. The general over-all rate and mass transfer relations can be expressed as follows:

and

where dwldt is the rate of reaction per unit of external surface; L)rree is the diffusion coefficient of the reactant through the "stagnant film" of thickness 6 ; k, is the rate constant per unit of reacting surface; S , is the specific internal surface area expressed per unit volume; m is the true order of reaction; n = 1, 2, 3, for a plane, cylinder, or sphere, respectively; and f is the roughness factor for the external surface. The first term on the right-hand side of Equation (28) represents reaction occurring within the solid, while the second term represents reaction occurring on the exposed external face. Since carbons have internal surface IA

I

0 Z'I)

09

GAS STREAM WITH CONCENTRATION OF REACTING GAS' Cg

."STAGNANT FILM" OF EFFECTIVE THICKNESS 6

0

I

REACTING SOLID OF DEPTH R

FIG.6. Illustration of general case of gas-solid reaction.

172

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

areas of the order of a t least several m?/cc., while the second term involves an area of only a few cm.‘/cc. (unless the solid is in a very finely divided form), the second term can be neglected except a t very high rates of reaction and almost zero penetration into the carbon. If the effect of volume change within the reacting specimen is small, the following formulas represent q to a sufficiently close approximation : Plane,

v=-

tanh qi 9

9 =

Bessel function of

Rod, q5

(see Appendix)

(30)

Sphere,

Clearly, the elimination of the unknown concentration C R between Equations (27), (28), and (29-31) is difficult. However, since the effective diffusion coefficient within the pores of carbon is considerably smaller than the free diffusion coefficient in the stagnant film (109) and fiince the thickness of the stagnant film is usually much smaller than R, it can be assumed that for large specimens the reaction in the solid will be mainly in Zone I1 before ( C , - C n ) becomes appreciable. Therefore, at low rates of reaction

-

dw -= dt

n

k,S,Crq

where 1 1 in Zone I and is given by Equations (29-31) in the transition region between Zones I and 11. When the reaction is in Zone 11, = n/qi and Equations (27) and (28) can be expressed as

or

Eliminating C n from the three terms in Equation (34) gives

GAS REACTIONS O F CARBON

173

At thc high temperatures required to enter the stagnant film-controlled zone (Zone 111),many reactions will tend to first order. Therefore, substituting m = 1 in Equation (35) and rearranging,

When reaction occurs a t an appreciable penetration into the solid, ksf is negligible compared with d]c8svDeti and

However, for very high rates of reaction, dk8svDeit is negligible compared with 1cj and

dw - -dt

CO . 6

-k s+f - Dime 1

Equation (38) will also apply when the carbon is nonporous, that is, Deli = 0. As k, becomes very large, Equations (37) and (38) will give

dw dt

-

CoDiree 6

(39)

Equation (39) represents the reaction rate in Zone 111. The reaction is clearly first order with respect to the reactant concentration in the main gas stream. This is clearly shown by Day (24) for the carbon-oxygen reaction, as shown in Fig. 7. Depending on the specific surface area of the carbon and the effective diffusion coefficient of the reactant through the carbon, it is not necessary for the reaction t o be represented by Equation (37) goingto Equation (36), Equation (36) going t o Equation (38), and Equation (38) going t o Equation (39) as the rate of reaction increases. In some cases, Equation (37) goes directly t o Equation (39) without reaction on the exterior surface area becoming an appreciable rate controlling factor.

E. R.4TES OF GAS-CARBON REACTIONS IN ZONE111 Using heat transfer data, Rice (110) shows that the film thickness of a fluid flowing over an object can be expressed as

174

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. I

I

=

I

1500

AUSTIN

I

OXYGEN CONCCNTRATION OF BLOW 0 9b.SX OXVpEN A 92.5 Y 0 04.0 X

0 6%0K

01

a.

I

I600

A

63.0%

0

12.OX 36.7 X

1

.. '

I I I I I 2000 2100 I700 I800 1900 TEMPERATURE OF CARBON SURFACE, O K .

I

2200

I

2300

FIG.7 . Illuatratiori of first-order kinetics for the carbon-oxygen reaction in Zone 111. [After R. J . Day, Ph.D. Thesis, The Pennsylvania State University, 1949.1

where p and p are the viscosity and density of the fluid, V is the linear flow velocity of fluid over the surface, and R is the radius of the solid. Using C , a T-'.", this relation and the additional relations, p a 1'.', p a Ti,', and Dlreea Ti.76 , Equation (39) can be expressed us

d_W dt

0:

(9)""

which states that the reaction rate is predicted to be independent of temperature. Actually, there is some doubt as to the variation of viscosity and diffusivity with temperature; but in any case, the reaction rate in Zone I11 varies only slightly with temperature. Many workers have attempted to confirm the variation of reaction rate for the carbon-oxygen reaction with linear gas flow rate, as expressed by Equation (41). Parker and Hottel ( I l l ) , reacting brush carbon with air at 1227O, find the rate varies with the 0.37 power of velocity. Mayers (112), using 40- by 60-mesh coke in l-in. high beds, obtains a value of 0.5 for the exponent; Chukhanov and Karahavina ( l l S ) ,in their high-velocity experiments using beds of particles 3 by 5.5 mm. in diameter, find a value of 0.4; Kuchta and co-workers ( l l d ) , using carbon rods, report an exponent of 0.45; Day (24),using carbon and graphite rods, reports a value of 0.5; and Tu et al. (116) report a value of 0.49. Graham et al. ( I I 6 ) , studying the variation in reaction rate of the carbon-steam reaction under high ve-

175

GAS REACTIONS OF CARBON

5.0

I

I I I VELOClllES CALCULATED AT 24.CAND I A T M

I

1

6ODOO FTIYIN.

I 0.50

I

0.55 IOOO/T, OK:'

I

0.60

I 0.65

I a70

FIG.8. Arrhenius plots for the carbon-oxygenreaction at different linear gas velocities in Zone 111. [After R. J. Day, Ph.D. Thesis, The Pennsylvania State University, 1949.1

locity conditions, find that the rate varies with the power of velocity ranging from 0.23 to 0.33. They conclude that when the power is less than 0.5, the reaction is not in Zone I11 but is in the transition region. Day ( 2 4 ) , who apparently is completely in Zone I11 for his studies on the carbon-oxygen reaction, confirms the small dependence of reaction rate on temperature, as shown in Fig. 8. Between 1227 and 2027", the activation energy is less than 8 kcal./mole at all flow velocities used. For a particular gas-carbon reaction, Equation (39), with one reservation, leads to the conclusion that under identical reaction conditions (i.e., C, , Dfree,and 6 are constant), the rate of reaction in Zone I11 is independent of the type of carbon reacted. The reservation is that in the carbon-oxygen reaction, the nature of the carbon may affect the CO-CO2 ratio leaving the surface and hence the reaction rate per unit of oxygen diffusing to the surface. Unfortunately, little data are available on reactivities of different carbons where the reaction has been conducted completely in Zonc 111. Day ( 2 4 ) reports that the reaction rates of petroleum coke, graphitized lampblack, and graphitized anthracite rods agree within 12% at a temperature of 1827" and at a constant gas velocity.

176

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

For reaction at the same temperature, it is of interest to predict the relative rates of the different gas-carbon reactions in Zone 111, when using a sample of fixed dimensions, a constant linear gas velocity, and a fixed concentration of reacting gas in the main stream.* Under these conditions, Equation (39) can be expressed as

where n = 1 for the carbon-carbon dioxide and carbon-steam reactions and n = 2 for the carbon-oxygen reaction. The relative value of n is based on the assumption that at the high temperatures encountered in Zone 111, the CO-CO2 primary product ratio for the carbon-oxygen reaction becomes large (SO). Consequently, each molecule of oxygen reaching the surface will result in the gasification of ca. two carbon atoms, whereas each molecule of carbon dioxide or steam reaching the surface will result in the gasification of one carbon atom. To simplify the calculation of relative values of Dfrep , p, and p for the gas in the stagnant film, the following average gas compositions in the film are assumed : C-02 : 34% Oz,66% CO C-COI : 34% C02,66% CO C-Hz0: 34% Hz0,33% H2 , 33% CO

Relative diffusivities for the mixtures are calculated assuming

Dfreea M+"' Relative viscosities are calculated from viscosities for the individual components at 0' ( l l 7 ) , weighting them on a mole fraction basis. The change in diffusivities and viscosities with temperature and pressure is assumed to be independent of gas mixture. If desired, more accurate calculations of diffusivities and viscosities of gas mixtures can be made using the upproaches of Wilke (118) and Bromley and Wilke (119), respectively. Table V presents relative values for Dfree,p , and p across the stagnant film for the gas-carbon reactions. Substituting these values in Equation (42), the relative reaction rates in Zone I11 for the gas-carbon reactions are calculated and also presented in Table V. Qualitatively, the rates of the carbonoxygen and carbon-steam reactions are predicted to be about twice the rate

+

* The reaction C 2H2 + CH, is not included in this consideration because, as discussed in Sec. 11, at high temperatures and atmospheric pressure, equilibrium greatly restricts this gasification reaction. That is, CR never approaches zero and, to the contrary, approaches C, closely. This means that the concentration gradient across the stagnant film is small and dw/dt is correspondingly small.

177

GAS REACTIONS OF CARBON

TABLE V Predicted Relative Rates of Carbon Gasification in Reaction Zone 111for Similar Shapes of Carbon Specimens and Constant Linear Gas Flow Rate

Reaction

c-0, c-co, C-H20

Relative physical data across stagnant film

DfW

B

1 0.9 1.9

1 0.9 0.6

P

Relative reaction rate in Zone I11

1 1.2 0.6

2 .o 1.o 1.Y

of the carbon-carbon dioxide reaction. The rate of the carbon-oxygen reaction is high because of the removal of ca. two carbon atoms from the surface for each molecule of reacting gas. The rate of the carbon-steam reaction is high because of the relatively high dzusivity value for the steam molecule across the stagnant film. Figure 9 graphically shows the marked effect which temperature level is

t

w

I4

a

z 0

I-

V

4

w

a W

s

A

T FIQ.9. Ideally, the predicted variation in the relative rates of the carbon-oxygen and carbon-steam reactions with temperature for a porous carbon.

178

P. L. WALKER, JR., FRANK RUSINKO, JR.,

AND L.

a.

AUSTIN

expected to have on the relative rates of the carbon-oxygen and carbonsteam reactions. At low temperatures, in Zone I, as discussed in Sec. IV, the carbon-oxygen reaction is many times more rapid than the carbonsteam reaction. Because of the higher true activation energy of the carbonsteam reaction and the higher temperature at which this reaction enters the comparable temperature zones, this difference in reaction rates rapidly decreases. Finally in line with the prediction presented in Table V, the reaction rates for these two reactions should be quite comparable in Zone 111.

VI. Use of Density and Area Profiles on Reacted Carbon Rods for Better Understanding of Gas-Carbon Reactions A. INTRODUCTION The availability of data on the change in physical structure of carbons after different degrees of burnoff at different temperatures can aid in the understanding of gas-carbon reactions. In the broadest sense, use of profile data after fractional burnoff enables a clear determination to be made of the temperature zone in which the reaction has occurred, as follows: 1. If the density profile is uniform through the sample, the reaction occurred in Zone I. 2. If the density at some depth into the sample equals the starting density, the reaction occurred in Zone I1 or 111. Petersen (87, I&)) discusses the use of profile data to understand better the mechanism of the carbon-carbon dioxide reaction. He reacted >$-in. diameter rod samples in an apparatus previously described (86). Profile data were determined on the reacted rods as follows: A %-in. hole was drilled through the center of the rod prior to placing it on an ordinary screwcutting engine lathe. Following incremental cuts of approximately 0.25 mm. from the exterior surface, the rod was removed from the lathe and weighed, and its diameter was determined by a micrometer caliper. For each cut, the apparent density of the material removed was calculated from the weight loss and volume of carbon removed. Profile data reported in this section were determined in a similar way, following reaction of spectroscopic carbon rods (National Carbon's L113SP) with carbon dioxide in the apparatus previously described (86). Briefly, the apparatus consisted of a vertical mullite reactor tube l>$-in. i.d. Carbon samples 2 in. long by W in. in diameter with a %-in. hole through their center (the rods weighing ca. 8.8 g.) were suspended in the reactor by connecting them through a %-in. mullite rod to a balance. Reaction at the top and bottom of the carbon rods was minimized by >$in. diameter mullite plates. Following reaction to ca. 11% burnoff (1 g.) at temperatures of 925, 1O00, 1200, and 1305", density and surface area profile data were determined. The area data were determined in a conventional

GAS REACTIONS OF CARBON

179

B.E.T. apparatus ( l a l ) ,using nitrogen as the adsorbate at liquid nitrogen temperatures. Reactivity data were also determined at a number of other temperatures between 900 and 1350°, but subsequent profile data are lacking. The experimental results obtained from the measurement of surface area remaining after each lathe cut can be plotted as cumulative surface area against radius. If S,' is the surface area per cm. of radial distance at radius T , the cumulative surface area is given by

S, =

\

72

S,'dr

(43)

TI

If S, is the specific surface area at T in cm.2/cc., S,' = S , ( r / R ) A , where A is the external surface area of the rod (excluding the ends) in cm.2 and R is the external radius. Therefore,

or

Thus, the specific surface area at any radius in the rod can be estimated from the dimensions of the rod and the slopes of the cumulative surface area curve. In a similar manner, the porosity at any radius in the rod can be estimated from the corresponding slopes of the curve of cumulative weight vs. radius by the equation Pr =

R

where pr is the apparent density of the carbon at T and w eis the cumulative weight. Then

where 8, is the porosity of the carbon at r and p t is the true density of carbon, which, in this case, equals 2.268 g./cc.

PROFILE DATATO DETERMINE RATE OF REACTION B. USE OF DENSITY AT ANY RADIUS IN THE CARBON ROD The most accurate way t o obtain the rate of reaction at any radius in the rod would be to react a series of identical rods under identical conditions to different burnoffs, followed by the clitting of each rod as described. The

180

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

1

1.0 I

porosity a t which particles blow off

at

i

c

I

v)

0

a

8

unreacted parosi t y

0

-!AR

0

I

1

R2

RI

RADIUS WITHIN RODFIG.10. Illustration of the porosity profile through a rod at two times, t~ and t 2 , when reaction is occurring at a constant rate in temperature Zone 11.

rate of reaction, ( d n l d t ) , , at any radius could then be estimated from the changes in porosity with time. This would be a tedious process. In this study, ( d n l d l ) , is determined more simply and probably about as accuratcly by an alternative method. At sufficiently high temperatures, the reaction in t,he rod will proceed so fast that the carbon dioxide concentration will be zero a t some point in the rod (Zone 11).After an initial burnoff, the porosity a t the surface will reach a value at which the carbon no longer has sufficient structural strength to remain attached t o the rod. Carbon then will be lost by particles blowing off in the reacting gas stream. When this point is reached, it is obvious by intuition that the rate of reaction will be constant for a small decrease in external radius; and the profile functions through the rod will be duplicated after a time interval At but moved in a radius AR (equilibrium burning). The condition of the rod a t two different times is illustrated in Fig. 10. Clearly, the over-all rate of reaction per cm.2 of external surface b is given by

b = -AR At

pu

(48)

where pu is the apparent density of the unreacted carbon and b is constant for a small change in external surface area. Considering I cm.' of e x t r r r d rod surface,

but

Therefore,

QAS REACTIONS OF CARBON

181

Determination of ( d n l d t ) , is possible, since ( d 0 / d r ) , can be found from the slope of the 0 us. r plot and b can be found from the experimental reactivity curve. It should be noted that (&/&), is the rate of reaction per cm. thickness of section, whereas the actual rate of reaction in an infinitesimal section of thickness dr is ( d n l d t ) , dr. From profile data t o be discussed shortly, it was found that Zone I1 was approximated only at reaction temperatures of 1305" and higher. The overall rate of reaction curve for this temperature is given in Fig. 11. If it is assumed that the abrupt change in reaction rate after 4-min. reaction time occurs at the onset of equilibrium burning, the measured decrease in external radius of the rod can be assumed to have occurred between 4 and 8 min., and b can be calculated. The value of b is found to agree well with the rate calculated from Fig. 11. It is of interest to note that several workers (99, 116) have assumed AR/At to represent the rate of reaction of a carbon specimen only when the reaction is proceeding entirely on the external surface. The above reasoning shows that AR/At can be a constant and represent the over-all rate of reaction when the reaction is occurring internally and the utilized surface area is far greater than the external surface area. Graham and co-workers (116), studying the carbon-steam reaction under high-velocity conditions,

REACTION TIME, minutes

FIG. 11. Plot of weight loss vs. time for reaction of spectroscopic carbon rod with carbon dioxide at 1305" (Zone 11) to 11% burnoff.

182

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

I-&+

REACTION TIME-

FIG.12. Typical plot of weight loss us. time for reaction of spectroscopic carbon rod with carbon dioxide at temperatures below Zone 11.

determine reaction rates from the change in external sample radius with time, using Equation (48). On the assumption that Equation (48) holds only when reaction takes place solely on the exterior surface, they calculate reaction probabilities, which they acknowledge to be about a thousandfold too high on the basis of other workers’ findings. Since they report that their reactant concentration at the exterior surface of the carbon does not go to zero but only to ca. C,/2, there is no doubt that some internal reaction is occurring. Using the formulas developed in Sec. V to estimate the degree of internal reaction and the true surface area undergoing reaction, reaction probabilities some thousandfold lower are calculated, in agreement with accepted values. At temperatures below Zone 11, equilibrium burning (as illustrated in Fig. 10) obviously is not obtained. It is found, however, that after some burnoff (usually less than 5 %) the reaction rate is essentially constant over a wide burnoff range. A typical reactivity plot is shown in Fig. 12. If it is assumed that the porosity measured at the close of the run is derived from uniform burning over time At, then

where Aer is the increase in porosity above the unreacted porosity at r . This estimation can only be used where the rate of reaction has been constant over most of the reaction time.

C. MASSTRANSPORT AND REACTANT CONCENTRATION PROFILES THROUGH THE ROD From a knowledge of the rates of reaction through the rod and the effective diffusion coefficient at any radius, it is possible t o determine the con-

GAS REACTIONS OF CARBON

183

centration profile through the rod without making any assumptions regarding the order of the reaction or the surface areas taking part in the reaction. Three limiting cases are possible depending on the manner in which mass transport is occurring through the rod. Equations for the three cases for the carbon-carbon dioxide reaction are derived in the Appendix and presented below. 1. Knudsen diffusion only is occurring (Case 1 ) :

where

At high rates of reaction, the reactant concentration in the center of the specimen, Co , approaches zero closely. 2. Bulk diffusion occurring but pores too fine to allow Poiseuille flow (Case 2) :

where Co/CR = L, p = C R

+ CR’ in g. of carbon per cc. of gas, and

[see Appendix for definition of (D:ff)r].CR‘ is the concentration of carbon monoxide at the surface. When the reaction is not in the stagnant filmcontrolled zone, C R = p ; and a t sufficiently high reaction rates, L ‘v 0. Therefore, Equation (53) can be simplified t o

cr= 2CR(1 -

e-F(‘)’CR)

(54)

3. Bulk diffusion occurring with a maximum of Poiseuille flow under conditions where Co ‘v 0 and p = C R (Case 3 ) :

Cr = C R ( eF ( r ) / C R - 1 ) where

(55)

184

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

D. DISCUSSION OF EXPERIMENTAL RESULTS 1. Density and Area Projiles. Figure 13 presents porosity profiles for the original carbon rod and for the carbon rods after reaction to ca. 11% burnoff at four different temperatures. The samples could only be cut down to a radius of ca. 0.35 cm. Attempts a t cutting to a smaller radius resulted in breaking through the thin carbon wall. The porosity point a t ca. 0.25 cni. represents the mean porosity of the carbon remaining after the last cut. At 1305", reaction occurred a t a considerable penetration into the rod, even though equilibrium burning was obtained. Reaction was effectively zero 0

.

6

as0.64-

0.62-

-

0.60

0.50-

A A

- UNAEACTED

- 9 2 5 * C. 0 - 1000 0 -1200 0 - 1305

I, 0

III

4

PIG.13. Porosity profiles through spectroscopic carbon rods before and after ra. 11% burnoff at different temperatures.

~

GAS REACTIONS OF CARBON

185

towards the center of the rod. Extrapolation indicates that the porosity a t the external surface is ca. 0.7 to 0.8. Since the external radius was decreased significantly during reaction, this suggests that the maximum porosity reached a t the surface before carbon particles dropped from the rod ranged from 0.7 to 0.8. In this experiment, carbon deposits were found in the top of the reaction tube. Apparently only about 70 % of the total weight loss a t this temperature was a direct result of carbon gasification. At 1200", no decrease in external radius occurred, with the surface porosity reaching a value of only 0.56. At this temperature, there wasa significant increase in porosity even near the center hole in the rod; consequently, it may be assumed that the carbon dioxide concentration was not zero in this part of the rod. Therefore, reaction was in the transition region between Zones I and 11. The reaction should be in Zone I1 when 4'7 = (R/CRD,rf)dw/dt> 4, as previously discussed. Since R is ca. 0.48 cm., and a t 1200", C Ris 1 X lo-' g. of carbon per cc., dw/dt is 0.22 X lo-' g. of carbon/min./cm.' and the mean Deff (as discussed shortly) is ca. 0.1 cm.'/sec., ~ + ~=q 1.7. Thus, the reaction should be near, but not in, Zone 11, in agreement with the interpretation of the porosity profile. It is seen that at 1000" the reaction is much more uniform through the rod but is still not in the chemical control zone. At this low rate of reaction, it appears that carbon dioxide is diffusing sufficiently rapidly between the inner wall of the carbon rod and the ceramic support rod to maintain : ~ n apprcciable concentration of reactant at the inner exposed surface of the rod. As expected, the minimum porosity (smallest amount of reaction) is found about half-way between the inner and outer radius, that is, at 0.4-cm. radius. Even at 925", the reaction is not uniform through the rod. This is difficult t o explain because the criterion for chemical control indicates that for the reaction rate at this temperature, the reaction should be well within Zone I. It can hardly be ascribed to a temperature gradient within the rod, since heat-transfer calculations show that the gradient through the rod t o supply the necessary heat of reaction at this low reaction rate is negligible ( 8 5 ) .Furthermore, since heat is being supplied to the sample from the outside, a minimum in temperature at an intermediate radius ( t o explain the minimum in reaction rate at a radius of ca. 0.4 cm.) is not conceivable. Possibily, the assumption of a complete interconnection of the pores within carbon rods is not correct. If the interior of the carbon rod was being supplied with reactant gas through both large and small pores which are not greatly interconnected, the nonuniformity of the profile a t 925" could be caused by the reaction still being in Zone I1 in the small pore system. The experimental Deff used to estimate the reaction zone would be determined almost entirely by diffusion through the large pores in the system and would

186

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

be considerably too high to be used t o calculate the temperature zone operative for the small pore system. If it is postulated that the gas-carrying pore system within the rod behaves as a series of pores with effective diffusion radii ranging over a complete distribution from small to large, with effective diffusion coefficients for the smallest radii group of diffusing pores being, perhaps, one-hundredth of that for the largest, then the nonuniform porosity found at low rates of reaction is clearly explained. Much of the work described in the following pages would need to be recalculated using distributions of surface area and porosity with diffusion coefficients and integrating the effects of the systems. It should be emphasized, how-

0 1 0.2 int rrna I radius

I

0.3

I

0.4 RADIUS, cm.

I

0.5

I

0.6 external radius

FIG.14. Specific surface area profiles through spectroscopic carbon rods before and after ca. 11% burnoff at different temperatures.

GAS REACTIONS OF CARBON

187

ever, that it is difficult on the basis of our present understanding of the physical structure of carbon rods (106) to envision anything but an interconnected pore system. Figure 14 presents specific surface area profiles on the same carbon rods on which porosity profile data were determined. Aa expected, the specific surface areas of the samples reacted at 1305 and 1200" decrease markedly as the radius decreases. For the rod reacted a t 1305", it is seen that a negligible increase in porosity at the internal radius results in a 60% increase in specific surface area at the same radius. This can be attributed to a significant amount of closed pore volume being opened up at small burnoffs (122, 123). The additional volume is negligible, but the additional surface area provided by the micropores is comparatively large. Again, looking at the profile for the rod reacted at 1305", it is seen that the specific surface area goes through a maximum at a radius of cu. 0.5 cm. This is in line with the findings of Walker and Raats (106) and Wicke (31) that the specific surface area goes through a maximum as a function of burnoff or sample porosity. The area profiles for the rods reacted at 925 and 1000"are, in general, as expected. They show relatively little variation in area with radius. By cross-plotting the data in Figs. 13 and 14,the relation between the specific surface area and the porosity of the carbon rods after reaction at different temperatures can be presented, as in Fig. 15. It is seen that the surface area developed in the rods is not only a function of the porosity developed but also a function of the reaction temperature. The development

POROSITY,

FIQ.15. Relation between specific surface area and porosity for spectroscopic carbon rods after ca. 11% burnoff at different temperatures.

188

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

01 0.3

I

0.4

I 0.5 POROSITY,

I 0.6

g,

(

7

FIQ.16. Variation of effective diffusion coefficient of carbon dioxide through carbon monoxide at N.T.P. with porosity of spectrosropic rarbon rods. PoroHit y developed by reaction with carbon dioxide at 950".

of an increasing surface area after constant burnoff as the reaction temperature is increased in the range of about 900 to 1200"has been previously discussed (106,124). It has been shown that variation in the over-all specific surface area developed in rods after reaction at different temperatures cannot be attributed only t o variations in porosity, as again shown in Fig. 15. 2. Variation of Doffwith Porosity of Carbon Rods. Before being able to calculate reactant concentrations through the rods at different reaction temperatures, it was necessary to determine experimentally values for Dell in the rods as a function of porosity. It has not been established that Dell is only a function of porosity for a given carbon material and independent of the temperature a t which this porosity is produced, but for simplicity this has been assumed t o be the case. Carbon rods % in. in diameter and in. long were cut from the original rods (the axis of the small rods being perpendicular to the axis of the original rods, since Detf perpendicular to the axis is the value desired) and reacted at 950" to various degrees of burnoff. The samples were then mounted in the diffusion apparatus described by Weisz and Prater (103)and Defffor hydrogen through nitrogen were determined at room temperature.* Deff values for three samples a t each burnoff were determined, and the values agreed within f 3 % at burn-

* The writers are indebted to P. B. Weisz of the Socony-Mobil Laboratories for determining the D.,r data.

GAS REACTIONS OF CARBON

0

0.1

0.3

0.2

0.4

(POROSITY I*

FIG.17. Relation between effective diffusion coefficient of carbon dioxide through carbon monoxide at N.T.P. and square of porosity for spectroscopic carbon rods.

offs below 33%. At higher burnoffs, the agreement between values was within *lo%. The diffusion coefficients, corrected t o carbon dioxide through carbon monoxide by multiplying by m 4 , are presented in Fig. 16 as a function of porosity. At porosities greater than 65 %, the pellets become too fragile t o handle. Contrary t o expectations, it is found that initial, small amounts of burnoff do not greatly increase Deft.This indicates that the marked increase in surface area for small amounts of burnoff occurs primarily by unblocking of pores which are not part of the main system of macropores through which the majority of diffusion is occurring. Apart from the data a t very low burnoffs, it is found that Defris directly proportional t o the square of the porosity, as shown in Pig. 17, or

Deff = AB2

(56)

where A = 0.0'35 crn.'/sec. a t N.T.P. This can be compared with the wellknown formula Deff =

where

Dfree

-

Y

(57)

is the tortuosity factor.* Possibly after a small initial burnoff,

* It is relevant to note that tortuosity defined by Equation (57) is by no means the same as that defined by ( L , / L ) , where L, is the effective tortuous pat,h length

190

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

y = l / e ( y = 1 w h e n 8 = 1) and Deff

=

Dfreeez

(58)

From Equation (58), D f r e e has the value of 0.095 cm.2/sec. at N.T.P., compared to the value of 0.14 cm.'/sec. found from free-diffusion experiments (126).As discussed, the values of Doff were obtained from diffusion of hydrogen through nitrogen converted to carbon dioxide through carbon monoxide by multiplying by m 4 . Since hydrogen is much smaller than nitrogen, carbon dioxide, or carbon monoxide, it is probably more accurate t o correct Deff using the factor

When this conversion factor is used, Dfmecalculated from Equation (58) is 0.147 cm.*/sec., in good agreement with the experimental value. Since porosity-radius curves have been obtained, it is possible t o plot curves of Dell against radius for the reacted rods. T o correct to the temperature of reaction, it is assumed that D,ff is proportional to T',*(128). 3. Reactant Concentration ProJile through Rod during Reaction at 1200". Equations (52), (54), and (55) are used to calculate the reactant concentration profile through the rod during reaction at 1200". Co is initially asbut as will be seen later, its value can be approximated. The sumed 4, most significant conclusion that can be drawn from the concentration data in Table VI is that there is no major difference in the decrease of concentration through the rod for the three different cases, even at high rates of reaction. At low rates of reaction, Equations (52), (54), and (55) all give the same result, since there is little pressure build-up or forced flow in the rod. As would be expected, Cases 2 and 3 require that the concentration gradient be somewhat steeper to diffuse in the required amount of carbon dioxide for reaction. I n Table VI, the concentration through the rods also is expressed as a percentage of the surface concentration. It is probable that the actual mass-transport process is a combination of all three cases. However, since the percentage falloff of concentration in the rod is not much different in the three cases, the results may be used on a and L the measured thickness of the sample. As discussed by Carman ( I d b ) ,it is difficult to justify theoretically values of ( L J L ) which depart much from G.The value of tortuosity defined by Equation (57) must be considered as a correction factor which includes ( L J L ) , but it is also a function of how the various-sized pores in a solid are interconnected. The tortuosity factor equals (L./L) only when the pores available for diffusion are not of widely different size and the interconnections between them are not constrictions. This can best be seen by noting that as the pore interconnections becomc small, Dell tends to zero; therefore, y tends to infinity even though the diffusion path and the porosity do not necessarily change very much.

QAS REACTIONS OF CARBON

191

comparative basis as long as it is remembered that the absolute magnitude of the concentrations is in doubt. It will be noted that the predicted concentrations of carbon dioxide at the surface of the rod ((2,) are 0.35,0.56,and 0.39 X g. of carbon per cc. for Cases 1, 2, and 3, respectively. Under the conditions of the reaction, C, is about 1.0 X lo-*. From Equation (27) (where 6 is calculated using Equation (40) with y5 = 2.0), it is estimated that C, - C R = 0.04 X lod4g. of carbon per cc. Therefore,

CR

N

1.0

x

indicating that the discrepancy between the values of carbon dioxide concentration calculated from concentration profiles and Equation (27 ) cannot be attributed to significant control of the reaction by mass transport resistance across the "stagnant film." A minor part of the difference can be attributed to COnot being zero, which can be seen by extrapolating the original rate-concentration curves t o zero concentration, as discussed shortly. The major part of the discrepancy is almost certainly caused by If the assumed variation between Dell and temperature (D.ff a Delf were to vary with temperature to about the 0.9 power, the correct carbon dioxide concentration at the outside of the rod would be calculated. Unfortunately, no data are available for this relation for these particulai carbon samples. 4. Variation of Reaction Rate with Temperature for Spectroscopic Carbon Reacting with Carbon Dioxide. Figure 18 presents the Arrhenius plot showing the variation in reaction rate with temperature for the spectroscopic carbon reacting with carbon dioxide. At temperatures below 950°, an activation energy of 93 kcal./mole is obtained, the value probably approaching E l reasonably closely (32, 68) . Between 950 and lOOO", there is an abrupt change in apparent activation energy, which might be interpreted as the entire transition region between Zones I and 11. However, this interpretation is not valid. The porosity profile for the carbon reacted at 1200" (Fig. 13) indicates that the reaction is still in the transition region. Further, using Equation (25), the start of Zone I1 is calculated to occur when dwldt is ca. 6 g. of carbon reacting per hour. Although this value is approximate, it is over 50 times the value of dwldt at 1000". Also presented in Fig. 18 is the ideal change in reaction rate of the spectroscopic carbon with temperature, assuming a true activation energy of 93 kcal./mole. Zone I1 should start at a reaction rate of ca. 6 g. of carbon per hour and knowing that 7 'v 0.5 at thestart of Zone 11, the temperature can be approximated. It is of interest to note that the ideal activation energy in Zone 11, 46.5 kcal./mole, is closely approximated by the change in experimental reaction rate with temperature above ca. 1250". It might be expected that the smaller values of experimental reaction

192

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN 100

I

I

low/ T.

I

I

j

OK.-'

FIG. 18. Arrhenius plot for reaction of spectroscopic carbon rods with carbon dioxide at 1 atm. pressure.

rate which are observed in the transition region between Zones I and I1 are a result of forced flow or pressure buildup in the rod opposing the entry of carbon dioxide. However, the concentrations listed in Table VI indicate that these factors cannot account for such a marked discrepancy in reaction rate. It is probable that the buildup of carbon monoxide concentration in the rod at temperatures above 10oO" results in retardation of the reaction.

GAS REACTIONS OF CARBON

193

5 . T r u e and Apparent Order of Reaction. From a knowledge of ( d n l d t ) , through the rod, the over-all rate of reaction can be determined by graphical integration. When this is done for the rod reacted a t 1200", it is found that the integrated rate of reaction in the rod (0.127 g. of carbon reacting per hour per cm.' of external area) agrees well with the total rate of reaction determined from the experimental rate curve (0.131), The corresponding values for the rod reacted at 1305" are 0.30 and 0.41, which indicates that 28 % of the over-all reaction is a result of carbon blowing from the external surface. This agrees well with the extent of mechanical loss of carbon predicted from the 1305" porosity profile (Fig. 13). At any radius r , the rate of reaction per unit area can be calculated from the quotient, ( d n / d t ) , / S , . Consequently, the specific rate of reaction and calculated carbon dioxide concentration (both taken a t the same value of r ) can be plotted to determine the true order of reaction, independent of diffusion control. Figure 19 presents such data for the carbon rod reacted at 1200", assuming the relative concentrations for Case 3 in Table VI to be applicable. From an auxiliary plot similar to Fig. 19, a finite reaction rate a t zero carbon dioxide concentration is found. Since the concentrations of carbon dioxide were calculated assuming COto be zero, it is clear that this reaction rate is due t80a finite Co concentration a t the center of the rod. The actual values of concentmtion at values of r were estimated by extrapolat-

C02 CONCENTRATION,

0. of carbon x 10' CC

FIG.19. Relation between specific reaction rate and carbon dioxide concentration in rod undergoing reaction at 1200".

194

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. a. AUSTIN

TABLE VI Concentrations of Carbon Diozide through Spectroscopic Carbon Rod Reacting at 1600" Baaed on Denaity Profile and Deft Data

C, , (9. of carbon in

0.35 0.40 0.45

0.60 0.55 0.60 R = 0.6225

C, as % of CR

C O * ) / ~x ~ . 104

Radius, cm. Case 1

Case 2

Case 3

Case 1

Case 2

Case 3

0.06 0.08 0.12 0.18 0.24 0.31 0.36

0.09 0.15 0.22 0.31 0.41 0.51 0.56

0.05 0.08 0.13 0.19 0.26 0.35 0.39

14 23 36 57 69 89 100

13 20 32 47 66 88 100

16 26 40 55 73 91 100

ing the rate vs. concentration plot to zero rate, taking the negative intercept on the concentration ordinate as CO, and adding this constant concentration term to the concentrations in Table VI, thereby arriving at Fig. 19. By this method, Cois estimated as 0.14 X lo-' g. of carbon per cc. at 1200". A similar plot for the rod reacted at 1305" is given in Fig. 20. Clearly, the reaction is not first order at either temperature, nor do the data fit a Langb C ) . The data fit an expresmuir expression of the form dn/dt = aC/( 1 sion of the form l/(dn/dt) = ( a / C ) - ( l / b ) , as seen in Fig. 21. Such an expression is consistent with the idea of carbon monoxide inhibition, as discussed below. Figure 22 presents the change of over-all reaction rate with change in partial pressure of carbon dioxide in the main gas stream. Nitrogen was used as the diluent, and the total flow rate was maintained constant. The over-all order of reaction is found to be ca. 0.5 from 950 to 1200". An overall order of reaction of cu. 0.5 close to the start of Zone I1 has been interpreted to mean a true reaction order of zero (70,79).In this case, however, as has been shown in Fig. 19, the true order is not zero at 1200".Therefore, the above reasoning is not valid. An over-all order of 0.5 would be expected (for reactionin Zone 11) if the mechanism of the reaction is represented by

+

as suggested by Ergun (46). Both a and b vary exponentially with temperature, a increasing and b decreasing with increase in temperature. By similar reasoning to that used to derive Equation (A18), it can be shown that for

GAS REACTIONS OF CARBON

195

o a0eo.w ~ 0 m 6 e aio QIZ a4 a16 0.18 QZO co, CONCENTRATION,W~* cc. carbon ~10' FIG.20. Relation between specific reaction rate and carbon dioxide concentration in rod undergoing reaction a t 1305".

reaction through a specimen,

cco= r ( p - cco,>

(60)

with r varying between 1 and 2 depending upon the pressure buildup which occurs in the material. Assuming I' = 1, for simplicity, cco cco,

-

Go,

- 1

(61)

The carbon dioxide concentration can be expressed as a fraction 'of the exterior gas concentration; that is, Cco,/p = f, with f varying from 1 to 0 from the exterior to the interior of the carbon. For any given value off, Cco/Cc,, is fixed, and therefore, dn/dt is a function off and not of CcOp alone, or

196

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

I

I CO, CONCENTRATION 2 3

I I

-

0-

1200

- 1305

'

['.of carbon x IoI-' cc. 4

5

6

7

c.

0 -

I 0

p.

10 I

COI CONCENTRATION

'

I

20 of carbon x IOJ-' cc.

3

FIG.21. Relation between reciprocal of specific reaction rate and reciprocal of carbon dioxide concentration in rods undergoing reaction at 1200 and 1305".

Consider now a plane specimen of carbon (of uniform specific internal surface area and uniform D,ff) reacted in Zone I1 at two exterior carbon dioxide concentrations p1 and pz . For a given temperature, Equation (62) shows that the specific reaction rate a t a given value of Cco,/p is fixed. Therefore, in the two cases, since Cco,/p covers the same range of values, the specific reaction rates will cover the same range of values. However, in going from one fixed value of Cco,/p to another fixed value, the change in concentration and, therefore, the diffusional mass transport, will not be the samc in both cases even though the specific reaction rate covers the same range of values. Clearly, for the higher concentration case, penetration will occur deeper into the specimen and the given specific reaction rate range will apply over a larger section of carbon. The fall in Ccoz/p through

GAS REACTIONS OF CARBON

n

197

0 - 1000

0.5

0

0

K

- 1200

k z 0

I-

0.1 -

2 0.075LL

0.I Pco,

IN

0.25 Q5 0.75 1.0 GAS FLOW OVER ROD, otm.

FIG.22. Apparent order of reaction for whole rod at reaction temperatures between 950 and 1200".

the material is as illustrated in Fig. 23, where the curves are of the same shape, but the penetration scale at the higher external concentration is expanded uniformly. Consider the two infinitesimal sections AL and SL a t the same Cc,,/p value. Let SL = M L , then from Equation (62): Reaction in volume element SL = S , (dn/dt) SL

(63)

where 8,is the internal surface area per unit volume and the specimen is considered to be of unit cross sectional area. Also Reaction in volume element AL Therefore,

= S,

(dn/dt) AL

=

l/X (reaction in 6L) (64)

198

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

AL 8L R ADlU S -+

0

L

FIQ.23. Illustrations of fractional carbon dioxide concentrations through reacting specimen, when exterior face is exposed to reacting gas of concentration P I or p~ , Pl

>PI.

That is,

Considering the gradient of Cco,/p at the external reacting face

[

d(Cco,lPi)

dx

1 d(C

=x[

"d","

/

) p2 1 2

That is,

Now the overall rate of reaction is equal to Deff

therefore

Eliminating X from Equations (69) and (66) and rearranging

GAS REACTIONS OF CARBON

199

Thus, the over-all rate should be proportional to the gas concentration in the main gas stream t o the half power. The above derivation applies to reaction completely in Zone 11. From Equation (59) it is seen that in Zone I, where Cco is small, the reaction should be zero order. Hence, over the transition region the apparent order of the over-all reaction should range from 0 to 0.5. This is not the case in the present work, as shown in Fig. 21, where the order is approximately 0.5 from 950 t o 1200".The discrepancy may be due to Equation (59) not being the correct equation for small values of carbon monoxide concentration. As was discussed in Sec. 111, the rate of the carbon-carbon dioxide reaction can be expressed as dt

1

+

+ c,, + il cco, '

33

33

which can be written in the form

by substituting for the Cco, Equation (60) with I' = 1. Equation (72) is of the form found to correlate the reaction rate us. concentration data presented in Fig. 21. When the pressure of carbon monoxide becomes appreciable, Equation (71) can take the form

which is of the form of Equation (59). Substituting for Cco Equation (60) with r = 1,

dn

x -e. + -

P

33

ilCC0,

cco,

[($>- (91

(74)

which again can be arranged in a form to satisfy the reaction rate versus concentration correlation presented in Fig. 21. Substituting K = i J j 1 and simplifying

200

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. 0. AUSTIN

-

COO CONCENTRATION WITHIN ROD Fro. 21. EfTcrt.of K [Eq. (75)j on the relation between the specific reaction rate m d carbon dioxide concentration in a rod. K < 1 (order > first) ; K = 1 (order = first) ; K > 1 (order < first).

The order of the reaction through the rod will depend on the value of K with respect t o 1, the various cases being illustrated in Fig. 24. In all ciises the order is approximately first when the carbon dioxide concentration is small. It should be kept in mind that all cases are derived from Equation (73), where the over-all order of reaction in the rod has been shown to be 0.5 when reaction is proceeding in Zone 11. At 1200 and 1305", K is found to be less than 1, whereas Ergun ( 4 6 ) quotes values of 1.8 and 2.4 (see Fig. 4) for reaction at 1200 and 130Oo, respectively.

E. SUMMARY The primary purpose of this section has been to show the possibilities for using density and area profile data to aid in the better understanding of gas-carbon reactions. In order to determine specific reaction rates and carbon dioxide concentrations a t given penetrations, it has been necessiiry to make assuinpt#ionswhich can only be approximations t o the truth. Several major anomalies in the results have been found, however. The calculated concentrations of carbon dioxide a t the external surface of rods rcacted a t 1200 (Table VI) and 1305" are not in agreement with the known carbon dioxide concentrations. Clearly, more information is required on the variation of Deff with temperature and its variation with porosity produced a t different reaction temperatures. It is feasible that a t high temperatures, considerable porosity may be produced without increasing D e f f to such a marked extent as found a t 900". Ariot,her anomaly is the nonuniformity of reaction found at 925O, when it, would be expected that the reaction should be in Zone I. The preliminary assumption of a completely interconnecting pore system may not be valid. It should also be noted that neither the value of K in Equation (75) nor the low-temperature activa-

GAS REACTIONS OF CARBON

201

tion energy of 93 kcal./mole agree with the values found by Ergun ( 4 5 ) . The activation energy value agrees much better with that suggested by Rossberg ( 3 2 ) and those found by Armington ( 6 2 ) .

VII. Some Factors, Other than Mass Transport, Which Affect the Rate of Gas-Carbon Reactions

The hope of attaining a quantitative understanding of the factors affecting the reactivity of carbons to gases has been the stimulus behind much work on gas-carbon reactions. At present, however, there is no clear understanding of why a given carbon reacts at a particular rate with a given gas under a fixed set of operating conditions. In this section, the possible effects on carbon reactivity of crystallite orientation, crystallite size, surface area, impurities in the carbon, heat treatment of the carbon, addition of halogens to the reacting gas, and irradiation are discussed briefly.

A. CRYSTALLITE ORIENTATION I n catalysis, one does not expect the activity of a catalyst t o be proportional t o its surface area, since there is good evidence that in many instarices cat,alytic action is limited to certain active regions which may constitute only a small fraction of the total surface area (129). As would be expected, the same reasoning holds true for gas-carbon reactions. Carbon is a multicrystallirie material, which can present varying degrees of surface heterogeneity depending upon the size and orientation of the crystallites. In the broadest sense, two main orientations of crystallites in the carbon surface need be considered-( 1) crystallites with their basal planes parallel to the surface and (2) crystallites with their basal planes perpendicular to the surface. According to Grisdale ( I S O ) , the rate of oxidation of carbon crystallites is ca. 17 times faster in the direction parallel to the basal planes (along their edges) than perpendicular to them. Therefore, it would be expected that the specific reactivity of a carbon would be at a minimum when its surface contains a maximum of crystallites with their basal planes parallel to the surface. Smith and Polley (131 ) showed this to be the case. They compared the oxidation rates of original and graphitized* (2700”) samples of Sterling FT carbon black, which have very close to the same surface areas (15.4 and 16.6 m.’/g.) and particle sizes (2,094 and 1,940 A. from electron micrographs). Figure 25 shows what they envision the orientation of the crys-

* It is to be emphasized that “graphitized” is used in this section to mean “heated to an elevated temperature above ca. 2200”.” This is in line with the popular usage of the word, and should not be interpreted to mean that after graphitization the carbon has a 100% graphitic structure. As discussed by Walker and Imperial (I%), artificial “graphite” approaches closely but does not have a 100% graphitic structure even after heat treatment to 3600”.

202

P. L. WALKER, JR., FRANK RUBINKO, JR., AND L.

0 RIGINAL CARBON BLACK

(f. AUSTIN

GRAPHITIZED CARBON BLACK

FIG. 25. Arrangement of crystallites in an original and graphitized (2700") particle of Sterling FT carbon black. [After W. R. Smith, and M. H. Polley, J . phy8. Chem.

80, 689 (1956).]

tallites in the two samples t o be. In the original carbon black, there are a number of exposed edges in the surface for reaction t o occur at a relatively high rate. On the other hand, they picture the graphitized carbon black as being in the shape of a polyhedron with its entire surface composed of crystallites with their basal planes parallel to the surface.* Smith and Polley find comparable rates of oxidation for the original and graphitized carbon blacks at temperatures of ca. 600 and BOO", respectively. If an activation is assumed for the oxidation of both carbons, energy of 50 kcal./mole ( 3 ) the ratio of reaction rates at the same temperature is ca. 200. Walker and co-workers (194)investigated the reactivity of a series of graphitized carbon plates to carbon dioxide in the apparatus previously described (86).The majority of the plates were fabricated from mixtures of 65 % petroleum coke (produced by delayed coking) and 35 % coal tar pitch, using standard techniques (136).Using X-ray diffraction, they determined the relative tendency of the different petroleum cokes to orient with their basal planes parallel to the surface of the carbon plates. A qualitative correlation is found showing that the reactivity of the plates decreases as the percentage of basal plane structure in the surface increases. Plates produced from a fluid coke (136, IS?') are found to have a gas reactivity lower than all plates produced from the delayed cokes, which is attributed to the fluid coke particles graphitizing in a manner similar to the Sterling FT carbon black, as previously discussed.

* The authors (131) and Kmetko (133) have confirmed definitely, from electron micrographs, that graphitization of carbon blacks of low surface area produces polyhedral particles.

GAS REACTIONS O F CARBON

203

B. IMPURITIES IN THE CARBON Much work has been done on the effect of the addition of impurities (salts and metals, chiefly) on the reactivity of carbon. Quantitatively, the effects are difficult to understand, since they are functions of the location of the impurity in the carbon matrix and the extent of interaction of the impurity with the matrix. Long and Sykes (94) suggest that impurities affect carbon reactivity by interaction with the r-electrons of the carbon basal plane. This interaction is thought t o change the bond order of surface carbon atoms, which affects the ease with which they can leave the surface with a chemisorbed species. Since the ?r-electrons in carbon are known to have high mobility in the basal plane, it is not necessary that the impurity be adjacent t o the reacting carbon atom. Indeed, it is thought that the presence of the impurity a t any location on the basal plane is sufficient for it t o affect the reaction. Impurities can either accelerate or retard carbon reactivity. Day (138) studied the effect of impurities on the oxidation of acetylene black by mixing equal weights of black and metallic oxides. He finds that a number of the impurities, including boron, titanium, and tungsten, inhibit oxidation, whereas iron, cobalt, nickel, copper, and manganese, among other metals, accelerate oxidation. Perhaps of greater significance is the finding that different methods of adding the impurity can affect markedly the degree of oxidation acceleration or retardation. For example, the addition of nickel originally as the nitrate is more effective than the addition of nickel originally as the hydroxide. Earp and Hill (99) find that the addition of salts t o graphite usually accelerates oxidation markedly; the notable exceptions being most of the borates and phosphates. Sat0 and Akamatu (139) report that alkali metals enhance the chemisorption of oxygen on carbon and weaken the carbon-carbon bonds a t the surface so as t o accelerate combustion. On the other hand, they report that phosphorus, while catalyzing the adsorption of oxygen on carbon, has a retarding effect on the release of the surface oxide. Nebel and Cramer (140) show that the addition of a series of lead compounds t o carbon at a concentration of ca. 5 wt. % lowers the ignition temperature (raises the combustion rate) of the carbon. Of importance is the finding that the extent of the catalytic effect depends on the particular salt. Lead acetate is the most effective, lowering the ignition temperature 293"; lead sulfate is the least effective, lowering the ignition temperature only 39". Lead pyrophosphate and lend orthophosphate are found not to lower the ignition temperature. Tuddenham and Hill (72) investigated the effect of addition of cobalt, iron, nickel, and vanadium to spectroscopic graphite on its gasification with

204

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

steam at 1100". They added the impurities as the nitrate. They report that relative gasification rates increase from 19-fold for nickel to 32-fold for iron. Gulbransen and Andrew (141) investigated the effect of iron on the reactivity of spectroscopic graphite to carbon dioxide. The porous graphite was impregnated with an iron nitrate solution and then heated to a relatively low temperature t o convert the iron nitrate to iron oxide. In one run, mm. they then pretreated the sample by holding it at 850" for 1 hr. at Hg, prior to reacting the sample at 700" in 76 mm. Hg carbon dioxide pressure. Over the 10 min. of reaction time, the impregnated sample (contnining 0.078% iron) is reported t o have a reaction ratc 530 times that of the original graphite. In another run, they pretreated under similar vacuum conditions but at a temperature of 700" for 16 hrs. They find negligible change in gasification rate following this pretreatment. Gulbransen and Andrew conclude that the iron impurity must be present as either the re1

-

0 0 A -

0 A

I

ti

E! I

i

0" V

t

B

s 10 a

8

E +

-

I

E

h

I

I

-

ORIGINAL GRAPHITE 1.61 g./cc. APPARENT DENSITY I, I, 1.83 'I II ,I 1.99 'I HEAT TREATED GRAPHITE 1.41g./Cc. APPARENT DE!SITY ,I 1.63 " I1 II 1.76 "

\\-

=

-

-

--

Sa E a v) 10'4

I

0.75

I

0.80 1000 I T ,

I

0.85

I

0.90

0.95

OK.-'

FIQ.26. Arrhenius plots for reaction of raw and heat-treated rods of Ceylon graphite with carbon dioxide at 1 atm. pressure. [After F. Rnsinko, Jr., Ph.D. Thesis, The Pennsylvania State University, 1968.1

GAS REACTIONS OF CARBON

205

duced metal or as the carbide t o catalyze the carbon dioxide reaction. They suggest that their vacuum pretreatment at 700" did not effectively reduce the iron oxides. Rusinko (89)investigated the reactivity of pelletized natural graphite rods, before and after heat treatment at 2600", with carbon dioxide a t a series of temperatures from 800 to 1100".On heat treatment, the ash content is found to decrease from ca. 2 % to less than 0.1 %. The crystallite size and specific surface area are found to undergo negligible change. Figure 2G correlates reactivity data on rods of various densities with temperature using an Arrhenius plot. Heat treatment is seen to reduce the specific reactivity of the rods by a factor of ca. 10 but not to change the activation energy (42 kcal./mole). The implication in this case is that the removal of impurities decreases the number of carbon sites able to participate in the reaction, but the removal does not change the mechanism by which the active sites react. It is known that the activation energy obtained is less than E c , since the specific reactivity of the rods increases as the diameter of the rods reacted is decreased. It appears that reaction is proceeding in the transition region between Zone I and Zone 11.

C. CRYSTALLITE SIZE Usually it is difficult to separate the effect of crystallite size on carbon reactivity from the effects of crystallite orientation and impurity content. However, Armington ( 6 2 ) attempted t o do so by reacting a series of graphitized carbon blacks with oxygen and carbon dioxide, as discussed earlier in this article. Assuming that upon graphitization all the carbon blacks are converted t o polyhedral particles with the surface composed almost completely of basal plane structure, it is possible to eliminate crystallite orientation as a variable. Spectroscopically, the total impurity content of all the graphitized carbon blacks is quite low; and to a first approximation, the analyses of the individual constituents are similar. By selecting carbon blacks of a wide range of particle size, Armington was able t o control the extent of crystallite growth upon graphitization, since crystallites only grow t o a fraction (usually from $6 to > i o ) of the carbon particle size. The graphitized carbon blacks range in crystallite size from ca. 20 t o 130 A. and in particle size from ca. 130 to 2000 A. Particle sizes calculated from B.E.T. surface areas (assuming no internal porosity) agree well with particle sizes approximated from electron micrographs. Figure 27 presents data on the specific reactivity of the series of carbon blacks in 0.1 atm. of oxygen at 600"vs. the specific surface area of the blacks. Since the crystallite size is an inverse function of surface area, the conclusion t o be drawn from Fig. 27 is that the specific reactivity of carbons increases with increase in crystallite size. Armington reports similar results for the

206

P. L. WALKER, JR., FRANK RUSINKO, J R . ,

I

0

50

I

AND L. G . AUSTIN

I

100 150 SURFACE AREA, m 2 4 .

I 200

FIG.27. Relation between specific reactivity to oxygen and specific surface urea of a series of graphitized carbon blacks. [After A . F. Armington, P1i.D. Thesis, The Pennsylvania State University, 1980.1

reaction of the same graphitized carbon blacks with carbon dioxide. He suggests that catalysis of the reaction by the impurities still present in the blacks is responsible for this effect of crystallite size on reactivity. That is, assuming the same quantitative and qualitiative impurity content in all blacks, the larger the crystallite size the greater the number of edge carbon atoms which can be affected by a given impurity atom by ?r-electron transfer through the basal plane. Edges of crystallites will serve as zones of high resistance to electron flow. Consequently, an impurity atom associated with one crystallite will have little effect on the reaction rate of edge carbon atoms on other crystallites in the matrix.

D. EFFECT OF HEAT TREATMENT OF CARBONS ON THEIRSUBSEQUENT TO GASES REACTIVITY

Several cases of the effect of heat treatment on the subsequent reactivity of carbon have already been discussed. In both the work of Rusinko (89)

207

GAS REACTIONS OF CARBON

2.01

I

I

0

2

4

I

I

6 8 REACTION TIME, hours

I

10

I2

FIG.28. Effect of heat treatment on the reactivity of carbon derived from petroleum pitch. Reaction of 2 g. of 6 X 8-mesh carbon with carbon dioxide at 1100". [After P. L. Walker, Jr., and J. R . Nichols, "Industrial Carbon and Graphite," Society of Chemical Industry, p. 334. London, 1957.1

and Smith and Polley (lSI),heat treatment at elevated temperatures produces a marked decrease in reactivity of the carbon. It is to be emphasized, however, that heat treatment to elevated temperatures also can increase the subsequent reactivity of carbon. Walker and Nichols (142) investigated the reactivity of cokes produced from coal tar pitch and petroleum pitch. Particle samples (2 g. of 6 X 8-mesh material) having seen maximum temperatures of either 1100 or ca. 2750" were reacted with carbon dioxide at 1100" in the apparatus previously described (86). Figure 28 presents the reaction rate curves for the samples derived from the petroleum pitch. The graphitized sample has a reaction rate some fivefold higher than the sample which has not seen a temperature above 1100". Similar results are found for the samples produced from the coal tar pitch with the graphitized sample having a reactivity over threefold higher than the ungraphitized sample. For both materials, graphitization produced a marked increase in crystallite size, a marked decrease in impurity content, and only a minor change in surface area. As a follow-up to this work, Walker and Baumbach (143) investigated the effect of heat treatment on the reactivities of carbons produced from 20 different coal tar pitches and one delayed petroleum coke. Heat treatment again produced a marked increase in crystallite size, a marked decrease in impurity content, and only a minor change in surface area. They use the

208

P. L. WALKER, JR., PRANK RUSINKO, J I ~ . ,AND L. G. AUSTIN

1 000

1100

I 2000 HEAT TREATMENT TEMPERATURE, C.

I 2500

FIQ.29. Effect of heat treatment on the reactivity of carbons derived from coal tar pitch and delayed petroleum coke. Reaction with carbon dioxide at 1150'. [After P. L. Walker, Jr., and D. 0.Baumbach, unpublished results 1969.1

same apparatus and procedure as that used by Walker and Nichols ( I @ ) , while studying reactivities in carbon dioxide at 1150". Of the 20 samples derived from coal tar pitch, in 19 cases the graphitized sample (2660') has a considerably higher reactivity than the samples which have seen a maximum temperature of only 1150'. On the other hand, the reactivity of the graphitized petroleum coke is about one-half that of the coke having fieen a maximum temperature of 1150". Of even more interest is the effect of heat treatment to different elevated maximum temperatures on subsequent reactivity to carbon dioxide. In Fig. 29, results on a typical sample produced from coal tar pitch and a sample produced from delayed petroleum coke are given. Pronounced effects of graphitization temperatures in the range 2570 to 2680" are found. As noted, two separate heat treatment runs at temperatures of ca. 2655" were made on the sample from coal tar pitch to confirm the maximum in the reactivity. There is no doubt that the maximum exists. The relative values of temperatures reported agree well with the temperatures estimated from electrical resistivity data on the heattJreatedsamples. That is, room temperature electrical resistivities of carbons heated in this temperature range are known to increase with increasing heat treatment temperature.

GAS REACTIONS O F CARBON

209

The authors feel that these preliminary results of Walker and Baumbach on the effect of heat treatment of carbon on subsequent gas reactivity serve t o indicate the complexity of the problem. At the same time, the results indicate the necessity of much additional work in this area if an understanding of the factors decting the rate of gas-carbon reactions are t o be understood. These results emphasize that total impurity content in carbons is not the decisive factor determining gas reactivities. Of more importance is the location of the impurity in the carbon matrix and its particular chemical form. It is suggested that heat treatment can bring the impurity into more intimate contact with the carbon matrix through high-temperature reactions so that a small amount of impurity can serve as a more efficient catalyst. Also t o be kept in mind is the fact that the crystallite size of the carbon can increase with increasing temperature-at least up to a point. As discussed previously, the size of the crystallite determines, in part, how effectively the catalytic impurities are used. I t is suggested that a detailed cxaminat,ion of the effect of heat-treatment temperature on the gas reactivity of the carbons studied by Walker and Bauinbach (143) might show a series of reactivity maxima which correspond t o temperatures a t which different catalytic impurities first begin t o show sigiiificant solid state diffusion and reaction with the carbon matrix followed a t higher temperatures by their complete volatilization from the sample. The advent of significmt diffusion and reaction of the impurity with the carbon could result in a subsequent increase in gas reactivity. Complete volatilization of the impurity from the sample could result in a subsequent decrease in gas reactivity.

E. ADDITION OF HALOGENS TO

THE

REACTING GAS

The role which halogens play in raising the CO-CO2 ratio of the product gas in the carbon-oxygen reaction has been discussed in Sec. 111. Halogens can also affect markedly the rate of carbon burnoff. Day (24), for example, investigated the effect of chlorine on the carbon-oxygen reaction under high velocity conditions. The carbon was heated solely by the energy supplied by the reaction, and a t 20,000 ft./min. in pure oxygen, a surface temperature of lGG0" was maintained. The introduction of 0.15% chlorine t o the oxygen stream lowers the surface temperature by 280"; 0.25 % chlorine immediately extinguishes the reaction. The chlorine is thought to be dissociating and chemisorbing on the carbon sites preventing the formation of a carbon-oxygen complex. If the chlorine has not extinguished the reaction, subsequent removal of the chlorine from the oxygen stream results in the surface returning t o its original temperature. However, the return t o normal does not occur as rapidly as the poisoning, which is almost instantaneous. Wicke ( 3 1 ) investigated the effect of POCL on the carbon-oxygen reac-

210

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

tion. He finds a normal ignition temperature of 650" in dry air. With 1 % by volume of POC1, added to the air, there is no reaction a t 650"; and it proves impossible to remove the inhibiting material from the carbon by subsequently passing in pure air. Even a t 900",the removal is found to take several minutes. The ignition temperature of the carbon is raised 200". Hedden (144) investigated the effect of the addition of POCla and chlorine on the rate of the carbon-carbon dioxide reaction at a temperature of 1100". After achieving a constant rate for the reaction in the absence of halogen-containing gas, he finds upon addition of impurity gas that there is an initial sharp increase in reaction rate. This is followed by a decrease in reaction rate. For POCL , the rate falls below the normal rate; for chlorine it remains above this rate. When the halogen-containing gas is stopped, the reaction rate in both cases increases sharply above the normal rate, followed by a continuing decrease back to the same value as that when no halogen-containing gas is added. The initial increase in reaction rate following halogen treatment t o a value greater than the normal value is ascribed to excessive surface roughening while the halogen is present in the reacting gas. The degree of surface roughening gradually decreases after the halogen gas flow is stopped until reaching the normal value. It can be concluded that the halogen-containing gases offer unusual possibilities for affecting the rate of attack of carbon surfaces by oxygen-containing gases.

F. IRRADIATION With the use of graphite as a moderator in nuclear reactors becoming of increasing importance, there is concern about the effects of irradiation on the rate of reaction of the graphite with gases. Aside from the practical importance of irradiation effects, high-energy irradiation of carbons provides a powerful tool for studying the relation between imperfections in the carbon lattice and rates of gas-carbon reactions. Relatively large and controlled concentrations of imperfections can be introduced into graphite by high energy particle bombardment, Kosiba and Dienes (146) investigated the effects of neutron irradiation on the rate of reaction of spectroscopic graphite rods with air. Figure 30 shows the effects of exposure of the graphite t o ca. 4 X 10" neutrons/cm.' at temperatures under 50" on its subsequent reaction rate over the temperature range 250 to 450". Prior irradiation increases the oxidation rate by a factor of ca. 5 to 6 at reaction temperatures of 300 to 350". The effect of irradiation decreases with further increase in reaction temperature, as evidenced by a larger activation energy of oxidation for the unirradiated graphite. Kosiba and Dienes estimate that at the reaction temperatures studied there is at most about 1% displaced carbon atoms remaining from the

GAS REACTIONS OF CARBON 470

4 7

3;)o

370

0 0

0.01

1.3

1.4

1.5

1.6

1000/T,

211

,

250 '1

- IRRADIATED - UNIRRADIATED

1.7

'K.-'

1.8

1.9

1

FIG.30. Arrhenius plots for the rate of oxidation in air of both unirradiated and previously irradiated spectroscopic graphite. [After W. L. Kosiba, and G. J. Dienes, Advances in Calalysis 9, 398 (1957).]

previous irradiation with neutrons. On the other hand, they observe that at W " ,for example, the higher oxidation rate of the irradiated sample persists even when 20 to 25% of the sample has been oxidized. They conclude, therefore, that the displaced carbon atoms are not themselves being oxidized preferentially but facilitate in some way the over-all oxidation. They further observe that this increase in reaction rate on irradiation is not brought about by an increase in surface area, since it is known from the recent work of Spalaris (1%') that the surface area decreases significantly upon irradiation at room temperature. Kosiba and Dienes (1.46)also investigated the effect of exposure of the graphite to gamma-irradiation during reaction on oxidation rates. On the

212

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

unirradiated graphite, they find that the reactivity at 300" is hardly a1tered at a gamma-flux of 2 X 10' r./hr., whereas a significant increase in reactivity is observed at a flux of G X 10' r./hr. On an irradiated graphite, a flux of 2 X 10' r./hr, increases the reactivity at 300" by a factor of three over the irradiated graphite not reacted in a gamma-field. This means that the irradiated graphite subsequently reacted in a gamma-flux of 2 X 10' r./hr. at 300" had an over-all oxidation rate some 18-fold higher than the unirradiated graphite whether or not the unirradiated graphite was exposed to the above gamma-flux during reaction. Kosiba and Dienes conclude that the gamma-ray effect is probably due to the ionization of oxygen molecules, since gamma-rays have not been observed to have any effect on the properties of graphite at the exposures used. ACKNOWLEDGMENT We wish to express our appreciation t o the following groups for support of our studiccl on gas-carbon reactions either through direct financial assistance or through the supplying and processing of carbon materials: U.S. Atomic Energy Commission through Contract No. AT(30-1) -1710; The Commoriwealth of Pennsylvania through its continuing support, of r o d research; Consolidation Coal Co. ; Godfrey 1,. Cabot, Inc.; National Carbon Co.; Plastics and Coal Chemicals Division of the Allied Chemical Corp.; Socony Mobil Oil Co.; Speer Carbon Co.; and Stackpole Carbon Co. Their support has made the writing of this article possible.

Appendix A. SOLUTION OF DIFFERENTIAL EQUATION (21) Equation (21) is a Bessel equation of the general solution However, the function Y otends to infinity as r tends to zero, while the function Jo remains finite; and as C must be finite at r = 0, B must be zero. Thus

or

By computation or by using tables of Bessel functions, values of CIA can values. Let be found for a range of (r/2)4Then, by plotting loglo ( C / A) us. 9/2, it was found that for values of + > 4 loglo ( C I A ) = 0.84(+/2)

- 0.75

(A5)

GAS REACTIONS OF CARBON

213

Hence, loglo [5.62(C/A)l

= 2(4/2)

loglo e

(A61

or

5.62(C/A) = e'

(A7)

When C = C R , 4 = R d m i and, therefore,

A = 5.62CRexp - R d m i

(A8)

or

C, = CHexp r

d m

exp - R d m f

(A91

From the plot of log ( C / A ) against 4/2, it is found that C, is approximately constant throughout the rod for 4 < 0.2. (At 4 = 0.2, the concentration at the center of the rod is ca. 20%less than CI1.)

B. DERIVATION OF EQUATIONS FOR REACTANT CONCENTRATION PROFILE THROUQH CARBON RODSDEPENDING UPON TYPE OF MASSTRANSPORT 1. Knudsen Di$usion Only I s Occurring. For a very fine pore material in which the effective pore diameter is less than the mean free path of the molecules, bulk diffusion and Poiseuille flow do not occur. For this case, the change in volume given when C COz 2CO has no influence on the rate of diffusion of carbon dioxide into the rod, and Dpffis not dependent on the total pressure in the pores. Considering a wedge of carbon (Fig. A l ) ,

+

-

I

I I I

I

\ \ \ 'k-

i

I I

UNIT AREA

OF

EXTERNAL SURFACE EXPOSED TO I REACTING GAS AT I CONCENTRATION CR I

/ /'

- 0

FIG.A l . Section of rod of radius R undergoing reaction.

214

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

the amount of C02which diffuses through a plane at r must equal the C02 reacted from 0 to r. That is,

From the values of (dnldt), obtained experimentally,

can be obtained by graphical integration. Then

and integrating,

where C, is the concentration of COZat r in g. of carbon per cc. and COis the concentration of COS in the center of the rod. At high reaction rates, CoN 0. F ( r ) can be determined by graphical integration. 2. Bulk Di$USion Occurring But Pores Too Fine to Allow Poiseuille Flow. For bulk diffusion, Doffa (1/P) ,where P is the total pressure. If Poiseuille Row is negligible, then the concentration profiles of COz and CO through the rod can be shown as in Fig. A2. For the reaction C COZ+ 2C0,the increase in volume of CO over COZis two; and at any point, the diffusion gradient for CO must be double that for C o t . If C' is the concentration of CO at a point where the concentration of COz is C,

+

dC' - = -2- dC dr dr

a

v)

u)

W

RADIUS -e

FIQ. A2. Illustration of the pressure profiles through a rod when substantial bulk diffusion and negligible Poiseuille flow are occurring.

215

GAS REACTIONS OF CARBON

or

C'

+A

-2C

=

At the external surface, let the total pressure P R be made up of q inerts, Cg carbon dioxide and CfRcarbon monoxide. Then P R

= CR

+ +q

(A151

C'R

Therefore, CR

+

= p =

C'R

P R

-q

where p is in g. of carbon per cc. WhenC = CR C' )

p

- CR =

-2cR

c'

+

(A161 =

CIR = p -

+A

C R

and

(A171

or = p

C R

- 2c

(A181

At any point in the carbon, the total pressure P r is given by

Pr

= q

+ C + C'

= q

+p +

C R

-C

(A191

Now the diffusion coefficient Doifat this point under a pressure of P, is obtained from the diffusion coefficient Deli at the same point but measured nt P Rby the relation

If q

= 0)

or

or

If CO/C, = L, where L is clearly

< 1,

216

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G . AUSTIN

t-

A CO, REACTED IN INTERIOR ‘ X - 2 CC. PER SEC. X cc. PER SEC OF CO

DIFFUSE OUT

CC. OF COO PER SEC DIFFUSE IN

X

Z CC. OF

COa PER SEC. SWEPT BACK IN POISEUILLE FLOW

1

X - 2 Z cc PER SEC. OF CO TRANSPORTED POISEUILLE FLOW BY

A

FIG.A3. Representation of flow conditions at a plane in a rod when bulk diffusion and a maximum of Poiseuille flow are occurring.

When the reaction is not in the stagnant film-controlled zone, C R = p and at high rates of reaction, L N 0. Therefore,

C,

=

2cR(1 -

e-P‘r)’CR)

( A25 1

3. Bulk Diflusion Occurring with a Maximum of Poiseuille Flow. The third limiting case is where the pores are so large that negligible absolute pressure differential builds up within the pores and Poiseuille flow carries the extra volume of CO to the exterior. Under these conditions, CO will diffuse out at the same rate as CO, diffuses in (that is, dC/dr = -dC’/dr), while CO is carried out by forced flow. It is clear, however, that the forced flow will also carry out some of the COz which diffuses in. This situation is represented by Pig. A3, where A A is a plane in the solid (after Thiele (100)). The total mass flow outwards in cc./sec. is given as Q = X - 22

+Z = X

-2

( A26 1

But Z = Q X C” where C” equals the concentration of COz in gas in cc. per cc. Therefore,

z

=

( X - Z)C

If

=

c”x

~

1

+ Cff

If 1 cc. of C o nweighs p g. of carbon a t the temperature and pressure applying, then considering 1 om.* of external surface of rod, the rate of reaction in g. of carbon per sec. is given by

i’g

dr = ( X - Z ) p =

(x - - -C”X )p 1

=

( - ) X P1 1

+ C”

+ C”

GAS REACTIONS OF CARBON

217

Now

Theref ore,

As C is in g. of carbon per cc., C” f(r)

=

=

C/p. Substituting from this and

1‘

(dn/dt),dr,

That is,

-dr

=

[‘-”dC 0

P+C

( A33

or (A34) For high rates of reaction, Co ‘v 0. If there is 100 % CO, in the reacting gas stream

c, = C , ( e

P(r)ICR

- 1)

(A35)

REFERENCES 1. Hougen, 0. A., and Watson, I<. M . , “Chemical Process Principles,” Part I, p. 254. Wiley, New York, 1950. 8. Based on data from: “Tables of Selected Values of Chemical Thermodynamic Properties.” National Bureau of Standards, Washington, D. C., 1949. 3. Trapnell, B. M. W. “Chemisorption.” Academic Press, New York, 1955. 4 . Loebenstein, W. V., and Deitz, V . R., J . Phys. Chem. 69, 481 (1955). 6. Savage, R. H., Ann. N . Y . Acad. Sci. 63, 862 (1951). 6. Gadsby, J . , Long, F. J . , Sleightholm, P., and Sykes, K. W., Proc. Roy. SOC. A193, 357 (1948). 7. KeIer, N. P . , and Man’ko, N . M . , Doklady Akad. Nauk S.S.S.R. 83, 713 (1952). 8. Zelinski, J . J . , Ph. D. thesis, The Pennsylvania State University, 1950. 9. Studehaker, M. I,., Huffman, E. W. D., Wolfe, A. C., and Nabors, L. G., Znd. Eng. Chem. 48, 162 (1956). 10. Garten, V. A,, and Weiss, D. lC.,Australian J . Chem. 8 , 68 (1955). 11. Graham, D . , J . Phys. Chem. 69,896 (1955). 18. Anderson, R. B., and Emmett, P. H., J . Phys. Chem. 66, 753 (1952). 13. Carter, R. L., and Greening, W. J., Preprint 319, American Nuclear Society, Nuclear Eng. and Sci. Congress, Dec. 1955.

218

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. Q. AUSTIN

14. Norton, J. F., and Marshall, A. L., Trans. A m . Znsl. Mining Met. Engrs. 166,

351 (1944). 16. Ingram, D. J. E., and Austen, D. E . G., i n “Industrial Carbon and Graphite,”

p. 19. Society of Chemical Industry, Londdn, 1957. 16. Winslow, F. H., Baker, W. O., and Yager, W. A., Proc. 1st and 2nd Conf. on Carbon, Univ. of Bufalo, p. 93 (1956). 17. Hennig, G. R., and Smaller, B., Proc. 1st and 2nd Conf. on Carbon, Univ. of Buflalo, p. 113 (1956). 18. Ingram, D. J. E., and Bennett, J. E., Philos. Mag. 46,545 (1954). 19. Langmuir, I., J. A m . Chem. SOC.37, 1139 (1915). 80. Sihvonen, V.,Trans. Faraday SOC.34, 1062 (1938). df. Meyer, L., Trans. Faraday SOC.34, 1056 (1938). 8.8. Strickland-Constable, R . F., Trans. Faraday SOC.40,333 (1944). 83. Duval, X.,J . chim. phya. 47, 339 (1950). 84. Day, R. J., Ph.D. thesis, The Pennsylvania State University, 1949. 86. Chukhanov, Z.F., Fuel 19, 17,49, 64 (1940). 86. Grodozovskii, M. K., and Chukhanov, Z. F., Fuel 16,321 (1936). 87. Rhead, T. F. E., and Wheeler, R . V., J. Chem. SOC.103, 461 (1913). 88. Arthur, J. R., Nature 167, 732 (1946). 88. Bridger, G. W., Nature 167, 236 (1946). 30. Arthur, J. R., Trans. Faraday Soc. 47, 164 (1951). 31. Wicke, E.,“Fifth Symposium on Combustion,” p. 245. Reinhold, New York,

1955.

3.8. Rossberg, M., Z. Eleklrochem. 60,952 (1956). 33. Lewis, W. K., Gilliland, E. R., and Paxton, R. R., Znd. Eng. Chem. 46. 1327

(1954). 34. Arthur, J. R., Bangham, D. H., and Bowring, J. R., “Third Symposium on

Flame, Combustion, and Explosion Phenomena,” p. 466.Williams and Wilkins, Baltimore, 1949. 36. Walker, 1’. L., Jr., and Wright, C. C., Fuel 31, 45 (1952). 36. Arthur, J. R., and Bowring, J. R., Znd. Eng. Chem. 43,528 (1951). 37. Reif, A. E., J. Phya. Chem. 66, 785 (1952). 38. Marsh, J.D. F., Publication No. 393,The Institution of Gas Engineers, London, 1951, 39. Lewis, W. K., Gilliland, E. R., and McBride, G. T., Jr., Znd. Eng. Chem. 41, 1213 (1949). 40. Wu, P. C., D.Sc. thesis, Mass. Inst. Technol., Cambridge, Mass., 1949. 4 1 . Graham, H. S., D.Sc. thesis, Mass. Inst. Technol., Cambridge, Mass., 1947. 4.8. Semechkova, A. F., and Frank-Kamenetskii, D. A., Acta Physicochim. U.R.S.S. 12, 879 (1940). @. Long, F. J., and Sykes, K. W., Proc. Roy. SOC.A193,377 (1948). 44. Reif, A. E., J. Phys. Chem. 66.778 (1952). 46. Ergun, S.,J. Phys. Chem. 80, 480 (1956). 46. Key, A., Gas Research Board Commun. (London) No. G.R.B. 40 (1948). 47. Strickland-Constable, R. F., J. chim. phye. 47, 356 (1950). 48. Strickland-Constable, R . F., Proc. Roy. SOC.Al89, 1 (1947). @. Paxton, R.R., 8c.D. thesis, Mass. Inst. Technol., Cambridge, Mass., 1949. 60. Bonner, F.,and Turkevich, J., J. A m . Chem. SOC.73, 561 (1951). 6f. Brown, F., Trans. Faraday SOC.48, 1005 (1952). 68. Orning, A. A., and Sterling, E., J. Phya. Chem. 68, 1044 (1954).

GAS REACTIONS OF CARBON

219

63. Gadsby, J., Hinshelwood, C. N., and Sykes, K. W., Proc. Roy. SOC.A187, 129

(1946). 64. Johnstone, H. F., Chen, C. Y., and Scott, D . S., Znd. Eng. Chem. 44,1564 (1952). 66. Jolley, L. J., and Poll, A., J . Znst. Fuel 28,33 (1953). 66. Dolch, P., Brennstofl-Chem. 14. 261 (1933). 67. Neumann, B., Kroger, C., and Fingas, E., Gas- u. Wasserfach 74,565 (1931). 68. Ingles, 0.G., Trans. Faraday SOC.48, 706 (1952). 69. Zielke, C. W., and Gorin, E., Ind. Eng. Chem. 47, 820 (1955). 60. Vulis, L. A., and Vitman, L. A., J. Tech. Phys. (U.S.S.R.)11, 509 (1941). 61. Thring, M. W., and Price, P . H., Iron and Coal Trades Rev. 169, 347 (1954). 62. Armington, A. F., Ph.D. thesis, The Pennsylvania State University, 1960. 63. Vastola, F. J., Ph.D. thesis, The Pennsylvania State University, 1959. 64. Duval, X.,J. chim. phys. 44, 296 (1947). 66. Karzhavina, N.A., Doklady Akad. Nauk S.S.S.R. 73, 971 (1950). 66. Batchelder, H.R., Busche, R. M., and Armstrong, W. P., Znd. Eng. Chem. 46, 1856 (1953). 67. Mayers, M. A., J. A m . Chem. SOC.66, 70 (1934). 68. Pilcher, J. M., Walker, P. L., Jr., and Wright, C. C., Znd. Eng. Chem. 47, 1742 (1955). 69. Key, A., and Cobb, J. W., J. SOC.Chem. Znd. 40, 440T (1930). 70. James, V. E.,Ph.D. thesis, West Virginia University, 1957. 71. Goring, G. E., Curran, G. P., Tarbox, R. P., and Gorin, E., Znd. Eng. Chem. 44, 1057 (1952). 72. Tuddenham, W. M., and Hill, G. R., Ind. Eng. Chem. 47, 2129 (1955). 73. Binford, J. S.,Jr., and Eyring, H., J . Phys. Chem. 60, 486 (1956). 74. Mayers, M. A., Trans. A m . Znst. Mining M e t . Engrs. 119, 304 (1936). 76. Scott, G. S.,and Jones, G. W., U.S. Bur. Mines, Rept. Invest. 3405 (1938). 76. Sihvonen, V., Z.Electrochem. 36, 806 (1930). 77. Chen, M. C.,Christensen, C. J., and Eyring, H., J, Phys. Chem. 69,1146 (1955). 78. Gulbransen, E.A., and Andrew, K. F., Znd. Eng. Chem. 44, 1034 (1952). 79. Blyholder, G.,and Eyring, H., J. Phys. Chem. 61,682 (1957). 80. Gilliland, E.R., and Harriott, P., Znd. Eng. Chem. 46, 2195 (1954). 81. Wicke, E., and Hedden, K., Z. Elektrochem. 67, 636 (1953). 82. Rossberg, M., and Wicke, E., Chem.-1ngr.-Tech.28, 181 (1956). 83. Arthur, J. R., Newitt, E . J., and Raftery, M. M., Nature 178, 157 (1956). 84. Lambert, J. D., Trans. Faraday Soc. 84, 1080 (1938). 86. Walker, P. L., Jr., Foresti, R. J., Jr., and Wright, C. C., Ind. Eng. Chem. 46, 1703 (1953). 86. Walker, P. I,., Jr., and Rusinko, F., Jr., J. Phys. Chem. 69, 241 (1955). 87. Petersen, E. E., and Wright, C. C., Znd. Eng. Chem. 47, 1624 (1955). 88. Walker, P. L., Jr., and Raats, E., J . Phys. Chem. 60, 370 (1956). 89. Rusinko, F., Jr., Ph. D. thesis, The Pennsylvania State University, 1958. 80. Hedden K., private communication, 1957. 91. Mayers, M.A., J . A m . Chem. Soc. 61, 2053 (1939). 92. Hunt, B. E., Mori, S., Katz, S., and Peck, R. E., Znd. Eng. Chem. 46, 677 (1953). 93. Dotson, J. A., Holden, J. H., and Koehler, W. A., Znd. Eng. Chem. 49, 148 (1957). 94. Long, F. J., and Sykes, K. W., J. chim. phys. 47,361 (1950). 96. Lambert, J . D., TTans. Faraday soc. 32,452 (1936). 96. Letort, M., and Magrone, R., J. chim. phys. 47,576 (1950). 97. Golovina, E.S., Izuest, Akad. Nauk S.S.S.R., Otdel, Tekh. Nauk p. 1343 (1949).

220

P. L. WALKER, JR., FRANK RUSINKO, JR., AND L. G. AUSTIN

98. Klibanova, T. M., and Frank-Kamenetskii, D. A., Acta Physicorhim. U . R . S . S . 18, 387 (1943). 99. Earp, F. K., and Hill, M. W., i n “Industrial Carbon and Graphite,” p. 326. Society of Chemical Industry, London, 1957. 100. Thiele, E. W., Ind. Eng. Chem. 31, 916 (1939). 101. Wheeler, A., Advances i n Catalysis 3 , 249-327 (1951). 106. Wheeler, A., i n “Catalysis” (P. H. Emmett, ed.), Vol. 2, pp. 105165. Reinhold, New York, 1955. 103. Weisz, P. B., and Prater, C. D., Advances i n Catalysis 6, 143-196 (1954). 10.4. Frank-Kamenetskii, D. A., “Diffusion and Heat Exchange in Chemical Kinetics.” Princeton University Press, Princeton, N. J., 1955. 106. Petersen, E. E., A m . Inst. Chem. Eng. J . 3, 443 (1957). 106. Walker, P. I,., Jr., and Raats, E., J . Phys. Chevn. 60, 364 (1956). 107. Aris, R., Chevn. Ens. Sci. 6, 262 (1957). 108. Weisz, P. B . , Z . physik. Chem. (Frankfurt) 11, 1 (1957). 109. Walker, 1’. I,., Jr., Rusinko, F.,Jr., and Raats, E., J . Phys. Chem. 69,245 (1955). 110. Rice, C. W., Ind. Eng. Chem. 16, 460 (1924). 111. Parker, A. S., and Hottel, H. C., Ind. Eng. Chem. 28, 1334 (1936). 118. Mayers, M. A . , Trans. A m . Znst. Mining Met. Engrs. 130, 408 (1938). 113. Chukhanov, Z. F., and Karzhavina, N. A., Fuel 20.73 (1941). 114. Kuchta, J . M., Kant, A., and Damon, G. H., Znd. Eng. Chem. 44, 1559 (1952). 116. T u , C. M., Davis, H., and Hottel, H. C., Ind. Eng. Chem. 26, 749 (1934). 116. Graham, J . A., Brown, A. R. G., Hall, A. R., and Watt, W., i n “Industrinl Carbon and Graphite,” p. 309. Society of Chemical Industry, London, 1957. 117. “Handbook of Chemistry and Physics,” 33rd ed., p. 1837. Chemical Rubber Publishing Co., Cleveland, 1951. 118. Wilke, C. R., Chem. Eng. Progr. 46,95 (1950). 119. Bromley, L. A., and Wilke, C. R., Ind. Eng. Chem. 43, 1641 (1951). 160. Petersen, E. E., Ph.D. thesis, The Pennsylvania State University, 1953. 161. Emmett, P. H., A m . SOC.Testing Materials Tech. Publ. 61, 95 (1941). 168. Spalaris, C. N., J. Phys. Chem. 80, 1480 (1956). 19.9. Dresel, 13. M., and Roberts, L. E . J., Nature 171, 170 (1953). 184. Petersen, E. E., Wright, C. C., and Walker, P. L., Jr., Znd. Eng. Chem. 47, 1629 (1955). 166. Carman, P. C., “Flow of Gases Through Porous Media.” Academic Press, New

York, 1956. 186. Kaye, G. W. C., and Laby, T. H., “Physical and Chemical Constants,” p. 37. Longmans, Green, London, 1936. 187. O’hern, H . A., Jr., andMartin, J . , Jr., Ind. Eng. Chem. 47, 2081 (1955). 188. Golovina, E. S., Doklady Akad. Nauk S . S . S . R .86, 141 (1952). 169. Emmett, P. H., ed., “Catalysis,” Vol. I, p. 53. Reinhold, New York, 1954. 130. Grisdale, R . O . , J . A p p l . Phys. 24, 1288 (1953). 131. Smith, W. R., and Polley, M . H., J. Phys. Chem. 60,689 (1956). 1%. Walker, P. L., Jr., and Imperial, G., Nature 180, 1184 (1957). 19.9. Kmetko, E. A., Proc. 1st and 2nd Conf. on Carbon, Univ. of Buflalo, p. 21 (1956). 134. Walker, P. I,., Jr., Rusinko, F., Jr., Rakszawski, J . F., and Liggett, I,. M . , Proc. 3rd Conf. on Carbon, Univ. of Buflalo, p. 643 (1958). 136. Abbott, H. W., “Encyclopedia of Chemical Technology,” Vol. 3, pp. 1-23. Interscience Encyclopedia, Inc., New York, 1949. 136. Johnson, F. B., and Wood, R. E., Petrol. Refiner 33, 157 (1954). 137. Molstedt, B. V., and Moser, J . F., Jr., Znd. Eng. Chem. 60, 21 (1958).

GAS REACTIONS OF CARBON

22 1

138. Day, J. E., Znd. Eng. Chem. 28, 234 (1936). 138. Sato, H., and Akamatu, H., Fuel 33, 195 (1954). 140. Nebel, G.J., and Cramer, P. L., Znd. Eng. Chem. 47, 2393 (1955). 141. Gulbransen, E.A., and Andrew, K. F., Znd. Eng. Chem. 44, 1048 (1952). 14s. Walker, P. L. Jr., and Nichols, J . R., i n “Industrial Carbon and Graphite,” p. 334. Society of Chemical Industry, London, 1957. 14% Walker, P. L., Jr., and Baumbach, D . O.,unpublished results (1959). 1.54. Hedden, K., 87th Intern. Congr. Znd. Chem. 2,68 (1954). 146. Kosiba, W.L., and Dienes, G. J., Advances i n Catalysis 9,398 (1957).

This Page Intentionally Left Blank

The Catalytic Exchange of Hydrocarbons with Deuterium C . KEMBALL Department of Chemistry, The Queen’s University of Belfast, Belfast, N . Ireland Page

I. Introduction.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. General Aspects of Exchange Reactions ............ Kinetics.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Classificatio Possible Me A Theory for Calculating the Initial Distributions of Products in Multiple Exchange Reactions.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. The Exchange of Molecules Possessing a High Degree of Symmetry.. . . . . . A. Methane. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Ethane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................... C. Cyclopentane and Cyclohexane. . . . . . . . . . I). Cycloheptane and Cyclo-octane.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Neopentane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. The Exchange of Molecules Possessing a Low Degree of Symmetry. . . . . . . . C. D. E. F.

.................................................... .......................................................... C. Compounds Containing Quaternary Carbon Atoms. . . . . . . . . D . Other Straight and Branched Hydrocarbons. . . . . . . . . . . . . . . . . . . . . . . . . . E. (+) 3-Methylhexane.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Branched Ring Compounds.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Some Results with Unsaturated Molecules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Concluding Remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

223

230

238 239 243 249 250 252 253 256 257 259 261

1. Introduction In 1946, H. S. Taylor ( 1 ) wrote, “The year 1933 was a red letter year for the science of surface catalysis since, in that year, the deuterium isotopes became available for catalytic studies.” Although it was soon realized that the catalytic exchange of hydrocarbons with deuterium was a field which was worthy of study, progress was limited until the mass spectrometer was adapted as an analytical tool by chemists. The wealth of detail about the nature of the reactions taking place on catalysts which can be obtained by the use of this instrument has resulted in extensive develop223

224

C. KEMBALL

ments in the last decade, particularly in the study of the exchange reactions of hydrocarbons. In the early work, the techniques used to follow exchange reactions with deuterium were measurements of the dilution of the deuterium by hydrogen by means of thermal conductivity, the examination of hydrocarbons by infrared spectroscopy and the determination of the total deuterium in the hydrocarbon by combustion. Morikawa et al. ( 2 ) investigated the exchange of CH4 arid D2and also CHI and CD4on a nickel-kieselguhr catalyst at temperatures above 138". The same authors (3) showed that the exchange of C2Haand Dz occurred between 100 and 130" which was approximately 60" lower than the temperature necessary to crack ethane to methane. Similar results were found for CsHe and D2 ( 4 ) . Farkas and Farkas ( 6 ) and Farkas ( 6 ) studied the exchange of a range of hydrocarbons, Cd&,, C & , , n-C4HlO , n-C6Hl4, and CyClO-CeHlz with D2 on platinum catalysts using the thermal conductivity technique. The three chief points established by these early investigations were: 1. The exchange of hydrocarbons with deuterium occurs much more readily than the cracking of hydrocarbons despite the fact that C-H bonds are substantially stronger than C-C bonds. 2. A dissociative mechanism is probably involved. 3. The ease of exchange depends on the hydrocarbon and the order of reactivity is roughly C Y C I O - C ~ H IC&l4 ~

C4Hlo

> CsHs > ClHs >> CH4

With the techniques available at that time, it was not possible to discover whether a single hydrogen atom is replaced by a deuterium atom or whether more extensive exchange occurs during the lifetime of the hydrocarbon molecule on the surface of the catalyst. This type of information became available only after the mass-spectrometric technique of following exchange reactions was introduced. The estimation of the relative amounts of the different isotopic species of a hydrocarbon in a mixture, by means of a mass spectrometer, is usually based on the assumption that the probability of ionization of each species to form the corresponding molecule-ion is not affected by the number of hydrogen or deuterium atoms in the molecule. For example, it is assumed that equal pressures of CzHa , CzDa, and all the intermediate species will give equal amounts of C2H6+, CzDa', etc., ions in the mass spectrometer. Field and Franklin (7)quote a considerable body of evidence which justifies the use of this assumption. The mass spectra observed have t o be corrected for contributions owing to the naturally occurring heavy carbon, 13 C, and the deuterium present in all hydrocarbons. Allowances have also to be made for the formation of fragmentary ions when these have the

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

225

same mass as molecule-ions of less highly deuterated species; e.g., the ions CaD6+ and C2D4+formed from C2Da have the same mass as the moleculeions CzHzD4+and C2H4D2+.The extent of the formation of fragmentary ions depends on the voltage applied to the electrons responsible for ionizing the molecule in the mass spectrometer, fragmentation decreasing with lower voltage. Consequently, better accuracy and greater ease of analysis are usually obtainable with electron voltages only a few volts in excess of value required to form the molecule-ion than with 50-v.or 75-v. electrons, which are often used as standard voltages in mass spectrometric analysis. A number of the early applications of the mass-spectrometric technique to catalytic exchange reactions emanated from Princeton and some of the other chemists, notably Kemball and Bond, who have worked in this field have clearly derived considerable inspiration from the time that they spent at Princeton University. One of the first publications appeared in 1948, when Parravano et al. (8) examined the exchange between CHI and CD4 on a silica-alumina cracking catalyst at temperatures between 345 and 384", and it is interesting that part of this research was carried out as early as 1941 by Hammel. In 1949, Wright and Taylor (9) examined the same reaction in more detail over a nickel-chromia catalyst at temperatures between 100 and 225" and found evidence of a stepwise exchange involving dissociation of the methanes. I n 1950, Turkevich et al. (10) reported on the exchange and deuteration of ethylene on a nickel wire and obtained the striking result that the ethane produced initially contained a substantial fraction of CsHa. Interest in this kind of work developed and in 1951 a number of papers were published. These included a note by Kauder and Taylor ( 1 1 ) on the exchange of propane and deuterium on platinized-platinum, a detailed investigation of the exchange of methane on evaporated nickel films by Kemball ( l a ) ,results covering a range of hydrocarbons from methane to n-butane on a cobalt-thoria Fischer-Tropsch catalyst by Thompson et al. (IS),and an investigation of the complex reactions occurring with the butenes and deuterium on nickel wires by Taylor and Dibeler (14). A preliminary report on this last topic was published in 1948 by the same authors ( 1 5 ) . It will be clear from the above that the work which has been carried out can be divided into two main classes. The first of these covers investigations of the exchange between saturated hydrocarbons and deuterium or the exchange between Merent isotopic species of saturated hydrocarbons in the absence of added deuterium. In this type of work, apart from a certain amount of cracking which may occur at higher temperatures, one is dealing with a pure exchange reaction, and the only effect of the catalyst is the production of different isotopic species, with no over-all change in the chemical composition of the reactant gases.

226

C. KEMBALL

In the second class of work, unsaturated hydrocarbons and deuterium have been used. With the majority of these substances deuteration occurs simultaneously with exchange and, in the case of larger molecules, doublebond migration and cis-trans isomeriaation may also occur. An excellent review of this whole field has been given by T. I. Taylor (16), and this review covers both classes of work up to 1956. In this chapter the discussion will be confined mainly to the progress that has been made in the period 1951 to 1958 in the study of the exchange reactions of saturated hydrocarbons. An attempt will be made to classify both the methods that have been used to study such reactions and the results that have been found by these investigations. Some aspects of the exchange of saturated hydrocarbons and other saturated hydrides have been reviewed recently by Anderson (17).

II. General Aspects of Exchange Reactions As there are a number of features which are common to all exchange reactions, it is of value to consider these in some detail before discussing the results which have been obtained for the exchange of individual hydrocarbons. Exchange reactions are a unique class of chemical reactions, and attention will be directed to the methods of interpretation of experimental data which are relevant to the study of exchange reactions and to the ways in which these may be classified. A. THEFINALEQUILIBRIUM

EXCHANGE REACTION Let us consider the reaction between a typical hydrocarbon, C,H, , and Dz . As a result of the exchange reaction, two kinds of equilibria will be OF AN

established. There will firstly be an equilibrium distribution between the total amount of deuterium in the “hydrocarbon” and the total amount of deuterium in the “hydrogen.”* Secondly, the relative amounts of the different isotopic species of hydrocarbon will be in equilibrium, and the same will apply to the different isotopic species of the hydrogen. In other words, the following interconversion equilibria for the hydrocarbon will be established,

+ CnHm-zDz * 2CnHm4I) CnHm-ID + CnHm-aDi * 2CnHm-zDz ............ ....... CnH1Dm-z + CnDm 2CnHDm-I CnHm

(Kd

(Kz)

(1)

(Km-1)

In those cases where equilibrium constants for reactions of this type have

* Throughout this chapter, terms such as “hydrocarbon” and ‘‘hydrogen’’ will be used a8 generic terms referring to all isotopic species, when the meaning of statements might otherwise not be clear.

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

227

been measured for hydrocarbons, it is found that the values obtained for the equilibrium constants are fairly close to the values expected for a random or classical distribution of the hydrogen or deuterium atoms between the various isotopic species of the hydrocarbon. This approximation does not apply to the interconversion equilibria between the three isotopic species of “hydrogen,” i.e., Hz , HD, and Dz , except at high temperatures because of the effect of the high relative changes in weight as H is replaced by D on the partition functions of the molecules. Likewise, the distribution of the deuterium between the “hydrocarbon” and the “hydrogen” usually differs slightly from the value expected in terms of a random distribution. The values of the equilibrium constants for the reactions shown in Equation ( 1 ) calculated by classical theory correspond to combinations of terms in the appr0priat.e binomial expansion and all the equilibrium constants are given by the general equation

(2) where the symbol (L) represents the number of ways of selecting i objects from a group of m identical objects. The validity of Equation ( 2 ) as a means of calculating the interconversion equilibrium constants for hydrocarbons may be illustrated by the results given in Table I for the exchange of propane and deuterium catalyzed by evaporated tungsten films (18). The errors in the experimental values of the equilibrium constants are about 5 %, and the agreement between the experimental and the calculated values is satisfactory. The differences between the two sets of values in Table I do not correspond to substantial discrepancies when the amounts of the various propanes are expressed in percentages of the total propane. This may be seen in Table 11, where the observed amounts of the nine propanes obtained using a ratio of deuterium to propane of 2: 1 are contrasted with percentages calculated by means of Equation (2) and the assumption that the mean percentage of deuterium in the propane is equal to the experimental value of 40.9%. TABLE I Equilibrium Constants for the Inlerconversion of the Propanes Equilibrium constant Experimental Calculated by Equation (2)

Ki

Ka

Ka

KI

Ks

Ks

Ki

2.20 2.29

1.80 1.75

1.75 1.80

1.50 1.56

1.66 1.60

1.96 1.75

2.40 2.29

228

C. KEMBALL

TABLE I1 Equilibrium Percentages of Propanes using a Ratio of Deuterium: Propane of 8 :1 Mean percentage of deuterium in the propanes for both distributions = 40.9% ds

Isotopic species

do

di

da

da

dc

Experimental Calculated by Equation (2)

1.8

7.8 8.3

19.3 19.9

29.2 27.6

23.6

1.5

24.0

ds

12.7 4 . 5 13.2 4.5

dl

ds

1.0 0.1 0.9 0.1

Similar results have been obtained for methane ( l a ) and for ethane (19). The values quoted in Table I1 also illustrate the point that the distribution of deuterium between “hydrogen” and “propane” differs from the value expected for a random distribution. With the ratio of pressures used, the expected percentage for the mean deuterium content of the hydrocarbon would be 33.3, which is substantially less than the experimental value of 40.9%. This type of deviation is also found with other hydrocarbons, but it does not affect the validity of using classical theory for the calculation of the interconversion equilibrium constants in studies of mechanism of exchange reactions. More accurate values for these equilibrium constants are necessary, however, if one is interested in the separation of isotopes by chemical processes.

B. THE RATE CONSTANTS OF

AN

EXCHANGE REACTION

The course of an exchange reaction may be followed either by the mean deuterium content of the hydrocarbon or by some parameter related t o this quantity. Let xi be the percentage of the total “hydrocarbon” present as the isotopic species C,H,-iDi a t time t . If a parameter 4 is defined by

4 = x1

+ 2x2 + 3x3 +

+ mx,

(3)

the mean deuterium content at any stage in the reaction is given by 4/m. Now provided that all the hydrogen atoms in the molecule are equally susceptible to exchange and provided that the influence of isotopes on the rate of reaction is ignored, the course of the exchange reaction will be given by the first-order equation

g

=

IC+ (1 -

&)

(4)

where k+ is a rate constant equivalent to the number of deuterium atoms entering 100 molecules of the hydrocarbon in unit time a t the start of the reaction and 6, is the equilibrium value of 4. Integration of Equation (4) gives

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

229

where 40 is the initial value of 4. Equations (4)and ( 5 ) are similar to equations developed by Harris (20)and, although they are only approximately true because of the assumption that all isotopic species react at the same rate, they are found to be obeyed in a great variety of exchange reactions. The results of a typical experiment for the exchange of ethane and deuterium (19) are shown in Fig. 1, and the appropriate plot using Equation (5) is given in Fig. 2.

m 90

80

70

bO

20

10

30

40

50

bO

T i m e (.in3

FIG.1. The percentages of the various ethanes observed during the exchange of an 8 : l deuterium:ethane mixture on 3.0 mg. of rhodium at 42.7" C . 1,CIHs ; A , C ~ H I D; A ~ ,C I H ~ D; 0 ~ , C2H2DI ; 0 , CzHDs ; 0 , CzDs. The percentage of C2HsD was too small to be measured accurately.

1

I

I

10

20

I 30

Time

I

I

40

50

(mi..)

FIG.2. The results of Fig. 1 plotted according to Equations (5) and (6).

230

C. KEMBALL

If the exchange of any hydrocarbon obeys Equation ( 5 ) , one can deduce that all the hydrogen atoms in the molecule are equally susceptible t o exchange. Consequently, failure to obey this equation may provide a useful indication of differences in reactivity between the various types of hydrogen atoms in the hydrocarbon, but this must be confirmed by devising and testing new equations because poisoning of the catalyst may also lead to failure of the equation. In order to gain a fuller insight into the nature of any exchange reaction, it is necessary to determine a second rate constant k, representing the initial rate of disappearance of the light hydrocarbon C,H, in percentage per unit time. This may be obtained by using the empirical first-order equation -log(za

- 2,)

=

kt

- log( 100 - 2,)

2.303(100 - z ~ )

(6)

where $0 is the percentage of C,H, present at time t and 100 and 2, are the initial and final percentages of this species. The application of this equation to the results given in Fig. 1 is shown in Fig. 2. The ratio of the two rate constants, defined by

M = -k? k

(7)

is an important quantity because it represents the mean number of hydrogen atoms replaced by deuterium atoms in each molecule of the hydrocarbon undergoing exchange in the initial stages of the reaction. C. KINETICS The rate constants which have just been defined differ from the rate constants of an ordinary chemical reaction. They are constant only for the course of an exchange reaction with a single mixture of reacting gases, but they are dependent on pressure and assume different values if the initial pressures of the reactants are altered. The true kinetics of an exchange reaction can be determined only by means of a series of experiments with different mixtures of the reactants because the course of a reaction with a single mixture follows the apparent first-order Equation ( 5 ) . It is important to understand why this apparent first-order behavior is found for the course of an exchange reaction with time whatever the true kinetics of the reaction. A failure to understand this feature of exchange reactions has sometimes led to unjustifiable statements about the ratedetermining step in such reactions. It is convenient to discuss a specific example-the exchange of ethane with deuterium. Suppose that the only adsorbed species taking part in the reaction are ( a ) physically adsorbed

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

231

“ethane” molecules, ( b ) chemisorbed “ethyl” radicals, (c) chemisorbed “ethylene” molecules, and ( d ) chemisorbed “hydrogen” atoms. When a mixture of ethane and deuterium is admitted to a catalyst, two things will happen : 1. The surface concentrations of each of the four kinds of “species” will build up, reach their equilibrium values and thereafter remain constant. 2. The exchange reaction will commence and will eventually lead t o isotopic equilibrium between all species in the system including molecules in the gas phase as well as adsorbed species. Let us suppose that the extent of the catalyst surface is such as t o accommodate only a small percentage of the molecules available in the system, say, 1%. Process 1 will be near completion after 1 % of the molecules present have been adsorbed and desorbed. Process 2 cannot be completed until at least 100% of the gas phase has suffered adsorption and desorption. It follows that throughout the bulk of the time required for the exchange reaction there will be equilibrium concentrations of the four kinds of “species” on the surface. The only factor which reduces the rate of the exchange from the initial value is the approach of the isotopic content of the “hydrogen” and the “hydrocarbon” to their equilibrium values, and this factor leads to the apparent first-order behavior. In this argument we have assumed that isotopic content does not affect the extent of adsorption or the rate of adsorption or desorption of any species, because these assumptions were implied in the derivation of Equation ( 5 ) . It follows that it is not possible t o say that the rate-determining step of an exchange reaction is the adsorption of the molecules as opposed to the desorption of the molecules, or vice versa. Both processes occur at the same rate, and it is convenient t o use the term “rate of adsorption/desorption” introduced by Anderson ( 1 7 ) . Similar considerations apply to processes occurring on the surface of the catalyst; i.e., the rate of dissociation of ethyl radicals to ethylene molecules will be equal to the rate of the reverse reaction. An alternative method of describing the kinetic behavior of an exchange reaction is to treat it as an example of a catalytic reaction where the products inhibit the reaction as they compete on equal terms with the reactants for the available surface. The study of the kinetics of exchange reactions has focused attention on the way in which the surface concentrations of various kinds of adsorbed radicals depend on the pressure of the reacting gases. Let us consider a situation where the adsorption of all species is weak; i.e., the extent of the surface is large compared with the fraction covered by adsorbed species. We shall assume that the strength of the adsorption is constant, which will be true for small variations in surface coverages. It is well known that, under these conditions, the fraction of the surface covered by hydrogen

232

C. KEMBALL

atoms, if there are hydrogen molecules in the gas phase, is given by OH

=

112 CHpEa

(8)

where CHis a constant. If now a more complicated example is considered, such as the formation of adsorbed ethyl radicals and ethylene molecules from ethane and hydrogen in the gas phase, the application of thermodynamic principles gives the following expressions for the fractions of the surface covered: PCsHs

and

PCxH6

(9) pg," PHa When the adsorption is not weak, Equation (8) may be expanded and written as a Langmuir adsorption isotherm: eC,H,

=

CCaH,

__

eC,H,

eH = c H p g : ( l - 0,)

=

=

CCaH, __

112 CHpH28F

(10)

where t9F is the fraction of the surface free from adsorbed species. This equation will hold only over a limited range of pressure because it is based on the assumption that the heat of adsorption is independent of coverage. For such a limited range of pressure, it is permissible to express the fraction of the surface covered in terms of a fixed power of the pressure of hydrogen, i.e., OH

(11)

a

and under these conditions it follows that

eF a

pZ1l2

(12)

The same principles may be applied to the adsorption of mixtures of ethane and hydrogen. Let us assume that ethyl radicals and hydrogen atoms require a single site and ethylene molecules require two sites. The following relationships will hold :

eH

0:

pZa2.eII;

and

except in this case e F I is dependent on the pressures of both gases, but over a limited range of pressure it may be expressed as OF'

a p&p&2H,

(14)

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

233

In this treatment the surface coverage of any species is expressed as the product of the appropriate function of the pressures of the gases and terms representing the chance of existence of the required number of free sites on the surface to accommodate the species. This type of approach is of value not only in the study of the mechanism of exchange reactions but it can also be applied with profit to a great many kinds of reactions in the field of heterogeneous catalysis. One example will be given to illustrate the importance of this approach to other types of catalytic reactions. There is good evidence that the rate-determining step in the decomposition of ammonia on a number of catalysts is the desorption of two nitrogen atoms to form a nitrogen molecule (21,2 2 ) . Furthermore, it has been shown that the rate of reaction in the presence of added hydrogen is often given by

This result has been explained by means of a complex theory due to Temkin but it can be explained equally well and more simply and Pyzhev (EL?), by the methods just described. The fraction of the surface covered by nitrogen atoms will be

Now, it is known that the adsorption of nitrogen is not greatly influenced by the presence of hydrogen, and so it is allowable to treat Equation (16) in a similar way to the treatment of Equation (10); i.e., over a limited range of pressure OF is a function depending solely on some power of ( p N A 8 / p g f )Assume . this power is ( m / 2 - 1 ) ; then we have

which is in accord with the experimental results.

D. CLASSIFICATION 1. Simple OT Stepwise Exchange Reactions. I n this class of exchange reaction, only a single hydrogen atom is replaced by a deuterium atom in each molecule which reacts on the surface of the catalyst. Isotopic species containing two or more deuterium atoms are formed only by successive reactions. This class of reaction can be recognized in three ways: a. The value of M defined by Equation ( 7 ) will be unity. b. The only initial product will be the monodeutero species, C,Hm-lD. c. The interconversion equilibria, Equations ( 1 ) and (2), will be satis-

234

C. KEMBALL

TABLE 111 Comparison of Distributions Observed during Simple Exchange Reactions with Values Calculated by the Appropriate Form of Equation (2) Reactant

Catalyst CrlOa gel

C(CHs), (ana- Pd film lyzed as CIXB+ ions) CHaNHz Pd film

Percentage of isotopic species

Reference

I Obs. Calc. Obs. Calc.

do

I

di

I I 1 I dz

ds

95.6 4.2 0.2 0.1 95.6 4.3 40.00 37.68 16.80 4.70 40.00 38.59 16.53 4.14

Obs. 70.0 Calc. 70.0

27.4 27.3

d4

ds

0.80 0.67

0.12 0.07

2.6 2.7

fied throughout the course of reaction, provided that the isotopic species of the hydrocarbon used as reactant are in equilibrium and provided that all hydrogen atoms in the molecule are equally susceptible t o exchange. The first two of these criteria are obvious, and the third may be proved quite readily. A proof for the specific case of the exchange between ammonia and deuterium has been given by Kemball (84),and Matsen and Franklin (26)have considered the general case. The third criterion can be useful in diagnosing the existence of a simple exchange reaction when the only experimental data available are a single set of analyses of products at one time during the course of the reaction. It is necessary to compare the observed distribution of species with a calculated distribution based on Equation (2) and having the same percentage of light hydrocarbon, C,H, , as the observed distribution. Some examples are shown in Table 111. The results for the exchange of neopentane are quoted in terms of the C&+ ions (X representing H or D) because the amounts of the molecule-ions formed in the mass spectrometer are too small for estimation. This factor does not prevent the application of the third criterion, provided the appropriate interconversion equilibrium constants are used. If the amounts of the parent molecules satisfy the equilibria, the ions formed by the loss of a methyl group will also exhibit a random distribution of species. Results for the exchange of methylamine and deuterium are also included in this table because they illustrate another important principle. Over palladium, in the temperature range -20 to 60°, only two of the five hydrogen atoms in the molecule, those in the amino group, are readily exchangeable. Provided the appropriate interconversion equilibrium constants are used to allow for this factor, the third criterion can be applied in this type of case where only a certain type of hydrogen atom in a complex molecule is exchangeable and the remaining hydrogen atoms are not reactive.

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

235

2. Multiple Exchange Rea,ctions. An exchange reaction in which more than one deuterium atom is introduced into the hydrocarbon molecule on each interaction of the molecule with the catalyst is classified as a multiple exchange reaction. The existence of such a process may be recognized in three ways, exactly analogous to those described above for simple exchange reactions: a. The value of M defined by Equation (7) will be greater than unity and will give the average number of deuterium atoms introduced t o each molecule reacting at the beginning of the reaction. b. The first products will cont,ain species having more than one deuterium atom, and the initial distribution of products will be important as a means of deciding the type of process occurring. c. The distributions of iso1,opic species during the course of the reaction will be richer in more highly deuterated species than calculated distributions based on Equation (2). It should be noted that the third criterion becomes progressively less valuable the greater the extent of the conversion because a multiple exchange process must eventually produce the same equilibrium distribution of isotopic species as a simple exchange process. In order to use the second criterion, distributions of products must be measured at low conversions in order to exclude species formed by successive interactions with the catalyst. At low conversions, if the fraction of molecules exchanged is y, the fraction which is exchanged twice will be approximately y2/2, and the distributions must be obtained before 10% has reacted to avoid significant contributions from successive exchanges. The technique ( I d ) in which the reaction vessel is linked directly t o the mass spectrometer is extremely useful for this purpose. The construction of a suitable capillary leak to control the amount of the reacting gases bleeding through into the spectrometer requires care. However, as only very small amounts of gas are needed for the mass spectrometric analysis, the rate of leak can be made sufficiently small to avoid any effect of the withdrawal of gas on the rate of the exchange process. The technique permits analyses to be made a t any stage in the reaction and is more powerful than the conventional technique involving the withdrawal of comparatively large samples for subsequent analysis. I n the study of multiple exchange reactions, it is desirable t o work with a high ratio of deuterium to hydrocarbon. This minimizes the influence of isotopic dilution of the deuterium on the rate of production of the more highly deuterated species during the early part of the reaction. The use of deuterium containing as small a percentage of hydrogen as possible is also important if true initial distributions of products are to be obtained. Some useful terminology has been introduced by Rowlinson et al. (29)

236

C. KEMBALL

for describing the types of multiple exchange occurring with hydrocarbons. If the exchange is limited to the hydrogen atoms on a single carbon atom, it is called a-a. Similarly, an exchange process involving the hydrogen atoms on two adjacent carbon atoms is known as a+, and if next-to-adjacent carbon atoms are involved, the term a-y is used.

E. POSSIBLE MECHANISMS 1. Simple Exchange Reactions. If nn examination of the features of the exchange of a particular hydrocarbon indicates that a simple exchange process is operating, the following possible mechanisms must be considered. a. The hydrocarbon molecule is not chemisorbed except during the actual exchange which takes place with a chemisorbed deuterium atom (or ion). Thus, for C,H, , the reaction may be depicted as -ID 1

I

B

S

4

1

I

B

B

where S represents the surface of the catalyst. b. The mechanism may be dissociative and involve adsorbed radicals (or ions) of the type C,H,-I . c. The mechanism may be associative and involve adsorbed species of the type C,,H,,,+I, which may be radicals or possibly ions. Further data on the kinetics of the reaction, the nature of the adsorption of the reacting gases, and the probable stability of the different kinds of intermediate species are needed to choose between these three possibilities. The evidence, some of which will be given later, suggests that mechanism b applies for the exchange of saturated hydrocarbons and mechanism c for the exchange of many unsaturated hydrocarbons; there is no evidence that mechanism a applies t o the exchange of any specific compound, but it is a possibility which must be considered. As with other types of catalytic reaction, it is possible to represent the detailed operation of either the dissociative mechanism or associative mechanism in more than one way. Even if it is established that the dissociative process operates, it may not be easy t o formulate the precise mechanism. It may consist of a Langmuir mechanism where all the species are chemisorbed, i.e., D I

c?Hm I

B sIs

+

DC,Hm-lH I

s B

I

I

s Bs B

C nHn-1 D ~

s lIi s

H

I s i s "sB

or alternatively, of an Eley-Rideal mechanism where one of the reacting species is gaseous or physically adsorbed, i.e.,

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

237

Detailed arguments have been put forward in support of each of these types of mechanism for the exchange of hydrogen and deuterium, but there is no clear evidence that one of these mechanisms operates to the total exclusion of the other. The existence of a simple exchange process implies that the adsorbed intermediate, whether it is of the dissociative type, C,Hm-l, or the associative type, C,Hm+I, must be a comparatively stable entity on the surface of the catalyst. The species must have little chance of undergoing further reaction involving the introduction of a second deuterium atom into the molecule during its lifetime on the surface, because this would give rise t o the formation of products with more than one deuterium atom and the exchange would cease to be a simple exchange. 2. Multiple Exchange Reactions. The possible mechanisms which may lead t o multiple exchange are necessarily more complicated than those responsible for simple exchange. Four alternatives may be considered. a. No chemisorption of the hydrocarbon except during the actual exchange

y

S

CnHm

I

k

D +

H

k

s‘

Cn

H m-2

i

s1

Dz H 1

s

(21)

b. The formation of a single type of dissociated species involving the loss of a t least two hydrogen atoms from the original molecule. c. The formation of a single type of associated species having a t least two hydrogen atoms more than the origiiial molecule. d. The existence of two or more types of adsorbed species of different states of hydrogenation with multiple exchange resulting from the interconversion of these species on the surface of the catalyst. On general grounds one would expect mechanism d to be the most common type because chemical changes are usually more rapid, the smaller the number of bonds which have to be broken and re-formed in the reaction. On this basis, mechanism a may be regarded as highly improbable. The determination of whet,her the multiple exchange occurs by an a-a, a-p, or a-y process is obviously an important step in the elucidation of the mechanism. Careful examinartion of the initial distributions of products can be most useful both in this connection and in providing evidence about the types of adsorbed species responsible for the exchange reaction. This examination may be carried out in two ways, either ( a ) the comparison

238

C. KEMBALL

of the distribution of products for the exchange of a single hydrocarbon with the calculated distributions based on assumed mechanisms, or ( b ) the comparison of the distributions of products obtained with a series of hydrocarbons on the same catalyst. Examples of both kinds of comparison will be given in Secs. I11 and IV.

F. A THEORY FOR CALCULATING THE INITIAL DISTRIBUTIONS OF PRODUCTS IN MULTIPLE EXCHANGE REACTIONS We shall now consider in outline a general method devised by Anderson and Kemball (19)for the interpretation of the initial distributions of products obtained in multiple-exchange processes. The method was devised, in the first instance, to apply to the exchange of ethane and deuterium, but it can be extended quite simply to cover other types of molecules. It involves the adoption of a specific mechanism from which calculated distributions are then obtained for comparison with observed distributions. The method will be illustrated for the exchange of ethane. The mechanism adopted is as follows. 1. Adsorbed ethyl radicals are formed by the dissociative adsorption of ethane. Every ethyl radical may either leave the surface with a deuterium atom to form an ethane molecule or lose one of the three hydrogen atoms of the methyl group t o form adsorbed ethylene. The chances of these two P ) and P/(1 P ) , respectively, P being a constant events are 1/(1 for a given catalyst. Equal chance is assumed for the loss of each of the three hydrogen atoms of the methyl group in the second process. 2. Every adsorbed ethylene molecule, so formed, will revert to an ndsorbed ethyl radical by the addition of a deuterium atom to one of the two carbon atoms, both of which have equal probability of gaining the deuterium atom. 3. The only type of adsorbed ethyl radical present initially is ClHl. A quantity Qz,yris now defined for each type of ethyl radical. The value of x (1 or 2) indicates the point of attachment of the radical to the surface. The values of y and z may range from 0 to 3, and they indicate the number of deuterium atoms which will be present on each carbon atom of the ethane molecule formed when the ethyl radical leaves the surface. The value of Q2,y., or Qm in abbreviated notation, represents the total chance of formation of the specific ethane molecule as an initial product from the corresponding ethyl radical, Thus, Ql,2a represents the chance of formation of CHD,.CDs from -CHD-CDs, and represents the chance of formation of CHs.CH2D from CHa CH-. Equations for each individual type of Q may be derived from one general equation

+

+

-

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

239

where a,(O) represents the initial chance of existence of an adsorbed ethyl radical of type m, pmnis the ratio of the chance of an ethyl radical of type m becoming an ethyl radical of type n to the chance of the radical of type m forming an ethane molecule, and pirnhas a similar meaning. The values of the quantities p,, and p ~ ,are readily obtained in terms of the parameter P defined above. Equations for each type of Q can thus be obtained as a function of this single parameter and consequently, theoretical initial distributions of products can be evaluated for any selected value of the parameter. The results derived by this approach will be discussed in Sec. IV,B. It must be stressed that Equation (22) is one example of a general equation from which initial distributions of products can be calculated for the exchange of a great variety of compounds with appropriate assumed mechanisms. Cases where more than two kinds of adsorbed radicals are involved can also be covered, but this necessarily involves the introduction of further parameters.

111. The Exchange of Molecules Possessing a High Degree of Symmetry In this section, we shall examine the results which have been obtained for the exchange of the saturated hydrocarbons methane, ethane, cyclopentane, cyclohexane, cycloheptane, cyclo-octane, and neopentane. The common characteristic of this group is that all the carbon-hydrogen bonds in each individual molecule are similar in nature. An attempt will be made to indicate how the results fit, into the classifications outlined in Sec. 11.

A. METHANE The exchange of methane with deuterium has been followed on nickel films (I$), on a cobalt-thoria Fischer-Tropsch catalyst (IS) and on films of rhodium, platinum, palladium, and tungsten (30). The important features of the exchange over the metal films may be summarized as follows: 1. All four deuteromethanes are found as initial products, as may be seen from the distributions given in Table IV. 2. Increasing the ratio of deuterium to methane reduces the amounts of CHzDz and CHD, formed initially, and this indicates that CHaD and CDI are the two main initial products. 3. The relative amounts of CH2D2, CHDa and CD4 formed as initial products approximate to values that would be expected if the interconversion equilibrium for these three compounds had been established. 4. Over nickel films, the activation energies for the production of the three compounds CH2Dz, C€ID3, and CD4 are similar; the values are 34 (subject to greatest error), 31, and 31 kcal./mole. The activation energy for the production of CHsD is only 24 kcal./mole.

240

C. KEMBALL

TABLE IV Initial Distributions of Products for the Exchange of Methane on Films Ratio of DZ/CH4 in reaction mixture

Rh Rh Ni Pt W Pd

1 .o 3.4 0.75 1 .o 1 .o 1 .o

Proportions of isotopic species oc.

162.0 171.6 237.0 259.3 150.6 254.3

di

dz

da

d4

0.21 0.21 0.12 0.36 0.76 0.95

0.045

0.29 0.16 0.24 0.25 0.09 0.02

0.45 0.63 0.61 0.27 0.14 0.02

0.03 0.12 0.01 0.006

These facts suggest that there are two mechanisms operating-a simple exchange mechanism producing CH3D and a multiple exchange mechanism (necessarily of the a-a type) producing mainly CD4 with CHD3 and CH2D2as minor products. The results obtained over the cobalt-thoria catalyst support these suggestions, since the percentages of the products found after a contact time of 11 hr. at 183"were CH3D, 9 %; CH2Dz,0%;CHD8, 0.1 %; and CDI , 1.7 %. There can be little doubt that the simple exchange mechanism operates by the formation of adsorbed methyl radicals. The multiple exchange mechanism might involve interconversion between methyl and methylene radicals or between methylene radicals and a more highly dissociated species. Further discussion of these possible alternatives will be given below. A difficulty arises in the derivation of activation energies and orders of reaction for the two mechanisms of exchange. One set of values is found if the two mechanisms are assumed t o be independent, and a second set is obtained from the same experimental data if the two mechanisms are coupled together in that the entities taking part in multiple exchange are formed via adsorbed methyl radicals and not directly from the gas phase. A detailed discussion of this problem has been given by Kemball (SO). The second of these alternatives may be considered rather more likely, since it implies stepwise removal of hydrogen atoms from the carbon atom in the formation of adsorbed methylene radicals, while the first alternative would involve the dissociation of methane into a hydrogen molecule and an adsorbed methylene radical. For this reason, the values of activation energies and the orders of reaction shown in Tables V and VI are those derived on the assumption that the two mechanisms are coupled, The subscript A refem to the rate of disappearance of light methane, CHI, i.e., the total rate of the adsorptionldesorption process. The subscript B refers to the rate at which the multiple exchange process occurs on the surface of

241

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

TABLE V Activation Energies and Frequency Factors for the Exchange of Methane on Films Temperature range,^^. Ni Pd Pt Rh W

206-255 243-308 159-275 138-217 92-174

.

9

kcal mole

28.4 23.5 22.8 23.6 9.1

kcal. EB ’ mole 33.8 35.7 26.4 28.0 11.7

loglOAA loglOAB (A in mol./ ( A in mol./ sec.cm.2) sec.cm.2) 24.9 23.2 23.5 25.2 17.5

28.0 27.2 25.1 27.9 18.3

TABLE VI Kinetic Data for the Exchange of Methane on Films and the Ratio of the Rates of Production of CDd and CHD3 with Equal Pressures of Reactants Catalyst Ni

Rh W Pt Pd

1.0 1.0 0.6 0.4 0.3

--0.9 -0 . 8 --0.4 -0.3 -0.1

1.0 1.0 0.9 0.5 0.3

-1.4 -1.2 -0.9 -0.9 -0.6

3.3 3.1 1.5 1.3 1 .o

the catalyst and corresponds t o the dissociation/recombination process of adsorbed methyl radicals. The energies of activation on tungsten are substantially lower than on the other four metals, and on each metal the breakdown of the methyl radicals ( B ) requires a greater activation energy than the adsorption/desorption of methane ( A ) . These results are in excellent agreement with recent work by Wright et al. (31). They showed that the adsorption of methane occurs more readily on tungsten than on nickel or iron and that, on all three metals, adsorption occurs more readily than decomposition of the methane t o form adsorbed radicals and gaseous hydrogen. Both reactions exhibit a similar order with respect t o methane pressure on each metal and the constant difference of -0.5 in the order with respect to deuterium pressure suggests that reaction B involves one less “hydrogen” atom t,han reaction A . The orders are consistent with the following mechanisms for the rate-determining steps

242

C. KEMBALL

The chances of finding bare sites on the metals (Sec. I1,C) will be given by (Nil

OF a

p;;.Ib

(Rh)

OF a

PD,

OF a

-0.376 -0.2

-0.126

-0.176

-0.276 PCH‘

pD,

-0.06

O F a PD,

(25)

-0.35 pCH 4

Kemball (12,SO) has argued that the multiple exchange must occur by interconversion of adsorbed methylene radicals and less-hydrogenated species. The fact that the activation energies for the production of all the highly deuterated species are equal implies that the rate of production of each of these compounds is controlled by a single rate-determining step, the formation of adsorbed methylene radicals. Anderson ( 1 7 ) has suggested that successive interconversion between adsorbed methyl and methylene radicals might be sufficient t o account for the multiple exchange. This does not seem likely, because it would imply that increase of temperature should favor the production of CD4relative to that of C H 2 D z , which is not found. The substantial amounts of C D 4 formed and the approximate equilibrium observed in the relative amounts of C H 2 D 2 , C H D 3 , and CD4indicate that the methylene radical, once formed, must undergo extensive reaction before returning to the gas phase as a methane molecule. Dissociation to methine radicals, or even perhaps to carbon atoms, must be occurring on the surface. A point which requires elucidation is the reason for the appreciable amounts of C H z D 2 and C H D 3 formed. Miyahara (32) carried out rather complex calculations of the initial distributions of products, based essentially on the mechanism given by Equations (23) and (24). He suggested that hydrogen atoms were present on the surface as well as deuterium atoms. He was then able t o account for the production of CHzD2 and CHD3simultaneously with the production of CD, . Now one would expect that any chemisorbed hydrogen atom formed by the dissociation of methane would be very rapidly replaced by a deuterium atom through exchange with the deuterium in the gas phase and, consequently, Miyahara’s postulate seems to be improbable. An alternative possibility is that a reacting species on the surface may

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

243

be able to acquire either a hydrogen atom directly from a neighboring radical or a deuterium atom from the surface. The relative chances of these two methods of acquiring “hydrogen” atoms would then govern the amounts of CH2D2 , CHDI and CD, formed initially. Evidence in support of this tentative hypothesis may be obtained from Equations (25) and the last column of Table VI. It is clear that the strength of the adsorption of methane relative to that of deuterium depends on the metal and increases from nickel to palladium. The production of CHD3 relative t o the production of CD, also increases from nickel to palladium, which is just what would be expected if hydrogen transfer between adsorbed radicals was possible. It must be emphasized that the interpretation of the kinetic data in terms of Equations (23) to (25) is based on the assumption that both kinds of exchange process involve adsorbed methyl radicals. Details of the alternative interpretation, assuming that the two processes are independent, may be found in the original literature (SO),in a theoretical paper by Markham et al. ( 3 3 ) and in the review by Taylor (IS).

B. ETHANE Thompson et al. ( 1 3 ) showed that exchange of ethane occurred a t 183” on a cobalt-thoria catalyst. With flow rates approximating to those used for the Fischer-Tropsch synthesis, 6.1 % of ethane exchanged to give 2.4 %, CzD6 ; 0.4%, C2HDs ; 0.3 %, C2H3D3 ; and 2.7 %, C2H4D2. More detailed results were obtained by Anderson and Kemball ( 1 9 ) ,who used metallic films as catalysts. The main points established may be summarized briefly as follows: 1. Over some metals, C2HsD is the major initial product, and the amounts of the other deuteroethanes formed decrease steadily with increasing deuterium content, The mean numbers of hydrogen atoms exchanging initially [see Equation (7)] for the metals showing this type of distribution of products are tungsten 1.30, molybdenum 1.16, and tantalum 1.15. 2. Palladium and rhodium form C2D6 as the major initial product with steadily decreasing amounts of the less deuterated species. The values of M for these metals are 4.8 and 5.0, respectively. 3. Other metals give initial distributions of products equivalent to a combination of the distributions described in 1 and 2; i.e., substantial amounts of both C2HaD and C2Da are formed, and the distribution plotted against deuterium content is U-shaped. The values of M for the metals showing this behavior are zirconium 2.3, chromium 2.5, vanadium 2.6, and platinum 3.5. 4. Exceptional amounts of C2H4Dz,such as were found over the cobaltthoria catalyst, are not observed over metal films.

244

C . KEMBALL

TABLE VII Observed and Calculated Initial Distributions of Products for the Exchange , of Ethane on Films Proportion of isotopic species __

Catalyst dl Mo Pd Pt

Ni (unoricnted)

d2

d8

dr

0.14 0.172 0.06 0,073 0.17 0.143 0.036

0.030 0.025 0.08 0.087 0.12 0.095 0.044

0.007 0.003 0.11 0.112 0.10 0.057 0.056

- -__

obs. calc. P = 0.25 obs. calc. P = 18 obs . calc. P = 2.0 (50%) calc. P = 18 (50%)

0.82 0.800 0.05 0.053 0.20 0.166 0.026

Total calc. obs. calc. P = 0.115

0.857 0.093 0.010 0.004 0.90 0.10 0.897 0.096 0.007

d6

ds

- -

0.19 0.169 0.15 0.029 0.084

0.51 0.506 0.26 0.010 0.254 - -___ __ -Total calc. 0.192 0.179 0.139 0.113 0.113 0.264 obs. 0.88 0.085 0.006 0.030 calc. P = O.llS(9501,) 0.852 0.091 0.007 0.002 0.002 0.003 0.004 0.006 0.032 calc. P = 30 (5%)

-

Ni (oriented

0.006

0.032

5. No substantial influence of temperature on the initial distributions of

products is observed with any metal, except nickel, which behaves anomalously. Typical initial distributions of products are shown in Table VII. The chief difference between the behavior of ethane and that of methane is that there is no evidence for the existence of a simple exchange mechanism with ethane. Even over molybdenum and tantalum, which have the lowest values for M, appreciable amounts of dz-ethane and d3-ethane are produced with the same activation energy that governs the production of &-ethane. There can be little doubt that the first step in the exchange of ethane is the formation of an adsorbed ethyl radical. It is reasonable to suppose that the multiple exchange occurs by the interconversion of adsorbed ethyl radicals and adsorbed ethylene molecules, i.e., an a-8 multiple exchange process. This supposition accounts for the absence of any substantial effect of temperature on the initial distributions of products because both the adsorption/desorption of ethane and the interconversion of adsorbed ethyl and adsorbed ethylene involve similar processes-the dissociation of a carbon-hydrogen bond of a methyl group in the ethane molecule. Anderson and Kemball (19) derived convincing support for this mechanism by using the theory outlined in Sec. 11, F. The observed distributions

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

245

of products found over molybdenum and palladium are in good agreement with distributions calculated for an appropriate value of the parameter P (Table VII). It was not possible to account for the U-shaped distributions such as that found over platinum by choosing a single value of this parameter. Reasonable agreement was found between theory and experiment, however, if it was assumed that two types of reaction were occurring-one with a low and the other with a high value of P . It was suggested that these two types of reaction were occurring on different crystal faces of the metals, and support for this suggestion was obtained by studying both unoriented and oriented nickel films (Table VII), A dual distribution (U-shaped) was found over unoriented nickel, but a n oriented film prepared by the method described by Beeck et al. (34) gave a simpler distribution. Beeck and his colleagues ( 3 4 , 3 5 ) believed that the method of preparation of the oriented films exposed 110 crystal faces t o the gas phase. Recent work by Sachtler et al. ( 3 6 ) , however, has suggested that this is not correct. It is possible that the differences in the behavior of the two types of film may be attributed to differences in the size of the crystallites. Burwell and Tuxworth (37') have shown that multiple exchange of hydrocarbons is favored by an increase in the particle size of nickel catalysts, and the absence of the extensive (high-P) multiple exchange over the oriented nickel films would be expected because these films consist of smaller crystallites than the unoriented films. Miyahara (38) has interpreted the experimental distributions of products found by Anderson and Kemball (19) by a theory assuming a much more complex array of adsorbed radicals on the surface. He postulated that CzHzand CzH3were involved in addition t o CzH4 and CrHb, and he was able t o account for the experimental results, but only by using three parameters and a further simplifying assumption. Kemball (39) has argued against the inclusion of CzHzand CzH3in the reaction scheme on the grounds that very few of these species will be present on the surface in the presence of excess hydrogen or deuterium, and also on the evidence of Jenkins and Rideal ( 4 0 ) , who showed that, although these species are formed when ethylene is admitted to a nickel surface, they cannot be rehydrogenated rapidly. The temperature ranges required for the exchange of ethane on the various metal films are shown in Table VIII together with activation energies and frequency factors. These relate t o the rate of disappearance of light ethane [see Equation (6)] and consequently apply t o the adsorption/desorption process. Nickel shows some activity between 0 and 75", no activity between 75 and 160", and then normal behavior a t higher temperatures. This may be due t o poisoning by highly dissociated species, like CzH3and CzH2, which are rehydrogenated a t higher temperatures.

246

C. KEMBALL

TABLE VIII Activation Energies and Frequency Factor8 for the Exchange of Ethane on Films Catalyst W Mo Ta

Rh Ni Ni

V

Pt Cr Pd Zr

Temperature range, “C. -80 to -29

-504

-44-0 0-70 0-75 162-195 102-180 134-192 149-215 145-207 158-192

E , kcal./mole 8.2 7 .O 7.8 11.7 6.5 to 2 18.0 20.7 12.5 13.9 21.4 15.4

log,oA ( A in mol./ sec. 10 mg.)

logioA ( A in mol./ sec. cm.*)

23.7 21.5 21.9 24.0

20.2

23.8 26.8 22.3 23.4 25.8 23.5

20.8

21.1

19.5 22.9

C. CYCLOPENTANE AND CYCLOHEXANE Anderson and Kemball (41) investigated the exchange of both these compounds on metal films a t low temperatures, and Bunvell and his colleagues ($9, 3’7,42) have used bulk catalysts and also films over a wider range of temperature. There is now a good measure of agreement between the views of the two schools. The initial distributions of products which are observed at low temperatures with metals producing extensive multiple exchange indicate that half the hydrogen atoms in these molecules are more readily exchanged than the remaining hydrogen atoms. Some typical distributions of products are shown in Table IX. Anderson and Kemball suggested that, owing to the approximately planar shape of these molecules, the hydrogen atoms on the side of the ring first attached t o the surface are the ones which are readily exchangeable. This hypothesis is now generally accepted and furthermore it is believed that these five or six atoms are exchanged by an a-/3 multiple exchange process. This involves interconversion between adsorbed cycloalkyl radicals and cycloalkene molecules, and it is analogous to the mechanism described for ethane except that the exchange can propagate round the ring and is not limited merely to two carbon atoms. Evidence in support of this mechanism has been obtained by calculating distributions of products by a theory similar t o that described in Sec. I1,F. The remaining hydrogen atoms in the molecule become as easily exchangeable as the first half if the molecules can “turn over” on the surface. Once this difficult step has been accomplished, the remaining hydrogen atoms can be exchanged by the facile a-/3 mechanism. This explains the second rise with deuterium content up to the fully deuterated species in

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

247

TABLE IX Initial Distributions of Products for the Exchange of Cyclopentane and Cyclohexane over Palladium Films Isotopic species

CIsHio

CsHn

O0

25 O

18.5’

44 O

0.240 0.127 0.066 0.048 0.265 0.007 0.007 0.048 0.039 0.153

0.182 0.096 0.038 0.053 0.177 0.003

0.124 0.040 0.034 0.036 0.052 0.277 0.031 0.043 0.042 0.094 0.071 0.154

0.034 0.019

0.008 0.058 0.062 0.323

0.009 0.015 0.033 0.118 0.028 0.029 0.042 0.102 0.136 0.435

the distributions of products shown in Table IX. Burwell and Tuxworth ($7’) have used the term “interchange reaction” to describe the difficult step which permits the exchange process to propagate from the first set of hydrogen atoms to the second set. Increase of temperature reduces and finally obliterates the discontinuity in the distributions of products at the point corresponding to the completion of the exchange for the first half of the hydrogen atoms. This may be seen from the result,s in Table IX, and it indicates that the interchange reaction must require a greater activation energy than either the adsorptionldesorption or the a-/3 multiple exchange process which have similar activation energies (cf. ethane). Anderson and Kemball ( 4 1 ) believe that the interchange reaction is analogous t o the a-a multiple exchange of methane and involves the formation of a species such as CHz-CHz I

I

oriented with the plane of the ring perpendicular to the surface. Burwell et al. (42) favor a more complex intermediate involving resonance structures of the type

248

C. KEMBALL

TABLE X Activation Energies and Frequency Factors for the Exchange of Cyclopentane and Cyclohexane on Films Catalyst

Pd W Rh Ni Pt Pd

Reactant CsHio CsH12 CeH12 CeHia CsHiz CsHiz

logid

Temperature range, "C.

E , kcal./mole

0-37 -69 to -48 -4 8 4 -35-0 0-3 1 18-82

14.2 11 10.4 10.8 12 13.0

( A in mol./ sec. cm.2) 22.7 23.4 22.5 21.9 22.5 21.4

because the interchange reaction is stereochemically similar to racemizution of an adsorbed species (Sec. IV,E) . Some data on the activation energies and the frequency factors for the adsorption/desorption of cyclopentane and cyclohexane are given in Table X.

D. CYCLOHEPTANE AND CYCLO-OCTANE Burwell el al. (42) have examined the exchange of these higher cyclic paraffins over palladium catalysts supported on y-alumina. The pattern of products at low amounts of exchange shows only a slight discontinuity after d7-cycloheptaneand negligible discontinuity after ds-cyclo-octane.This indicates that the interchange reaction must be easier the larger the size of the ring. Burwell et al. suggested that the normal a-p multiple exchange will permit the interchange reaction to occur with these larger and more flexible cycloalkanes, and they make some interesting deductions about the stereochemistry of the di-adsorbed alkane taking part in an a-0 exchange process. Two possible conformations for this species are shown in Fig. 3. As cyclopentane is nearly planar, the 1,2-diadsorbed cyclopentane must have a structure similar to the eclipsed conformation shown in Fig. 3b, and this leads to the exchange of the five hydrogen atoms in &-positions, i.e., on one side of the ring, by the a-0 exchange process. Considerations of the geometry of cyclohexane also suggest that the eclipsed conformation is more likely with this molecule and probably with other molecules as well. Now for exchange to propagate to both sides of the ring by the a-8process, it follows that an eclipsed trans-l,2-diadsorbed cycloalkane must be formed. Examination of models showed that the eclipsed trans-1 ,2-diadsorbed cyclo-octane should be formed as readily as the corresponding cis-structure, in agreement with the experimental observation that all hydrogen atoms on the Cs-ring are equally susceptible to

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

a . STAGGERED

249

b. E C L I P S E D

FIG.3. Conformations of 1,2-diadsorbed alkane.

exchange. The intermediate behavior with cycloheptane, which shows a small discontinuity in the pattern of products after the d7 compound, is attributed t o the slight strain involved in the formation of eclipsed trans1,2-diadsorbed cycloheptane. The patterns of products observed with all the four ring compounds from Cs to Cs are thus consistent with the hypothesis that a-8 multiple exchange involves an eclipsed 1,2-diadsorbed alkane. Further support for this hypothesis will be given in Sec. IV,F.

E. NEOPENTANE The exchange of neopentane with deuterium has been examined over I number of metal films by Kemball ($7) and ovcr nickel films a t higher temperatures by Rowlinson et al. ($9). This compound has attracted interest because a n a-@ multiple exchange process cannot operate, and an a-y process is necessary if more than three hydrogen atoms are t o be exchanged during a single sojourn of the molecule on the catalyst. Very little multiple exchange is found with neopentane. A simple exchange mechanism operates over palladium (see Table 111),presumably involving the formation of adsorbed neopentyl radicals. The absence of any a-a multiple exchange over palladium is not surprising, since this metal is the least efficient catalyst for this type of process with methane (see Table IV). Some a-a multiple exchange occurs with rhodium at low temperatures and to a lesser extent with tungsten and nickel; i.e., the multiple exchange is limited almost entirely to the introduction of two or three deuterium atoms into the molecule. As with methane, the activation energy for the a-a multiple exchange is greater than that for the simple exchange. At higher temperatures (30-99"), rhodium causes a more extensive multiple exchange to occur with the initial production of species containing more than three deuterium atoms. This process requires an even greater activation energy than the cr-a multiple exchange, and it must be attrib-

250

C. KEMBALL

uted to the onset of an a-y mechanism, presumably involving interconversion between the adsorbed neopentyl radical and the 1,3-diadsorbed 2,2dimethylpropane. Thus, the results with neopentane indicate that the a-y type of multiple exchange process is not of major importance for the exchange of hydrocarbons except possibly at rather high temperatures. This conclusion is in excellent agreement with the fact that the exchange process cannot propagate past a quaternary carbon atom in complex hydrocarbons, which was discovered by Burwell and Briggs (43) and will be discussed in Sec. IV,C.

IV. The Exchange of Molecules Possessing a Low Degree of Symmetry In this section we shall consider work on the exchange of propane, i-butane, and a series of larger hydrocarbon molecules with deuterium. Some of the investigations on the exchange of propane and i-butane have been carried out to examine the relative reactivity of hydrogen atoms attached to primary, secondary, and tertiary carbon atoms. The work on the larger hydrocarbons has been developed mainly by Burwell and his colleagues (26, 29, 37, 42, 43). They have shown that comparative studies of the distributions of products obtained from the exchange of a series of hydrocarbons can yield interesting conclusions about the mechanism of the reactions. This approach is complementary t o that adopted by Anderson and Kemball (19, 41 ), who concentrated on the detailed interpretation of the distributions of products obtained in the exchange of individual symmetrical hydrocarbons. A. PROPANE The exchange of this compound has been examined quite thoroughly. Kauder and Taylor ( 11 ) used platinized platinum as a catalyst and Thompson et al. ( 1 3 ) used a cobalt-thoria Fischer-Tropsch catalyst. The reaction has also been studied over a series of metal films by Kemball (18) and more recently over a number of bulk metal catalysts by Addy and Bond (44). Kauder and Taylor (11 ) found that the secondary hydrogen atoms were exchanged more rapidly than the primary hydrogen atoms between 20 and 40" on platinized platinum and also that the total amount of multiple exchange was not extensive. On the other hand, Thompson et al. (13) found that very extensive exchange occurred over cobalt-thoria at 183";the only products observed were 4.7 % C3Ds and 1.4 % C3HDl. Further evidence that the type of exchange of propane depends on the nature of the catalyst was obtained by Kemball (18). Films of tungsten gave distributions of products similar to those obtained with ethane over this metal and the course of the reaction followed Equation ( 5 ) . This in-

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

251

dicated that all the hydrogen atoms in the molecule exchanged a t approximately the same rate over tungsten. The course of exchange over nickel did not follow Equation ( 5 ) , and devising appropriate modifications of this equation and also Equation ( 6 ) made it possible to show that the secondary hydrogen atoms exchange about 10 times as fast as the primary hydrogen atoms on this metal. Further evidence t o confirm this behavior was obtained from the distributions of products during the reaction. There were higher percentages of the dl and dl compounds than would have been expected on the basis of a simple exchange mechanism (Sec. II,D,l). Clearly the i-propyl radical must be formed more readily than the n-propyl radical on a nickel surface. This leads to successive exchange of the two secondary hydrogen atoms accompanied by minor amounts of multiple exchange of the remaining hydrogen atoms. Rhodium films gave extensive multiple exchange with propane, as with ethane, and ds-propane was the most abundant initial product. Similar results were found by Addy and Bond (44) over bulk metal catalysts of palladium, rhodium, iridium, and platinum. The close similarity between the pattern of products obtained with propane and with ethane suggests that the multiple exchange occurs by an a-j3 type of process, i.e., the successive interconversion of adsorbed propyl radicals (n-or i-) and adsorbed propylene molecules. The results found for the exchange of neopentane (Sec. II1,E) indicate that it is unlikely that an a-y mechanism plays any substantial part in the exchange of propane. Any difference in the reactivity of primary and secondary hydrogen atoms is masked on these metals which cause extensive multiple exchange t o occur. In this connection, the results obtained by Addy and Bond ( 4 5 ) for the distributions of propanes formed in the deuterolysis of n-propgl chloride and i-propyl chloride over palladium are interesting. The distributions of propanes from both reactions were almost identical and closely similar to the distribution found for the exchange of propane over the same metal, despite the fact that the i-propyl chloride reacted four times as fast as the n-propyl chloride. Addy and Bond (44) attempted to calculate initial distributions of products for comparison with their experimental results on the assumption that propane molecules were being produced with a random collection of hydrogen and deuterium atoms. I n order t o obtain agreement with the experimental results, they had to assume that three types of reaction were occurring over most of their catalysts. Fraction A of the propane molecules was assumed to be leaving the surface with a random collection of hydrogen and deuterium atoms but rich in deuterium. Fraction B was assumed t o have a random collection containing roughly equal amounts of the two isotopes, and Fraction C was the reverse of Fraction A , rich in hydrogen. This method of calculating distributions of products is based on a ques-

252

C. KEMBALL

tionable assumption, namely, that the propane molecules produced under conditions where an average number of P interconversions between propyl radicals and propylene molecules occurs will acquire a random collection of hydrogen and deuterium atoms. The calculated distributions for ethane with P = 18 or P = 30, shown in Table VII, are obvious illustrations of this point because they differ substantially from random distributions. A better method of calculating distributions of products for the exchange of propane is by means of the appropriate modification of the general theory outlined in Sec. II,F. The agreement between calculated and observed distributions which Addy and Bond obtained is to a certain extent dependent on the choice of five arbitrary parameters to account for seven independent experimental observations. Data on the energies of activation and frequency factors for the adsorption/desorption in the exchange of propane over bulk metals (44) and over films (18) are shown in Tables XI and XII, respectively. B. &BUTANE The results for the exchange of i-butane on metal films (IS) were similar to those for the exchange of propane. Tungsten causes a slight amount of TABLE XI Activation Energies and Frequency Factors for the Exchange of Propane on Bulk Metal Catalysts Catalyst

Rh Pd

logl,A ( A in

Temperature range, "C.

catalyst h.)

60-200 100-200 100-200 100-250

Ir Pt

70reaction/g.

E , kcal./mole 17.3 17.2 17.7 18.2

10.7 9.5 11.8 10.2

TABLE XI1 Activation Energies and Frequency Factors for the Exchange of Propane and i-Bu1an.e on Films Catalyst

Reactant ~~

W W Ni Ni Rh

Temperature range, "C.

E, logloA ( A in kcal./mole mol./sec. cm.*)

~

C A i-CdHlo CsH.s(2'H) i-CdHio(3'H) C:Hs

-82 to -24 -80 to -27 47-0 -47-0 -25-16

-

9.0 7.9 10.4

9.0 13.3

21.8 20.1 21.7 20.5 25.0

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

253

multiple exchange t o occur, and all the hydrogen atoms exchange at approximately the same rate. Nickel exchanges the tertiary hydrogen atom about 10 times as rapidly as the primary hydrogen atoms. Thus, taking into account the relative numbers of the different kinds of hydrogen atoms in propane and i-butane, the relative reactivities over nickel are in the ratio 90:30: 1 for 3":2": 1" hydrogen atoms. Extensive multiple exchange was observed over rhodium with d d - b u t a n e as the most abundant initial product. Activation energies and frequency factors for the adsorption/desorption process are given in Table XII. C. COMPOUNDS CONTAINING QUATERNARY CARBONATOMS Burwell and Briggs (43)examined the exchange of 3,3-dimethylhexane

( I ) and 2,2,3-trimethylbutane (11) with deuterium over a nickel-kieselguhr catalyst in the temperature range, 90" t o 130". The exchange was followed both by measuring the total deuterium in the hydrocarbon by combustion and also by examining the various carbonium ions formed from the molecules by loss of alkyl groups in the mass spectrometer: CHa-CHz-

x""

-CHZ-CH~-CH~

AH3 I

CH3-

8""

-CH-CH3

&Ha AHa

I1

The maximum number of hydrogen atoms exchanged initially in both cases was seven. I n an experiment with I, it was found that the products showed 2.61 % exchange of the propyl group, 1.06 % exchange of the ethyl group, negligible exchange of the methyl groups, and also a negligible fraction of the molecules with exchange in both the ethyl and propyl groups. These results indicated that the exchange process could not propagate past the quaternary carbon atoms in I and I1 but that the tertiary carbon atom in I1 was not a barrier. Similar results were obtained by Rowlinson et al. (29)for the exchange of 3,3-dimethylpentane and 1 ,1-dimethylcyclohexane over various nickel catalysts a t temperatures around 200". Only five hydrogen atoms in the former and 10 in the latter were exchanged in the initial process.

D. OTHERSTRAIGHT AND BRANCHED HYDROCARBONS The exchange of a number of compounds in this category with deuterium has been examined by Burwell and his colleagues. n-Heptane has been exchanged over nickel-keiselguhr (43),reduced nickel oxide (29),a series of nickel catalysts of varying crystallite size (S7), and over palladium supported on y-alumina (42).Less extensive studies were also made with 2,3-dimethylbutane (29,42) and n-hexane (42).

254

C. KEMBALL

The chief conclusions from this work can be summarized briefly: 1. All hydrogen atoms in these molecules are exchangeable and extensive multiple exchange occurs. 2. With nickel catalysts, the extent of the multiple exchange increases with increase of temperature, reduction of the ratio of deuterium to hydrocarbon, and increased size of crystallite. 3. More extensive multiple exchange is found over palladium than over nickel, and the influence of temperature on the extent of the exchange is less marked. These results are in excellent accord with the results described for the simpler hydrocarbons, ethane, Sec. III,B, and propane, Sec. IV,A. The work of Burwell and Littlewood (26') on the exchange of n-hexane with deuterium over chromium oxide gel is important because it is one of the few cases where an oxide catalyst has been used for the exchange of saturated hydrocarbons and deuterium. A simple exchange process occurred at 201" (see Table 111), but at 350" a certain amount of multiple exchange was found, although the extent of this was much less than that observed with metal catalysts. Burwell and Littlewood suggested that, at the higher temperature, the production of the more highly deuterated species might occur through dehydrogenation to hexene followed by the addition of deuterium.

E. (+) 3-METHYLHEXANE Very interesting results have been obtained on the simultaneous exchange and racemization of optically active 3-methylhexane by Burwell and his colleagues (29, 37, 42, @). The nature of the exchange reaction is similar to that for other hydrocarbons just discussed, i.e., the occurrence of extensive multiple exchange with all hydrogen atoms exchangeable. The important point established by this work, however, was that, over a wide range of temperature and with a number of different catalysts, the rates of the exchange and of racemization were approximately the same. The ratio exchangelracemization varies from 2.1 to 1.1 over nickel catalysts (29, 37, 43) and from 1.3 t o 0.9 over palladium catalysts (&), and the highest ratios are always associated with the production in the exchange process of molecules containing only a few deuterium atoms. Burwell and his colleagues suggested that every molecule which undergoes exchange a t the optically active tertiary carbon atom is also racemizing. The departure of the ratio of exchangelracemization from unity was attributed to the formation of molecules which had exchanged hydrogen atoms not in the neighborhood of the optically active carbon atom. They suggested that a special mechanism such as that depicted in Equation (27) must operate to bring about racemization because they held the view that

C-4TALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

255

the ordinary a-/3 multiple exchange process, i.e., successive interconversion hetween adsorbed alkyl radicals and alkene molecules, would not permit racemization t o occur. They pointed out that it should be possible to observe exchange without racemization a t low temperatures because, under these conditions, the normal a-p mechanism should operate in preference t o the special mechanism responsible for racemization. The behavior at low temperatures should be comparable to that found for cyclopentane and cyclohexane (Sec. III,C), where the a-/3mechanism occurs more readily than the “interchange reaction.” Unfortunately, there is difficulty in carrying out simultaneous measurements of exchange and racemization at low temperatures because of the high contact time required t o give measurable racemization. Burwell el al. (4.2) did, however, establish that the ratio of exchange/racemization is 3.1 on a rhodium film at 100”. Anderson (17) has criticized the mechanism for racemization proposed by Burwell and his colleagues and suggested that the formation of an adsorbed alkyl radical might lead t o racemization if the desorption of the radical can occur by a deuterium atom approaching either the ‘(top” or the “bottom” of the adsorbed radical (Fig. 4).Anderson pointed out that, in order to preserve the principle of microscopic reversibility, there would have to be two kinds of adsorption/desorption process. It is tempting to elaborate Anderson’s mechanism by suggesting that adsorption/desorption involving the removal/addition of the hydrogen atom a t the lower part of the molecule takes place by a Langmuir type of mechanism [see Equation (19)] and the other process by an Eley-Rideal mechanism [see Equation (2011.

More experimental evidence is required on this interesting topic because neither of the suggested mechanisms for racemization seems entirely satisfactory. That proposed by Burwell involves the existence of adsorbed

M

I M

FIQ.4. Anderson’s mechanism for racemieation.

256

C. KEMBALL

species of an unstable type. The drawback to Anderson's mechanism is that it implies that two different kinds of adsorption/desorption process occur a t approximately the same rate on a number of different metals.

F. BRANCHED RING COMPOUNDS Some very striking results have been obtained by Burwell et al. (42) for the exchange of methylcyclopentane with deuterium a t 50" over palladium on y-alumina. Maxima in the distribution of products a t an early stage in the reaction were observed at the d4, de , and dlz compounds. The behavior is analogous to that found with cyclopentane and cyclohexane except that the hydrogen atoms in the molecule consist of a set of four and a set of eight instead of two equal sets each of which is readily exchangeable. A diagram of the molecule is shown in Fig. 5, and there can be little doubt that the first set consists of the four hydrogen atoms cis to the methyl group and that the second set consists of the remaining hydrogen atoms. The pattern of products thus provides supporting evidence that the normal a-p multiple exchange process involves an eclipsed 1,&diadsorbed alkane (Fig. 3b) and not the staggered conformation (Fig. 3a). The production of the d12-compoundindicates that an interchange reaction must also be operating to allow the exchange to propagate from one set of hydrogen atoms t o the second set. Earlier work on the exchange of methylcyclopentane and methylcyclohexane over nickel catalysts at 150°-200" had not shown any discontinuities in the distribution pattern of the products ( 2 9 ) . There was a uniform rise up to a maximum for the fully deuterated species. This is not surprising, since similar behavior was noted with cyclopentane and cyclohexane under the same conditions. At these high temperatures the interchange reaction occurs sufficiently readily to mask any division of the hydrogen atoms into sets. A maximum of two hydrogen atoms was found to be exchanged initially I

I

I FIQ.5. Methyl cyclopen tane.

\

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

257

FIG.8 . Bicyclo[2.2.l]heptane.

in bicyclo [2.2.1] heptane over palladium on y-alumina by Bunvell (42). A diagram of the molecule is shown in Fig. 6, and it can be seen that the four pairs of hydrogen atoms on the two ethylene bridges are rigidly held in eclipsed conformities. The authors suggested that the two hydrogen atoms exchanged initially were one of these pairs and that this behavior was further evidence for the stereochemistry of the 1,2-diadsorbed alkane involved in a-p multiple exchange.

V. Some Results with Unsaturated Molecules In this section we shall consider a few results of work on the exchange of unsaturated molecules under conditions where the rate of addition of hydrogen or deuterium is negligible compared with the rate of exchange; i.e., the molecules are behaving in a way similar to the behavior of saturated molecules. There is not sufficient space to discuss the important and extensive work involving the simultaneous exchange and deuteration of unsaturated molecules, but this subject has been reviewed by Taylor (16). The exchange of benzene and deuterium has been examined in a static system by Anderson and Kemball (46'). The exchange reaction was found to be too fast to be measured over nickel films a t -45" and over irons films a t 0", although the rates of deuteration under these conditions were slow and measurable. Silver films catalyzed exchange a t 293" to 373" without causing deuteration, but the absence of deuteration was expected at these high temperatures with the low pressures of reactants used because of thermodynamic considerations. Palladium and platinum gave rise t o exchange and deuterationin the temperature ranges 18"to 80"and - 44"to 0",respectively,

258

C. KEMBALL

but there was evidence that the two reactions were completely independent. The distributions of the cyclohexanes formed at any stage in the course of the reactions were consistent with the hypothesis that deuteration involved pure addition without exchange. The initial distributions of the deuterobenzenes resulting from the exchange reaction over these two catalysts were similar t o the initial distributions of the dl- t o da-cyclohexanes found in the exchange of cyclohexane over the same metals (41) except that the amounts of dl-benzene were found to be greater than the corresponding amounts of dl-cyclohexane. The distributions of benzenes obtained over silver films were similar to those given by palladium and platinum. Comparison between observed distributions and distributions calculated by the appropriate modification of the theory outlined in Sec. II,F suggested that an 0-8 multiple exchange process was operating. Anderson and Kemball formulated the mechanism for this as interconversion between adsorbed phenyl radicals and adsorbed phenylene biradicals. It should be noted that this mechanism does not involve the destruction of the resonance stability of the benzene ring, and therefore it is consistent with the fact that the benzene molecule appears to undergo exchange as though it were a saturated molecule. Some important observations have been made by Flanagan and Rabinovitch ( 4 7 ) on the exchange and isomerization of Irans-ds-ethylene and on the exchange of do-ethylene and d4-ethylene over nickel wire. The reactions were carried out in the absence of added hydrogen or deuterium and very little self-hydrogenation leading to the formation of ethanes was observed. The results showed that the reactions were occurring by a simple or stepwise mechanism and that both the exchange and the isomerization involved the formation of adsorbed ethyl radicals. The formation of these radicals implies that a certain amount of dissociation adsorption of the ethylene occurs, but Flanagan and Rabinovitch stress the point that a small amount of chemisorbed hydrogen or deuterium arising in this manner can catalyze the entire reaction. It is interesting that the same adsorbed hydrocarbon species, i.e., ethyl radicals and ethylene molecules, are found to be important in these reactions as well as in the exchange of ethane and deuterium. Flanagan and Rabinovitch were able to establish another point of general interest. They deduced from the relative rates of exchange and isomerization of trans-dz-ethylene that there is an isotope effect in the rupture of the carbon-"hydrogen" bond when adsorbed ethyl radicals dissociate to form adsorbed ethylene molecules. The ratio of the rupture probabilities of C-H and C-D decreased from 15.9 a t -78" to 1.4 at 429". More evidence of this kind would obviously be valuable because it suggests that some revision may be necessary of the theory for calculating initial distributions of

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

259

products in exchange reactions (Sec. I1,F) in which isotopic effects were ignored.

VI. Concluding Remarks It is convenient to summarize the conclusions about the catalytic activity of the metals for breaking C-H bonds and about the mechanism of the exchange of different types of hydrocarbons in a single section because the two topics are closely related. The simplest method of estimating activities of different metals for the same reaction is to compare the temperature or range of temperatures required to bring about similar rates of reaction on each of the metals. The comparison of rates of reaction at a single fixed temperature is less satisfactory because it often involves considerable extrapolation of the experimental data and the unjustifiable assumption that the activation energy remains constant over a wide range of temperature. A complete account of catalytic activity covering the variation of both activation energy and frequency factor on different catalysts cannot be attempted until considerably more information is available about the relative strengths of adsorption of the reactants and the various species formed from them. The following general points may be made about the catalytic activity of metals for the exchange of hydrocarbons: 1. Metal films are usually more active than bulk metal catalysts. 2. The reaction which has been studied over the greatest range of metals is the exchange of ethane and the order of activity of the different metals (Table VIII) is W(Mo, Ta)

> Rh > Pt(V, Cr, Zr) > Pd

Nickel shows some activity at low temperatures approximately equivalent to the activity of rhodium, but normal catalytic behavior with a constant activation energy is observed only at substantially higher temperatures similar to those necessary for activity with palladium. 3. The pattern of activity of the metals for the exchange of other hydrocarbons appears to be similar to that established for the exchange of ethane. For the exchange of methane (Table V), nickel behaves like palladium, but for the exchange of the cyclic hydrocarbons (Table X), nickel exhibits activity similar to that of rhodium. 4. Although the data for complete comparisons are not available, there is a constant pattern for the relative ease of exchange of different hydrocarbons on each metal. The hydrocarbons may be grouped in descending order of reactivity as follows: a. cyclic hydrocarbons, cyclopentane being slightly more reactive than cyclohexane,

260

C. KEMBALL

6. ethane, propane, and i-butane, the larger molecules being slightly more reactive than ethane, c. neopentane, d. methane. Although it has been stressed that the rate of an exchange reaction is governed by an adsorption/desorption process, it is probable that the main single factor governing the reactivity of different hydrocarbons on different metals is the ease of chemisorption of the hydrocarbon on the catalyst. Evidence for this may be found in the work of Wright et al. (31) and of Trapnell ( 4 8 ) .Anderson (1'7) has shown that there is a correlation between theeficiency of metals for the exchange of ethane and the strength of the bonds between the atoms of the metal. As the strength of the adsorption of hydrogen does not vary greatly on different transition metals, provided the surface is fairly well covered, it is possible that the ease of adsorption of a hydrocarbon molecule on different metals depends chiefly on the strength of the metal-carbon bond which is formed. Anderson's correlation may result from parallel trends in the strength of metal-metal and metal-carbon bonds. A point which has not been examined is the nature of the surface during exchange reactions carried out at high temperatures such as those required for the exchange of methane. Surface carbides may be formed under these conditions. The inactivity of iron films and the comparatively small activity of cobalt films at 300" for the exchange of ethane (19) may possibly be due to the tendency of these metals to form not only surface but also bulk carbides. The main facts about the mechanism of the exchange reactions of saturated hydrocarbons are as follows: 1. Simple exchange is found only with methane and with molecules like neopentane which have isolated methyl groups attached to quaternary carbon atoms. These compounds are less reactive than hydrocarbons which have two or more adjacent primary, secondary, or tertiary carbon atoms. 2. The most prominent type of multiple exchange is the a-P type, and it is probable that this involves interconversion between adsorbed alkyl radicals and eclipsed 1,2-diadsorbed alkane molecules. Any molecule capable of undergoing this type of exchange is usually fairly readily exchangeable. 3. The a-a type of multiple exchange always requires a greater activa t'ion energy than simple exchange or a-/3multiple exchange with the same hydrocarbon. It follows that it is more difficult to remove a second hydrogen atom from one carbon atom than t o dissociate the first hydrogen atom. 4. The a-y type of multiple exchange appears to be of little importance except a t high temperatures.

CATALYTIC EXCHANGE OF HYDROCARBONS WITH DEUTERIUM

261

5. The character of the exchange reaction of most hydrocarbons is similar on films and on bulk catalysts of the same metal. 6. The extent of each kind of multiple exchange depends on the nature of the catalyst, but the same metal shows comparable behavior with different hydrocarbons. Palladium and rhodium exhibit extensive a-/3 exchange, but tungsten does not. Rhodium and, to a lesser extent, nickel exhibit extensive a-a exchange, but palladium is a poor catalyst for this type of exchange. Anderson (i7)and Addy and Bond (44) have pointed out that the extent of the a-/3 type of exchange appears t o increase with the % d-bond character of the intermetallic bonding of the metal (cf. Pauling, 43). This correlation may result from the dependence of the a-/3process on the interatomic spacing of the metals. The feature of the study of the exchange reactions of hydrocarbons which is of the greatest interest t o those concerned with more complex catalytic reactions is the knowledge gained about the nature and reactivity of the radicals formed on the surface of catalysts. Such information will lead ultimately t o a better understanding of the mechanism of more complex catalytic reactions. It is important t o stress that the amounts of the adsorbed species responsible for the reaction may be small. A species covering only one-thousandth part of surface may play an important part in a reaction, provided that it is formed readily and is also reactive. This point means that the type of information gained by the study of exchange reactions may be a better guide to the important species taking part in the mechanism of catalytic reactions than studies on the adsorption of individual, or even mixed, reactants. Conversely, it follows that the kinds of adsorbed species shown to exist by an examination of exchange reactions are not the only types of species present on the surface.

REFERENCES 1 . Taylor, H. S., Am. Scienlisl 34, 553 (1946). 8. Morikawa, K., Benedict, W. S., and Taylor, H . S., J . Am. Cheni. SOC.68, 1445 (1936). 3. Morikawa, K., Benedict, W. S., and Taylor, H. S., J . Am. Chem. SOC.68, 1795 (1936). 4 . Morikawa, K.,Trenner, N . R., and Taylor, H. S., J . Am. Chem. SOC.69, 1103 (1937). 6. Farkas, A., and Farkas, L., Trans. Faraday SOC.36, 917 (1939). 6. Farkas, A., Trans. Faraday SOC.36, 522 (1940). 7. Field, F . H., and Franklin, J. L., “Electron Impact Phenomena” p. 214. Academic Press, New York, 1957. 8 . Parravano, G., Hammel, E. F., and Taylor, H. S., J . Am. Chem. SOC.7 0 , 2269 (1948). 9. Wright, M. M., and Taylor, H. S., Can. J . Research B27, 303 (1949).

262

C. KEMBALL

10. Turkevich, J., Bonner, F., Schissler, D., and h a , A. P., Di8CUS8iOflS Faraday Soc. 8, 352 (1950). 11. Kauder, L. N., and Taylor, T. I., Science 113, 238 (1951). 18. Kemball, C., Proc. Roy. Soc. A207, 539 (1951). 13. Thompson, S. O., Turkevich, J., and h a , A. P., J. A m . Chem. SOC.73, 5213 (1951). 14. Taylor, T. I., and Dibeler, V. H., J. Phys. Chem. 66, 1036 (1951). 16. Dibeler, V. H., and Taylor, T. I., J. Chem. Phys. 16, 1008 (1948). 16. Taylor, T. I., in “Catalysis” (P. H. Emmett, ed.) Vol. V, p. 257. Reinhold, New York, 1957. 17. Anderson, J. R., Rev. Pure and Appl. Chem. 7, 165 (1957). 18. Kemball, C., Proc. Roy. SOC.A W , 377 (1954). 19. Anderson, J. R., and Kemball, C., Proc. Roy. SOC.A223, 361 (1954). do. Harris, G. M., Trans. Faraday SOC.47, 716 (1951). 81. Frankenburg, W. G., in “Catalysis” (P. H. Emmett, ed.) Vol. 111, p. 171. Reinhold, New York, 1955. 88. Bokhoven, C., van Heerden, C., Westrik, R., and Zwietering, P., i n “Catalysis” (P. H. Emmett, ed.) Vol. 111, p. 265. Reinhold, New York, 1955. 83. Temkin, M. I., and Pyzhev, V., Acta Physicochim. U.S.S.R. 12, 327 (1940). 84. Kemball, C., Proc. Roy. SOC.A214, 413 (1952). 86. Matsen, F. A., and Franklin, J. L., J . A m . Chem. SOC.72,3334 (1950). 86. Burwell, R. L., Jr., and Littlewood, A. B., J. A m . Chem. Soc. 78, 4170 (1956). 87. Kemball, C., Trans. Faraday SOC.60, 1344 (1954). 88. Kemball, C., and Wolf, F. J., Trans. Faraday SOC.61, 1111 (1955).

99. Rowlinson, H. C., Burwell, R. L., Jr., andTuxworth, R. H., J. Phys. Chem. 69,

225 (1955). SO. Kemball, C., Proc. Roy. SOC.A217, 376 (1953). 31. Wright, P. G., Ashmore, P. G., and Kemball, C., Trans. Faraday Soe. 64. 1692 (1958). 38. Miyahara, K., J. Research Znsl. Catalysis, Hokkaido Univ. 4, 177 (1957). 33. Markham, M. C., Wall, M. C., and Laidler, K. J., J. Phys. Chem. 67, 321 (1953). 34. Beeck, O., Smith, A. E., and Wheeler, A., Proc. Roy. SOC.A177, 62 (1940). 36. Beeck, and Ritchie, A. W., Di8CU8SiOfl8 Faraday SOC.8 , 159 (1950). 36. Sachtler, W. M. H., Dorgelo, G., and van der Knapp, W., J. chim. phys. 61, 491 (1954). $7. Burwell, R. L., Jr., and Tuxworth, R. H., J. Phya. Chem. 60,1043 (1956). 38. Miyahara, K., J . Research Znst. Calalyeis, Hokkaido Univ. 4, 143 (1956). 39. Kemball, C., J . Research Znsl. Catalysis, Hokkaido Univ. 4, 222 (1957). 40. Jenkins, G. I., and Rideal, E. K., J. Chem. SOC.p. 2490 (1955). 41. Anderson, J. R., and Kemball, C., Proc. Roy. Soc. A l l , 472 (1954). 0.Burwell, R. L., Jr., Shim, B. K. S., and Rowlinson, H. C., J. Am. Chem. SOC.70, 5142 (1957). 43. Burwell, R. L., Jr., and Briggs, W . S., J. Am. Chem. SOC.74, 5096 (1952). 4.4. Addy, J., and Bond, G. C., Trans. Faraday SOC.63, 368,383,388 (1957). 4.5. Addy, J., and Bond, G. C., Trans. Faraday SOC.63, 377 (1957). 46. Anderson, J. R., and Kemball, C., Advances in Catalysis 9, 51 (1957). 47. Flanagan, T. B., and Rabinovitch, B. S . , J. Phys. Chem. 60,724, 730 (1956). 48. Trapnell, B. M. W., Trans. Faraday SOC.62, 1618 (1956). 49. Pauling, L., Proc. Boy. Soc. A196, 343 (1949).

o.,

lmmersional Heats and the Nature of Solid Surfaces J. J. CHESSICK

AND

A. C. ZETTLEMOYER

Surface Chemistry Laboratory, Lehigh University, Bethlehem, Pa. page

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

263

111. Wetting and Adsorption Thermodynamics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....................... A. The Wetting of Solids by Liquids.

268

C. The Surface Energy of Solids. . . .

..........................

273

A. Homogeneous and Heterogeneous Polar Solids

V.

VI.

VII. VIII.

C. Clay-Water Systems D. Natural and Modified Organic Fibers.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 E. Utility of Heat-of-Immersion Values for Samples Equilibrated a t High Equilibrium Pressures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 Wetting by Organic Liquids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 A. Comparative Immersional Heat Values for Hydrophobic and Hydro280 philic Solids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Interaction of Polar Solids with Organic Liquids. . . . . . . C. The Interaction of Hydrophobic Solids with Organic Liquids. . . . . . . . 283 Average Polarity of Solids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 A. The Rating of Solid Surfaces by Immersional Heats in Organic Liq284 uids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Use of Water to Rate the Polarity of Solid Surfaces. . . . . . . . . . . . 286 C. The Degree of Surface Hydrophilicity of Solids.. . . . . . . . . . . . . . . . . . . . 287 D. The Immersion of Solids in Liquid Nitrogen.. . . . . . . . . . . . . . . . . . . . . . . 288 Site Energy Distribution.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 tting D a t a . . . . . . . 288 A. Heats of Adsorptio . . . . . . . . . . . . . . . . . . 291 Solution Adsorption. . A. Preferential Adsorption from Solution by Polar and Nonpolar Solids. . . 291 B. The Wettability of Surfactants in Aqueous Solutions.. . . . . . . . . . . . . . . 294 References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296

1. Introduction In interpreting chemisorption and catalysis on solid surfaces, interest largely centers on the intimate nature of the forces emanating from the surface. Practically always involved is a heterogeneity of sites for adsorp203

264

J. J. CHESSICK AND A. C. ZETTLEMOYER

tion due to a variety of causes. Of course, the adsorption isotherm when it can be measured reflects the specific interactions which occur. In physical adsorption, the adsorption energies may often be elucidated by the calculation of heats of adsorption from isotherms at several temperatures or by direct calorimetric measurements. I n chemisorption, on the other hand, heats of adsorption may be impossible to obtain either calorimetrically or from isotherms a t different temperatures; this is so for adsorbate-adsorbent systems having very long equilibrium times or very low equilibrium pressures. Moreover, the catalyst or other solids of interest may be poor conductors of heat; or, more important, nonspecific adsorption may occur, particularly as gas is admitted, if the temperature of the experiment is low and sites of high energy are present. These experimental difficulties are principally responsible for the lack of important thermodynamic data for many adsorbate-catalyst systems. Most of these problems are surmounted by the heat-of-immersion technique, except where heat effects are slow, or even when the adsorbate can only be studied as a liquid. Twin calorimeters can be used to get satisfactory results for some slow reactions with surfaces. Increasing amounts of adsorbate may be added to the surface and sealed off indefinitely if necessary to wait for equilibrium. The equilibrium pressures need not be known if the amounts adsorbed can be determined gravimetrically or otherwise. If wetting measurements are carried out near room temperature, the problem of nonspecificity of adsorption can be overcome by saturating the sample with adsorbate and then removing known amounts of the chemisorbed species at increasing temperatures under vacuum. An alternate method is to adsorb at elevated temperature prior t o immersion, which is usually carried out near room temperature. Then, the heats of immersion of the series of prepared samples in the liquid used as the adsorbate spell out the site energy distribution. Of the various calorimetric approaches to the measurement of interaction heats, the measurements of heats of immersion of fine powders in liquids are the simplest and most highly developed both from the technical and theoretical points of view. The present discussion of the theory and applications should point the way to broader uses and likely developments in the immediate future. Heats of immersion* are usually small exothermic quantities of the order of magnitude of heats of adsorption, but are quite significant. For example, the heat of immersion into water of a clean, polar, high-surface-energy solid like an insoluble inorganic salt interacting physically may be -400 ergs pcr sq. cm.; that of a clean, nonpolar, low-energy solid-like graphite may

* Harkins introduced the term “heat of emersion” to avoid carrying the negative sign.

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

265

be -30 ergs per sq. cm. The heat effect is less than the total surface energy of the solid, but they are related; the heat of immersion is also related to the energy of adhesion. These interrelations will be discussed in Sec. 111. The phenomenon of the evolution of heat on wetting has been known since the report by Pouillet (the “Pouillet effect”) in 1822 (an even earlier observation was recorded by Leslie in 1802), and 27 more references on the subject are to be found in the literature up to 1926. By the 1930’s rather good calorimeters were developed and attention was being given to outgassing techniques prior to immersion. Yet, until the advent of the B.E.T. method (1936-1940) for estimating specific surface areas, the heat effects could not be put on a sound basis. The first paper which did so was that by Boyd and Harkins (1 ) . Thereafter, the Harkins, Bartell, and Il’in groups made significant contributions. Nevertheless, the influence of immersional calorimetry in surface chemistry waned. One of the chief reasons was the difficulty of manipulating the ever more complex calorimeters. In the 1950’s the confluence of two factors gave new impetus to the field: one was the appearance of simple calorimeters based on thermistors as the temperature sensing elements, and the other was the impact of other surface chemistry developments which pointed the way to what could be achieved through heat-of-immersion measurements. The number of purposes to which immersional studies can be put is on the increase. The list which follows may suggest other applications: 1. To obtain fundamental information concerning interactions of surfaces with adsorbate molecules, particularly where other techniques are not suitable. Included are organics, molten metals, or solutions. It should be stressed now that heat-of-immersion values for a given solid in a variety of liquids can be misleading and that heat values as a function of the amount of preadsorbed wetting liquid are usually necessary. 2. To rate the polarity of solid surfaces from their heats of immersion in simple organic liquids having different peripheral dipole moments. For the first time, this technique allows an experimentally derived number to be put on the average force field emanating from solid surfaces. 3. To measure the site energy distribution or other surface properties of powders by measuring heats of immersion as a function of the amount of preadsorbed wetting liquid. Heats of immersion of the partly covered surfaces reveal the site energy distributions. For acid sites on cracking catalysts, for example, adsorbates of different basicity can be used to develop a topographical map of the surface activity. 4. To study the nature and extent of solution adsorption. A small amount of work has been done in studying the interfacial region at solid-solution interface by heat-of-immersion measurements. Few other methods show

266

J. J. CHESSICK AND A . C. ZETTLEMOYER

promise for examining the interactions of solutions of two or more components with solid surfaces where the adsorption of both components must be taken into account. 5 . To rate the wetting of powders by water; i.e. their hydrophilicity. A definite value can be assigned even in cases where the contact angle is zero.* 6 . To rate the wetting tendency of surfactants for hydrophobic surfaces. A graphitic powder, for example, with its low heat of wetting in water, yields much higher heat effects if immersed in solutions of surface active agents. Heats of dilution and of demicellization can be taken into account, if desired, to arrive directly a t energies of interaction. Each of these areas of application of heat-of-immersion measurements will be discussed in that which follows.

It. Experimental For reliable measurements of heats of immersion with simple calorimeters, the surface area of the powder should be at least 5 sq. m./g. and might need to be larger depending on its density, bulk density, and polarity. With modern microcalorimeters, on the other hand, the surface area of the solid need be only a fraction of a square meter per gram. Details for the construction and operation of calorimeters are given in basic references by Boyd and Harkins ( I ) , by Zettlemoyer el al. ( 2 ) and by Berghausen (S),and will not be repeated here. Instead, the general characteristics of the different types and corrections to be applied to the measurements will be discussed. The Boyd and Harkins reference might be considered the fist careful calorimetric work taking into account the special problems of immersional calorimetry; it was also the first such work in which surface areas could be assessed (the B.E.T. method) as a necessary auxiliary for putting values on a unit area basis; temperature changes were measured with a 36-junction thermocouple and a White double potentiometer with a sensitive galvanometer, Zettlemoyer et al. provided a detailed study of the use of a thermistor connected to a Mueller bridge as the temperature-sensing element in immersional calorimetry. Thermistor calorimeters have since gained wide use because of the simple construction and operation they make possible. Brief

* The term “wetting” has been much maligned. Sometimes a contact angle of 90” has been called the dividing line between wetting and non-wetting. Sometimes the dividing line bas been put at 0”. Of course, even among a group of systems having iero contact angles, the energy of interaction and thus the degree of wetting can vary. Wetting is not a rcgo”or “no-go” proposition, but a definite value can be assigned from the heat of immersion per unit surface area.

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

267

mention of such use for thermistors was given first by Hutchinson ( 4 ) and was also expanded later by Bartell and Suggitt (6) to twin calorimetry. Instead of B resistance bridge, the latter workers used two thermistors in opposite arms of a Wheatstone bridge with a White double potentiometer across it. Pierce, Mooi, and Harris ( 6 ) , with a similar single-calorimeter ) that a student bridge with design to that of Young and Smith ( 7 ~ )showed a sensitive galvanometer with two thermistors in opposite arms of the bridge circuit or with a high potential drop across the bridge could suffice for some purposes. A limitation in the behavior of thermistors is imposed by the electrical energy which must be introduced during operation; for this reason, the thermistors operate at a slightly higher temperature (ca. 1 X lo-' deg.) than the immersional liquid in the calorimeter. Hutchinson and Manchester ( 7 a ) and Kiselev et al. ( 7 b ) also recently designed calorimeters for wetting measurements based on platinum resistance thermometers. Twin differential microcalorimeters have been described by Berghausen el al. (3), by Hackerman (8),and by Whalen and Johnson (9). Hackerman employs thermistors, whereas the other two are based on thermocouples and in addition are run adiabatically. These calorimeters appear to have about 10 times the sensitivity of simpler designs, but for many purposes the large additional difficulties in design, construction, and operation do not seem to be warranted. Berghausen and coworkers, however, have shown that they can estimate slow heat evolutions, after the first few minutes, due to surface reactions. In developing a suitable calorimeter, factors of primary importance include: a steady and sufficient stirring rate, accurate measurement of temperature changes, accurate measurement of the electrical energy equivalent, sufficiently rapid attainment of thermal equilibrium, attainment of suitable rating or steady-state periods, and small changes in the latter due to the increase in viscosity when the powder is broken into the liquid. A point of considerable dispute in the literature is the correction to be made for sample bulb breaking. Contributions could arise from: residual strains in the glass, lowering of the breaker rod, the energy imparted by the plunger, the pAv work as the bulb space is filled with the immersional liquid, the condensation of any vapor enclosed in the bulb, and any evaporation of liquid into the increased space above the immersional liquid. Investigators report exothermic corrections of 0.3 to 1.6 joules (8, 6),corrections which can be accounted for by considering only the pAv work (S), or even corrections which are vanishingly small (9). Some of this controversy can be resolved by considering the magnitudes of the corrections for water us. organic liquids. For example, if 0.05-ohm change in resistance

268

J. J. CHESSICK AND A. C. ZE'M'LEMOYER

is sought in a thermistor calorimeter, then a correction of 0.0005 ohm for breaking a bulb into 300 ml. of water is practically negligible. On the other hand, when the calorimeter contains an organic liquid, the heat capacity of the system is much lower, so that for the same resistance change the correction may amount to 0.007 ohm. In any case, corrections should be estimated directly on the calorimeter and system employed. The measurement of the heats of wetting of solid powders is complicated by the needs for using a large enough sample to give a reasonably high area for wetting, for some mechanism of introducing the solid powder into the wetting liquid, and for a sufficient volume of liquid to insure complete wetting with little increase in the viscosity of the wetting liquid by the suspended particles. In spite of the above mentioned difficulties, sensitivities of the order of 0.01 cal. can be obtained in heats of wetting measurements. The precision and accuracy of such measurements are determined mainly by the surface area of the powder, as mentioned earlier, its surface energy and cleanliness, and the polarity and purity of the wetting liquid.

111. Wetting and Adsorption Thermodynamics A. THEWETTINGOF SOLIDSBY LIQUIDS Contact angle, reversible work of adhesion, or heat of wetting values all are physical quantities connected with the particular wetting property of a given solid-liquid system. Each, however, is limited in application and sometimes difficult to obtain experimentally. For example, contact angle measurements are useless to make distinctions between solid-liquid interactions when the liquid contact angles are all zero. Too often, work-ofadhesion values must be calculated from adsorption data difficult to obtain with precision, particularly at very low equilibrium pressures or pressures near saturation. Heats of immersional wetting, although limited to powders with sufficient area, are both easy to measure and applicable in principle to all solids. The types of film formation on liquid surfaces either by adsorption or spreading have been defined and systematized by Harkins and co-workers (10).The same organization does not exist for the wetting of solid surfaces by liquids. There exist two types of spreading of liquids on insoluble solids which resemble, respectively, duplex and nonduplex spreading in liquidliquid systems described by Harkins. Spreading wetting occurs with most liquids on solids of high free surface energy. The spreading of a liquid into a stable, multimolecular film on the solid has been observed most often and is analogous to the initial duplex spreading of a liquid into a multimolecular film on another liquid. Such a film on liquids is not stable, however, and

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

269

reverts to a monolayer and lens for the numerous systems studied (10). Nonduplex spreading of a liquid into a monolayer on another liquid has its counterpart in liquid-solid systems, where an adsorbed, oriented, and close-packed monolayer is formed upon which the liquid cannot spread. This situation undoubtedly develops when certain n-butyl monofunctional derivatives adsorb on polar solids such as rutile (TiOn) or calcium fluoride (Sec. V,B) or when monolayers of paraffin derivatives are deposited on glass or platinum as demonstrated by Zisman and co-workers (11) . Hare and Zisman (12) have termed liquids which cannot spread over their own adsorbed monolayers “autophobic.” Nonspreading results when a liquid of sufficiently high surface tension is placed on a low-free-surface-energy solid. Such solids, designated lowenergy solids ( 1 3 ) , are generally organic compounds, waxes, or polymers. Nonspreading liquids placed on such surfaces form drops with measurable contact angles. The distinction between these nonspreading systems and solid-autophobic liquid systems can be subtle, since drops with finite angles are found on both. Present evidence suggests that a solid surface in equilibrium with a nonspreading liquid drop at its Saturation pressure is not covered by a continuous, adsorbed film (Sec. IV,B) . Indeed, the adsorptivity of vapors on low-energy organic surfaces can often be negligible (13). On the other hand, autophobic liquids on contacting high-energy solids essentially create new nonwettable surfaces. Free energy values are most suitable for describing the wettability of a solid by a liquid. Actually, the initial spreading coefficient, SLV.lS”

=

YS4YSL

+

YLVO)

(1)

where ysO, y L v o, and ysL are the surface free energies of the solid in vacuo, of the liquid and of the solid-liquid interface, respectively, is just such a measure, since it is the free-energy change for extending a film 1 sq. cm. over the solid. When a duplex film forms, the spreading coefficient is equal to the film pressure a t saturation, p e , according to the expression: S L V ~ I=S ~Q c

Ysv.

(2) where (pS is equal to the difference in the surface free energy of the solid, yso, and of the solid covered with an adsorbed film, y s v o , in equilibrium with its saturated vapor. This equation is not valid for nonduplex spreading, since in this case y s v a for the monolayer is not equal to the term (ysL y L v o )Nevertheless, . q e is a good measure of wettability for both systems and can be calculated from adsorption data by the Gibbs equation. Unfortunately, accurate values for qe are difficult to obtain because of the need for accurate low-pressure adsorption data and because capillary condensation, particularly between particles of small diameter, cannot be = 7s.-

+

270

J. J. CHEBSICK AND A. C. ZETTLEMOYER

avoided at high-equilibrium pressures unless special precautions are taken (14). For nonspreading systems, the final spreading coefficient is related to the equilibrium contact angle by the equation SLV.,*V.

=

YLVO

cos 8 -

YLVO

(3)

Hence, contact-angle measurements are suitable for rating the wettsbility of low-energy solids. Such measurements for liquid drops on low-energy solids of negligible area have proved to be a useful technique in the hands of Zisman and collaborators (11, 12, 13). Aside from advances in the understanding of wetting of such surfaces in terms of molecular structure, these workers have prepared deposited monolayer surfaces whose surface chemical stoichiometry, at least, is much better defined than any of the solids normally studied by adsorption techniques. Nevertheless, contact angles as a measure of wettability are limited to low-area and low-energy solids. Powdered samples, which form the bulk of the systems investigatcd by adsorption techniques, are excluded. Low-energy powders have sometimes been compressed into buttons, which after cleaning and polishing are apparently suitable for contact-angle measurements. This was done for the first two substances in Table I. Immersional heat values are listed in Table I for a variety of solids from low-energy Teflon to the high-energy solids, CaFl and rutile, TiOz. The hydrophobic solids are characterized by higher heat values for immersion in the organic liquids then water; the reverse is true for the hydrophilic solids. Immersional heats can be a direct measure of the wettability of solids, although care must be exercised in their use. Chemisorption of water or the organic wetting liquid, as well as sintering or other structural changes of the solid during sample preparation, is one of the factors which must be recognized when comparing the wettability of solid surfaces. Even in the absence of such complications, heat of immersion can be misleading. For example, nitrogen and water adsorption measurements (16) show that the surface of Aerosil, a h e silica, is predominantly hydrophobic, although it would be classified weakly hydrophilic on the basis of the measured heat values. Nevertheless, the major differences in the surface properties of these solids are revealed by the heat values listed in Table I. The data of Table I1 for the hydrophilic solid anatase (TiOS) and a hydrophobic graphite reveal the parellelism between the free-energy data for these systems and heat data for comparable systems of Table I. This parallelism is certainly not general. A more fundamental discussion of solid-liquid interactions and a method for estimating the surface polarity of solids are given in Sec. V and VI.

271

HEATS O F IMMERSION OF SOLIDS I N LIQUIDS

TABLE I Heals of Immersion of Solids in Liquids at 96"

Solid

I

References

Water

-

Teflon (poly CF,) Graphon (C) Aerosil (SiOl) Calcium fluoride (CaFd Rutile (TiOl)

I

n-butyl n-heptane chloride

I

angle, water

32 165 463

56 106 228 430

58 112 118 126

110" 82"

550

502

144

0

0 0

TABLE I1 Spreading Pressure Values, qr ,for the Adsorption of Vapors 012 Anatase and Graphite' P.

(ergs/cm.*)

Liquid

Water n-propyl alcohol n-butane n-heptane

Anatme (TiOd

Graphite

190 108 43

ca. 19

46

69

-

0 Harkins, W. D., "The Physical Chemistry of Surface Films," pp. 280 ff. Reinhold, New York, 1952.

B. THE THERMODYNAMICS OF ADSORPTION AND IMMERSIONAL WEWING In most instances, heats of immersion have been obtained by immersing an evacuated (clean) solid into a carefully purified liquid. More informative data are obtained by comparing the immersion of samples of the powder possessing known increasing amounts of preadsorbed wetting liquid. The small exothermic heat values are generally reported on a per unit area basis except for porous solids, swelling systems which have large internal areas, or agglomerates of small particles where capillary condensation can occur between primary particles. In such cases, it is difficult to establish the amount of the available surfaces remaining at any time as the amount preadsorbed increases.

272

J. J. CHESSICK AND A. C. ZETTLEMOYER

If the surface area Z of a solid is known, the heat of immersion can be put on unit area basis and

where eso and esL are energies of the solid surface and the solid-liquid interface, respectively. The energy changes are essentially the same as changes in enthalpy because volume changes during the immersion process are usually negligible. The energy of adhesion of the liquid to the solid is defined as eA(sL)

=

+ (es- -

eLV0

(5)

e8L)

or eAtsL) =

- hrcsL)

eLvo

(6)

Therefore, besides heats of immersion, only the surface energy of the liquid is required to obtain the energy of adhesion. Adsorption and immersion processes can also be related through the heat effects. The integral heat of adsorption of N A moles of adsorbate in the vapor state at equilibrium pressure p and temperature T is

-

(7) where h z ( S I L ) is the heat liberated on immersion of a solid precovered with N A adsorbed molecules at a surface concentration r = N A / Z and AHL is the molar heat of liquefaction. The net heat of adsorption hada.

=

[~z(sL)

[hrw

~z(s~L)]

r A H L

- hZ(S/L)l

is equal to the difference in the integral molar energy of the liquid and the energy En' of the adsorbate-solid system. Then [ h r c s ~)

hz(sfL)]

=

Ahads.

-rAHL

=

I'(E4' - E L )

E L

(8)

The term h z ( B ~can L ) be less than, equal to, or greater than the enthnlpy of the liquid surface h L v o a t surface coverages beyond the monolayer. Too often the values of the heat of immersion hz(sL)of a solid into a variety of liquids are compared directly when the additional knowledge of the value of h z ( s f L ) at a specified surface coverage is necessary for proper interpretation of wetting phenomena (see Sec. IV and V). The relationship

1 r

- hr(sm1 = r(EA' - E L ) =

[~z(JL)

0

qlt d r

- rAHL

(9)

can be used to obtain the isosteric heat of adsorption, q a t , from heat of

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

273

immersion data. These isosteric heats should agree with those obtained from the Clausius-Clapeyron equation from adsorption equilibrium pressure data at constant coverage at different temperatures. If we assume as Hill (16) that changes in the solid surface being covered by physical adsorption can be neglected, then the molar energy EA or other pertinent thermodynamic functions of the adsorbate itself can be calculated from adsorption and calorimetric data. This assumption, of course, cannot be entirely correct; but the assessment of the contribution to the value of a given thermodynamic function due to perturbation of the solid surface is a difficult task. While it is presently assumed to be a second-order effect, it may become important to the future development of our understanding of surface phenomena. In chemisorption where severe surface perturbations can occur, the Clausius-Clapeyron equation cannot be applied, since equilibrium pressures are low and often unobtainable. Nonetheless, a differential heat analogous to the isosteric heat can be obtained from heats of immersion without recourse to pressure data where the amounts adsorbed prior to immersion can be measured gravimetrically (Sec. VI1,A). The state of the adsorbed film on solids could be elucidated most simply from entropy data. If the entire entropy change can be attributed to the adsorbate, Jura and Hill ( 1 7 ) showed that the difference in entropy of the adsorbed film S A and the entropy of the bulk liquid S L is given by the expression Here, X is the relative equilibrium pressure. Again, this equation for the change in A S is exact if the solid perturbation is not negligible, although the necessity for precise adsorption data over the entire range of equilibrium relative pressures severely limits its use. ENERGYOF SOLIDS C. THE SURFACE Surface energy, or, more precisely, surface free energy, is one of the most important parameters determining the adsorptive and wetting characteristics of a solid in the presence of a vapor or liquid. These in turn influence flocculation aggregation, crystal growth, and indeed most other colloidal behavior exhibited by solid particles. Despite the importance of these energy values, only scanty data exist in the literature. Lack of knowledge of the nature of real surfaces precludes accurate calculations of these energy values from intermolecular potentials. Surface-energy values can be obtained from measured differences in the heats of solution of finely divided particles and large crystals of a substance. These measurements are difficult to obtain. In addition, the strain imparted to a solid as particle

274

J. J. CHESSICK AND A. C. ZETTLEMOYER

TABLE 111 Surjace Energy of Solids at I6"

Surface Surface free energy, energy ergs/cm.S

Solid Graphite Pol ytetrafluoroethylene Rutile (Ti031 the

120 55

9tW

70 69

-

A value of 6 = 1 wa8 used but the lack of experimental bases for this value makes value for rutile rather uncertain. Personal communication from R.J. Good.

ego

size is decreased and effects due to surface lattice imperfections cannot be assessed by this approach. Giriialco and co-workers (It?), on the other hand, derived the following expression which can be used to estimate the surface energies of solids from the increasingly available heats of immersion :

Here, e L y a the , surface energy of the wetting liquid, and the heat of immersion hr(gL) are the measurable quantities. The quantity r-D is given by the expression :

and may be assumed equal to unity for certain solid-liquid systems. Table 111 contains values for the surface energy of a variety of solids calculated from Equation (11).

IV. Wetting by Water A. HOMOGENEOUS AND HETEROGENEOUS POLAR SOLIDS Water is the most important substance in immersional calorimetry met whether as the wetting liquid itself or as an impurity. Indeed, rigorous efforts must be employed to remove water from organic liquids when wetting measurements are made; even trace amounts can drastically influence the heat values obtained on immersion of hydrophilic solids (19). It is no accident, therefore, that the more comprehensive wetting studies, where enthalpy changes have been obtained for solids with increasing known amounts of preadsorbed wetting liquid, have been confined almost exclusively to solid-water systems. The various wetting curves for dif-

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

275

pri V

L

0

t

w

I

2om 10 0

P/P.

FIG. 1. Known types of heat-of-immersion curves obtained for different solids as a function of pre-adsorbed wetting liquid. (a) Homogeneous surfaces, e.g., water on some hydrophilic surfaces (chrysotile asbestos). (b) Heterogeneous surfaces, e.g., water on Ti02 . (c) Lyophobic surfaces, e.g., water on Graphon (graphitic surface). (d) Swelling solids yielding internal surface, e.g., water on Wyoming bentonite (montmorillonite). (e) Porous solids or organic fibers, e.g., methyl alcohol in charcoal or water in wool keratin.

ferent solids are illustrated in Fig. 1, where heat values are plotted against either the volume preadsorbed or the relative pressure of water at which the solids were equilibrated at 25" before immersion into water. Figures l a and l b are typical curves found for the heat of immersion of polar solids in water (and are also found for polar and nonpolar solids in organic liquids). An example of Fig. la is found in the immersion of chrysotile asbestos having known and increasing amounts of physically adsorbed water on its surface (22).The linear relationship between the heat of wetting and the volume adsorbed up to about a monolayer is significant and indicates surface homogeneity since the heat evolved is proportional to the amount of bare surface present. In accord with this finding, the isosteric heat values calculated from adsorption isotherms increased with coverage to a maximum near the monolayer as expected for adsorption on a surface possessing nearly uniform sites. The more frequently encountered exponential decrease in the heat of wetting with increasing surface coverage illustrated in Fig. l b was first

270

J. J. CHESSICK A N D A . C. ZETTLEMOYER

observed by Harkins and Jura (19)for the system water-anatase (Tion). Such curves are indicative of a heterogeneous surface structure. B. THEIMMERSIONAL

WETTING OF

HYDROPHOBIC SOLIDS

Figure l c differs markedly from those obtained for the immersion of polar solids in water; initially the heat values are small but increase with increasing amounts of preadsorbed water. Thus far, only one such curve has been reported in the literature: for the system Graphon-water ( 2 0 ) . Graphon is a graphitized carbon black which has an essentially homogeneous, homopolar surface (21 ). Nevertheless, a small fraction of heterogeneous sites is responsible for the limited adsorption of water on the surface of this solid. Similar curves can be expected for other hydrophobic solids. The very low water adsorption by Graphon precludes reliable calculations of thermodynamic quantities from isotherms at two temperatures. By combining one adsorption isotherm with measurements of the heats of immersion, however, it is possible to calculate both the isosteric heat and entropy change on adsorption with Equations (9) and (10). If the surface is assumed to be unperturbed by the adsorption, the absolute entropy of the water in the adsorbed state can be calculated. The isosteric heat values are much less than the heat of liquefaction with a minimum of 6 kcal./mole near the B.E.T. V , ; the entropy values are much greater than for liquid water. The formation of a two-dimensional gaseous film could account for the high entropy and low heat values, but the total evidence (22) indicates that water molecules adsorb on isolated sites ( 1 in 1,500), so that patchwise adsorption takes place. Comparison of water and nitrogen adsorption measurements reveal that Teflon, a polytetrafluorethylene, can have more polar sites per unit area each possessing a higher adsorption potential for water than the hydrophilic sites on Graphon (23). On the other hand, both heat of immersion and contact angle measurements listed in Table 1, definitely establish that, overall, Teflon is the more hydrophobic of the two solids. Obviously, the chemical nature of the nonpolar portion of the surfaces rather than the infrequent polar sites govern the gross wettability of these solids. Most hydrophobic or low-energy surfaces possess such active sites which may be critical in their behavior or use properties. Savage has demonstrated ( 2 4 ) , for example, that the lubricating ability of graphite is greatly enhanced by adsorbed water on these isolated sites. Also, the failure of Teflon as an organophobic covering for paint-mixer blades is probably due to such Eites. If the number of polar sites on a hydrophobic surface is increased, at some point, a solid of profoundly different surface characteristics should emerge. The nonporous silica, Aerosil, with 75 % of its surface made up of

HEATS OF IMMERSION OF SOLIDS I N LIQUIDS

277

nonpolar portions as evidenced by the differences between nitrogen and water surface areas (see Sec. V1,B) behaves as a hydrophilic solid in wetting measurements (Table I) and in other studies (see Sec. VIII). Bartell and Ruch ( 2 5 ) studied the effect of depletion of oriented monolayers of noctadecylamine on platinum and found abrupt changes in contact angle only after about 60% of the monolayer is removed.

C. CLAY-WATER SYSTEMS Swelling phenomena, ion exchange reactions, and changes due to chemical, thermal, or mechanical modification of clays and other minerals can be followed readily by heat-of-immersion measurements in water. As a matter of fact, soil chemists recognized early the importance of this experimental approach (26) even when estimations of surface areas were nonexistent or unreliable. Too often, however, the complexity of the minerals or soils, or the lack of knowledge of surface areas, exchange site constituents, soluble or other impurities, etc., have made interpretation of experimental data impossible or at best subject to serious doubts. However, the influence of particle diameter (by), firing temperature ( 2 8 ) and constituents in exchange positions (29) on heats of wetting of certain clays in water have been determined. For example, Seifert found small but regular increases in the heat of wetting of a series of ion-exchanged monoionic kaolinite samples in the order K < Na < H < Ca The principal reason for this effect is attributable to differences in hydration of the various cations, since differences in area can be neglected. Seifert observed the same trend for cation-exchanged montmorillonite, although the differences in the heat values were much greater. The energy of swelling of the platelets comprising the primary particles of this material can be significant (30)and, no doubt, differs from sample to sample to different degrees. Ideally, a calorimetric study of the surface properties of a clay mineral would require the use of carefully identified, fractionated, mono-ionic samples for which adsorption data are available. A knowledge of both the internal and external area as well as swelling energy data are additional pieces of information necessary for clays which swell, such as bentonite. Unfortunately, no such extensive study has been made, nor has any method been developed to measure quantitatively the endothermic heat of swelling, Unlike most natural organic fibers, the swelling energy of a Wyoming bentonite (30) is large enough to influence the shape of the immersional heat curve as shown in Fig. Id, although the magnitude of the swelling cannot be subtracted from the composite heat curve. In the absence of

278

J. J. CHESSICK AND A. C. ZETTLEMOYER

swelling, the heat curve would be expected to rise continuously from p / p o = 1 to p / p ~= 0. The two inflections near 0.3 and 0.6 relative pressure correspond to the relative pressures a t which the first and second water layers enter between the bentonite platelets as shown by X-ray diffraction data. An isosteric heat curve for this sample calculated from heat of wetting data is plotted in Fig. 2 and compared with one obtained by Mooney, Keenan, and Wood (31) from desorption data for a sodium montmorillonite. Though the two samples are not directly comparable, the large differences in the heat values and, particularly, in the shape of the curves apparently reflect changes in the solid due to swelling. An estimation of the swelling energy might be possible if heat data would be obtained by the two methods for a given sample. The interactions of the acid sites on clay cracking catalysts with water can be studied readily by immersional calorimetry even if chemisorption occurs and yields information concerning the type and energy distribution of these sites as we shall see in Sec. VI1,A.

D. NATURAL AND MODIFIED ORGANICFIBERS One of the principal objectives of any immersional calorimetric investigation of organic fibrous materials is to assess the thermodynamic functions

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

279

for the interaction of the first water molecules with the functional groups on the surface of the solid. However, such studies are also complicated by the lack of knowledge of the true extent of the wetted areas (largely internal) and the heats of swelling and of solution. Water and nitrogen adsorption probably can be utilized best to estimate external and internal areas. Heats of solution can be measured directly. Strauss (32) found that endothermic heats of solution of collagen and gelatin in water contribute significantly to the total heat effect on immersion. Indeed, measured heats changed from exothermic to endothermic a t about 50" as the immersion temperature was increased from 30 to 80". Potential heats of swelling can be estimated from isosteric heats obtained from both calorimetric and isotherm data as suggested in Sec. IV,C. Rarely have all these factors been considered by workers in this field. Immersional heats must be corrected for solution and swelling (or these effects must be proved absent) to obtain correct energies or other thermodynamic functions for the adsorption or absorption process. Morrison and co-workers (33) measured the heat of wetting of silk fibroin and wool keratin in water as a function of their water content. Their data received proper thermodynamic treatment except that heats of solution are presumed absent. It is likely that solution effects are absent or small if zero heats are found for the immersion of samples equilibrated near the saturation pressure of water. While the integral heat values do not reflect strong swelling for these two materials as they do for bentonite, for example, both the differential heats and entropies give evidence for this phenomenon. The first water molecules enter between the fibers and are responsible for most of the swelling energy. Multilayer adsorption occurs causing further intermolecular swelling and eventually intermicellar swelling. The heat of immersion curve in Fig. l e is typical of those obtained by Morrison (33), Kanagy ( 3 4 ) , Dumanski ( 3 6 ) ,and others for similar solids. Dumanski and co-workers (36) investigated starch, gelatin, agar, sodium stearate, cellulose, and silica gel-water systems and found that the heat evolved owing to all the (bound) water interacting with the solid in the immersion process is very nearly a constant independent of the solid and equal to about -80 cal./g. water. This value is essentially the net integral energy of adsorption for this amount of water and suggests that swelling and solution heats balance or are negligible a t the immersion temperature employed. This constant heat value cannot be used to assess the interaction energies of the first water molecules with the functional groupings of these solids, since these interactions are masked when energies of adsorption at maximum [ (multilayer) (%)I coverage are compared.

280

J. J. CHESSICK AND A. C. ZETTLEMOYER

E. UTILITY OF HEAT-OF-IMMERSION VALUESFOR SAMPLES EQUILIBRATED AT HIGHEQUILIBRIUM PRESSURES The heat values for samples equilibrated with water vapor at pressures near saturation can approach 118.5 ergs/cma2,the surface enthalpy of water a t 25". Indeed, this is common for nonporous polar solids of low specific area. Here, the heat evolved represents the destruction of an area of liquidlike surface very nearly equal to the surface area of the solid and is the basis for the absolute method of area determination of Harkins and Jura ( 3 6 ) . The external area of bentonite in this region of relative pressure calculated by this method was found to be 30 % greater than the nitrogen area of the dry sample and could be accounted for by the imbibition of two water layers between the individual platelets comprising the primary particles of the clay ( 3 0 ) . The loss of external area due to the filling of capillaries between the primary fibers of agglomerates of attapulgite clay ( S T ) , as well as the loss of internal area of fibrous organic materials (SS), can also be followed. Young and Healey (38) have postulated a hollowtube structure for chrysolite asbestos on the basis of such heat measurements.

V. Wetting by Organic Liquids A. COMPARATIVE IMMERSIONAL HEAT VALUESFOR HYDROPHOBIC AND HYDROPHILIC SOLIDS Heats of immersion of two solids with widely different surface characteristics (39) are given in Table IV. Rutile ( TiOz) is a heteropolar hydrophilic solid, and Graphon is homopolar and hydrophobic. A homologous series of alcohols and hydrocarbons and several n-butyl derivatives are the wetting liquids. Water is included for comparison. The heat values are markedly higher for the polar solid immersed in polar liquids; they also vary considerably with the functional group of the liquid. For Graphon, however, the heats are almost unaffected by the structural features of the wetting liquid. This nonpolar solid, despite the interacts I), presence of a small amount of hydrophilic sites on its surface (& with the liquids primarily through London dispersion forces. Because of the additive nature of these forces, each adsorbed molecule tends to lie flat on such a surface (40). In the case of a polar molecule the functional group is oriented somewhat away from the nonpolar surface toward the liquid.

B. THEINTERACTION OF POLAR SOLIDS WITH ORGANIC LIQUIDS The electrostatic force field emanating from the surface of a polar solid exerts a strong orienting influence on molecules possessing peripheral

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

28 1

TABLE IV Heats of Immersion of Solids in Organic Liquids at 26"

Water Ethyl alcohol n-butyl alcohol n-amyl alcohol n-butyl iodide n-butyl aldehyde n-nitropropane n-butyl amine n-butyl chloride Butyric acid Hexane Heptane Octane

550 397 410 413 395 556 664 330 502 506 135 144 140

32 110 114 120

106 106 115 103 112 127

dipoles (40). In treating the wetting results in Table IV, the butyl derivatives are assumed to adsorb in a close-packed monolayer with their polar groups directed toward the surface of the rutile; thus, Equation (8) for the net integral energy of adsorption from the liquid state can be written hl(SL)- h,

=

r(1gA- E ~ )

(13)

where h, is the heat of immersion of the monolayer-covered solid. The quantity h, should be about the same from one butyl derivative to another, since the monolayer covered surface is essentially paraffinic in nature. In addition, calculations show (41 ) that the influence of the solid surface on second layer molecules is small for the monolayer orientation postulated because of the large distance between layers. Ellison and Zisman (42) found that the nature of oriented monolayers of perchloro-pentadienoic acid is independent of the solid substrate, and Levine and Zisman (43) concluded from their studies of higher homologues of some of these butyl derivatives that every film of a polar-paraffin derivative on polar solids prepared by vapor condensation comprises a close-packed oriented monolayer. Thus, the assumption of a close-packed, oriented monolayer of these butyl derivates whose terminal methyl groups govern the wettability of the film covered rutile appears justified. The area occupied per molecule is taken as 20 A2.; hence, the surface concentration averages about 5 X lo'* mol./crn.'. An estimation of the average electrostatic field of the rutile surface is possible, since the net heat of adsorption, or the heat of wetting directly

282

tz I

J. J. CHESSICK AND A. C. ZETTLEMOYER

.

0

I

2

I

4

I

3

4

5

DIPOLE MOMENT

FIQ. 3. Net integral heat of adsorption at 26" as a function of the dipole moment of various n-butyl derivatives on rutile (TiOl).

(since hi is a constant), is a linear function of the dipole moment of the wetting liquid. If the differences in the heats of wetting shown in Fig. 3 arise from the interaction between a permanent dipole, p and the electrostatic field of the dielectric, then the slope of the line in Fig. 3 is a measure of the average electrostatic field according to the equation

E,

=

F,

(14)

The slope of the line yields the average electrostatic field strength F of 2.7 X 10' e,s.u./cm,* (at the operative distance from the rutile surface to the center of the dipole) ; this value is in excellent agreement with that calculated for rutile by de Boer (44) from the results for argon adsorption by Morrison, Los, and Drain ( 4 ) . That the linear relationship of Fig. 3 holds for the molecules other than the alcohol, amine, and acid is surprising. These three molecules possess dipoles pointing with their positive end toward the outside of the functional group and located near the periphery of the molecules. These peripheral dipoles can approach closely enough to interact strongly and be oriented by the negative ions which are thought to comprise the outer layer of most ionic dielectric substances (40).The higher and consistently increasing heat values found for the chloride, aldehyde, and nitro-compound whose nonperipheral dipoles point with their negative end away from the functional group are difficult to explain. I l k and co-workers (46) also found

283

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

that the increase in the heat of wetting of barium sulfate by chlorobenzene, phenol, and nitrobenzene could be attributed to the dipolar component of the adsorption energy. Obviously, more work is needed in this field. C. THEINTERACTION OF HYDROPHOBIC SOLIDSWITH ORGANICLIQUIDS Unlike polar solids, the heats of immersion of hydrophobic solids in mono-substituted paraffins and nonpolar liquids differ only slightly ( 4 7 ) . Such data for the immersion of graphite samples, specially treated to remove surface oxide, in a variety of liquids are listed in Table V. Heat values for samples 1 and 2 were obtained by Bartell and Suggitt (5). Heat (48) and free-energy (14) values for the immersion of bare and monolayer-covered samples of graphite sample 3 in the respective liquids were also measured. The agreement in heat values for the three different graphites is, in general, good. A comparison of hzcsL,and hZtsfL,values for sample 3 are particularly instructive. The heat values for the immersion of monolayer-covered samples in toluene, cyclohexane, and carbon tetrachloride are three to four times less than the heat values for the bare samples and are also significantly less than the surface enthalpies of the respective liquids. Apparently, the attractive forces of the covalent carbon atoms in the surface are highly localized. In addition, the interaction of first-layer molecules with those in the second layer is less than between the liquid molecules; in adsorbed monolayer, as indicated by the Vm’s (14). The net energy of adsorption TABLE V Heats of Immersional Wetting of Graphite Samples in Organic Liquids at b6’a Liquid

Water Toluene p-xylene Cyclohexane Carbon tetrachloride n-heptane n-octane Methanol n-propanol n-butanol

-hz(srL) 9 :rgs/cm.s, Sample 1 Sample 2 3ample 3 Sample 3 48

-

124 101 115

-

120 119 106

56

101 -

123 116

-

102

-

-

123.0 102.0 113.0 122.5 -

29.6

90.6

-

( E L- E A ) , cal./mole, Sample 3

-

32.8 38.3 78.2

-

a Monolayer and equilibrium pressure data for sample 3 were obtained from Professor F. E. Bartell (see also Craig, R. G . , Van Voorhis, J. J., and Bartell, F. E., J . Phys. Chem., 80, 1226 (1966).

284

J. J. CHESSICK AND A. C. ZETTLEMOYER

is greatest for toluene which has a planar structure so that it can lie flat on the surface. The energy changes are next highest for carbon tetrachloride and cyclohexane: the symmetrical nature of carbon tetrachloride and the distortion of cyclohexane a t the surface permitting close approach of the four carbon atoms to the surface are possible causes. The quantity h r ( B , L ) in the case of heptane is greater than the value of the surface enthalpy of the liquid, and the net integral energy of adsorption of the monolayer is lower than any of the other liquids except propyl alcohol. These molecules may be less restricted on the surface and may interact more strongly with second-layer molecules. It is possible, however, that the calculated B.E.T. V , is low and h r ( 8 , L ) reflects the remainder of the surface-heptane interactions. While heats of wetting for a solid in a variety of pure liquids can be informative, heat values as a function of the amount of preadsorbed wetting liquid are more desirable. The data of Table V for the immersion of bare and monolayer-covered samples of graphite illustrate the limitations of single heat measurements. The more comprehensive studies applied to the immersion of rutile in the n-butyl derivatives should furnish answers to questions concerning the nature of the adsorbed film on this solid. Indeed, preliminary measurements substantiate the assumption of an oriented monolayer of adsorbed alcohol on rutile made in Sec. V,B. Unlike watersolid systems, almost no comprehensive heat measurements have been reported for solid-organic liquid systems except that of Itazouk (49) for the immersion of bare and film-covered samples of a porous charcoal in methyl alcohol and the recent work of Pierce et al. (60) on carbon-benzene systems. Such information would be most instructive.

VI. Average Polarity of Solids A. THERATINGOF SOLID SURFACES BY IMMERSIONAL HEATSI N ORGANIC LIQUIDS The concept of rating the polarity of solid surfaces arose from the observation (Sec. V,B) that the net integral energies of adsorption for the various butyl derivatives onto rutile (or, directly, the heats of immersion hrcsL,) form a very nearly linear plot against the dipole moment of the wetting liquid. The linearity persists for similar heat values obtained with other solids as shown in Fig. 4; therefore, an average electrostatic field F for these solids can be estimated from the slope of these curves (16). Field strength values for these solids and a series of commerical pigments (51) are tabulated in Table VI. The other solid adsorbents are all less polar than rutile. As stated previously, Aerosil is a partly hydrophobic, nonporous, fine silica; thus, the slope of the curve for this solid represents

HEATS OF IMMERSION OF SOLIDS I N LIQUIDS

285

-800

5-700 '0

5 Q a

g- 600 2

0

gY -500 s

f

,-400 0

I-

a

:-SO0

-200

-100

I

0

DIPOLE

2 MOMENT

3

4

FIG.4. Heats of immersion at 25' of various solids as a function of the dipole moment of the wetting liquid. TABLE VI Average Electrostatic Field Strengths for Solid Adsorbents and Commercial Pigments from Heats of Immersion Solid Adsorbents Calcium fluoride Aerosil Graphon Teflon Pigments Rutile Iron blue (high strength) Rutile (general purpose) Chrome yellow (medium) Barium lithol (strg. med. toner) Carbon black (short channel)

F X 1C6, e.s.u./cm.a 2.5 1.1 0 0 2.7 2.2 2.1 1.9 (1.7) 0.7

286

J. J. CHESSICK A N D A . C. ZEWLEMOYER

mostly the contribution due to the hydrophilic portions of the surface. Teflon has the lowest surface free energy of all the solids in agreement with the findings of Schulman and Zisman (62)that surfaces containing terminal fluorine atoms are the most nonwetting known. Graphon and Teflon no doubt possess small electrostatic fields, but their small magnitudes are undetectable by these measurements. The influence of the surface polarity of powders on their adsorption and dispersion properties can be profound, as is discussed in Sec. VII1,A. The values of F are likely to be put to many uses as more of them are measured. The electrostatic surface fields are doubtless involved in the phenomena of chemisorption and catalysis, capable of inducing polarization or electron shift of adsorbing molecules. For silica-alumina catalysts, the production of active M-0-M surface groups must be considered the most important factor responsible for chemisorption and catalytic activity.

B. THE USE OF WATERTO RATETHE POLARITY OF SOLID SURFACES For simplicity, a systematic classification of solid surfaces by heat of immersion values in a single liquid is desirable. Il'in (65) predicted on theoretical grounds a decrease in the heat of wetting with increase in the size of the cation in salts or oxides. However, Il'in and Kiselev (64) found that the decrease in the heat of immersion from SrS04 to PbS04 to BaS04 predicted by theory is not observed. Heat data for the immersion of a variety of solids in water are listed in Table VII. Here, too, the changes in heat values are not systematic. Lack of order is not surprising, since such solids are polycrystalline and also have different tendencies to react with water forming surface hydroxides or hydrates. For example, the sensitivity of the heat of immersion of SnOzto activation temperature, listed in Table VII, suggests chemisorption of water even for a 400" activated sample. Too often changes in surface stoichiometry, due to loss of hydroxyl water, occur on heating oxides. These active surfaces chemisorb water or organic molecules such as alcohols and amines during adsorption or immersion. Similar behavior for silicas, aluminas, and silica-alumina cracking catalysts is expected. Obviously, the use of these current immersional heat values for the systematic classification of this type of solid is limited. Nonetheless, important information about the strength of water adsorption can be obtained from such measurements. Care must also be exercised in the use of organic wetting liquids, since Schreiner and Kemball (66) and others have shown that chemisorption can occur on the surfaces of dielectric and ionic surfaces outgassed a t elevated temperatures.

287

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

TABLE VII Heat of Immersional Wetting ' Solids in Water al86" Solid

,

Activation" temp., "C.

ergs/cm.'

200 400 200 400

-hmL)

Crystal structure

a.

645 f 25 665 f 12 539 & 15 550 f 7

Orthorhombic KNOa

8.42

400 200 400 200 400

550 f 18 491 f 6 613 f 10 351 f.8 364 f 9

Tetragonal SnO2

CaFa BaFl

200 200

463 f 10 435 f 11

Fluorite

5.45 6.19

SnOo

200 400 200 400

407 451 570 550

PbO arrangement

3.24

200 200

646 f 7 532 f 9

SrCOs BaCOs Ti02 (rutile) SnOz PbOi

CrzOa FetOa Activated at mm. Hg. (I

f 17 f 32

8.83 4.49 4.72 4.93

3.95

f 35 f 21 Hexagonal

5.35 5.42

T "C. on an adsorption apparatus to an ultimate vacuum of 10-'

C. THEDEGREE OF SURFACE HYDROPHILICITY OF SOLIDS A solid surface may possess various fractions of both hydrophilic and hydrophobic areas. For example, the surface treatment of clays or pigments to make them hydrophobic and oleophilic is an important technological process. The completeness of the treatment depends not only on the extent but also on the nature of the organic adsorbate film. Also, the surfaces of the industrially important high-area silicas, such as Aerosil, made by flame hydrolysis of silicon tetrachloride, can possess both silanol (Si-OH) and siloxane (Si-0-Si) surface groups according to Iler (66). Evidence has been presented by Young (67) that only the silanol groups physically adsorb water and that the silicon-oxygen bonds are essentially homopolar. When water vapor adsorbs only on a portion of a surface, the degree of surface hydrophilicity of a solid can be rated from the ratio of the effective water area to the nitrogen area or by immersional heat values in water. Adsorption measurements were used by Young to follow the conversion of

288

J. J. CHESSICK AND A. C. ZETTLEMOYER

silanol groups to siloxane groups by dehydration a t elevated temperatures. Patrick (68) measured the heats of wetting of similar silicas in water after dehydrations up to 900" and found a nearly linear decrease in the heat values as the surface hydroxyl content decreased. From these data, Iler estimated the heat of wetting of a siloxane and silanol surface as -130 and - 190 ergs/cm.*, respectively. The value of - 130 ergs/cm.2 appears too large if dispersion forces alone are responsible for adsorption on siloxane surfaces as proposed by Young (67). A rating of the polarity of a completely dehydrated silica by immersion studies in polar organic liquids would be revealing.

D. THEIMMERSION OF SOLIDS IN LIQUID NITROGEN An interesting observation is the fact that the heats of immersion of Graphon, Silene EF ( a precipitated calcium silicate), alumina, and magnesia in liquid nitrogen are practically the same. This wetting liquid tends to rate all these solids the same (69) and probably for the same reason that allows the almost universal use of this adsorbate for nitrogen area determinations. While the last three solids probably present oxygen surfaces, the inclusion of the hydrophobic Graphon in the group introduces a quite different surface. This apparent universality suggests that the heat evolved when 1 g. of a solid is immersed in liquid nitrogen can be taken as a measure of the specific area of the solid. It also supports the presently accepted contention that nitrogen gas adsorption is the most reliable technique for estimating the specific surface area of powders.

VII. Site Energy Distribution A. HEATBOF ADSORPTION FROM HEATOF IMMERSIONAL WETTINGDATA

It has already been pointed out that differential or integral heats of adsorption can be calculated from heat-of-immersion values without recourse to two or more isotherms where the amounts preadsorbed on the sample before immersion are measured gravimetrically. This technique is particularly useful where chemisorption occurs a t very low and difficult to measure equilibrium pressures. In comparison, too, direct calorimetric determination of heats of adsorption can be less accurate, although less tedious, than heat-of-immersion determinations. Lack of accuracy can occur with poorly conducting, nonmetallic adsorbents, where long equilibrium times are required for vapor phase adsorption or where surface sites do not fill in strict accordance with the site energy distribution of the solid surface. Atoms or molecules can be expected to stick to the first part of the surface they strike when strong chemisorption occurs and then molecules are likely not to move freely over

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

289

the surface at the temperature of the experiment. Harkins (60) has pointed out that wetting heat effects are generally liberated within a few minutes and that sufficient time can always be taken to insure that solid and vapor equilibrium is reached before immersion of the sample. To avoid the lack of even distribution mentioned above, the chemisorbed species often can be removed from initially saturated samples by desorption at carefully regulated evacuation temperatures. This technique has been particularily suited to the study of the hydration of sites on bentonite (SO), attapulgite (37), FeaOa (61 ) , and molybdenum oxide surfaces ( 6 2 ) . A topographical picture of the number and strength of acid sites on the surfaces of solids, for example, clays and cracking catalysts, is highly important to an understanding of the use properties of these materials and can be obtained readily from such heat of immersion measurements. This approach was used to investigate the heterogeneous nature of acid-exchange sites on attapulgite clay (63) and, more recently, on a commercial cracking catalyst. For this purpose, attapulgite initially out-gassed at 400" was treated with butyl amine vapor at 25" for a period sufficient to insure saturation. When this sample was outgassed a t 25" to constant weight, some of the amine remained adsorbed on acid sites of varying high strengths. The total amount of amine chemisorbed and amounts desorbed between 25 and 400" were measured gravimetrically. Then, heats of wetting were determined, as a function of these amounts of preadsorbed amine N A . The differential heat curves shown in Fig. 5 were obtained from the integral heats of adsorption according to Equation ( 9 ) ; these values are analogous to isosteric heats, although the heat of liquefaction has not been added to m d

*

The heat curves, themselves, are informative. The kaolin-based pellet catalyst has a few more active sites then attapulgite, but its site activity decreases rapidly and to values only about 3 kcal./mole above the heat of liquefaction of the liquid at maximum coverage. Obviously, a distinction cannot be made between physical adsorption and chemisorption for some of the amine adsorbed a t full coverage on the cracking catalyst. On the other hand, attapulgite has a much narrower distribution of adsorption energies, and the lowest heats are about double the heat of liquefaction of butyl amine. Therefore, it appears safe to conclude that the amount remaining after evacuation at 25" is chemisorbed. The surface coverage f3is the ratio of the amine remaining adsorbed after evacuation at temperature T to the total adsorbed after saturation and evacuation a t 25" to an ultimate vacuum between and lo-' mm. Hg. The maximum adsorption at 25" on attapulgite and the catalyst sample is large and covers about one-half of the total available area of the solids if 20.8 A2. is used as the cross-sectional area of adsorbed butyl amine.

290

J. J. CHESSICK AND A. C. ZETTLEMOYER

o

0.1

a2

0,s

04

SURFACE

03

0.8

COVERAGE

0.7

0.0

e

FIG.5. Differential heats of adsorption of n-butyl amine at 25' versus surface coverage 8. (1) Attapulgite clay. (2) Kaolin-based cracking catalyst.

The acid site distribution function g(e) given by the equation

d N n - g(e) m -

+

can be defined so that the sites with energy between e and e de can accommodate g ( e ) d e moles at S.T.P. of adsorbed gas ( 6 4 ) and is depicted in Fig. 6 for attapulgite clay. The distribution might be expected to be different for different initial activation temperature or for adsorbates of different basicities. This method should be particularily useful for comparing the behavior of cracking catalysts of varying activities or varying selectivities. The question as to whether Lewis or Bronsted acid sites are most effective in cracking remains a moot point. In any case, it is suspected by some investigators that active sites develop as the adsorption occurs, for example, by the shift of a six-coordinated t o a four-coordinated aluminum. The findings of Richardson and Benson (66) on pyridine adsorption on silica-alumina

HEATS OF IMMERSION OF SOLIDS I N LIQUIDS

291

E, KCAL./MOLE FIG.6. The distribution of the energies of acid sites on attapulgite clay.

gels, for example, can obviously be extended by the immersion technique as described above. The shape of the curves and the magnitudes of the site energies vary with the history of the catalyst, whether pellet or fluid type, whether steam deactivated or not, and whether fresh or used.

VIII. Solution Adsorption A. PREFERENTIAL ADSORPTION FROM SOLUTION BY POLAR AND NONPOLAR SOLID Future important contributions of heats of immersion will be made in the field of solution adsorption despite the necessity for more exacting experimentation. The common problem in solution adsorption has been to define the nature and extent of the interface between solid particles and mixed liquids. Specifically, more information is needed concerning the orientation and solvation of adsorbed molecules as well as the composition and practical boundary of the adsorbed phase. Direct adsorption measurements yield only net changes in concentration and indirect approaches must be taken (66). Much can be learned, however, by measuring the heats of immersion of powders into two component solutions of varying composition where the adsorption of one component is predominant. This technique, also, is the only available method for measuring the heat of adsorption of

292

J. J. CHESSICK AND A. C. ZElTLEMOYER

viscous, low-vapor-pressure liquids or solids from solution onto an adsorbent. Several rather simple systems have been elucidated by this technique. Hutchinson ( 4 ) and Crisp (67)investigated adsorption from solutions of polar organic molecules from nonpolar solvents onto polar solids. Young, Chessick, and Healey (68) measured the adsorption of n-butyl alcohol from aqueous solution onto the hydrophobic solid, Graphon. In this last system particularly, it is reasonable to assume preferential adsorption and eventual formation of a discrete, oriented monolayer of nonsolvated solute molecules. An equation was developed for the alcohol-water-Graphon system from a model based on the adsorption of the alcohol on uniform sites on the surface of Graphon (see Sec. IV,B) with the hydrocarbon chain of the alcohol oriented flat on the surface and the polar hydroxyl group directed toward the liquid. Calculated heats of wetting were. in excellent agreement with experimental values. Graham and Hansen (69) also found systems such as these ideal for analysis by other techniques. The results of both studies were essentially in agreement. However, these latter workers offer evidence that the alcohol is adsorbed without hydration and further postulate that an additional -CHgroup is elevated from the surface near monolayer coverage. Typical of the sort of data needed to determine whether additives affect the interface is that provided by a study of the influence of n-heptyl compounds on the gel structure of dispersions containing polar solids in nonpolar vehicles (70). The influence of the polar heptyl compounds on the fluidity of dispersions of rutile and a fine silica (HiSil) in a dibasic ester, Plexol 201, is shown in Fig. 7. Apparently, the more polar rutile adsorbs all except the chloride and in these cases thinning results. HiSil has a lower F value and adsorbs only the amine and alcohol preferentially. Greases prepared from the least polar solid, Aerosil, are also least influenced by these additives (or even by more complex ones). Measurements of the solution isotherms for HiSil and Aerosil reveal significant adsorption of heptyl alcohol, but no detectable chloride adsorption in the same concentration range. Since all the heptyl derivatives are completely miscible with Plexol 201, and the alcohol and the amine are not the most polar of these additives, the question arises as to the cause of their relative effectiveness. One reason for the adsorption of the alcohol is shown in Fig. 8. Its heat of solution is endothermic, whereas the heat of solution of the chloride, the least effective additive, is actually exothermic. The heat of immersion of a solid into a solution at monolayer coverage can be divided into the following steps:

PH'r

= AH.d.

4-P H d 4- mi

(16)

293

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

ll

IN PLEXOL-201

I

2 I 2 3 WEIGHT PERCENT ADDITIVE FIQ.7. Effect of n-heptyl additives on fluidity as measured with a precision penetrometer (arbitrary numbers) for Rutile-Plexol 201 and HiSil-Plexol 201 Gels.

I

0

14'

12

I

I

I

-

-

0

-2

I

- -CI

0.5 I .o 1.5 2.0 WEIGHT PERCENT ADDITIVE. FIQ. 8. Heats of solution of n-heptyl compounds in Plexol 201 (di-2-ethylhexyl sebacate).

0

294

J. J. CHESSICK AND A. C. ZE'TTLEMOYER

TABLE VIII Heats of Adsorption of Heptyl Alcohol onto Aerosil and H i S i l from Parafin Oil Solutions at 86" (Monolayer Coverage)

Solid Aerosil

A H , , ergs/cm.* AH* , ergs/cm.' AHi , ergs/cm.* AH.d. , ergs/cm.* AH,,d. , kcal./mole

- 150 - 13 -78

-59

-1.7

HiSil

-227 - 12 -78 - 137 -4.0

where AHedmis the enthalpy change for the adsorption of a monolayer of adsorbate vapor onto the solid, A H d is the heat of dilution of the additive from solution, and AHi is the enthalpy change for the formation of the interface between the adsorbed monolayer and the solution. In the application of this relationship, mixed adsorption is assumed absent at monolayer coverage. Heats of adsorption of heptyl alcohol on to Aerosil and HiSil from paraffin oil solutions are listed in Table VIII. The heat of immersion and heat of dilution terms were measured calorimetrically. Assuming monolayer coverage and orientation of the polar end of the additives toward the surface, the enthalpy of formation of the adsorbate-solution interface was estimated to be -78 ergs/cm.* by measuring the heat liberated when a sample covered with a monolayer of heptane is immersed in heptane at 25". The heptyl additives have a similar influence on greases built with either Plexol 201 or paraffin oil, and, indeed, the influence was much greater for the more polar solid HiSil. Although free-energy changes, not enthalpy changes, are the true driving forces, the heat effects appear to give the proper perspective of the mechanism.

B. THEWETTABILITY OF SURFACTANTS IN AQUEOUS SOLUTIONS Heats of immersion provide a new tool for rating the wetting ability of surfactants from aqueous solution. If a high-area graphite, Graphon, or carbon black is used, the increased heat effect obtained with the surfactant solution over that obtained with water allows the wetting tendency of the surfactant to be rated (7'1). Typical heats of immersion in surfactant solutions are listed in Table IX. If the heat of adsorption is desired, then the heat effect caused by breaking up the micelles to provide the surfactant ions for adsorption must also

HEATS OF IMMERSION OF SOLIDS I N LIQUIDS

295

TABLE I X Heats of Immersion of Graphon in 0.60% Aqueous Solutions of Surfactants Approx. purity

Surfactant

Heat of immersion -hl(sL)

,

ergs/cm.* Water Sodium lauryl sulfateo Purified sodium dodecylbenzene sulfonate Sodium dodecylbenzene sulfonate (Santomerse-3)b Monobutyl biphenyl sodium monosulfonate (Areskap-100)b Monobutyl phenylphenol sodium monosulfonate (Areskap-lOO)b Sodium decylbenzene sulfonate (Santomerse-D)b Sulphate ester of an alkyl phenoxy polyoxyethylene ethanol (Alipal CO-436)' Polyoxyethylated nonyl phenolsc Igepal CO-650 Igepal CO-850 Igepal CO-530

b

100 100

32 142 120

Anionic Anionic

99

73

Anionic

-

78

Anionic

-

51

Anionic

90

47

Anionic

-

68

Anionic

99 99 99

88 60 40

Nonionic No n ion ic Nonionic

Supplied by National Bureau of Standards, Washington, D. C. Supplied by Monsanto Chemical Co., St. Louis, Mo. Supplied by General Aniline and Film Corp., Easton, Pa.

be added in Equation (16). The effect of added salts on the adsorption of surfactant can also be investigated by heat of immersion techniques ('71).

List of Symbols erm)

Heat of immersion per gram of evacuated solid in a pure liquid Heat of immersion per unit area of an evacuated solid in a pure liquid Heat of immersion per unit area of a solid containing a known amount of preadsorbed wetting liquid

eA(sL)

AHL Ahad,

Energy of immersion per unit area of an evacuated solid in a pure liquid Energy of adhesion, liquid t o solid Molar heat of liquefaction of bulk liquid Integral heat of adsorption of a gas or vapor a t equilibrium pressure pa and temperature T

J. J. CHESSICK A N D A . C. ZETTLEMOYER

Isosteric heat of adsorption Contribution t o total energy of adsorption due t o interaction of a permanent dipole with the electrostatic field of a solid Average electrostatic field of a solid surface Dipole moment Molar internal energy of the bulk liquid

SLV%VC

z

Molar internal energy of the adsorbed layer Molar internal energy of an adsorbed layer perturbed by the solid surface Molar entropy of the bulk liquid Molar entropy of the adsorbate layer

V,

SA’

Surface free energy of a solid in uucuo Surface free energy of the liquid Surface free energy of a solid-liquid interface Surface free energy of a solid in equilibrium with vapor a t saturation pressure

a

d

AH,

Surface energy of a solid in vacuo Surface energy of the liquid Surface energy of a solidliquid interface

Film pressure at saturation Initial spreading coefficient -liquid over an evacuated solid Final spreading coefficientliquid over a solid in equilibrium with vapor a t saturation Specific surface area per gram of solid Contact angle Moles adsorbed per gram Surface concentration Bol tzman ’s constant Relative equilibrium pressure of adsorbate Monolayer volume of adsorbed gas or vapor per gram of solid Molar entropy of an adsorbed layer perturbed by the solid surface Total enthalpy change for the immersion of an evacuated solid in a solution a t a concentration a t which monolayer adsorption occurs Heat of dilution of a solute from a solution Enthalpy change for the formation of an interface between an adsorbed monolayer and solution Integral heat of adsorption of a monolayer of adsorbate vapor onto the solid surface

REFERENCES 1 . Boyd, G. E., and Harkins, W. D., J. Am. Chem. SOC.64, 1190-94 (1942);see also

Harkins, W. D., and Dahlstrom, R., Znd. Ens. Chem. !42, 897 (1930);Clark, A., and Thomas, B. D., J. Phys. Chem. 43, 579 (1939). 2 . Zettlemoyer, A. C., Young, G. J., Chessick, J. J., and Healey, F. H., J . Phys. Chem. 67, 649 (1953). 3. Berghausen, P. E., i n “Adhesion and Adhesives” (Clark, Rutzer, and Savage, eds.), p. 225. Wiley, 1954.

HEATS OF IMMERSION OF SOLIDS I N LIQUIDS

297

4. Hutchinson, E., T r a m . Faraday SOC.43,443 (1947). 6 . Bartell, F. E., and Suggitt, R. M., J . Phys. Chem. 68, 36 (1954). 6. Pierce, W. C., Mooi, J., and Harris, R. E., J. Phys. Chem. 62, 655 (1958).

7. Young, T. F., and Smith, M. B., J. Phys. Chem. 68,716 (1954). 7a. Hutchinson, E., and Manchester, K. E., Rev. Sci. Instr. 28, 364 (1955). 7b. Kiselev, A. V., Kiselev, V, F., Mikos, N. N., Muttik, G. G., Runov, A. D., and Shcherbakova, K. D., J . Phys. Chen. (U.S.S.R.)23, 577-94 (1949). 8. Hackerman, N., API Project 47d Reports, University of Texas, 1957. 9 . Whalen, J. W., and Johnson, W. F., Magnolia Petroleum Company, Dallas, Texas, private communication, in press. 10. Harkins, W. D., “The Physical Chemistry of Surface Films,” pp. 94 ff. Reinhold, New York, 1952. 11. Shafrin, E. G., and Zisman, W. A., J . Phys. Chem. 61, 1046 (1957); Levine, O., and Zisman, W. A,, ibid. 61, 1068 (1957). 18. Hare, E. F., and Zisman, W. A., J. Phys. Chem. 69,335 (1955). 13. Fox, W. H., and Zisman, W. A., J . Colloid Sci. 6 , 514 (1950); 7, 109 (1952). 14. Craig, R. G., Van Voorhis, J. J., and Bartell, F. E., J . Phys. Chem. 60, 1225 (1956). 16. Zettlemoyer, A. C., Chessick, J. J., and Hollabaugh, C. H., J. Phys. Chem. 62, 489 (1958). 16. Hill, T. L., J . Chem. Phys. 17, 520 (1949). 17. Jura, G., and Hill, T. L., J. Am. Chem. SOC.74, 1598 (1952). 18. Girifalco, L. A., Kraus, G., and Good, R. J., J. Phys. Chem. 62, 418 (1958). 19. Harkins, W. D., and Jura, G., in “The Physical Chemistry of Surface Films” (W. 13. Harkins, ed.), p. 256. Reinhold, New York, 1952. 80. Young, G. J., Chessick, J. J., Healey, F. H., and Zettlemoyer, A. C., J. Phys. Chem. 68, 313 (1954). 81. Hill, T. L., Emmett, P. H., and Joyner, L. G., J . Am. Chem. SOC.73, 1105 (1951). 88. Pierce, C., and Smith, R . N., J . Phys. Chem. 64, 784 (1950) ;Young, G. J., Chessick, J. J., Healey, F. H., and Zettlemoyer, A. C., ibid. 68, 313 (1954); Graham, D., ibid. 60, 1022 (1956). 23. Chessick, J. J., Healey, F. H., and Zettlemoyer, A. C., J. Phys. Chem. 60, 1345 (1956). 24. Savage, R. H., Ann. N . Y . Acad. Sci. 63, 862 (1951). 26. Bartell, L. S., and Ruch, R. J., Presented a t the 30th National Colloid Symposium, Madison, Wisconsin, June, 1956. 26. Pate, W. N., Soil Sci. 20, 329 (1925); Anderson, M. S., J. Agr. Research 38, 565 (1929). 87. Parmele, C. W., and Frechette, D., J . Am. Ceram. SOC.26, 108 (1942). 88. Harmon, C. G., and Fraulini, F., J. A m . Ceram. SOC.23, 252 (1940). 29. Stifert, A. C., Ph.D. thesis, Pennsylvania State University, 1942. 30. Zettlemoyer, A. C., Young, G. J., and Chessick, J. J., J. Phys. Chem. 69, 962 (1955). 31. Mooney, R. W., Keenan, A. G., and Wood, L. A., J . Am. Chem. SOC.74. 1367 (1952). 38. Strauss, G. R., Ph.D. thesis, Lehigh University, Bethlehem, Pa., 1955. 33. Dunford, H. B., and Morrison, J. L., Can. J . Chem. 33, 904 (1955); Morrison, J. L., and Hanlan, J. F., Nature 179, 528 (1957). 34. Kanagy, J. R., J . A m . Leather Chemists Assoc. 49,646 (1954).

298

J. J. CHESSICK AND A. C. ZETTLEMOYER

36. Dumanskii, A. V., Kolloid Zhr. 12,319 (1950); Dumanskii, A. V., and Mekryach, E. F., ibid. 17, 171 (1955); Dumanskii, A. V., Zzvest. Akad. Nauk S.S.S.R., Otdel. Khim. Nauk No. 3, 270 (1956). 36. Harkins, W. D., and Jura, G., J . Chem. Phys. 11, 430 (1943). 37. Chessick, J. J., and Zettlemoyer, A. C., J . Phys. Chem. 80, 1181 (1956). 38. Young, G. J., and Healey, F. H., J . Phys. Chem. 68, 881 (1954). SO. Healey, F. H., Chessick, J. J., Zettlemoyer, A. C., and Young, G. J., J. Phys. Chem. 68, 887 (1954). 40. de Boer, J. H., Advances i n Catalysis 8 , 3 0 4 0 (1956). 41. Chessick, J. J., Zettlemoyer, A. C., Healey, F. H., and Young, G. J., Can. J . Chem. 85, 251 (1955). 42. Ellison, A. H., and Zisman, W. A., J . Phys. Chem. 67, 622 (1953). 4.9. Levine, O., and Zisman, W. A,, J . Phys. Chem. 81.1068 (1957). .64. de Boer, J. H., Advances i n Catalysis 8 , 102 (1956). 46. Morrison, J. A., Los, J. M., and Drain, L. E., Trans. Faraday SOC.47,1023 (1951). 46. Il’in, B. V., Kiselev, V. F., and Aleksandrova, G. I., Doklady Akad. Nauk S.S.S.R. 102, 1155 (1955). 47. Bartell, F. E., and Fu, Y., J . Phys. Chem. 59, 1758 (1929); Culbertson, J. L., and Winter, L. L., J . A m . Chem. SOC.69, 308 (1937); Laporte, F., Ann. Phyaik [12] 6 , 5 (1950). 48. Yu, Y.-F., Zettlemoyer, A. C., and Chessick, J. J., “Free Energies, Heats and Entropies of Wetting of Graphite,” presented before the Colloid Division, American Chemical Society, New York, Sept. 1957. 40. Rasouk, R. I., J . Phys. Chem. 46, 190 (1951). 60. Pierce, W. C., Mooi, J., and Harris, R. E., J . Phys. Chem. 82, 655 (1958). bf. Zettlemoyer, A. C., O$cial Digest 28, 1238 (1957). 68. Schulman, F., and Zisman, W. A,, J . Colloid Sci. 7,465 (1952); J . A m . Chem. SOC. 74, 2123 (1952). 63. Il’in, B. V., Doklady Akad. Nauk S.S.S.R. 66, 269 (1947). 64. Il’in, B. V., and Kiselev, V. F., Doklady Akad. Nauk S.S.S.R. 82, 85 (1952). 66. Schreiner, G. D. L. and Kemball, C., Trans. Faraday SOC.49, 190 (1953). 66. Iler, R. K., ed., “The Colloid Chemistry of Silica and Silicates,” p. 234. Cornell University Press, Ithaca, N. Y., 1955. 67. Young, G. J., J . Colloid Sci. 15, 67 (1958). 68. Patrick, W. A., i n “The Colloid Chemistry of Silica and Silicates” (R. K. Iler, ed.), p. 241. Cornell University Press, Ithaca, N. Y., 1955. 60. Chessick, J. J., Young, G. J., and Zettlemoyer, A. C., Trans. Faraday SOC.60, 587 (1954). 60. Harkins, W. D., “The Physical Chemistry of SurfaceFilms,” pp.. 268 ff. Reinhold, New York, 1952. 61. Healey, F. H., Chessick, J. J., and Fraioli, A. V., J . Phys. Chem. 60, 1001 (1956). 68. Zettlemoyer, A. C., and Chessick, J. J., J. Phys. Chem. 68,242 (1954). 63. Chessick, J. J., and Zettlemoyer, A. C., “Study of Silicate Minerals. V. A. Quan-

titative Determination of the Acid Strength of Exchange Sites on Attapulgite,” presented a t the National Colloid Symposium, Urbana, Illinois, June 1958, J . Phys. Chem., 62, 1217 (1958). 64. Drain, L. E., and Morrison, J. A., Trans. Faraduy SOC.48, 316 (1952). 66. Richardson, R. L., and Benson, 5.W., J . Phys. Chem. 81,405 (1957). 66. Kipling, J. J. and Tester, D. A,, J . Chem. Soe. pp. 4123-4133 (1952).

HEATS OF IMMERSION OF SOLIDS IN LIQUIDS

299

Crisp, D. J., J . Colloid Sci. 11,356 (1956). Young, G. J., Chessick, J. J., and Healey, F. H. , J . Phys. Chsm. 60,394 (1956). Graham, D. and Hansen, R. S., J . Phys. Chem. 60, 1153 (1956). Chessick, J. J., and Young, G. J., J . Colloid Sci. 13, 358 (1958); Chessick, J. J., Zettlemoyer, A. C., and Young, G. J., J . Colloid Sci. 13, 372 (1958). 7f. Zettlemoyer, A. C., Schneider, C. H., and Skewis, J. D., “Second International Congress of Surface Activity,” Vol. 111, p. 472. Academic Presa, New York, 1957.

67. 68. 69. 70.

This Page Intentionally Left Blank

The Catalytic Activation of Hydrogen in Homogeneous, Heterogeneous, and Biological Systems J. HALPERN Department of Chemistry, University of British Columbia, Vancouver, B.C., Canada Paps

I. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. Homogeneous Systems ...................................... 303 A. Metal Ions in Aqueous Solution, . . . . . . . . . . B . Effects of Complexing.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 . . . . . . . . . . . . . . . . . . 314 C . Effects of Solvent. . . . . . . . . . . . . . . . D. Cobalt Carbonyls and Related Co E. Base-Catalyzed Exchange Reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 F. Discussion.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 ... . . . . . . . . . . . . . . . . . 329 111. Heterogeneous Systems. . . . . . . . . . . . . . . . . . . A. Metals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................... 329 B. Oxides.,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 C. Other Solids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Biological Systems. . . . . . . . . . . . . .............................. V . Concluding Remarks. . . . . . . . . . . ....................... Acknowledgements.. . . . .................................. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365

1. Introduction A number of factors contribute to the special interest which is associated with the study of reactions of hydrogen. Among these is the fact that hydrogen is the simplest stable molecule and its properties have been thoroughly elucidated both experimentally and theoretically. Furthermore, the hydrogen molecule is relatively inert from the kinetic standpoint, and its reactions are susceptible to catalysis by a variety of substances. Because of these properties, the study of reactions of hydrogen affords exceptional opportunities to gain a better understanding of chemical reactivity and catalytic phenomena in general. In addition, many catalytic hydrogenation reactions are of great industrial importance. A great variety of substances are capable of catalyzing the reactions of hydrogen. The best known and most important of these are solids, including metals (especially transition metals), oxides (ZnO, Cr2O3, A1203, etc.), and certain salts. Thus, the apparent activation energy for the hydrogena301

302

J. HALPERN

tion of ethylene or the exchange between molecular hydrogen and deuterium is of the order of 50 kcal./mole in the gas phase but only about 10 kca1.l mole on the surface of a transition metal such as nickel or palladium. Since 1931 (Stephenson and Strickland, I ) it has also been recognized that certain microorganisms contain an enzyme or system of enzymes, to which the name hydrogenase has been given, which can catalyze various reactions of hydrogen. More recently, a number of metal ions and complexes have been found to catalyze hydrogenation reactions homogeneously in solution. Because of their inherent simplicity, both from the chemical and kinetic standpoints, the catalytic mechanisms in these systems have proved pnrticularly susceptible to detailed elucidation. In considering the function of the catalyst in reactions of hydrogen, it is useful to distinguish between three types of reactions. 1. Exchange with deuterium or with protolytic substances:

+ Ds + 2HD Hc + Dz0 + HD + HDO Hs

2. Reduction of substances which are also reduced readily by “reversible” electron-donors or at electrodes:

+ Hz aAg(8) + 2H+ Crz07’ + 3H2 + 8H+ + 2Cr+++ + 7H20 Quinone + Hz Hydroquinone 2Ag+

-t

-

(31 (4)

(5)

3. Reactions with other “inert” molecules (particularly reactions in which new carbon-hydrogen bonds are formed) :

In a rather loose sense it can be stated that these successive types of reactions impose increasingly stringent requirements on a catalyst. Thus, catalysts which are active in reactions of the last type are normally effective also for the first two types, but the converse of this is not as generally valid. There are, however, in each class of reaction, examples of catalysis by all three types of catalysts (homogeneous, heterogeneous, and enzymic). Observations that a given catalyst is frequently active in a wide variety of hydrogenation reactions suggest that the function of the catalyst is to activate the hydrogen molecule, in the sense of forming a reactive complex with it, which can then enter into reactions with a variety of substrates (Polanyi, 2 ) . This concept appears to have validity particularly in relation to the first two types of reactions considered above. Reactions of the third type frequently require also “activation” of another reactant, however, and

CATALYTIC ACTIVATION O F HYDROGEN

303

hence it is not surprising that they exhibit a more complex pattern of catalytic effects Insofar as it is possible to isolate this aspect of the subject, this review will be concerned primarily with the concept of the "activation of hydrogen" as a common feature of catalytic hydrogenation reactions. Emphasis will be placed on a consideration of the factors which determine catalytic activity and specificity and on the qualities which relate homogeneous, heterogeneous, and enzymic catalysts. Complex reaction mechanisms involving steps other than the activation of hydrogen, particularly where these are rate-determining, will be accorded only passing treatment. The catalytic activation of hydrogen was last reviewed in this series in 1948 by Eley (5). I n the intervening decade many important advances have occurred at both the theoretical and experimental levels of the subject. Some of these have resulted from studies on new types of catalysts, notably metallic alloys, semiconductors, and homogeneous catalysts and from related developments in the theory of the solid state. Other advances have been made possible by a variety of new experimental techniques for the study of surface phenomena, including magnetic, electrical, and spectroscopic measurements. In this article attention will be directed particularly to these developments. The various types of catalyst systems will be considered in the order (1) homogeneous, (2) heterogeneous, (3) biological. This does not reflect the historical development of the subject nor the relative importance of the three types of catalysts, but rather the order of their increasing complexity and of the uncertainty about the mechanisms by which they function.

II. Homogeneous Systems It is now well recognized that even under mild conditions, molecular hydrogen can undergo many reactions homogeneously in solution. This recognition is, however, of fairly recent origin and can be traced to Calvin's discovery (4, 6) in 1938 that at temperatures of about lOO", hydrogen reduces cupric acetate or benzoquinone homogeneously in quinoline solution, in the presence of dissolved cuprous acetate. The latter functions as a true homogeneous catalyst for these, as well as for other reactions of hydrogen, e.g., parahydrogen conversion. Subsequently, and particularly since 1953, many other systems in which hydrogen undergoes homogeneous reactions, have been discovered and studied. Most of these involve metal ions or complexes as reactants or catalysts. The literature on homogeneous hydrogenation reactions is now fairly extensive and is covered in several recent reviews (Weller and Mills, 6 ; Halpern, 7), to which the reader is referred for a more detailed account, particularly of the historical and experimental aspects of the subject.

304

J. HALPERN

A. METALIONS IN AQUEOUS SOLUTION Among the species which can activate hydrogen in solution are a number of simple metal ions, e.g., Cu++ (Halpern et al., 8, 9 ) , Ag+ (Webster and Halpern, 10-12), Hg++ and H g F (Korinek and Halpern, IS), and the oxyanion Mn04- (Webster and Halpern, 14). In aqueous solution these ions are reduced homogeneously by hydrogen a t temperatures below 100" to compounds of lower valence or to the metallic state. Cu* and Ag+ can also catalyze exchange of hydrogen with water and hydrogenation of dissolved substrates such as oxygen or dichromate, which do not react with hydrogen in the absence of a catalyst. The mechanisms which have been proposed for the activation or splitting of hydrogen in these systems, are summarized in Table I. The experimental evidence on which these mechanisms are based has been reviewed elsewhere (Halpern, 7, 15). 1. Cu++. Information about the activation of hydrogen by Cu++ is derived largely from kinetic measurements on the cupric perchlorate catalyzed hydrogenation of dichromate [Equation (4)].The rate-law for this reaction is of the form

providing convincing evidence for the following mechanism :

+H

CU++

+ Cu++ 6Cu+ + CrzO7- + 14H+ CuH+

ki t y CuH+

-

2Cu+

fast

2Cr-

+ H+

+ H+ + 6Cu* + 7H20

(9) (10)

(11)

Equation (8) also describes the kinetics of the reduction of the cupric salt, ultimately to metallic copper, which is observed in the absence of an added substrate (Macgregor and Halpern, 16).This reaction proceeds by a mechanism in which step (11) is replaced by the rapid disproportionation of Cu+ to give metallic copper: 2cu+ F: cup)

+ cu*

(12)

Splitting of the hydrogen in these reactions occurs in step (9). The splitting is heterolytic, a hydride ion combining with Cu++ to form CuHf while a proton combines with a water molecule to form H30+.An examination of the effects of complexing Cu++ on its catalytic activity, to be considered later, suggests that it is one of the ligand water molecules of the cupric ion which is involved and that the reaction with hydrogen results in its "re-

CATALYTIC ACTIVATION OF HYDROGEN

305

placement” by a hydride ion. In the light of this, a better representation of reaction (9) would be Cu(H20)F

+ Hz -+

CuH(HzO).-i+

+ HsO+

(13)

The rate of activation of hydrogen in this system is, of course, that of step (9), i.e., kl[H2][Cu++].The difference between this rate and that at which hydrogen is actually consumed is due to the back reaction which regenerates hydrogen. The magnitude of this difference depends on the ratio of the competing rate constants, k 2 / k 3 , whose value at 110” is 0.26. Conflicting indications as to the importance of the back reaction at higher temperatures have come from studies of the reduction of cupric salts to metallic copper (Macgregor and Halpern, IS) and of the Cu++-catalyzed reaction between H2 and O2 (McDuffie et al., 16a). The latter reaction has been studied from the standpoint of its possible application in reducing radiolytic decomposition of water in aqueous homogeneous reactors. 2. Ag+. The kinetics of the silver perchlorate-catalyzed reduction of dichromate are given by

The two terms in the rate expression apparently correspond to independent reaction paths. The second of these involves the heterolytic splitting of hydrogen by a mechanism analogous to that just described for Cu*, with the formation of AgH as an intermediate. The evidence for this is even more conclusive than in the case of Cu*, since measurements using D20enriched water have confirmed the formation of HD which the mechanism predicts because of back reaction of the hydrogen splitting step (Webster and Halpern, 11). The kinetic parameters for kl are listed in Table I. ( k z / k 3 ) is given by 2.3 X 107 exp (- 14,000/RT) ; the magnitude of both the temperature dependence and the pre-exponential factor in this expression are surprisingly large. The first term in the rate expression, which is pH-independent, apparently corresponds t o a reaction path in which hydrogen splits homolytically, i.e., 2Ag+

k + HZ---+ 2AgH+

(15)

followed by rapid reaction of AgH+ with the substrate. It is of interest that although this mechanism of activation of hydrogen differs radically from that for Cu++,analogous intermediates (CuH+ and AgH+) are formed in the two cases. Despite the fact that the homolytic splitting of hydrogen is a termolecular reaction involving two Ag+ ions, it competes with the bi-

TABLE I Summarv of Kinetics and Mechanisms of the Catalytic Activation o Hgdrogen i n Romogeneowr Systems0 Temp. range, "C.

Kinetics of hydrogen activation

-

Rate law

M%,

kcal.

26 14

Ass,

Proposed mechanism of hydrogen splitting

- 10

CUH+ H+ CU++ H2 2Ag+ Ht + 2AgH+ Ag+ Hs -+ AgH H+ Hg* H2 -* Hg 2H+ or Hg++ Hz + HgH+ H+ Ha* Hz + Hg2 2H+ or Ht + 2HgH+ H&* MnOr Ht + Mn(V) (see text). Ag+ Mn04Ht + AgH+ Mn04' H+ CuAcs Ht + (CuH)Ac HAc OHHt + HH2O NHsHs -+ HNHa CuHp2 H2 + (CUH)HP HHp

ex.

cu++ Ag+

Water Water

W-140 30-120

23

-6

Hg++

Water

65-100

18

- 12

HgF

Water

65-100

20

- 10

&Or Ag+

Water Water

30-70

14 9

- 17

CuAct* OHNHsCuHpzb

Water Water Ammonia Heptanoic Acid Heptanoic Acid Quinoline Pyridine

80-140 80-190

24 21-28

-7 -7

125-155

29

-2

12.5-155

20

- 16

+ MnOl

CuHpb CuAc AgAcb

30-60

-25

-26

13-16 12-14

+

-+

+

+

+

+ + + +

+

+

+

+

(--2x))

+ + + + + + + CuHp + Ht CuH + HHp SCUAC+ Hr -+ ~ ( C U H ) A C

(-25)

AgAc

-50

25-117 2.5-78

+ + + +

+

+

+ Hs

+

AgH

+ HAc

+

a This table includes only those systems for which the kinetics of hydrogen activation have been established; a number of other systems are referred to in the text. b Ac- = acetate; Hp- = heptanoate. c Peters, E., and Halpern, J., J. Phys. Chem. 69,793 (1955) ; Halpern, J., Macgregor, E. R. , and Peters, E. , ibid. 60, 1455 (1956). d Webster, A. H., and Halpern, J., J . Phys. Chem. 61, 1239 (1957). a Korinek, G. J., and Halpern, J.,J. Phys. Chem. 60,285 (1956). I Webster, A. H., and Halpern, J., Trans. Faraday SOC.63.51 (1957). 0 Peters, E., and Halpern, J., Can. J. Chem. 53.356 (1955). Wilmarth, W. K., Dayton, J. C., and Fluornoy, J. M., J. A m . Chem. SOC.76, 4549 (1953); Miller, S. L., and Rittenberg, D., ibid. 80, 64 (1958). Wilmarth, W. K., and Dayton, J. C., J. A m . Chern. Soc. 76,4553 (1953). i Chalk, A. J., and Halpern, unpublished work. Calvin, M., J. Am. Chem. SOC.61, 2230 (1939); Weller, S., and Mills, G. A., ibid. 76, 769 (1953); Wilmarth, W. K., and Barsh, M. K., ibid. 78, 1305 (1955). I Wright, L.W., Weller, S., and Mills, G. A., J. Phys. Chem. 69, 1060 (1955); Wilmarth, W. K., and Kapauan, A. F., J. Am. Chem.

'

rsoc. 78,1308 (1956).

2

3

2

2 %

$ 1

8 0 %¶

! I U

TI0 H

1:

308

J. HALPERN

molecular heterolytic splitting process because of its lower activation energy (15 vs. 24 kcal.) and predominates over the latter at low temperatures. Schematic potential energy diagrams depicting the two paths of activation of hydrogen by Ag+ and showing the quantities which determine the energetics of each path are given in Fig. 1. 3. MnO,. Another reaction in which AgH+ has been proposed as an intermediate is the Ag+-catalyzed reduction of MnO, by Ht in acid solution (Webster and Halpern, 14). The rate-determining step of this reaction, whose kinetics are of the form, k[Hz][Ag+][MnO4-], is probably

+ MnOr) + Ha

AgMn04 (or Ag+

--t

AgH+

+

+ H+

(16)

followed by fast reactions of AgH+ and MnO4- to yield the observed product (MnOz) and regenerate Ag+. The role of Mn04- in reaction (16) can

I

Ag

-H

MECHANISMI-2Ag**

Ag

-H

b

DISTANCE

H,

+

2AgH*

r

DISTANCE

MECHANlSMII-Ap** H,

+ AgH+ H*

FIG. 1. Schematic potential energy diagrams for the homolytic (I) and heterolytic (11) splitting of hydrogen by Ag+. All processes and energy terms are in solution. D = dissociation energy; Z = ionization potential. [Webster, A. H . , and Halpern, J., J . Phys. Chem. 61, 1239 (1967).]

CATALYTIC ACTIVATION OF HYDROGEN

309

be regarded as that of replacing one of the Ag+ ions in reaction (15); its effectiveness in doing so is connected with its high one-electron affinity. In this connection it is of interest that the activation energy of this reaction (-9 kcal.) is one of the lowest yet observed for a homogeneous hydrogenation process. Permanganate also undergoes an uncatalyzed homogeneous reaction with hydrogen in which it is reduced to MnOz in acid solution, or Mn04- in basic solution. The kinetics, which are pH-independent over a wide range, are of the form, k[Hz][Mn04-]. Energetic considerations (Halpern, 15) suggest that manganese (V) is formed as an intermediate by a rate-determining step which may involve transfer of a hydride ion or of two electrons from H2 to Mn04-:

+ HZ+ HMnO,' + H+ Mn04- + HZ+ MnO.' + 2H+ MnOd-

(17) (18)

or of an oxygen atom from Mn04- to H Z: Mn0,-

+ Hz + M n O s + H 2 0

The abnormally large negative entropy of activation (- 17 e.u.) suggests that one of the first two mechanisms of hydrogen splitting, rather than the last one, is correct. 4. Hg++and H g p . There is also some uncertainty about the mechanism of activation of hydrogen by these ions. In each case the kinetics are first order in H2 and in the metal ion and exhibit no pH-dependence in the range 0.025 to 0.1M H+. On energetic grounds the most likely intermediates in these reactions are Hg atoms, suggesting that the rate-determining step in each case is a two-electron reduction of the metal ion, with release of two protons to the solvent: Hg++ Hg?

+ Hz Hg + 2H+ + HZ-+ 2Hg (or Hgz) + 2H+ 4

(20)

(21)

Alternative possibilities, however, involving the formation of HgH+ as the initial product of the rate-determining step, through heterolytic splitting of hydrogen by Hg++ or homolytic splitting by H g p , cannot be excluded. This species, if it is formed, would probably dissociate in aqueous solution into Hg and H+. 5. Other Ions. A number of other metal ions including Caw, Mg++, Zn++, MnH, Co++, Ni++, Cd++, Pb++, Al+++,Fe+++,T l w , Cr-, Ce++++, U O F , VO,, and CrO4- in aqueous solution, have been tested for catalytic activity, in most cases at temperatures up to 150",but found to be inactive (Halpern, 17). Rittenberg (f8)has reported that the acetates of cadmium, zinc, and magnesium are catalytically active in aqueous solution, but no

310

J. HALPERN

experimental details were given. No autocatalysis due to Cu+ was noted during the reduction of aqueous solutions of cupric perchlorate or sulfate by hydrogen (Macgregor and Halpern, IS). The activity of Cu+, if any, is thus small compared with that of Cu++; this is of interest in view of the pronounced catalytic activity which cuprous salts display in certain organic media (Calvin, 4, 6). Similarly, no autocatalysis due to MnOd was noted in the reduction of Mn04- in basic solutions (Webster and Halpern, 14). Attempts to ascertain the activity of Pd++in aqueous solutions were unsuccessful because traces of metallic palladium, a powerful heterogeneous catalyst, always formed. It has been found, however, that the complex, PdClr’, does activate hydrogen homogeneously (Halpern, Harrod and Potter, 19).The kinetics are first order in the complex and in Hz (implying heterolytic splitting) with AHS = 20.0 kcal./mole and ASS = -6.8 e.u. RhCld exhibits similar behaviour (AH$ = 24.5; ASS = 9.4), while the chloro- complexes of Pt(IV), Ir(III), Ir(IV), Os(1V) and Au(II1) apparently are inactive (19a).

B. EFFECTS OF COMPLEXING The reactivities of metal ions toward hydrogen are greatly influenced by complexing (Table 11). The results for cupric complexes (Peters and Halpern, 20,$1) can be accounted for in terms of the heterolytic splitting mechanism proposed earlier. On the basis of this mechanism, the activity would be expected to depend on two properties of the complex (1) the basicity of the ligands and (2) the metal-ligand bond strength. Thus, increasing the basicity of the ligands surrounding the metal ion should facilitate heterolytic splitting of hydrogen [Equation (13)] by permitting increased stabilization of the proton released. This could account for the enhancement of the catalytic activity of the cupric ion by complexing with acetate and certain other basic anions and for the fact that the catalytic activities of the complexes increase in the same order as the basicities of the ligands; HzO < C1- < 804- < acetate < propionate, butyrate. On the other hand, with increasing strength of metal-ligand bonding (reflected in the magnitude of the formation constant), removal of the ligand or its replacement by a hydride ion becomes increasingly difficult, and hence low catalytic activity would be expected for very stable complexes even with basic ligands. This would account for the low activities of the cupric glycine and ethylenediamine complexes. Another way of looking at this is that in very stable complexes the low-lying orbitals of the metal ions are used in bonding with the ligands and are therefore not available to accept electrons from the hydrogen molecule. Either description of this “poisoning” effect is also applicable to the well-known poisoning of heterogeneous metallic catalysts by electron-donating substances such as sulfur compounds (Maxted, 2%).

311

CATALYTIC ACTIVATION O F HYDROGEN

TABLE I1 Effects of Complezing on Catalytic Activity of Metal Ions "Mean" formation constant *

Complex0

Relative catalytic activity

-

CUBUZ CuPrz CUACZ c u s o4 CuCl4cu++ CuGlz Cu(en)P

150 150 120 6.5 2.5 1 <0.5 0.1

30

1

5 1

x 102 -1 -

x x

107 10'0

22

HgSOi Hg++ HgtC HgAcz HgPr2 HgCIdHgClr HgBrz Hg(en)?'

2 x 10' 1.6 x 104

1.1 x lo-' 4 x 10-2

6 X 10'

3.2 x 10-3 2.5 x 10-3 1.7 x 10-3 1 x 10-8

4

4 x 10' 3.5 x 107 5.1 X 101'

C

C C

C

d e e d d d d d

x lo-'

d

f f f f

80 25 1 inactive

-3

7

C

C

1

-

AgAc Aden)%+ Ag+ Ag(CN12-

C C

1.8

-

x 10' -

2.4 X loo

Refs.

Ligand designations: Bu- = butyrate; Pr- = propionate; Ac- = acetate; G1- = glycinate; en = ethylenediamine. 0

*

x,

where K , = [MXmI/[MIIXIm Peters, E., and Halpern, J., Can. J . Chem. 34,554 (1956). d Korinek, G.J., and Halpern, J., Can. J. Chem. 34, 1372 (1956). Korinek, G. J., and Halpern, J., J. Phys. Chem. 60,285 (1956). f Webster, A. H., and Halpern, J., J. Phys. Chem. 61. 1245 (1957). c

The effect of pH on the apparent catalytic activity of cupric ions in a solution containing glycine is depicted in Fig. 2. The activity passes through a maximum at a pH of about 2. This has been interpreted (Peters and Halpern, 21) in terms of the following pH-dependent equilibria : CHI-NHI

0-C-0

0-c=o

AHa ~ H ZC d

2H+

o=

-0

' ( a

AH2

I

2H+

d

(22)

f;JH3

0 Cu++

+2HOk-cH~h

312

J. HALPERN

I

I

I

I

I

I

2

3

4

5

6

PH FIQ.2. Effect of pH on the Cu(I1)-catalyzed hydrogenation of dichromate at 130”, 20 atm. H) , in an aqueous solution containing0.05M Cu(C10,)2 ,0.20Mglycine. [Peters, E., and Halpern, J., Can. J . Chem. 34, 554 (1956).]

In the intermediate pH range (-2) the cupric ion is present predominantly as a “carboxylate” complex ( B ) which is more reactive than either the fully-complexed glycinate ( A ) or the uncomplexed ions which predominate at higher and lower pH, respectively. This behavior is reminiscent of that observed for many enzyme-catalyzed reactions whose rates pass through a maximum as the pH is varied. Similar explanations, involving the concept of a “bifunctional” catalyst comprising both an acidic and basic site, have been proposed for such systems (Laidler, 23,24). In general, the effects of complexing on the reactivities of the mercuric (Korinek arid Halpern, 26) and silver (Webster and Halpern, 12) ions are susceptible to similar interpretations, although it should be noted that the order of activity of the various complexes differs for the three metals. Thus, the chloride and acetate complexes of the mercuric ion (which are much more stable than the corresponding cupric complexes) are less reactive than the ayuo complex, while the relatively unstable sulfate complex is more reactive. In the case of silver, the acetate and ethylenediamine complexes are more reactive than the uncomplexed ion, while the very stable cyanide complex is inactive. Several effects connected with complexing in the mercury system are worthy of special note: 1. The reactivity of Hg$+ is much lower than that of Hg++. This can he

CATALYTIC ACTIVATION O F HYDROGEN

313

interpreted by considering H g p as a "complex" of Hg++, containing a covalently bonded Hg atom as a ligand. The low reactivity of this complex is consistent with its high stability (Table 11). 2. Although the reactivities of HgClZ and HgAcz toward Hz are much lower than that of the uncomplexed Hg++ ion, further complexing with C1or Ac- results in a slight increase in reactivity. Thus, the reactivity of HgCld-, while still much lower than that of Hg++, is about 30% higher than that of HgClz ;the corresponding effects for the acetate and propionate complexes are much greater. This behavior, at least for the chloride system, is consistent with the general interpretation of the effects of complexing, suggested earlier, in view of the marked decrease in the values of successive formation constants (k, = [HgCl',2-"']/[HgC1~3_~'][C1-])in going from HgC12 to the higher complexes, i.e., kl = 5.4 X lo6;kz = 3.1 X lo6;k 3 = 7; kq = 10 (Lindgren, Johnson, and Sill&, 26). 3. The addition of certain basic anions such as OH-, C03=, and Ac-, increases the rate a t which Hg(en)t+ reacts with H z (Fig. 3). This effect was observed under conditions of complete complexing of the mercuric ion with ethylenediamine, the rate being independent of the (excess) ethylenediamine concentration. It is clear that the effect of the added anions is not due to their complexing of mercury by replacement of ethylenediamine. A possible explanation of the effect of OH- [which was also observed for the

I

V-No Pr 0 -Na2C0, 0 -No ClO,

TV

w m

0-No OH

A- Na Ac

8

0 CONC.

0.5

1.0

OF ADDED SALT-MOLE LITER'

FIG. 3. Effect of various anions on the rate of reduction of Hg(I1) BC 123' in aqueous solutions containing initially 0.01M Hg(C10& , 0.05M ethylenediamine. [Korinek, G . J., and Halpern, J . , Can. J. Chem. 34, 1372 (1956).]

314

J. HALPERN

ethylenediamine complex of Ag+ (Webster and Halpern, 12) but not of Cu++]is an acid-base equilibrium of the type

which has been recognized for other amine and ethylenediamine complexes, e.g., Co(NH&+, Au(en);+, and Pt(NHa)&la+ (Basolo and Pearson, 27). The complex formed by this process would be expected to have enhanced reactivity in view of the replacement of ethylenediamine (NH2R) by a more basic ligand (-NHR). An alternative interpretation which could also be applicable to Ac- and COa- is that the enhanced activity is due to the formation of an ion-pair complex of the type Hg(en)P. X-, in which the anion replaces an outer-sphere water molecule.

C. EFFECTS OF SOLVENT In addition to water, a variety of organic liquids, including amines, carboxylic acids, and hydrocarbons, have been used as solvents in the study of the homogeneous reactions of hydrogen with metal salts. In general, there is more uncertainty about the nature of the species present in such systems than in aqueous solution and, correspondingly, it is usually more difficult to elucidate the reaction mechanisms in detail. The most extensive solvent effect studies have been made on cupric, cuprous, and silver salts. A number of the more important results are considered below. 1. Cupric and Cuprous Salts in Inert Solvents. The reduction of cupric heptanoate by hydrogen to the cuprous salt proceeds homogeneously in a variety of nonpolar solvents. In heptanoic acid solution, both the cupric and cuprous salt contribute to the activation of hydrogen, the latter being more active (Chalk and Halpern, 28). The reaction is thus autocatalytic (Fig. 4), the rate-law being of the form

+

-d[H~l/dt = ki[Hz][C~H~zl kz[HzI[C~Hpl (1)

(24)

(11)

The kinetic parameters for kl and kz are summarized in Table I. The two terms in the rate-expression have been identified with the following renction paths: Path I.

ki + Hn + (CuH)Hp + HHp fast (CuH)Hp + CuHpz 2CuHp + HHp CuHp + H2 -% CuH + HHp fast CUH + CuHpz (CuH)Hp + CuHp fast (CuH)Hp + CUHPZ+2CuHp + HHp

CuHpi

4

Path 11.

(25) (26)

(27) (28) (29)

CATALYTIC ACTIVATION O F HYDROGEN

315

0.N

I

O

w

m

K

0 v)

m 4

NO.0I

0

a 0

N

t

a

'0

FIG.4. Reduction of cupric heptanoate (initially 0.4M) a t 145" in various solvents: (1) diphenyl; (2) heptanoic acid; (3) octadecane; (4) heptanoic acid, using Dt instead of Ha.

The reactive species in this system are apparently the unionized C U H D ~ and CuHp molecules. CuHp undergoes some dimerimtion, but the dimer is inactive. Addition of sodium heptanoate to the solution results in the formation of higher cupric and cuprous complexes, probably Naz(CuHp4) and Naz(CuHp8) ; these species are also inactive. Heptanoic acid appears to function as an inert solvent in this system, since essentially similar results. apart from minor quantitative differences, were obtained in octadecane and diphenyl. Relative to the behavior observed in aqueous solution, the following features of these systems are particularly noteworthy : 1. In each case, hydrogen appears to be split heterolytically by the cupric ion or complex, with the formation of CuH+ as an intermediate.

316

J. HALPERN

2. Despite the high basicity of the heptanoate ion, the activation energy for the reaction of CuHp, in heptanoic acid solution (30 kcal.) is considerably higher than that for either Cu++ or CuAcz in aqueous solution (26 and 24 kcal., respectively). A plausible explanation for this is that formation of the transition-state involves some metal-ligand separation, which requires more energy in a nonpolar medium. 3. The high reactivity of the cuprous species, relative to that of the cupric species, is in marked contrast to the properties of the corresponding ions in aqueous solution. 4. The deactivation of cupric (and cuprous) heptanoate by further complexing with heptanoate contrasts with the effect of acetate complexing in aqueous solution. The reasons for this, as well as for the previous observation, are not clear. 5. Although thermodynamically favorable, reduction beyond the cuprous state, i.e., to metallic copper, proceeds only slowly in these solutions. Thus, the rate of uptake of hydrogen decreases abruptly when all the cupric heptanoate has been reduced to the cuprous state (Fig. 4).When deuterium is used instead of hydrogen, it begins to exchange with the heptanoic acid at this point, and HD is formed at a rate comparable to that of the earlier reduction of cupric heptanoate. This shows that the activation of hydrogen continues but that the reactions leading to the formation of metallic copper, e4.1 CuH

+ CuHp

--t

2Cu

+ HHp

(30)

are slow, so that in the absence of CuHpz ,most of the CuH which is formed in step (27) undergoes back reaction to regenerate hydrogen. I n contrast, metallic copper forms readily when cupric salts are reduced in aqueous solution. This is attributable, in part at least, to the higher equilibrium constant for the cuprous disproportionation reaction in the latter system. The reduction of cupric salts of various other organic acids (e.g., benzoate ; cinnamate ; o-toluate ; o- and m-chlorobenzoate) has been examined using either the corresponding acid or diphenyl as solvent (Chalk and Halpern, 38). In most cases the behavior was found to be qualitatively similar to that described for cupric heptanoate. For several salts, compared in diphenyl, there was an indication that the activity of the cuprous salt increased and that of the cupric salt decreased with increasing basicity of the anion. 2. Cuprous Salts in Quinoline and Other Amines. The first homogeneous hydrogenation reaction to be recognized and studied was the reduction of cupric acetate in quinoline solution. Following Calvin’s pioneering work in 1938 (4, 6 ) , this and related systems have been examined by a number of other workers including, Weller et al. (29, SO) and Wilmarth et al. (31-33).

CATALYTIC ACTIVATION OF HYDROGEN

317

Qualitatively this system has many features in common with the reduction of cupric heptanoate described earlier. The rate plots for the uptake of hydrogen by quinoline solutions of cupric acetate resemble those in Fig. 4, showing pronounced autocatalysis and an abrupt decrease in rate when reduction to the cuprous salt is complete. In relation to the systems already considered, the following features are of particular significance: 1. No activation of hydrogen by the cupric species can be detected. It is not possible, however, to rule out activities of the order of those exhibited by cupric salts in aqueous or heptanoic acid solution, since these would probably be masked by the very high activity of the cuprous species in this system. 2. The rate-law for the activation of hydrogen (Table I) is second order in the cuprous salt. This contrasts with the rate law for cuprous heptanoate in heptanoic acid but resembles that for the “low-temperature” path of activation of hydrogen by Ag+ in aqueous solution. As in the latter case, it seems likely that hydrogen is split homolytically in this system to give CuH+ as an intermediate. 3. The catalytic activity varies with the anion of the cuprous salt. Thus the activities of the acetate, salicylaldehyde, and 4-hydroxysalicylaldehyde are similar and somewhat higher than those of the stearate and benzoate. Cuprous nitrobenzoates and nitrosalicylaldehydes, as well as the cuprous complexes of certain Schiff bases, are inactive. The general trend appears to be for the catalytic activity to increase with the basicity of the anion. This has been interpreted in terms of a decrease in the energy required to promote electrons from the full 3d shell to empty upper orbitals (Calvin and Wilmarth, 32). There is some doubt about the kinetics of the activation of hydrogen by cuprous acetate in the closely related solvent, pyridine. Wright, Weller, and Mills (34) have reported that the rate-law in this solvent (and in dodecylamine) is first-order in cuprous acetate, suggesting heterolytic splitting of hydrogen. On the other hand, Wilmarth (36)has observed a second-order dependence similar to that in quinoline. The reasons for this discrepancy and for the difference between pyridine and quinoline, if real, are not clear. 3. Silver Salts in Pyridine. The activation of hydrogen by silver acetate in pyridine solution, leading to the formation of metallic silver, has been studied by Wright et al. (34) and by Wilmarth and Kapauan (36).The kinetics (Table I) suggest that hydrogen is split heterolytically in this system with the formation of AgH. The activation energy is much lower (13-16 us. 24 kcal.) than that of the corresponding process involving Ag+ in aqueous solution. This can be attributed in part to increased stabilization by the more basic solvent of the proton which is released when hydrogen splits heterolytically . Going from aqueous solution to pyridine (or complexing

318

J. HALPERN

Ag+ in aqueous solution by a basic ligand) thus causes the heterolytic mechanism to be increasingly favored over the homolytic one. AND RELATED COMPLEXES D. COBALTCARBONYLS This section is devoted to a brief discussion of a very interesting class of homogeneous catalysts which includes the metal carbonyls and a number of related covalent complexes containing metal-carbon bonds. These catalysts are characterized by much greater versatility and, in some cases, greater activity, than the ionic complexes already considered and in this respect come closer to resembling heterogeneous catalysts, particulraly metals, in their properties.* For example, dicobalt octacarbonyl is a catalyst for the addition of hydrogen to olefinic bonds and for the hydrogenolysis of alcohols, while the catalytic activity of metal ions is, on the whole, confined to the reaction of hydrogen with substances which are readily reduced by reversible electron donors. On the other hand, the greater complexity of these catalysts has made it more difficult to establish their structure and the mechanism of their action. 1. Dicobalt Octacarbonyl. The chemistry of this interesting and versatile catalyst has been the subject of several recent reviews (Orchin, 38;Wender, Sternberg, and Orchin, 39). Dicobalt octacarbonyl reacts readily with molecular hydrogen in solution to form cobalt hydrocarbonyl :

Coi(C0)s

+ Ha * 2HCo(CO)‘

(31)

It also catalyzes a variety of reactions of hydrogen with olefins, aldehydes, and carbinols. The best known of these is the hydroformylation (0x0) reaction, discovered by Roelen (40),in which a molecule of hydrogen and a molecule of carbon monoxide add simultaneously to an olefin (usually at temperatures of 75” to 200” and partial pressures of 50 to 150 atm. each of CO and H2) to form an aldehyde: RCH=CHa

+ H1 + CO + RCH&H&HO

(32)

Despite very extensive studies on this reaction, there is still considerable uncertainty about its mechanism. The reaction occurs at about the same rate in a wide variety of organic solvents, including benzene, heptane, and alcohol, suggesting that polar intermediates are not involved (Wender et al., 41). Reaction of the olefin with preformed cobalt hydrocarbonyl also gives the aldehyde product (Wender et al., 42). This, together with the observation that cobalt hydrocarbonyl is formed under hydroformylation conditions in the absence of olefin, but cannot be detected in the presence

* According to Wender and Sternberg (U),“metal carbonyls may be considered ns parts of the surface of a transition metal, cut off from the surface and stabilized by carbon monoxide molecules.”

CATALYTIC ACTIVATION OF HYDROGEN

319

of olefin (Orchin, Kirch, and Goldfarb, 43), suggests that the mechanism may involve the sequence of step (31) followed by reaction between the hydrocarbonyl and olefin (Adkins and Krsek, 4).Direct evidence for a reaction between cobalt hydrocarbonyl and olefins, leading to the formation of a complex having the composition (HCo(CO)&.olefin. CO, which decomposes with the evolution of CO and formation of aldehyde, has recently been obtained by Kirch and Orchin (46). On the other hand, kinetic evidence, notably the inverse dependence of the rate on the partial pressure of CO, has led to the formulation by Natta (46) and Martin (47) of an alternative type of mechanism which involves reaction of dicobalt octacarbonyl with olefin (cyclohexene)rather than with hydrogen :

+

Con(C0)~ CsHio Coa(C0)7.CsHio Coz(C0)a

+ HZ

+ 2co

4

Co,(CO)r.CaHio I

k2

ka

fast

Cog(C0)s

+ CO

+ C~HIICHO

CO,(C0)8

(33)

(34) (35)

This leads to a rate-law of the correct form, i.e.,

The accepted structure of dicobalt octacarbonyl is

8 Wender and Sternberg (37)have suggested that the structure of the olefincarbonyl complex (I) is RCH-CHR

While there is no direct evidence for this, they have succeeded in isolating an analogous cobalt carbonyl-acetylene complex (Sternberg et al., 48).

320

J. HALPERN

They have also proposed that reaction an intermediate complex

($4) proceeds via the formation of

RCH-CHR

b A

(C0)a OH H o(C0)r

\/

which decomposes to give COZ(CO)a and CaHl1CHO. The analogy between these complexes and the species which might be expected to form as intermediates in reactions on metal surfaces is striking. It should be noted that the Natta-Martin mechanism, while satisfactory from the kinetic standpoint, does not assign any role in the reaction to cobalt hydrocarbonyl. Hence, it is not readily reconciled with the evidence for the formation of the latter under hydroformylation conditions and its known reactions with olefins (Kirch and Orchin, 46). Sternberg, Markby, and Wender (49) have proposed a mechanism for the evolution of hydrogen which occurs when iron pentacarbonyl is treated with aqueous alkali, which involves the binuclear complex,

as an intermediate. It is possible that the isoelectronic complex

is an intermediate in reaction (31). An alternative interpretation of some features of the hydroformylation reaction (including the inverse CO dependence), in terms of heterogeneous catalysis by an (unidentified) insoluble cobalt component, has recently been advanced by Aldridge, Fasce and Jonassen (49a). The universal validity of this seems doubtful in the light of the considerable evidence favoring a homogeneous mechanism. 2. Cobaltous Cyanide. The behavior of this catalyst system is very com-

CATALYTIC ACTIVATION O F HYDROGEN

321

plex and the following discussion is based principally on a recent investigation by Mills, Weller, and Wheeler (50).Earlier related studies have been made by Iguchi (61)and by Winfield (c52-5.4). Aqueous solutions containing cobaltous cyanide complexes (the prevalent complex in a freshly prepared solution is believed to be Co(CN):-) absorb hydrogen rapidly (-5 min.) a t temperatures as low as O”, the total uptake corresponding to that required to reduce all the cobalt to the +1 state. It has been suggested (Mills et al., 50) that the mechanism of the reaction involves the heterolytic splitting of hydrogen, i.e.,

+ HZ+ X- + Co(CN)sH4- + HX Co(CN)6HC + Co(CN):- + 2Co(CN):- + H+ Co(CN):-

(37) (38)

X- is an anion, whose participation in the rate-determining step is suggested by the marked ionic-strength dependence of the rate. If deuterium is used, it is found that following the reduction, exchange with water occurs, but a t a considerably lower rate than that of the reduction reaction. On standing, a cobaltous cyanide solution gradually loses its reducibility (i.e., the amount of hydrogen which it can take up decreases). There is a parallel decrease in the paramagnetism of the solution, which corresponds initially to one unpaired electron for each cobalt (11) ion. This “aging” process has been attributed to dimerization. The dimer is apparently nonreducible but, like the reduced complex (Co’), activates hydrogen and catalyzes the D2-H20exchange. The activity of all these catalysts is very high relative to most of the homogeneous hydrogenation catalysts previously considered. It is of interest that a Co (11) dimer of the composition Co2(CN)G6contains the same number, 34, of electrons in the combined “outer shells” of the two cobalt atoms as does C O ~ ( C O.)In ~ the absence of strong cobalt-cobalt bonding in these complexes, it would appear that each is just two electrons short of a “closed-shell” configuration. Similarly, Co(CN):-, with 17 outer electrons is one electron short of an “inert-gas” configuration. 3. Ethyleneplatinous Chloride. Flynn and Hulburt (56) have shown that when toluene or acetone solutions of ethyleneplatinous chloride are reacted with hydrogen a t temperatures below - lo”, in the presence of an excess of ethylene, ethane may be formed without accompanying deposition of metallic platinum. The reaction under these conditions appears to be homogeneous, and the following mechanism, which, however, requires substantiation, has been proposed :

+ 2CzH4 2PtClz(CzHi)z + 2Hz + (PtCIzCtHJz + 2CzHs 2PtClz(C~H4)?

(PtClzCzH4)z

(39) (40)

322

J. HALPERN

E. BASE-CATALYZED EXCHANGE REACTIONS Hydroxide ion catalyzes the conversion of parahydrogen in aqueous solution and the exchange of hydrogen with the deuterium in heavy water. This reaction was first observed in 1936 by Wirtz and Bonhoeffer (66),and its kinetics have been examined in detail by Wilmarth, Dayton, and Fluornoy (67)and by Miller and Rittenberg (68). The rate-law given in Table I has been confirmed over the temperature range 80-190”,and OHconcentrations of to 1M. The two mechanisms which have been considered for the reaction involve, either intermediate formation of a solvated hydride ion,

+ OD- + H- + HOD (slow) DOD + H- + OD- + DH (fast) H,

(41) (42)

or, concerted attack on HI of OD- and D20leading to the formation of HD through synchronized proton transfers, i.e.,

DO-~GG~F! -ISDO-

+ DH + HOD

(43)

The distinction between the two representations not a rigid one, hinging on the lifetime of the intermediate hydride ion. If the latter is very short (of the order of the time of a single vibration or collision), the two mechanisms clearly become equivalent. In support of the first mechanism it has been pointed out (Wilmarth et al., 67) that step (41) is energetically consistent with the observed activation energy of the exchange reaction (24-28 kcal.) if the hydride ion which is formed is assigned a radius of 1.5 A. (based on crystal data for solid hydrides) and is assumed to be “normally” solvated. The assumption of normal solvation (Eley and Evans, 69),however, implies a configuration in which the anion is surrounded by a number of water molecules, each oriented with the positive end of its electrical moment directed inward, so that one of the H atoms just “touches” the central ion. In this configuration, the barrier for proton transfer, either by a classical or tunneling mechanism, from the water molecule to the hydride ion can hardly be very high. [That for proton transfer between HaO+ and HI0 in aqueous solution has been estimated (Conway, Bockris, and Linton, 60) at about 4 kcal.] Accordingly, the lifetime of a “solvated” hydride, if it is formed, must be very short. Indeed, there appears to be no experimental evidence for the existence of hydride ions in aqueous solution. Nor has there been detected, under the conditions of the OH--catalyzed exchange, any reduction of dissolved substrates such as CrO,“ which might be expected to react readily with H-. It has been suggested (Wilmarth et d., 67) that a mechanism involving concerted proton transfers, such as that represented by Equation (43),

CATALYTIC ACTIVATION OF HYDROGEN

323

would predict an analogous acid-catalyzed exchange path in which D 2 0 and DsO+ assume the roles of OD- and D20, respectively:

This analogy is plausible on energetic grounds, since the decreased base strength of the proton acceptor should be approximately compensated by the increased acid strength of the proton donor. In view of the different species involved, however, it is reasonable to expect appreciable differences in the configurations of the transition states and hence in the activation barriers for the two paths. Therefore, the failure to observe an acid-catalyzed exchange reaction cannot be taken as conclusive evidence in favor of the alternative (hydride ion) mechanism. At 100" the rate of the OH--catalyzed conversion of parahydrogen is, a t most, twice that of the D2-H20 exchange. The small magnitude of this isotope effect suggests that proton transfer in the rate-determining step occurs by a classical mechanism rather than by tunneling. This is also indicated by the "normal" value (8 X 10" l.m.-l set.-') of the frequency factor of the reaction. Analogous parahydrogen conversion and deuterium exchange reactions, catalyzed by N H 1 , have been observed in liquid ammonia (Wilmarth and Dayton, 61). The kinetics are of the same form as those of the OH--catalyzed reaction in water and the mechanism is open to similar interpretations. The NH2--catalyzed reaction is much faster, its rate constant at -50" being 10' times that of the OH--catalyzed reaction at 100". The assumption of equal frequency factors for the two reactions leads to a calculated activation energy for the NH2--catalyzed reaction of about 10 kcal. This low value has been attributed to the much greater base strength of NH, relative to OH-. The results provide some support for the hydride ion mechanism. Although bases have played important roles in many of homogeneous hydrogenation reactions discussed earlier, the OH-- and NHz-catalyzed exchange reactions are unique in that they do not involve any metal ions or metal-containing species in the hydrogen-activating process. At the aame time, it is significant that the reactions observed in these systems are of a much more restricted nature than in the earlier ones. The exchange which is observed appears to occur by a mechanism involving only proton shifts, without the formation of any reactive reducing intermediates which could lead to any reaction other than regeneration of a hydrogen molecule. In this sense, OH- and NH2- cannot be regarded as hydrogenation catalysts; nor is it clear that the concept of "activation of hydrogen," implying the formation of a hydrogen-containing species which is significantly more

324

J. HALPERN

reactive than the hydrogen molecule, can be applied to these systems. A t present it appears that the ability to activate hydrogen in this sense, at least in homogeneous systems, is restricted to metal ions and complexes.

F. DISCUSSION 1 . Eleclron ConJiguralionand Reactivity. The ability to activate hydrogen in solution appears to be restricted to a relatively small, but chemically diverse, group of species. It is therefore natural to inquire whether these possess some property in common to which their catalytic activity is related. If such a common basis for catalytic activity exists, it is clearly not to be found in a single mechanism for the hydrogen activation process. Thus, there is ample evidence that hydrogen can be activated in solution both by homolytic and heterolytic splitting and that superficially similar catalysts, or even a given catalyst (e.g., Ag+ or Cu+), under different conditions or in different solvents, can split hydrogen by different mechanisms. A comparison of the electron configurations of the various catalytic species shows more promise. Thus, it can be seen from Table I11 that all the metal ions which react with hydrogen in solution, have outer electron configurations corresponding to nearly filled or just filled d-shells ( d * d L O ;) ions with electron configurations outside this range are inactive. Some insight into the significance of this can be derived from an inquiry into the causes of the inertness of hydrogen. This property is a consequence of the high dissociation energy (ca. 103 kcal.) and the closed-shell electron configuration of the molecule, which result in strong repulsion forces when it approaches to within reacting distance of another electronically-saturated molecule (one in which the low-lying orbitals are filled). The activated complex in such a reaction generally contains more electrons than can be accommodated in low-energy orbitals, and consequently some electrons must be promoted to antibonding orbitals; this is reflected in a high activation energy. This type of situation is depicted by the schematic orbital splitting diagram in Fig. 5a. Little electronic promotion will be required, however, TABLE I11 Electron Configurations of Catalytic Ions Metal ion

cu++ cu+ Pd++ Ag+ Hg++

Outer electron configuration

‘I

Isoelectronic” metal

3d9

co

3d10 4d8 4dl0

Ni Ru Pd Pt

5d10

CATALYTIC ACTIVATION O F HYDROGEN

(a)

(b)

325

(c)

FIG. 5. Schematic orbital splitting diagrams.

if the reactant is a species ( M ) having an empty orbital of sufficiently low energy (Fig. 5b). Such a species can also act as a catalyst for the direct reaction of two electronically-saturated molecules by providing an orbital to accept “unwanted” electrons and thus lowering the energy of the activated complex (Fig. 5c). An interpretation of catalytic activity along these lines has previously been suggested by Eyring and Smith (6%). The presence of low-lying orbitals capable of accepting electrons should be reflected in a high electron-affinity, and this property would thus be expected to characterize all those species which activate hydrogen homogeneously. This is borne out by the data in Table IV, which lists the “electron affinities” (as measured by the sum of the first and second ionization potentials of the corresponding atoms*) of most of the common divalent metal ions. It is seen that the electron affinity increases along each transition series (because of increasing nuclear charge), passing through a maximum value for an ion with a filled or nearly filled d-shell. In each period the highest electron affinity is associated with that ion for which catalytic activity has been demonstrated. A closer examination of the relationships just considered suggests that the optimum configuration for catalytic activity is achieved when the d-shell is filled or nearly filled but when the ion has, or can make available by electron promotion, an empty d-orbital. In the case of ions with d”J configurations, this involves promotion of electrons to higher orbitals, and accordingly it would be expected that the reactivity of the ion should depend inversely on the energy of separation of the d- and s- levels. For an atom or ion of given electron configuratoin, this separation increases with nuclear charge; accordingly, it is not surprising to find that while Cu+, Ag+, and Hg++ are active, the more highly charged isoelectronic ions, Zn++, Cd++, and Tl+++, are inactive. (Even higher activity would be predicted for Au+.) Furthermore, a large decrease in the d-s separation in going from Zn++(3dlo)and Cd++(4dLo)to Hg++(5d1°) helps to explain why

* A similar pattern is obtained if the hydration energies of the ions are subtracted, to give the corresponding electron-a5nities in solution.

J. HALPERN

326

TABLE IV Two-Electron Afinitiee (E)o of Divalent Metal Z m and Work Functions (9) of “Ieoelectronic” Metals (All Values in Electron Volts)

-

Outer electron configuration

dQ

18.0

Metal

Metal

do

dlo

-

dI08

-- -

Ca++ Mn++ Fe++ Metal

da

-

-

23.1

V

24.3

Cr 4.1

4.4

co++

Ni++

25.2

26.8

Mn

Fe

4.0

4.6

Sr++ Tc++

Ru++ Rh++ Pd++

16.7

(24)

(26)

Mo

Te

-

-

-

Nb

4.2

-

Ba++ Re++ OS++ 15.2

-

3.8

27.9

Ru 4.5

dlOel -

Cu++ Zn++

Ga++ Ge++

27.9

27.4

26.5

24.1

co

Ni

cu

Zn

4.3

4.9

4.5

4.3

Cd++ In++

Sn++

29.0

25.9

24.7

21.9

Rh

Pd

Ag

Cd

4.7

5.0

4.7

4.0

Pb++

Ir++

Pt++ Au++ Hg++ T1++

(21)

(24)

(25)

(27)

29.2

29.2

26.4

22.4

Ta

W

Re

0s

Ir

Pt

AU

HII

4.1

4.6

6.0

4.6

4.6

6.4

4.7

4.5

-- - -- ---

M a+

0 E = exothermicity of the process: 2e + Mtp).Data are from LandoltBornstein, “Atom- und Molekularphysik,” Vol.1, p. 211 (1960). Values in parentheses are of uncertain accuracy. * Work function values are from Landolt-Bornstein, “Atom- und Molekularphysik,” Vol. 4, p. 759 (1955).

the latter ion is active and the former two inactive. Of related interest is the influence of the magnitude of the d-s separation on the bonding and stereochemistry of complexes of d10 ions, recently discussed by Orgel (62%). The same principles account for the catalytic activity of dicobalt octacarbonyl and the cobaltous cyanide complexes, since these species, as pointed out earlier, are each just one or two electrons short of a closed-shell configuration. Similarly, the high reactivity of MnOl- is in line with its high one- and two-electron affinities (Carrington and Symons, 6%) ; on the other hand, the isoelectronic C r O r ion, in which the gap between the highest occupied and lowest unoccupied orbitals is considerably larger (Carrington, Schonland, and Symons, 6%), is inactive. 2. Eflects of Complexing. For complexes of the univalent ions (Ag+ and Cu+) the catalytic activity generally increases with the basicity of the ligand or solvent. This has been attributed to increased stabilization of the proton which is released in the splitting of hydrogen or, in one instance

CATALYTIC ACTIVATION OF HYDROGEN

327

where hydrogen splits homolytically, to a reduction of the energy required for the promotion of electrons from the filled d-shell to higher orbitals (Calvin and Wilmarth, 32). For the divalent ions (Cu++ and Hg++), a similar correlation is observed, providing that the complexes formed are relatively unstable (weak metal-ligand bonding). Formation of very stable complexes, however, drastically reduces the activities of these ions, and this effect appears to be more pronounced in nonpolar solvents than in aqueous solution. Two slightly different, but not unrelated, interpretations may be placed on this behavior: (1) that excessive electron displacement from the ligand results in the filling of the low-lying orbitals of the metal ions, so that they are no longer able to accept electrons from hydrogen, or (2) that the reaction of the complex with hydrogen involves separation (possibly as a step in its replacement by a hydride ion) of a ligand from the metal ion, requiring more energy as the strength of the metal-ligand bonding increases or, for a given ligand, as the polarity of the solvent decreases. Both these views emphasize the very close analogy between the effects of complexing in these systems and the ordinary poisoning of heterogeneous catalysts (Maxted, 22) and inhibition of enzymes. In connection with the second interpretation, it is perhaps significant that “poisoning” due to complexing is much more pronounced for the divalent metal ions than for the univalent ones, which, in general, form much weaker metal-ligand bonds. It is also of interest that no catalytic activity has been observed for any metal ion with a charge greater than two. Because of their very high electron affinities, ions such as Tla+, Fea+, and Ce4+ might have been expected t o react readily with hydrogen; their failure to do so may be due to excessively strong metal-ligand bonding (even in the case of the aquo complex). 3. Bifunctional Catalysts. The view that the catalytic activation of hydrogen (and of other inert molecules) requires the concerted action of two adjacent metal atoms has often been expressed, and indeed there are many examples both in heterogeneous and homogeneous systems where this appears t o be the case [cf. Equations (15), (21), (31)]. At the same time, a number of reactions are now well-established in which hydrogen is activated, usually with heterolytic splitting, by catalysts containing only one metal atom (Cu++, Hg++, etc.). It is important to recognize, however, that also such catalysts are “bifunctional,” comprising usually a metal ion and a base which exert concerted action on the hydrogen molecule during the activation process, splitting it by a “push-pull” mechanism [cf. Equation (1311. The two functional groups of the catalyst contribute, to some degree independently, to determining its activity. Thus, the Cu* ion, a good electron and hydride acceptor, activates hydrogen even when coupled with a relatively weak base such as water. At the other extreme, a very strong

328

J. HALPERN

base such as OH- or NH2- appears to be able to activate hydrogen even in the absence of a metal ion, the role of the hydride acceptor being assumed by an H 2 0 or NHs molecule. Greatly enhanced activity, however, can be achieved with a catalyst such as cupric acetate, in which a metal ion possessing good electron- and hydride-accepting properties is coupled with a relatively strong base. This emphasizes a very important general principle in catalysis, with wide applications both in the understanding of catalytic phenomena and in the preparation of more powerful and selective catalysts. It appears, however, that this principle cannot be exploited very effectively in relatively simple species, such as ordinary metal complexes. Thus, metal ions such as Cu++ and Hg* in aqueous solution cannot be coupled, simply by complexing, with very strong bases, either because (as in the case of OH-) of solubility restrictions or because catalytic activity is destroyed by metal-ligand interaction. To be effective, the two functional groups must be so disposed that they can interact simultaneously with a hydrogen molecule, but at the same time are prevented from interacting with (neutralizing) each other. This involves anchoring the groups to some “rigid” framework such as a surface or a suitable molecule. An interesting model system of this type has been described by Swaiii and Brown (64, who examined the effectiveness of various catalysts in the mutarotation of tetramethylglucose in benzene, the rate determining step being the hydrolysis of the hemi-acetal link. This reaction requires simultaneous attack of a nucleophilic reagent (base) at the H position and of an electrophilic reagent (acid) a t the 0 position to give the open chain aldehyde form of the sugar. Thus, while neither pyridine nor phenol alone was very effective, a mixture of the two was found to be a powerful catalyst, the kinetics being third order, i.e., of the form, k[sugar][pyridine][phenol]. Furthermore, while 2-hydroxy pyridine is a weaker base than pyridine and a weaker acid than phenol, its catalytic activity was found to be much more pronounced (particularly at low concentrations) than an equivalent mixture of the two. This compound is believed to act as a bifunctional catalyst through a mechanism of the following type: H

b



11

H H

+

I

0

I

H

H

CATALYTIC ACTIVATION OF HYDROGEN

329

Similar explanations, supported by convincing evidence, have been proposed for the high catalytic efficiency and specificity of enzymes (particularly hydrolytic enzymes) relative to ordinary catalysts (Laidler, 23, 24; Wilson, 66, 66).

111. Heterogeneous Systems A. METALS 1 . Introduction. It is generally recognized that there is close connection between heterogeneous catalysis and chemisorption and that catalyzed reactions on solid surfaces proceed through chemisorbed intermediates. Unfortunately, the problems of elucidating the mechanisms of heterogeneous reactions are much more difficult than in homogeneous systems, and in most instances the connection referred to above is not understood in detail. Nevertheless, the study of chemisorption has provided one of the most fruitful approaches to the elucidation of catalytic phenomena (Trapnell, 67,68;de Boer, 69; Becker, 70). Chemisorption of a saturated molecule generally involves its dissociation and consequent bonding between the fragments and the surface. For the chemisorption of hydrogen on a metal this process can be represented by the familiar Lennard-Jones (71) diagram depicted in Fig. 6 . Most adsorption processes are exothermic, and hence the activation energy for desorption (Ed)is greater than that for adsorption (G);thus, desorption is likely

D I S T A N C E F R O M M E T A L SURFACE

FIG.6. Schematic potential energy diagram for the adsorption of hydrogen on a metal surface.

330

J. HALPERN

to be the rate-determining step in many heterogeneously catalyzed reactions. Since the activation energy for this step is given by Ed

=

Ea

+Q

(45)

where Q is the heat of adsorption, it is apparent that for high catalytic activity E , and Q must both be small; i.e., chemisorption must be both rapid and weak (Polanyi, S). This establishes rather narrow limits for the manifestation of catalytic activity, and the understanding of catalytic phenomena is seen to rest in large measure on an understanding of the factors which influence E, and Q. Some further insight into these relationships is provided by Schwab and Killman's ( n u ) recent attempt to calculate the activation energy of the HrDz exchange reaction in the gas phase and on a nickel surface using the semi-empirical method of Eyring and Polanyi (7lb).From this calculation it appears that the catalytic effect is related primarily to the lowering of the dissociation energy of the adsorbed molecules. This, in turn, is related, but not necessarily in a simple way, to the heat and activation energy of chemisorption. From the experimental standpoint, there is much evidence t o support the view that the ability of metals to chemisorb hydrogen readily and to function as efficient hydrogenation catalysts is related to the presence of an unfilled (but not too empty) d-band (Trapnell, 67). Thus, it is only the transition metals which chemisorb hydrogen rapidly below room temperature, and it is also these metals, particularly those belonging t o group VIII of the periodic table, which are the most active hydrogenation catalysts. Moderate activity above room temperature is exhibited by copper, silver, and gold, the elements immediately succeeding the transition groups, and this has been attributed to the ease of d-s promotion and consequent creation of d-band vacancies in these metals (Trapnell, 72; Mikowsky, Boudart, and Taylor, 73). Most other elements are inactive even at higher temperatures. This pattern of dependence of catalytic activity on electron configuration is strikingly similar to that which has already been described for metal ions. This is emphasized by Table 111, which shows that while most of the metals whose ions activate hydrogen homogeneously (e.g., Cu, Ag, Hg) are not themselves catalytically active, the elements whose atoms are isoelectronic (i.e., have the same number of outer electrons) with these ions are among the best metallic catalysts (Ni, Co, Pd, Pt, etc.). This suggests an important link, through electron configuration, between catalysis in homogeneous and heterogeneous systems. The interpretation proposed earlier for metal ions, namely, that catalytic activity is associated with high electron affinities,

CATALYTIC ACTIVATION OF HYDROGEN

33 1

also appears to have applicability to metallic catalysts, and may provide at least a partial explanation for this link. Thus, it can be seen from Table IV that the electron affinities (work functions) of metals follow a pattern resembling that of the “isoelectronic” ions, the maximum in each period occurring just prior to the filling up of the d-band and being associated with those metals which exhibit pronounced catalytic activity (i.e,, Ni, Pd, and Pt)* The explanation for the apparent correlation between catalytic activity and electron affinity of metals cannot be as simple as that which has been advanced for the homogeneous catalysh. This is because chemisorption on metals (unlike the splitting of hydrogen by metal ions in solution*) is an exothermic process and, hence, as shown earlier, catalytic activity depends not only on a low activation energy of adsorption but also on a low heat of adsorption. The interpretation applied earlier to homogeneous catalysts can account for an inverse dependence of E, on the work function, but does not suggest any obvious reason why Q should show a similar dependence. Actually, Conway and Bockris (75) have drawn attention to the existence of an inverse correlation between the heats of adsorption of hydrogen on metals and their work functions. Previously, Beeck (76) had observed an inverse dependence of the heat of adsorption of hydrogen on various metals, on the per cent d-character of the metallic bonding (Fig. 7), the latter being deduced from Pauling’s theory of metals (77). This has generally been interpreted as implying that bonding of the chemisorbed species involves atomic d-orbitals of the metal, the availability of which is inversely related to the degree of participation of d-electrons in the bonding between the metal atoms themselves. This dual dependence also implies that there must be some degree of mutual correlation between the work function of a metal and its per cent d-character; this is borne out by Fig. 8. At the present time, there does not appear to he any satisfactory theoretical interpretation of these observations, and the correlations described above must be considered of a semiempirical nature. It should also be noted that in the Pauling theory of metals, from which the concept of per cent d-character of metallic bonding is derived, this property is itself a function of the interatomic distance in the metal; this illustrates one of the difficulties in distinguishing between the influence of electronic and geometric factors on the catalytic properties of metals. Dowden (78) has recently discussed the factors affecting the strength of chemisorptive bonds on different types of solids. He has suggested that weak chemisorption, which is often associated with high catalytic activity,

* Processes such as those represented by Equations (9) and (16) may be likened to the examples of endothermic chemisorption discussed by de Boer (74).

332

J. HALPERN

Rh

20

46

40

PERCENT

d

-

I

50

CHARACTER

FIG.7. Dependence of the heat of chemisorption of hydrogen (Q)and the catalytic activity for ethylene hydrogenation ( K ) on the percent d-character of the metallic bonding [after Beeck, O., Discussions Faradau SOC.8, 126 (1950)l.

involves essentially d-orbitals, while strong chemisorption arises from dsphybrid orbitals. 2. Chemisorption of Hydrogen. Studies on the chemisorption of hydrogen on metals, prior to 1954, have been reviewed by Eley (3)and by Trapnell (68, 79), and only a few of the more important results will be mentioned here. I n subsequent sections, consideration will be given to more recent results obtained using a number of promising new experimental techniques including electrical, magnetic, and spectroscopic measurements. a. Classical adsorption studies. General techniques for the preparation of clean metal surfaces have been developed only comparatively recently, and most of the reliable earlier adsorption measurements were made on tungsten wires whose high melting point and low vapor pressure permits heating i n vacuo to very high temperatures (-2500" K) to remove surface contamination. On such wires, Roberts (80) found, using the accommodation coefficient method, that chemisorption of hydrogen proceeded rapidly even at 79" K and that a complete monolayer was apparently formed at hydrogen pressures as low as mm. He also showed (using resistance measurements to follow the increase in temperature of the wire resulting

333

CATALYTIC ACTIVATION OF HYDROGEN

5.5

gPt

I

5.0

ln t-

-I

0

> z

0

a

g Fe

t0 W

/

J

W

5 4.5 .8

OCU

4.0

25

I

I

I

29

33

37 ‘1. d -

/ I 41

I

I

45

49

53

CHARACTER

FIG.8. Relation between work function and percent &character. [Conway, B. E., and Bockris, J . O’M., J . Chent. Phys. 26, 532 (1957) .]

from the liberation of the heat of adsorption) that the heat of adsorption of hydrogen decreases with increasing surface coverage, from 45 kcal./mole for a bare surface to 18 kcal./mole for a nearly saturated one. Careful measurements by Frankenburg (&I), using tungsten powders, indicated an even more pronounced fall in the heat of adsorption, but gave lower values for the surface coverage than had been obtained by Roberts. Other important studies on tungsten wires include Bosworth’s (82) contact potential measurements, which revealed that the work function of a hydrogen-covered tungsten surface is about 1 volt higher than that of a clean one. (More recent measurements of Mignolet (83,84) on evaporated tungsten films give a somewhat lower value, -0.5 volt.) From this the surface dipole has been estimated to be about -0.4D, the negative value indicating that the adsorbed hydrogen layer is negative. There is, however, some uncertainty about the detailed interpretation of such measurements (Boudart, 85). Possibilities for the study of adsorption on clean metal surfaces have

334

J. HALPERN

been extended greatly in recent years by the development of techniques for the preparation of evaporated metal films (Beeck et al. 86-88; Trapnell, 72). Beeck and his coworkers (88, 89) have studied the adsorption of hydrogen and other gases on films of metals such as tungsten, nickel, and iron. Their results for hydrogen on tungsten were in good agreement with those obtained earlier by Roberts using tungsten filaments. They found that the activation energy for the adsorption of hydrogen on all these metals was very low and that the heat of adsorption decreased with increasing surface coverage (Fig. 9). The initial heats of adsorption of hydrogen on different metals were found to depend inversely on the per cent d-character as shown in Fig. 7; the significance of this has already been considered. Other important measurements on the chemisorption of hydrogen include those of Rideal and Trapnell (90-92) on tungsten films and those of Schuit and his coworkers (99, 94a, b ) on silica-supported metal catalysts. In each case the heat of adsorption was observed to decrease to very low values as complete coverage was approached. b. Electrical resistance of thin metal films. Techniques for the measurement of changes in the electrical resistance of thin metal films, when gases are adsorbed on their surfaces, have been developed by Suhrmsnn and his coworkers (96, 96) and employed to obtain information about electronic interactions between the metal and adsorbed substance. The change in resistance accompanying the adsorption of hydrogen on nickel at 293" K. is depicted in Fig. 10. The decrease in resistance is interpreted as denoting transfer of electrons from hydrogen to the metal, the hydrogen being adsorbed as a positive layer. On pumping off, only part of the hydrogen is removed, as shown by the increase in resistance to a limiting value. Even

I

0.2 04 0.6 0.8 1.0 FRACTION OF SURFACE COVERED

FIG.9. Dependence of the heat of adsorption of hydrogen on surface coverage. [Beeck, O., Discussions Faraday SOC.8, 118 (1950).]

CATALYTIC ACTIVATION OF HYDROGEN

335

Hq on Ni T = 293’K A = H2 admitted B = H2 pumped oft

0

60

120

180 Time

240

300 min.

FIG. 10. Changes in electrical resistance during the chemisorption of hydrogen on nickel films. [Suhrmann, R., in “Chemisorption” (W. E. Garner, ed.), p. 106. Academic Press, New York, 1957.1

at 90” K., admission of hydrogen over a nickel film caused an instantaneous decrease in resistance, indicating rapid chemisorption. The magnitude of the resistance change increased with the hydrogen pressure up to about mm. when complete surface coverage was apparently attained. No change in resistance was observed on pumping off at this temperature. The above results were obtained on tempered nickel films, i.e., films which had been heated to room temperature. On untempered films which had been condensed and kept a t 90” K., adsorption of hydrogen was accompanied by an increase in resistance. This was attributed to the presence of some crystalline spots with abnormally low work functions; it was suggested that adsorbed hydrogen atoms migrate to these spots, where they take up electrons and form a negative layer. In support of this, it was found that the work function of this film was lower by 0.45 v. than that of a tempered one. The hydrogen chemisorbed on such a film is apparently more weakly held, since some of it could be pumped off even a t 90°K. Sachtler and Dorgelo (96a,b) observed that the direction of change of resistance, when hydrogen is adsorbed on a nickel film, depends on the conditions under which the film is prepared. With films prepared by evaporation under very high vacuum (
336

J. HALPERN

the former and negative in the latter case. These differences were attributed to contamination of the latter film. Sachtler and Dorgelo (96c) also found that the direction of change of resistance varies with the amount of adsorbed hydrogen. As the hydrogen pressure is increased, the resistance of clean nickel films prepared under high vacuum first increases, passes through a maximum, and then falls.* This would appear to be related to Mignolet's (83) observations that a t low surface coverage the surface potentials of hydrogen films on nickel and other metals are negative but with increasing coverage they pass through minima and finally attain positive values. Sachtler and Dorgelo (96c) have interpreted these observations in terms of two forms of adsorbed hydrogen which appear successively and which affect the resistance (and surface potential) in opposite directions. The second form, which causes the resistance to decrease, is less stably bound than the first and is preferentially removed on pumping. The nature of the two forms of chemisorbed hydrogen, whose existence now appears to be well established, has also been ) by Mignolet (96d). One suggestion is that discussed by Dowden ( 7 8 ~and the second form consists of molecular hydrogen stabilized by charge transfer interaction (M-. . .Hz+)with the metal; another is that the two types of adsorption are associated with different lattice sites. Using platinum fdms, Suhrmann, Wedler and Gentsch (96'e) obtained results which, while differing in some respects from those for nickel, also provide evidence for at least two types of hydrogen adsorption. At room temperature and low surface coverage (0 < 1) hydrogen appeared to be adsorbed atomically with some further dissociation into protons and electrons which diffused into the bulk of the metal. Part of the resistance change was slow and apparently governed by this diffusion process. This adsorption was irreversible and was not accompanied by any change in the photoelectric emissivity of the surface. When the hydrogen pressure was increased beyond about loFanun., so that the surface coverage exceeded unity, another form of adsorption was observed which was accompanied not only by a decrease in resistance but also by an increase in the photoelectric emissivity. This change was reversible and was attributed to a molecular film polarized in the direction of Ht+. At lower temperatures (77 and 90°K.) the decrease in resistance resulting from the initial adsorption was instantaneous, (indicating the absence of diffusion effects), and was accompanied by a decrease in the photoelectric emissivity. The former effect was attributed to dissociation of some of the adsorbed hydrogen into

* This observation has been confirmed by Suhrmann (private communication). The results shown in Fig. 10 are characteristic of the higher pressure region, exceeding about 10-2 mm. Ho

.

CATALYTIC ACTIVATION OF HYDROGEN

337

protons and electrons and the latter to adsorbed atoms polarized in the direction of H-. At higher pressures and surface coverages (0 > l), reversible adsorption accompanied by a decrease in resistance and increase in photoelectric emissivity, and attributed to H2+, was again observed. c. Magnetic susceptibility measurements. In a series of papers published since 1956, Selwood and his co-workers (97-102) have described a method for making simultaneous measurements of the volume of gas adsorbed on a nickel surface and of the accompanying change in the magnetic susceptibility of the metal. Since the method (like that of electrical resistance measurements described above) depends on measurement of the change in a bulk property, due to a surface effect, it is only applicable to very finely divided metal (-100-A. particles). This condition is fulfilled for active supported nickel catalysts, and most of the measurements to be discussed were made using a Universal Oil Products Co. nickel-kieselguhr catalyst (52.8% Ni). Typical of the magnetic changes which are observed when hydrogen is adsorbed on such a catalyst a t room temperature, are those depicted in Fig. 11. If the hydrogen is introduced slowly, the magnetic susceptibility falls progressively as the pressure is increased, leveling off at a constant value as the pressure levels off a t one atmosphere. This decrease in magnetization is attributed to pairing of electrons in the nickel d-band by electrons transferred from the adsorbed hydrogen. If the hydrogen is admitted rapidly, there is a larger initial drop in the magnetization followed by a recovery to the same final value as in the previous case. This transient effect is due to heating up of the nickel by the liberated heat of chemisorption. On pumping off, the magnetization rises but does not reach its original value; apparently only part of the hydrogen is desorbed. Readmission of

TIME

- MINUTES

FIQ. 11. Changes in specific magnetization of a supported nickel catalyst during adsorption and desorption of hydrogen at room temperature. [Selwood, P. W . , J . Am. Chem. Soe. 78, 3893 (1956).]

338

J. HALPERN

one atmosphere of hydrogen causes the magnetization to fall again, but the transient effect is not observed; apparently the heat of adsorption associated with the “removable” fraction of the chemisorbed hydrogen is very low. Typical magnetization-volume and pressure-volume isotherms obtained by this method are shown in Fig. 12. The magnetization-volume isotherm is linear; i.e., each hydrogen molecule adsorbed causes the same decrease in magnetization. In the interpretation of these results, and those for other gases to which this method has been applied, it has been assumed on the basis of reasonable, but not conclusive, evidence that the slope of this isotherm corresponds to the entry of one electron into the nickel d-band for each chemisorbed H atom. From Fig. 12 it is evident that chemisorption of hydrogen is not complete a t one atmosphere, and recent measurements (Vaska and Selwood, 109) indicate that it continues to much higher pressures. One of the valuable features of this method is its ability to distinguish clearly between physical adsorption and chemisorption, since only the latter causes an appreciable change in magnetization. Thus, Selwood (97‘) has demonstrated conclusively that some hydrogen is rapidly chemisorbed by nickel even at -196”, but the amount is small compared with that at higher temperatures. At room temperature it was found (Lee, Sabatka, and Selwood, 98) that the rapid initial adsorption of hydrogen is followed by a very slow further uptake associated with a much smaller decrease in magnetization. It was suggested that this slow process is due to diffusion of hydrogen into the

0.88

0

4

8

I2

16

20

Cc. H2/g. Ni. FIQ.12. Magnetization-volume and pressure-volume adsorption isotherms for hydrogen on nickel-kieselguhr at 27”. [Selwood, P. W., J . Am. Chem. Soc. 79,3346 (1957).I

CATALYTIC ACTIVATION OF HYDROGEN

339

catalyst micropores with chemisorption on very small particles which are nonmagnetic at room temperature. The thermal transient effect has been used to estimate the heat of chemisorption of hydrogen and its variation with surface coverage, giving the results shown in Fig. 13 (Lee et al. 98). The initial heat of adsorption (-35 kcal./mole) is in good agreement with earlier estimates by other methods but the fall with surface coverage to a negligible value is more pronounced than most previous measurements indicated; this effect is of great importance in connection with the interpretation of catalytic phenomena. Selwood (100, 101) has also employed this technique to study the chemisorption of ethylene, ethane, benzene, and cyclohexane on supported nickel catalysts. Among the new information and important conclusions derived from these measurements are the following: 1. The catalyst surface appears to be heterogeneous, containing a t least two types of sites; thus, there are three to four times as many sites available to hydrogen as to ethylene and over six times as many as to ethane. 2. At room temperature, ethylene is adsorbed, without fragmentation, by opening of the double bond and formation of two Ni-C bonds; similarly, benzene appears to be adsorbed by six-site (Ni-C) attachment. Adsorption of saturated hydrocarbons involves fragmentation with the formation of Ni-H, as well as Ni--C, bonds. At 100" ethylene also undergoes fragmentation.

Cc. H,/g.Ni

FIQ. 13. Variation of the differential heat of chemisorption of hydrogen on nickelkieselguhr calculated from magnetic measurements. Relative magnetization (v/uo) are also shown. [Lee, E. L., Sabatka, J . A., and Selwood, P . W., J . Am. Chem. SOC. 79, 5391 (1957).]

340

J. HALPERN

3. There is some evidence that at room temperature, chemisorbed hydrogen cannot, migrate on the catalyst surface. 4. Hydrogen is chemisorbed on a surface partially covered with ethylene and reacts rapidly with the ethylene at room temperature. This is not consistent with observations thnt hydrogen-deuterium exchange on nickel is strongly inhibited by the presence of chemisorbed ethylene. The latter effect may be due to fragmentntion of ethylene which occurs only a t higher temperatures. 5. Hydrogen is chemisorbed on a henzene-covered surface, displacing the benzene into a van der Waals layer. Hydrogenation apparently involves reaction of this physically adsorbed benzene with chemisorbed hydrogen and proceeds only when the surface is nearly covered with hydrogen. This is attributed to weakening of the Ni-H bond with increasing surface coverage. Apparently the chemisorbed hydrogen is reactive only if its heat of adsorption, which depends inversely on the surface coverage, is below about 12 kcal./mole. A few measurements of the changes in magnetization of nickel resulting from the chemisorption of hydrogen, ethylene and acetylene have also been reported by Broeder and his co-workers (lo%, b). Their results and conclusions appear to be qualitatively in accord with those of Selwood. d . Infrared spectra of chemisorbed species. Techniques developed since 1954, principally by Eischens and his co-workers (10.4-107), permit the direct determination of the infrared absorption spectra of species chemisorbed on supported metal particles or on thin metal films. Such measurements have already provided important information about the nature of chemisorbed species and the mechanisms of catalyzed reactions. The infrared spectra of CO chemisorbed on Pd and Pt have bands a t 1,820-1,925 cm.-l and a t 2,070 cm.-’. By analogy with the spectra of metal carbonyls these have been attributed to “linear”

and “bridged”

structures, respectively (Eischens et al., 104, 108). The former structure predominates on Pt (Fig. 14) and the latter on Pd. Since the lattice parameters of Pt and Pd are almost identical, this difference is not readily

CATALYTIC ACTIVATION OF HYDROGEN

34 1

FREWENCY cm-I

Fro. 14. (A) Infrared spectrum of CO chemisorbed on silica-supported platinum; (B) after treatment with hydrogen. [Eischens, R. P., J . Chem. Educ. 36, 385 (1958).] ’

explained by geometrical concepts, and it has been attributed, instead, to chemical factors, reflecting differences in the availability of electrons a t the surfaces of the two metals. The chief concept involved here is that the bridged structure requires more electrons (and hence the contribution of more electrons by the metal) than the linear one. In line with this, it is of interest that the addition of Hz to CO adsorbed on a Pt surface causes an increase in the intensity of the band due to bridged CO and a shift in the band due to linear CO in the direction of lower frequencies. These effects suggest that the adsorption of Hz makes it easier for Pt to contribute electrons to bonding with CO. This, in turn, implies that the chemisorbed hydrogen has donated electrons to the metal and is adsorbed as positive ions. The suggestions that hydrogen is chemisorbed as covalently bonded atoms or as hydride ions are more difficult to reconcile with these observations. The failure to observe any bands clearly attributable to chemisorbed hydrogen has also been taken (Eischens, 108) as “negative” evidence against covalent bonding, since it is likely that hydrogen covalently bonded to a surface metal atom would have an absorption band in the 4,0001,250-cm.-’ region.* Using a somewhat different technique, involving multiple reflection of the infrared beam by the catalyst surface instead of, as above, transmission through a transparent catalyst sample, Pickering and Eckstrom (108a) did note a marked decrease in the reflectivity of a rhodium surface on admitting hydrogen, as well as the appearance of a number of absorption bands which they attributed to chemisorbed hydrogen. However they were unable to interpret the spectra in detail. Infrared measurements have also provided information about the catalytic hydrogenation of olefins (Eischens, 106). Thus, the spectrum of ethylene (Fig. 15) chemisorbed on a nickel surface (on which some hydrogen

* See Note Added in Proof (I), p. 370.

J. HALPERN

342

35

4.0

/A

65

20

Fro. 15. (A) Infrnrrtl spertrn of ethylene rhemisorhed on nirkel; (B) after treatnirrit with hydrogen. [ISischenn, It. P., 2. Elektrorhem. 60,782 (1956).]

has been preadsorbed) , suggests that the predominant species is H2C-CH2

* *

(where the asterisk denotes a bond between the carbon atom and the metal surface). On increasing the hydrogen pressure, a new band appears which can be identified with adsorbed ethyl radicals, H2C-CHs. Similar

*

evidence for half-hydrogenated intermediates was found for other olefins, providing direct support for hydrogenation mechanisms which postulate a step of the type HIC-CH, *

i

t

+ Hz

HC-CHs I)

+H

e. Bond type in chemisorbed hydrogen. Many of the experimental phenomena discussed above touch on the nature of the bonding between hydrogen in :I chemisorbed film and the surface of the underlying metal. This question, which is of great importance in relation to catalytic phenomena, has been the subject of some disagreement, and a brief review of the present position is therefore in order. The reader is also referred to recent discussions of this question by Broeder et al. (103b),Schuit and van Reijen (94b) and Takaishi (10%). The three extreme cases of bonding which can be considered, correspond to the adsorption of hydrogen as H+ ions, covalently bonded H atoms or H- ions. Dowden (109) hiis considered, from a theoretical standpoint,, the factors favoring each of these types of adsorption. The criteria for negative ion formation are opposite to and readily distinguishable from the other two; the distinction between the criteria for the formation of covalent bonds

CATALYTIC ACTIVATION OF HYDROGEX

343

and positive ions is, however, less well defined. This also applies to most experimental criteria. Evidence for the adsorption of hydrogen as negative ions is uncommon and, while not inconsistent with some of the electrical resistance and surface potential measurements quoted abovc, this type of bonding is not readily reconciled with the magnetic measurements and other considerations. It is therefore generally considered unlikely. Both covalent bonding and positive ion formation have, however, been considered as reasonable possibilities, the former generally being favored, but not conclusively in all cases. Among the evidence in favor of covalent bonding, the following may be cited: 1. The inverse correlation between the heat of chemisorption and the per cent d-character of bonding in the metal. 2. Values of the heat of adsorption of hydrogen on various metals calculated by Eley (1 10) using P:wling’s covalent bond energy formula, i.e., E(M-H)

=

f$[E(M-M)

+

E(H-H)]

+ 2 3 . 0 6 [ x ~-

xHI*

(47)

where E refers to bond energy and X to electronegativity, are in reasonable agreement with experimental ones. 3. Simplified calculations (Emmett and Teller, 111; Couper and Eley, 119) (which do not however adequately take account of polarization effects) indicate that adsorption of hydrogen as ions is prohibitively endothermic. 4. Contact-potential measurements usually indicate only a small surface dipole associated with chemisorbed hydrogen, and in the case of metals such as tungsten and nickel, the negative end of the dipole appears to be on the outside. While there is some uncertainty about the detailed interpretation of these measurements, they are more readily reconciled with covalent than with ionic bonding. On the other hand, some of Suhrmann’s electrical resistance measurements on nickel and platinum films and Eischens’ observations. of the effect of Hz 011 the infrared spectrum of chemisorbed CO on platinum, referred to earlier, suggest electron-transfer from hydrogen to the metal, i.e., adsorption of positive ions. It should be noted that Selwood’s observations that chemisorption of hydrogen lowers the magnetization of nickel are consistent with either covalent bonding or positive ion formation, since they demonstrate only pairing of d-band electrons (Selwood, 99). In view of the conflicting indications provided by different measurements, it seems likely that the bond type in chemisorbed hydrogen is actually of an intermediate and variable nature, approaching essentially

344

J. HALPERN

ionic or covalent character only under limiting conditions. Reference should also be made to the evidence which has accumulated in favor of a “second”, more weakly bound, form of chemisorbed hydrogen, observed under a variety of conditions, particularly at high surface coverage. While there may be grounds for distinguishing this from physical adsorption, it is nevertheless probable that the adsorbed species is molecular, probably hydrogen molecules polarized in the direction of Hz+. The significance of this type of adsorption in relation to catalytic phenomena is not altogether clear but there are indications that it may be important particularly at low temperatures. In connection with this it should be stressed that in the metal hydride intermediates (CuH+, AgH+, AgH, etc.), which are formed in the homogeneous reactions of hydrogen with metal ions in solution, the character of the metal-hydrogen bonding is believed to be essentially covalent. f. Decrease of the heat of chemisorption with surface coverage. This is a fairly general phenomenon in adsorption on metals and of great importance in relation to catalysis, since catalytic activity tends to depend inversely on the heat of adsorption. One of the factors which may contribute to this effect and which has long been recognized as having an important bearing on adsorption and catalysis, is the heterogeneity of the metal surface. This refers to the presence on the surface of different types of sites. Since adsorption tends to occur first on the more strongly adsorbing sites, the apparent heat of adsorption on a heterogeneous surface should fall with increasing surface coverage. Surface heterogeneities of this type, associated with the, presence of different crystallographic planes, plane edges and corners, dislocations, lattice defects, impurities, etc., are undoubtedly common. It is now believed, however, that this intrinsic or a priori surface heterogeneity is not the most general cause of the effect under consideration and that much of the observed decrease in the heat of adsorption with increasing surface coverage arises from effects associated with the adsorption process itself (Boudart, 86). Among these are repulsions between the adsorbed molecules and changes in the work function of the metal. The latter effect, which is due to the dipoles associated with the adsorbed species, is particularly important and has been termed induced heterogeneity (Boudart, 86) or the work function eflect (de Boer, 69, 123). Part of the variation in heat of adsorption with surface coverage is also undoubtedly associated with the occurrence of different forms of chemisorption, to which reference has already been made. This subject has been reviewed recently by de Boer (69, 113) and by Schuit and van Reijen (94b). 3. Catalytic Mechanisms on Metals. a. Parahydrogen conversion and

CATALYTIC ACTIVATION OF HYDROQEN

345

hydrogen-deuteriumexchange. The activation of hydrogen on metal surfaces is revealed most clearly by these reactions, since they involve no reactants or products other than hydrogen itself. On transition metals the parahydrogen conversion proceeds predominantly through the chemical mechanism (as distinct from the magnetic one which will not be considered here). Since this mechanism involves dissociation of hydrogen, it is also manifested in an exchange reaction when a hydrogen-deuterium mixture is used. Two types of mechanisms have been proposed by Bonhoeffer and Farkas (114) and by Rideal (116), respectively. For the exchange reaction these can be represented as

+ Hz 2M + Dz

2M

i= 2M-H

2M-D

M-H

+ M-D

2M

+ HD

(48)

and 2M

+ Hz + 2M-H

M-H+DzeM-D+HD

(49) (50)

where M is a catalyst atom. Thus, the Bonhoeffer-Farkas mechanism involves evaporation of hydrogen moleculesformed by random combination of atoms in the chemisorbed layer, while the Rideal mechanism involves reaction between a chemisorbed atom and a hydrogen molecule from the gas phase or from a van der Waals layer. Trapnell ('79) has reviewed the work on these systems prior to 1952 and has attempted to interpret all the available information in the light of these mechanisms. Most of the pertinent earlier measurements were made on tungsten or nickel. Two of the most important observations are (1) that exchange on these metals is rapid even at temperatures below - 100" and (2) that the order of the reaction with respect to the hydrogen pressure is considerably less than unity (0.3-0.6). At one time it appeared difficult to reconcile both these observations with either of the above mechanisms. Thus, Roberts' (80) studies, referred to earlier, suggested that the chemisorption of hydrogen on tungsten was rapid even a t liquid air temperatures but that evaporation of the chemisorbed film (on pumping) occurred only above 700" K. The latter observation appeared to rule out the Bonhoeffer-Farkas mechanism. On the other hand, the Rideal mechanism could not readily account for the low reaction order. Subsequent adsorption studies, however, with more sensitive techniques than those used by Roberts indicate that chemisorption of hydrogen on tungsten (Frankenburg, 81, Rideal and Trapnell, 90-92) and nickel (Selwood, 98, 102) is incomplete until much higher pressures and that the latter stages of chemisorption are associated with very low heats of adsorption, approaching zero for complete surface coverage. This means that the

346

J. HALPERN

reversible chemisorption of hydrogen proceeds even a t very low temperatures provided that the surface coverage is high enough, and the validity of the Bonhoeffer-Farkas mechanism thus appears to be established. The question of whether a mechanism of the Rideal type is simultaneously operative and contributes to the over-all rate of conversion or exchange remains open. Trapnell (92, 79) has attempted to answer this question by comparing the observed rate with that calculated for the BonhoefferFarkas mechanism alone. The agreement is satisfactory, suggesting that the Rideal mechanism does not make an important contribution; however, the calculation involves a number of assumptions, and the result cannot be considered as entirely conclusive. In the light of new data which fail to show the pressure dependence predicted for the Bonhoeffer-Farkas mechanism, Couper ~t al. (116a), while tigrcdng that, this mechanism probably applies to transition metals at higher teniperatures and to sp-metnls :it a11 temperatures, continue to favor the liideal mechanism for tungsten helow 108°K. Recent evidence, discussed earlier, for a “second” layer of hydrogen is quoted t o reconcile this mechanism with the observed low reaction order. Schuit and his co-workers (94a, b) have recently made a careful study of the hydrogen-deuterium exchange reaction on silica-supported metal catalysts. The catalytic activities and apparent activation energies, dctermined for various metals, are listed in Table V. The increase in activation energy was accompanied by a partially compensating increase in the frequency factor. Attempts to interpret the adsorption and kinetic data for nickel in terms of a Bonhoeffer-Farkas or a Rideal mechanism alone were unsuccessful, and it was suggested that both mechanisms might be operating simultaneously. TABLE V CaLalUtic rlctivilies of Various Silica-Supported Metals for H?-D? Exchangea Metal

Pt Rh Ru Ni co Fe cu

Relative catalytic activity

Apparent activation energy (kcal.)

1 1 1

1 3

10-2 10-2 10-4 10-4

-

8 8

Schuit, G. C. A., de Boer, N. H., Dorgelo, G. J . H., and van Iteijen, L I,., i n “Chemisorption” (W. E. Garner, ed.), p. 39. Academic Press, New York, 1957.

347

CATALYTIC ACTIVATION OF HYDROGEN

1.0

-

293 O K

/.

0.8 -/ 9

e

w

0

3 0.6-

E 4

w / g

195%

a --.

--

3- 170°K

/

= 0.4 0” \

&

a

02 -

140°K

-t-

*(l-’l-S

I

40

80

120

77%

, 160

FIG. 16. Exchange of hydrogen with prentlsorbed deuterium on s silica-supported nickel catalyst. [Schuit, Q . C. A . , de Boer, N . H., Dorgelo, G . J . H., and van Reijen, I,. L . , in “Chemisorption” (W. 14:. Garner, ed.), p. 39. Academic Press, New York, 1957.1

An interesting effect (Fig. 16) was observed when a catalyst, covered with preadsorbed deuterium, was brought into contact with gaseous hydrogen. Part of the adsorbed deuterium was found to exchange very rapidly, another part a t a measureable rate, and a residual part virtually failed to undergo exchange. The fraction which exchanged rapidly increased with the temperature; thus a t 77°K. there was almost no exchange, while a t 293” K. all the adsorbed deuterium exchanged rapidly. The composition of the gas phase always corresponded to the equilibrium HD :D,: H1: ratio. This behavior suggests that the metal surface is heterogeneous and possesses a variety of sites, of which only certain ones are active at any given temperature. It would also appear that the chemisorbed hydrogen atoms are immobile and cannot migrate from (‘inactive” to “active” sites. As mentioned earlier, Selwood’s magnetic studies (100) on supported nickel catalysts lead to similar conclusions. An analogous exchange experiment performed earlier by Eley (116), using a clean evaporated tungsten film, also showed a slight va,riation in activity over the surface.

348

J. HALPERN

ATOM

O h

Au.

Fro. 17. Activation energy for parahydrogen conversion on palladium-gold alloys. The broken line denotes the paramagnetic susceptibility in arbitrary unite. [Couper, A . , and Eley, D. D., Discussions Faraday Soc. 8, 172 (1950).]

Couper and Eley (118) studied the parahydrogen conversion reaction on palladium-gold alloys and determined the dependence of catalytic activity on alloy composition. They found (Fig. 17) that the activation energy increases abruptly at the composition (60 atom % gold) at which the d-band, which in pure palladium contains 0.6 hole per atom, is just filled by the s-electrons of the added gold. This filling of the d-band is also reflected in the decreasing paramagnetic susceptibility of the alloy, which approaches zero at about the same composition. These observations, as well as results of a number of other catalytic studies on alloy systems (Schwab, 11‘7; Dowden and Reynolds, 118) which are capable of similar interpretations, emphasize the importance of the electronic factor and particularly of an unfilled d-band, in relation to catalytic activity. b. Other hydrogenation reactions. Many other reactions of hydrogen, involving addition to or exchange with a variety of compounds, are also catalyzed by the transition metals. A consideration of the mechanisms of these reactions touches on many problems not directly related to the activation of hydrogen and is beyond the scope of this review. Fortunately, excellent recent reviews have been published on the most important types of reactions, including the hydrogenation of ethylene (Eley, 119) and of and acetylenic compounds olefinic (Corson, 120), aromatic (Smith, In), (Bond, I%), as well as the exchange of deuterium with saturated hydrocarbons (Anderson, 123; Bond and Addy, 124) and other compounds (Anderson, 183; Taylor, 186). Many of these reactions have fairly complex mechanisms, and in most cases these have not as yet been fully resolved, The pattern of results is however sufficiently developed that a few important generalizations can be noted.

349

CATALYTIC ACTIVATION OF HYDROGEN

1. In most cases the slow step of the reaction is not simply the activation or chemisorption of hydrogen, but involves other chemisorbed species. Thus, the exchange of deuterium with methane and with other saturated hydrocarbons is much slower than with hydrogen and probably proceeds through dissociative adsorption of the hydrocarbon. 2. In general, it is the same group of metals, i.e., the transition metals, which are active catalysts for all these reactions, while the nontransition metals are inactive. 3. The order of catalytic activity of different metals for the various reactions, is similar, although by no means identical, indicating that, while the reactions may involve different species and different mechanisms, the observed catalytic effects have a common basis. Thus, the following orders of decreasing catalytic activity have been observed : For hydrogen-deuterium exchange (Schuit et al., 94a) : Pt, Rh, Ru

> Ni, Co > Fe, Cu

For the hydrogenation of ethylene (Beeck, 89): Rh

> Pd > Pt > Ni > Fe > W > Cr > Ta

For the hydrogenation of acetylene (Sheridan and Reid, f26): Pd

> Pt > Ni, Rh > Fe, Cu, Co, Ir > Ru, 0 s

For the hydrogenation of benzene (Schuit and Van Reijen, 94b):

> Rh > Ru > Pd > Co > Ni > Fe For deuterium-ammonia exchange (Kembsll, fd7) : Pt > Ph > Pd > Ni > W > Fe > Cu > Ag Pt

For the exchange of deuterium with most saturated hydrocarbons (Anderson, 123): W

> Mo > Ta > Rh > V > Cr > Zr > Pt > Pd > Ni >

Co

> Fe

Some of the above results were obtained on evaporated films, others on supported catalysts; in the case of ethylene hydrogenation, a similar order of activities was observed for both types of catalysts (Schuit and Van Reijen, 94b). In connection with the interpretation of these trends it should be noted that in some reactions (e.g., ethylene hydrogenation) the activation energy remains substantially constant and the frequency factor changes as the metal is varied, while in other reactions (e.g., deuterium-ammonia exchange) the reverse is the case. In the exchange of deuterium with saturated hydrocarbons, a compensation effect (Cremer, 128) has been noted. The significance of these different patterns is not clear.

350

J. HALPERN

In general, the trends noted above correspond to an increase in catalytic activity with increasing per cent d-character of metallic bonding or, roughly, with increasing work function. As noted earlier, this also implies ail inverse dependence of catalytic activity on the heat of adsorption (i.e., on the strength of the chemisorptive bond). This is to he expected where desorption is a rate-determining process. It should be noted that the correlations being discussed here are far froni perfect and exceptions can be found in nearly each of the reaction series. (For the ethylene-hydrogen and deuterium-ammonia reactions, the correlation between catalytic activity and per cent d-character is nearly quantitntive.) This is to be expected in view of the experimental difficulties involved in preparing clean and reproducible metal surfaces, particularly where different metals are being compared. In :my ntkempt, to correlate catalytic properties with work functions, it shnuld iilso I)e recognized that, t,hc work functioii is affected by adsorption, and therefore that thc work functions of metals under catalytic conditions, or even their relative order, may he somewhat different than those of the clean metals. c. Electrolytic hydrogen evolution. The overpotential for electrolytic hydrogen evolution at a metal electrode is influenced by the same factors which determine its catalytic activity in other reactions of hydrogen. Thus, Conway and Bockris (75) have shown that for a large number of metals, the exchange current density (which is inversely related to the overpotential) increases with the work function (Fig. 18) or with the per cent d-character. They have attributed this to the accompanying decrease in the heat of adsorption of hydrogen and have concluded that the ratedetermining step on these metals is HaO+

+ M-H + e -+

Hz

+ M + H20

(51)

For T1, Pb, and Hg, the exchange’s current density shows an inverse dependence on the work function, and the rate-determining step on these metals is believed to be H80+

+M +e

-+

M-H

+ H20

(52)

B. OXIDES 1 . Inlroduction. There is much evidence, both of a11 experimental and theoretical nature, to suggest that at least for some oxides, there is a significant link between chemisorption, catalytic activity, and semiconducting properties. One theoretical approach knowii as the boundary-layer theor!j (Aigrain and Dugas, 129;Hauffe and Engell, ISO; Weisz, 131) treats this problem from the standpoint of electron-transfer between the semiconductor and chemisorbed layer. The resulting space charge which builds up in the boundary layer between the interior and surface of the semi-

CATALYTIC ACTIVATION O F HYDROGEN

351

-

TE

-9

0

-5

-8

.-

0

0 -7

Q

3 -6

-5

-4

’

-3 3.5

4.0

4.5

5 .O

5.5

@ INELECTRON VOLTS

FIQ. 18. Dependence of the exchange current density of elcrtrolytic hydrogen evolution on the electronic work function. [Conwny, B. E., and Bockris, J . (I’M., J . C h e w Phys. 26,532 (1957).]

conductor alters the density and potential energy of the surface electrons; this is reflected in a change in the heat of adsorption and in the “reactivity” of the chemisorbed species. An alternative theoretical approach, due to Wolkenstein (IS%’),emphasizes covalent bond formation between the semiconductor and :idsorbate using conduction electrons or frec valences of the semiconductor. The bonds arc treated as essentially localized, although the concentration of free valeiices a t the surface may be determined by the bulk properties of the solid. Still another approach (Dowden, Mackenzie, and Trapnell, 133) emphasizes covalent bonding between the adsorbed

352

J. HALPERN

species and the metal ions of the oxide, using atomic orbitals or electrons of the latter, without reference to the semiconducting properties of the solid. These views are not necessarily mutually exclusive. 2. Zinc Oxide. The chemisorption of hydrogen and the hydrogen-deuterium exchange reaction have probably been more extensively studied on this than on any other oxide. Zinc oxide adsorbs hydrogen only after a suitable activating pretreatment such as heating in vacuo or under hydrogen (Molinari t ~ n dParravuno, 134), which may result in a nonstoichiometric composition containing excess zinc. The activated oxide is II n-type semiconductor. Pnrravano and Boudart (156) have depicted chemisorption of hydrogen on zinc oxide, as occurring through hcterolytic splitting on a pair of adjacent zinc-oxygen sites, followed by proton-transfer, e.g., 0'

- Zn++- 0-

+ Hz + 0'

- (ZnH)+

- (OH)-

-+ (OH)-

- Zn - (OH)-

(53)

This process transforms the zinc oxide to an impurity semiconductor, since the loosely held valence electrons of the zinc atom are easily excited into the conduction band. Reversible chemisorption and hydrogen-deuterium exchange on zinc oxide have been observed (Taylor el al., 136, 1%; Harrison and McDowell, 138) over a wide temperature range extending from -190 to 200". The adsorption follows a complex pattern and shows evidence of at least two distinct types of chemisorption with different activation energies, which operate in different temperature regions (on either side of about 100"). While the cause of this is still not fully understood, the suggestion (Taylor, 139) that it is due to surface heterogeneity appears plausible. The possibility of different adsorption mechanisms has also been considered. Thus, Garner (140) has associated low-temperature and high-temperature adsorption of hydrogen on metal oxides with M-H and O - H bonding, suggesting that the latter type is irreversible. He has also considered (141) the possibility that the two types of chemisorption on zinc oxide occur on F centers and interstitial zinc atoms, respectively. Another interesting suggestion, advanced by Morrison (142), is that low-temperature adsorption occurs on 0-sites (arising from adsorbed oxygen): 0-

+ %Hi -+

OH-

(54)

while high-temperature adsorption occurs a t 0- sites : 0'

+ JIHz

--t

OH-

+e

(55)

Both types of adsorption form OH-, but only the latter type releases an electron to the zinc oxide. This interpretation is consistent with Kubokawa and Toyama's recent observation (143) that only the high-temperature

353

CATALYTIC ACTIVATION OF HYDROQEN

chemisorption on zinc oxide causes a (reversible) increase in electrical conductivity. Parravano et al. ( 1 4 3 ~have ) suggcstcd that thc t,hcrmal gencration of active centers (possibly surface 0- sites created by electron transfer to the bulk) is the rate-determining process in the chemisorption of hydrogen on ZnO. Evidence that hydrogen is not involved in the rate-determining step of the adsorptioii process is provided by their observations that the rates of chemisorption of Hz and TIz :trc ident,icd over :L considerable temperature range. To obtain further insight into the relation between catalytic activity and scmiconductivity, Molinari arid Parravano (134) examined the effect of adding foreign ions to the zinc oxide on the catalyzed hydrogen-deuterium reaction. They found (Fig. 19) that over the temperature range 40-170", the catalytic activity of ZnO WAS increased by the addition of Ga203 and A1203 but reduced by the addition of LLO. This corresponds to a direct dependence on the free electron concentration, since the lat70

1

i

PR E TREATMENT

40

60

00

TEMPERATURE

100

I20

3 5 0 OC

140

160

180

T ("C.1. FIG.19. Ha-D2exchange on zinc oxide containing various foreign cations. q = percent exchange, for a catalyst area of 0.1 sq. m. and a flow rate of 0.238 cc./sec. N o linari, E . , and Parravano, G . , J . A m . Chem. Soc. 76, 5233 (1953).]

354

J. HALPERN

ter is increased by Ga203 and Ah03 but decreased by Li20. This has led Hauffe (144) to conclude that desorption, involving electron transfer to H+ and D+ ions, is the rate-determining step of the reaction: H&w

+ Dw.: + 2e

-+

HD(,w

(56)

The implications of this are that, just as for the transition metals considered earlier, the catalytic activity depends inversely on the strength of bonding of the chemisorbed hydrogen. Harrison and McDowell (138) observed that while neither zinc oxide nor a ,a-dipbenyl-B-picryl hydrazyl (a solid free radical) alone, catalyze hydrogen-deuterium exchange measurably at 77” K, a mixture of the two Bolids possesses considerable catalytic activity. They suggested that the effective catalyst in this mixture is the zinc oxide and that its catalytic activity is enhanced by electron transfer to the a ,a-diphenyl-8-picryl hydrazyl. It should be noted, however, that this implies a dependence of catalytic activity on electron concentration which is opposite to that observed by Molinari and Parravano (13.4). To bring the results for this system into line with those obtained by the latter workers, it would be necessary to postulate electron transfer in the opposite direction, i.e., from the a ,a-diphenyl-p-picryl hydrazyl to the zinc oxide.* 3. Chromic Oxide. This appears to be a p-type semiconductor in oxygen but becomes n-type when heated in hydrogen (Weller and Voltz, 1 4 4 ~ )A. comparison of its catalytic properties with those of zinc oxide (n-type) is therefore of some interest. Volts and Weller (146) measured the activity of chromic oxide for hydrogen-deuterium exchange a t -78” and - 195”, after pretreatment at 500” in an atmosphere of oxygen or hydrogen. They found the catalytic activity of the “reduced” state to be higher than that of the ‘(oxidized” state, although the latter had a higher concentration of defects (positive holes) responsible for electrical conductivity. The relation between catalytic activity and conductivity is thus opposite to that for zinc oxide, although in both cases the activity appears to increase with the electron concentration. The interpretation advanced earlier for zinc oxide has also been extended to chromic oxide (Baker and Jenkins, 146). Another interesting observation made in the course of this work was that hydrogen chemisorbed on chromic oxide at high temperatures did not exchange with deuterium a t -78”, although the hydrogen-deuterium exchange reaction itself proceeds a t this temperature. This suggests that, just as in the case of zinc oxide, there are two types of chemisorption on chromic oxide. Voltz and Weller (146) found that both the oxidized and reduced catalysts were susceptible to poisoning by small amounts of water vapor and,

* See Note Added in Proof

(21, p. 370.

CATALYTIC ACTIVATION OF HYDROGEN

355

from the magnitude of this effect, estimated that the “active” surface areas of the two forms of the catalyst, did not exceed 2 % and 15%, respectively, of the total areas. 4. Transition Metal Oxides. Dowden et al. (133) measured the rates of hydrogen-deuterium exchange on the oxides of most of the metals of the first transition series and obtained the pattern of results shown in Fig. 20. In this figure, the rates shown for the more active oxides (Crz03,Co304, and NiO) were determined a t -78”, while most of the other rates were measured at 0”.A comparison of the rates at the same temperature would thus follow a similar pattern but show an even greater spread. The reaction order with respect to the hydrogen pressure was found to be close to unity for ZnO and Gaz03 and about 0.2 for CrzOl and NiO; the low values for the last two oxides are most readily explicable in terms of a BonhoefferFarkas mechanism. The quantitative significance of such a superficial comparison of catalytic activities of different materials is probably very limited, and considerable caution should be exercised in interpreting the results. Different procedures and pretreatment5 were employed in the preparation of the various oxides tested, and it is not known to what extent these influence the resulting catalytic activities. As far as the apparent pattern is concerned, there is no obvious correlation between the catalytic activities of these oxides and their semiconducting properties. On the other hand, the

0.02L’ ‘ I

\

0

TI02 V 2 0 5 V 2 O 3 Cr2O3 MnO h 2 0 3 Co304 NIO

CuO

Cup0 ZnO G a 2 0 3 0802

FIG.20. First-order rate constants for H,-Dt exchange on various metal oxides. [Dowden, D. A., Mackenzie, N., and Trapnell, B. M. W., Proc. Roy. SOC.A2S7.245 (1956) .I

356

J. HALPERN

activities do appear to vary in a regular manner (Fig. 20) with the electronic configurations of the metal ions, and this would suggest that localized interaction or bonding with individual surface metal ions is important in chemisorption and catalysis on these oxides. However, even approached from this standpoint, the detailed significance of the pattern of activities shown in Fig. 20 is not clear. The highest activities are associated with the electron configurations 3ds(Crz03),3ds or 3dl(C0304), and 3d8(Ni0); low activity appears to be associated with an empty or nearly empty d-shell (TiOz, VtOa), a nearly filled d-shell (CuO), and a 3d6 configuration (MnO and FeZO3).The latter is not surprising in view of the well-known stability of half-filled d-shells. The pattern as a whole bears little resemblance to that either for metals or for ions in solution, where the highest activities are generally associated with nearly filled or just-filled d-shells. The inactivity of CuO (3d3 is particularly surprising, as is the moderately high activity of CuzO, ZnO, and Ga203 (3d'O). It seems likely that a complete interpretation of the catalytic activities of these oxides must take account not only of the electron configurations of the ions, but also of the role of defects, including those which give rise to semiconductivity ; the importance of the latter has already been demonstrated in the cases of zinc oxide and chromic oxide. 6. Alumina. Weller and Hindin (147)have observed that y-alumina, after dehydration at high temperatures (450-650"), is an active catalyst for hydrogen-deuterium exchange a t temperatures as low as -100". At this temperature, deuterium does not exchange with the catalyst hydrogen. Increasing the dehydration temperature (in the range 450" to 650") or the time of dehydration at a given temperature reduces the residual water content of the alumina (to limiting values of the order of 1 %) and increases the catalytic activity. Although alumina is known to be an n-type semiconductor, its catalytic activity does not appear to be related to an oxygen-deficient structure, since it is not affected by pretreatment with oxygen at temperatures from -100" to 650". Instead it has been suggested that the catalytic activity is due to the strained, high-energy surface sites formed by the high-temperature dehydration, which involves removal of structural water. Rehydration results in poisoning of the catalyst, the magnitude increasing with the amount of water added and with the rehydration temperature. A suggested explanation is that adsorption of water at low temperatures is nonspecific, while at high temperatures, owing to increased mobility, there is redistribution to the sites of highest activity. Maximum poisoning can be achieved by adsorption of an amount of water which is sufficient to cover only 2 % of the surface. Hydrogen itself is also a poison, and its effect follows a similar pat tern. These catalysts are also active for the hydrogenation of ethylene (Hindin

CATALYTIC ACTIVATION OF HYDROGEN

357

and Weller, 148), but only at much higher temperatures, 350450". In general, different pretreatments of the catalyst affect both reactions in the same way, suggesting that they both occur at the same active sites. The slow step in the hydrogenation of ethylene must, however, involve the ethylene molecule and not simply dissociation of hydrogen, since the latter occurs readily at much lower temperatures. The enhancement of the catalytic activity of y-Al2O3a t -78", by y-irradiation at this temperature, has been reported by Kohn and Taylor (148a). '(Annealing" of the extra activity was observed when the irradiated catalyst was warmed to room temperature suggesting that the effect is linked to electrons trapped during irradiation. C. OTHER SOLIDS 1. Halides. Wright, Weller, and Mills (149) have found that aluminum chloride catalyzes hydrogen-deuterium exchange measurably at 200". They have suggested that the hydrogen is split heterolytically and that exchange occurs indirectly through the intermediate formation of a deuterated hydrogen donor species in the catalyst, as indicated schematically in the following equations:

+ H+[AICIs(OH)]- = D H + D+[AlCla(OH)]H+:H- + D+[AICls(OH)]H D + H+[AlCls(OH)]-

D+:D-

(57)

(58)

The presence of a hydrogen donor impurity in the aluminum chloride was confirmed by the observation that HD was formed when deuterium alone was contacted with the catalyst. Bonhoeffer, Farkas, and Rummel ( 1 6 0 ~found ) that parahydrogen conversion proceeds measurably on sodium chloride between 20 and 340" with an apparent activation energy of about 8 kcal./mole. De Boer (74) has suggested that this reaction may involve endothermic adsorption of hydrogen, the observed activation energy being that of the adsorption process. 2. Sulfides. Rittenberg (18) has reported that a number of metal sulfides, including MoSz , WSz , and CoS (but not FeS, CuS, C U ~ Sor , NiS), are active catalysts for hydrogen-exchange reactions. With MoSz , the initial product of exchange of HZwith D2O was found to be H D rather than Dz , suggesting that the H2 molecule is split heterolytically, so that only one H atom (probably the proton) exchanges rapidly with deuterium in the water, i.e.,

+ HH+ + DzO : D+ + HDO Ha= H+ 4t

x

4t

H-

+

+ D+= HD *

4t

358

J. WLPERN

A similar experiment using platinum as catalyst yielded D2 as the initial product suggesting homolytic splitting on this surface with both adsorbed atoms exchanging rapidly with the water. 9. Charcoal. Turkevich and Laroche (160b)have measured the rate of Ha-D2exchange at 50", on a graded set of charcoals, prepared from glucose and heat treated at various temperatures from 380" to 950". Catalytic activity was small for the charcoals heated below 600" but increased rapidly above this temperature. It was also found that the area under the electron spin resonance "line," due to unpaired electrons in the charcoal, passed through a maximum, while the width of the "line" passed through a minimum, for a heating temperature of about 605". This implies (1) that the concentration of unpaired electrons is highest in the charcoals heated to 605" and (2) that the interaction between electrons (electron delocalization), reflected in the broadening of the electron spin resonance "line," increases as the temperature at which the charcoal is heated is raised from 605" to 950". These results are taken to indicate that the catalytic activity of the charcoals for H2-D2exchange is not associated with the presence of individual electrons but rather with a pool of interacting electrons. It is of interest that the catalytic activity for ortho-parahydrogen conversion at - 195.8", which involves the magnetic, rather than the chemical mechanism, was found to be greatest for the 605" charcoal, i.e., that having the highest concentration of unpaired electrons (Fig. 21).

IV. Biological Systems The ability, evidenced by a variety of microorganisms (see Table VI), to activate molecular hydrogen and enable it to undergo oxidation and exchange reactions, has been attributed to an enzyme, or system of en-

TEMPERATURE OF MEATIW OF

-

FIQ.21. Catalytic activities of charcoals heated at various temperatures. k

= rate constant for o-pHI conversion at -195.8"; c = rate constant for H1-Dnexchange at 60". [Turkevich, J . , and Laroche, J . , 2. phyeik. Chem. (Frankfurt) [N.S.] 16, 399 (19%) .I

359

CATALYTIC ACTIVATION OF HYDROGEN

TABLE VI Comparison of Exchange and Reducing Activities of Diflerenl H ydrogenase-Containing Organisms"

I

Organism

E . coli P . vulgaris D . deaulfuricana C . pasteuranium Acetobacter peroxydans R . rubrum Azotobacter vinelandii

Rateb

I

Methylene Beneyl viologen/ blue/ exchangec exchangec

HrDlO Methylene exchange blue red'n.

I

41,000 6,000 40,000 2,500 170 24

I

24,000 4,500 58,000 9,m 670 163

Ratio of rates

~

I

0.60 0.75 1.45 3.6 3.9 6.4

0.069 0.039 0.45 1.57 0.28

Red. beneyl viologen/ exchanged

1 - 1 -

0.011 0.064 0.077 0

-

0

260 7,500 28.5 a Kraana, A. I . , and Rittenberg, D., Proc. Natl. Acad. Sci. U.S . 42, 180 (1956). Rate of exchange or reduction in microliters of H2 per milligram of nitrogen per hour. e Ratio of the rate of hydrogen uptake by the indicated dye to the rate of the exchange reaction. d Ratio of the rate of hydrogen evolution from reduced bensyl viologen to the rate of the exchange reaction.

zymes, called hydrogenase (Stephenson and Sttrickland, 1; Green and Strickland, 151). This enzyme has never been isolated in the pure state, and our knowledge about it is derived indirectly from studies on bacterial suspensions or extracts. The hydrogenase in such preparations appears to be in particulate form and shows some variation in its properties from one strain of bacteria to another. Attention here will be confined to a consideration of the chemical nature of hydrogenase and the mechanism of its catalytic action. For a discussion of the biological role of hydrogenase, the reader is referred to a recent review by Gest (162). Since hydrogen-activating ability, in nearly every other known catalyst, is associated with the presence of metal atoms or ions, it seems reasonable to assume that this is also the case for hydrogenase. Indeed, there is much evidence to suggest that hydrogenase is an iron complex and that iron plays an important role in its action. Thus, it has been reported that bacteria grown in an iron-free medium are devoid of hydrogenase activity (Waring and Werkman, 163) and that the activity of partially resolved preparations from certain bacteria is stimulated by iron (Peck et al., 154). Susceptibility of the enzyme to light-reversible inhibition by CO has been interpreted as evidence for an iron-porphyrin complex (Hoberman and

360

J. HALPERN

Rittenberg, 165); the possibility of an iron-sulfhydryl group has also been considered (Rittenberg, 18). The enzyme is also susceptible to inhibition by oxygen and other oxidizing agents. The inhibition by oxygen can be reversed by hydrogen, but treatment of the oxidized suspension with cyanide removes this reversibility (Krasna and Rittenberg, 156). These observations suggest that it is the ferrous state of iron which is active. It has been suggested (Gest, 167) that molybdenum is also an active component of hydrogenase. Beyond this, little appears to be known about the chemical nature of hydrogenase or the site of its activity. Among the reactions which hydrogenase has been found to catalyze (Krasna and Rittenberg, 166, 168) are parahydrogen conversioii; deuterium-water exchange; reduction by hydrogen of various oxidizing agents such as methylene blue and benzyl viologen; and evolution of hydrogen from reduced viologen. Although it is believed that the mechanisms of most of these reactions have a common basis, their rates vary considerably for any given hydrogenase preparation and the pattern of this variation differs for different preparations (Table VI). Accordingly, there is some uncertainty, and some disagreement, about which of these reactions provides the best primary assay of hydrogenase activity, i.e., a measure of the rate of activation of hydrogen. Just as with other catalysts, it seems reasonable to favor the conversion or exchange reactions for this purpose, since they involve no substrates other than hydrogen itself. On the other hand, it has been pointed out that for some hydrogenase preparations, the rates of these reactions are lower than that of the reduction of methylene blue (Gest ct al., 16.4, 157).

The conversion of parahydrogen and the deuterium-water exchange reaction have been studied by a number of workers (Farkas et al., 169, 160; Rittenberg et al., 166, 166, 161; Couper, Eley, and Hayward, 162). The two reactions proceed with comparable rates and are similarly affected by inhibitors. The rate of conversion or exchange was found to be proportional to the bacterial or extract concentration, in the region of low concentrations, with an apparent activation energy (for P. vulgaris arid E. coli) of 7-10 kcal./mole; in this region, chemical activation of hydrogen by hydrogenase appears to be rate-determining. At higher bacterial concentrations, the rate tended to level off and the apparent activation energy decreased; this was attributed to rate limitation by a transport process, the nature of which is not clear. Water appears to play an important role in the conversion, as well as the exchange reaction. Thus, no parahydrogen conversion was observed for a bacterial suspension (P. vulgaris) in DzO (Krasna and Rittenberg, 166; Farkas, 16S), suggesting that the mechanism involves exchange with the hydrogen in water. Dried bacterial films were found to be inactive for the

CATALYTIC ACTIVATION OF HYDROGEN

36 1

conversion reaction, but they regained their activity on rehydration; on the other hand, rehydration with DzO resulted in only very low activity for Hz-D~O exchange (Couper et al., 162). The inactivation caused by heavy water appeared to be to some degree irreversible, since redrying and rehydration with ordinary water only partially restored the activity of the bacteria for parahydrogen conversion. The cause of this is not clear. When methylene blue or ferricyanide was added to an aqueous suspension of P. vulgaris, there was an induction period before D2-H20 exchange commenced (Rittenberg and Krasna, 158, 16'1). During this stage, the added substrate underwent reduction, subsequent to which the exchange reaction proceeded at the normal rate. This indicates that both the reduction and exchange reactions occur through a common reducing intermediate formed by the activation of hydrogen. Some insight into the nature of this intermediate is provided by the observation (Rittenberg and Krasna, 161) that the maior (but not exclusive, cf. Beetlestone and Couper, 164) initial product of the exchange reaction between H z and a suspension of P. vulgaris in D 2 0 is HD rather than DP(Fig. 22). This implies, as first pointed out by Krasna and Rittenberg (156, 161), that Hz is split heterolytically by hydrogenase with the formation of an enzyme-hydride complex and the release of a proton which exchanges rapidly with hydrogen (or deuterium) of the water. The mechanism thus appears to be analogous to that which has already been considered (Equations 9 and 10) for the activation of hydrogen by cupric salts, i.e.,

FIG.22. Exchange of H2 with a D20 suspension of proletis vulgaris. [Rittenberg, D., and Krasna, A. I . , Discussions Faraday Soc. 20, 185 (1955).]

TIME t mln.)

J. HALPERN

362

E+Ha

k

EHE

EH-+S+H+

+ H+

+ SHt

(62)

(63)

where E stands for the enzyme, S for the substrate, and SHz for the product. As pointed out earlier, the principal requirement for an active catalyst for the heterolytic splitting of hydrogen is the presence of two suitably disposed functional groups-a metal atom to combine with the hydride ion and a base (: B) to act as a proton acceptor. In line with the evidence for the presence of a ferrous complex in hydrogenase, Rittenberg (18) has suggested the following model for the active site of the enzyme, m

+

I

:B H 1 = F e H + H:B+ (64) The observation that the activity of hydrogenase passes through a maximum with varying pH also finds explanation in terms of such a model (Rittenberg, 18). Inactivation a t low pH may be due to neutralization of the basic site and at high pH to hydrolysis of the ferrous ion, i.e., Feff

--

OHH+ F e O F :B I Fe* :B +Fe++ inactive active

H:B+ inactive

(65)

Thus, it is only in an intermediate pH range that both functional groups can coexist in active form. This type of pH dependence has also been observed for other enzyme reactions and explained in similar terms (Laidler, $4). In this connection, the pH dependence of the catalytic activity of cupric glycinate complexes, discussed earlier (Fig. 2) should also be recalled. An alternative mechanism, proposed by Gest (l67),involves the homolytic splitting of hydrogen on two adjacent Fe sites and transfer of the resulting H atoms to a flavein moiety. While the reduction of 2-electron acceptors, such as methylene blue, becomes possible at this stage, it is suggested that 1-electron transfer processes such as reduction of viologen or deuterium-water exchange require mediation by a second metal, possibly molybdenum. This has been used to explain the observation that certain microbial preparations (Azotobacter vinelandai and Rhodospirillum rubrum) which catalyze the reduction of methylene blue, are inactive for the deuterium-water exchange or hydrogen evolution from reduced viologen. On most other grounds, however, a mechanism involving the heterolytic splitting of hydrogen is to be preferred. Where the rate of reduction of a substrate is lower than the rate of parahydrogen conversion or deuterium-water exchange in the absence of substrate (Table VI), contributing factors may be (1) inhibition of hy-

CATALYTIC ACTIVATION OF HYDROGEN

363

drogenase activity by the substrate, particularly where the latter is an oxidizing agent, or (2) a slow reaction between the enzyme hydride intermediate and the substrate [kg < kz in Equations (62) and (63)]; in the latter case (just as for the Cu2+-or Ag+-catalyzed reduction of dichromate in acid solution), reduction by hydrogen in heavy water should be accompanied by some exchange. It is probable that some of the reactions of hydrogen (e.g., the hydrogenation of fumarate) which are observed in hydrogenase-containing bacterial suspensions involve, in addition to hydrogenase, other enzymes whose function is connected with activation of the substrate. It should be emphasized that there is no direct evidence that the active site in hydrogenase contains iron, and, in this connection, it is of interest that attempts to activate hydrogen in solution by iron salts or complexes have thus far been unsuccessful. Since the concentration of active enzyme in the bacterial preparations which have been used to study hydrogenase activity is unknown and may be very small, the possibility that some other metal ion (e.g., Cu++), which is present only in trace concentrations, is involved, cannot be ruled out. The possibility that the activity of hydrogenase is associated not with a metal, but with low-lying delocalized orbitals of the protein itself, arising from a periodic array of interacting peptide bonds, has also been considered (Rittenberg and Krasna, 161). It is unlikely that conclusive information on these points will be obtained until the enzyme is isolated and studied in pure form.

V. Concluding Remarks Because of the historical circumstance that heterogeneous hydrogenation catalysts were well known before any homogeneous ones were recognized, the view has been emphasized in most of the earlier theories that catalytic activity depends critically on some property characteristic of the solid state (e.g., crystal lattice geometry or electronic band structure) which cannot be realized in simpler species such as exist in the gas phase or in solution. The numerous examples cited of homogeneous activation of hydrogen by metal ions and complexes in solution have clearly forced a modification of this view and suggest that a comprehensive explanation of catalytic activity must be sought along different lines. Perhaps the most important general requirement for catalytic activity which has emerged from the great variety of systems examined, is the availability of low-lying unfilled orbitals, preferably with a high degree of d-character, which can accept electrons from hydrogen and form weak bonds with it. This property can be realized in a variety of metal ions and complexes, as well as in solids, and in each case is most pronounced just prior to the filling of a d-shell or d-band.

364

S. HALPERN

Many specific points of similarity have been noted between the various types of catalysts considered, Perhaps the most striking of these is the close parallel between the properties of metallic catalysts and those of the “isoelectronic” ions. Among the possible implications of this are (1) that the activation of hydrogen on metal surfaces involves a high degree of localized interaction with individual metal atoms and (2) that the catalytic activity of metals is predominantly governed by electronic factors, since geometric factors could not possibly carry over to the metal ions. Recent magnetic measurements by Deitz and Selwood (166) also support the suggestion of metal-hydrogen bond localization on metal surfaces. The higher activity which metallic catalysts display, relative to metal ions in solution, is associated in part with stronger metal-hydrogen bonding. Thus, chemisorption of hydrogen on catalytic metals is usually exothermic, while splitting of hydrogen by metal ions in solution appears, in every case, to be endothermic, Connected with this is the observation that while the rate-determining process in heterogeneously catalyzed reactions is frequently desorption, in homogeneous systems it always appears to be the hydrogen splitting step. At the present time it does not appear possible to account in detail for the catalytic properties of most nonmetallic solids. In the case of oxides, there are indications that semiconducting properties, as well as localized interactions of hydrogen with individual metal ions, are involved. The catalytic studies on alumina (Weller and Hindin, 147, 148) also emphasize another important and universal feature of heterogeneous catalysis, which has no counterpart in homogeneous systems. This refers to the existence on a solid surface of sites which exert catalytic activity simply by virtue of being in a strained, high-energy condition which has little connection with the recognizable electronic or geometrical properties of the solid or its surface as a whole. Such sites can be created on a solid surface by a variety of processes, and become “frozen in” because of the low mobility of the solid state. The concept of “active centers” in Catalysis is, of course, a very old one, but the emphasis placed on it has tended to decline as it has become generally recognized that many solids, because of their electronic properties, possess “inherent” catalytic activity which is manifested even on an ideal surface, free from heterogeneities or points of strain. The two concepts are by no means mutually exclusive, and even with catalysts of the latter type, the possibility of enhancement or modification of activity due to surface heterogeneities must always be considered. The great difficulties which this imposes on the quantitative interpretation of catalytic phenomena are readily appreciated. Valuable insight into certain catalytic phenomena has been gained from the study of the effects of complexing on the catalytic properties of metal

CATALYTIC ACTIVATION OF HYDROGEN

365

ions. Some of these effects are closely related to the ordinary poisoning of heterogeneous catalysts and to the inhibition of enzymes. Studies on metal complexes have also emphasized the important properties of “bifunctional” catalysts for the activation of hydrogen. In the absence of an adequate knowledge of the chemical nature of hydrogenase, it is not possible to interpret its properties in detail. I n a number of important respects, however, its catalytic behavior resembles that of some of the homogeneous catalysts that have been studied. Among these, the cupric complexes probably come closest to serving as models for this enzyme. Finally it should be noted, that many of the principles and concepts which have emerged from the study of the catalytic activation of hydrogen also have application to other “inert” molecules, including nitrogen and hydrocarbons. These substances are unreactive for much the same reasons as hydrogen, and it is reasonable to expect that the catalytic properties required to bring about their activation, while more stringent than for hydrogenation catalysts, will depend on similar factors. Up to the present, the only promising catalysts for reactions of nitrogen and of saturated hydrocarbons have been solids, but there is every reason to expect that further investigation will also reveal, as has been the case for hydrogen, homogeneous catalysts capable of activating these molecules. The recognition that there are microorganisms and enzymes which can accomplish the fixation of nitrogen a t low temperatures offers encouragement and direction toward this end (166). ACKNOWLEDGEMENTS

Grateful acknowledgement is made to the National Research Council of Canada, the Research Corporation, and to the donors of The Petroleum Research Fund, administered by the American Chemical Society, for support of this work.

REFERENCES I . Stephenson, M., anti Strickland, 1,. H., Biochem. J . 26, 205 (1031). 2. Polanyi, M., Sci. J . Roy. Coll. Sci. 7 , 21 (1937). 3. Eley, D.D., Advances i n Calalysis 1. I57 (1948). 4 . Calvin, M., Trans. Faraday SOC.34, 1181 (1938). 6. Calvin, M., J . A m . Chem. Soe. 61, 2230 (1939). 6. Weller, S.W . , and Mills, G. A., Advances i n Catalysis 8 , 163 (1956). 7. Halpern, J., Quart. Revs. (London) 10, 463 (1956).J . Phys. Chem. 63. 398 (1959). 8. Peters, E.and Halpern, J., J . Phys. Chem. 69, 793 (1955). 9. Halpern, J., Macgregor, E. R . , and Peters, E . , J . Phys. Chem. 80, 1455 (1956). 10. Webster, A. H . , and Halpern, J., J . Phys. Chem. 60,280 (1956). If. Webster, A. H., and Halpern, J., J . Phys. Chem. 61, 1239 (1957). 18. Webster, A. H., and Halpern, J . , J . Phys. Chem. 61,1245 (1957).

366

J. HALPERN

13. Korinek, G.

J., and Halpern, J., J. Phys. Chem. 60,285 (1956).

14. Webster, A. H., and Halpern, J., Trans. Faraday SOC.63, 51 (1957). 16. Halpern, J., Advances i n CataZysis 0, 302 (1957). 16. Macgregor, E. R., and Halpern, J., Trans. M e t . SOC.A.Z.M.E. 212, 244 (1958). 16a. McDuffie, H. F., Compere, E. L., Stone, H . H., Woo, L. F., and Secoy, C. H., J . Phys. Chem. 61, 1030 (1958). 17. Halpern, J., unpublished. 18. Rittenberg, D., Paper prevented a t International Conference on Enzyme Chemistry, Tokyo, 1957. 19. Halpern, J., Harrod, J. F., and Potter, P. E., Can. J . Chem., in press. f Q a .Halpern, J. and Harrod, J. F., unpublished. 20. Peters, E., and Halpern, J., Can. J. Chem. 33, 356 (1955). 21. Peters, E., and Halpern, J., Can. J. Chem. 34, 554 (1956). 21. Maxted, E. B., Advances i n Catalysis 8, 129 (1951). 23. Laidler, K . J., Discussion Faraday SOC.20, 83 (1955). 24. Laidler, K. J., “Introduction t o the Chemistry of the Enzymes,” pp. 17, 154. McGraw Hill, New York, 1954. 26. Korinek, G. J., and Halpern, J., Can. J . Chem. 34, 1372 (1956). 16. Lindgren, B., Johnson, A., and S i l l h , L. G., A d a . Chem. Scand. 1, 461 (1047). 27. Basolo, F., and Pearson, R. G., “Mechanisms of Inorganic Reactions,” p. 386. Wiley, New York, 1958. 28. Chalk, A. J., and Halpern, J., unpublished. 29. Weller, S., and Mills, G. A , , J. A m . Chem. SOC.76,769 (1953). 30. Wright, 1,. W., and Weller, S., J . A m . Chem. SOC.76, 3345 (1954). 31. Wilmarth, W. K., and Barsh, M. K., J . Am. Chem. SOC.76, 2237 (1953). 32, Calvin, M., and Wilmarth, W. K., J . A m . Chem. SOC.78, 1301 (1955). 33. Wilmarth, W. K., and Barsh, M. K., J. A m . Chem. SOC.78, 1305 (1955). 34. Wright, 1,. W., Weller, S., and Mills, G. A., J . Phys. Chem. 60, 1060 (1955). 36. Wilmarth, W. K., private communication. 36. Wilmarth, W. K., and Kapauan, A. F., J. A m . Chem. SOC.78, 1308 (1956). 37. Wender, I., and Sternberg, H . W., Advances i n Catalysis 0, 594 (1957). 38. Orchin, M., Advances in Catalysis 6,385 (1953). 39. Wender, I., Sternberg, H. W., and Orchin, M., i n “Catalysis” (P. H. Emmett, ed.), Vol. 5, p. 73. Reinhold, New York, 1957. 40. Roelen, O . , German Patent 103,362 (1938); U. S. Patent 2,327,066 (1943). 41. Wender, I., Mctlin, S., Ergun, S., Sternberg, H. W., and Grcenfield, H., J . Am. Chem. SOC.78, 5401 (1956). 4.2. Wender, I., Sternberg, H. W., and Orchin, M., J. Am. Chem. SOC.76,3042 (1953). 43. Orchin, M., Kirch, I,., and Goldfarb, I., J . A m . Chem. SOC.78, 5450 (1956). 44. Adkins, H., and Krsek, G., J. Am. Chem. SOC.70, 383 (1948). 46. Kirch, L., and Orchin, M., J . Am. Chem. SOC.80.4428 (1958). 46. Nntta, G., Ercoli, R., Castellano, S., and Barbieri, F. H . , J . A m . Chem. SOC.76, 4049 (1954).

47. Martin, A. R., Chem. & Znd. p. 1536 (1954). 48. Sternberg, H. W., Greenfield, H., Friedel, R. A., Wotiz, J., Markby, R., and Wender, I., J. A m . Chem. SOC.78, 5401 (1956). 49. Sternberg, H. W., Markby, R., and Wender, I., J . A m . Chem. SOC.78, 5704 (1956); 70, 6116 (1957).

49a. Aldridge, C. Id., Fasce, E. V., and Jonassen, H . B., J . Phys. Chem. 62, 869 (1958).

CATALYTIC ACTIVATION OF HYDROGEN

60. Mills, G. A., Weller, S. W, and Wheeler, A., J. Phys. Chcni. 63, 403 (1959). M.,J . Chem. SOC.Japan 63, 634, 1752 (1942). 62. Winfield, M. E., Revs. Pure Appl. Chem. (Australia) 6, 217 (1955).

367

61. Iguchi,

6s. Bayston, J., King, N. K., and Winfield, M. E., Advances in Catalysis Q, 312 (1957). 64. King, N. K., and Winfield, M. E., J. A m . Chem. SOC.80,2060 (1958). 66. Flynn, J. H., and Hulburt, H. M., J . A m . Chem. SOC.76, 3393, 3396 (1954). 66. Wirtz, K., and Bonhoeffer, K. F., 2.physik. Chem. (Leipzig) A177.1 (1936). 67. Wilmarth, W. K., Dayton, J. C., and Fluornoy, J. M., J . A m . Chem. SOC.76, 4549 (1953). 68. Miller, S. I,., and Rittenberg, D., J. A m . Chem. Soc. 80,64 (1958). 69. Eley, D. D., and Evans, M. G., Trans. Faraday Soc. 34, 1093 (1938). 60. Conway, B. E., Bockris, J. O’M., and Linton, H., J . Chem. Phys. 24, 834 (1956). 61. Wilmarth, W.K., and Dayton, J. C., J. A m . Chem. SOC.76, 4553 (1953). 62a. Eyring, H.,and Smith, R. P., J. Phys. Chem. 66, 972 (1952). 68b. Orgel, L.E., J. Chem. Soc. p. 486 (1958). 6Sa. Carrington A., and Symons, M. C. R., J. Chem. SOC.,p. 3373 (1956). 6Sb. Carrington, A., Schonland, D., and Symons, M. C. R., J. Chem. SOC.,p. 659 (1957). 64. Swain, G . G., and Brown, J. F., Jr., J. A m . Chem. Soc. 74,2538 (1952). 66. Wilson, I. B., in “The Mechanism of Enzyme Action” (W. D. McElroy and B. Glass, eds.), p. 642. Johns Hopkins Press, Baltimore, 1954. 66. Wilson I. B., Discussions Faraday SOC.20, 119 (1955). 67. Trapnell, B. M. W., Quart. Revs. (London) 8, 404 (1954). 68. Trapnell, B.M. W., “Chemisorption.” Butterworths, London, 1955. 69. de Boer, J. H., Advances i n Catalysis 8, 17 (1956). 70. Becker, J. A., Advances i n Calalysis 7, 135 (1955). 71. Lennard-Jones, J. E., Trans. Faraday Soc. 28, 333 (1932). 71a. Schwab, G.-M., and Killman, E., Bull. SOC. chim. Belg. 67, 305 (1958). 71b. Eyring, H.,and Polanyi, M., 2.physik. Chem. Bl2, 279 (1931). 72. Trapnell, B. M. W., Proc. Royal SOC.A218, 566 (1953). 73. Mikovsky, R . J., Boudart, M., and Taylor, H. S., J. Am. Chem. Sor. 76, 3814 (1954). 74. de Boer, J. H., Aduances in Catalysis 9, 472 (1957) 76. Conway, B. E., and Bockris, J. O’M., J . Chem. Phys. 26, 532 (1957). 76. Beeck, O.,Discussions Faraday SOC.8, 126 (1950). 77. Pauling, L., Proc. Roy. SOC.A196, 343 (1949). 78. Dowden, D . A., in “Chemisorption” (W. E . Garner, ed.), p. 3. Academic Press, New York, 1957. 78a. Dowden, D. A., Bull. SOC. chim. Belg. 67, 439 (1958). 79. Trapnell, B. M. W., in “Catalysis” (P. H . Emmett, ed.), Vol. 3,p. 3.Reinhold, New York, 1955. 80. Roberts, J. K., “Some Problems in Adsorption.” Cambridge University Press, London and New York, 1939. 81. Frankenburg, W. G., J. A m . Chem. SOC.66, 1827,1838 (1944). 82. Bosworth, R. C. L., Proc. Cambridge Phil. SOC.33, 394 (1937). 8s. Mignolet, J. C. P., Discussions Faraday SOC.8, 105 (1950). 84. Mignolet, J. C. P., Rec. trau. chim. 74, 685 (1955). 86. Boudart, M., J. A m . Chem. SOC.74, 3556 (1952). 86. Beeck, O.,Smith, A. E., and Wheeler, A., Proc. Roy. SOC.A177, 62 (1940).

368

J. HALPERN

8Y. Beeck, O., Revs. Modern Phys. 17, 61 (1945). 88. Beeck, O.,Advances in Catalysis 2, 151 (1950). 89. Beeck, O.,Discussions Faraday Soc. 8, 118 (1950). 90. Rideal, E.I<., and Trapnell, B. M. W., Discussions Faraday SOC.8, 114 (1950). 91. Rideal, E. I<.,and Trapnell, B . M. W., J. chim. phys. 47, 126 (1950). 98. Trapnell, B. M . W., Proc. Roy. SOC.A206, 39 (1951). 93. Schuit, G. C . A., and de Boer, N. H., Rec. trav. chim. 72, 909 (1953). 94a. Schuit, G. C. A., de Boer, N. H., Dorgelo, G . J. H., and van Reijen, L. I,., i n “Chernisorption” (W. E. Garner, ed.), p. 39. Butterworths, London, 1957. 94b. Schuit, Q. C . A., and van Reijen, 1,. L., Advances in Catalysis 10, 243 (1958). 96. Suhrrnann, R., Advances in Catalysis 7, 303 (1955). 96. Suhrrnann, R., in “Chernisorption” (W. E. Garner, ed.), p. 106. Academic Press, New York, 1957. 96a. Sachtler, W . M . H., J . Chem. Phys. 26, 751 (1956). 96b. Sachtler, W.M. H., and Dorgelo, G. J. H., J. chim. phys. 64, 27 (1957). 96c. Sachtler, W.M . H., and Dorgelo, G. J. H., Bull. soc. chim. Belg. 67, 465 (1958). 96d. Mignolet, J. C. P., Bull. soc. chim. Belg. 67, 358 (1958). 96e. Suhrmann, R., Wedler, G., and Gentsch, H., 2. phyzik. Chem. (Frankfurt) [N.S.]17, 350 (1958). 9Y. Selwood, P. W. , J. Am. Chem. SOC.78,3893 (1956). 98. Lee, E. L., Sabatka, J. A., and Selwood, P. W., J. Am. Chem. SOC.79,5391 (1957). 99. Selwood, P. W . , Advances in Catalysis 9,93 (1957). 100. Selwood, P.W., J. Am. Chem. SOC. 79, 3346 (1957). 101. Selwood, P. W., J. A m . Chem. SOC.70,4637 (1957). 108. Vaska, I,., and Selwood, P. W., J. A m . Chem. SOC.80, 1331 (1958). 10Sa. Broeder, J. J . , van Reijen, I,. I,., and Korswagen, A. R., J . chin&.phys. 64.37 (1057). 10Sb. Broeder, J. J., van Reijen, 1,. I.., Sachtler, W. M. H., and Schuit, G . C. A., 2. Blektrochern. 80, 838 (1956). 103c. Tukaishi, T., 2. physik. Chem. (Frankfurt) [ N . 5.1 14, 164 (1958). 104. Eischens, R. P., Pliskin, W. A., and Francis, 5. A., J . P h y s . Chem. 80,194 (1956). 106. Eischens, R. P., 2. Electrochem. 60, 782 (1956). 106. Eischens, R. P., and Pliskin, W. A., Advances in Catalysis 9, 662 (1957). 107. Eischens, R. P.,and Pliskin, A., Advances in Catalysis 10, 1 (1958). 108. Eischens, R . P., J . Chem. Erluc. 36, 385 (1958). 108a. Pickering, H.L., and Eckstrom, H. C., J. Phys. Chem. 63, 512 (1959). 109. Dowden, D. A , , J. Chem. SOC.,p. 242 (1950). 110. Eley, D. D., Discussions Faraday SOC.8,34 (1950). 111. Emmett, P. H., and Teller, E., 18th Rept. Comm. on Contact Catal?jsis, New York, p. 68 (1940). 222. Couper, A., and Eley, D. D., Discussions Faraday SOC.8, 172 (1950). 113. de Boer, J. H., in “Chernisorption” (W. E. Garner, ed.), p. 27. Academic Press, New York, 1957. 114. Bonhoeffer, K . F., and Farkas, A., 2. physik. Chem. (Leipzig) Bl2, 231 (1931). 116. Rideal, E. K., Proc. Cambridge P h i l . SOC.36, 130 (1939). 116a. Couper, A , , Eley, D. D., Hulatt, M. J., and Rossington, D. R., Bull. soc. chim. Belg. 67, 343 (1958). 116. Eley, D . D., Proc. R o y . Soc. A178, 452 (1941). 117. Schwab, G.-M., Discussions Faraday SOC.8, 166 (1950). 118. Dowcicn, D.A , , and Reynolds, P. W., Discussions Faraday SOC.8, 184 (1950).

CATALYTIC ACTIVATION OF HYDROGEN

369

119. Eley, D. D., i n “Catalysis” (P. H . Emmett, ed.), Vol. 3, p. 49. Reinhold, New York, 1955. 180. Corson, B. B., in “Catalysis” (P. H . Emmett, ed.), Vol. 3, p. 79. Reinhold, New York, 1955. 181. Smith, H. A , , in “Catalysis” (P. H. Emmett, ed.), Vol. 5, p. 175. Reinhold, New York, 1957. 12.8, Bond, G. C., i n “Catalysis” ( P . H. Emmett, ed.), Vol. 3, p. 109. Reinhold, New York, 1955. 123. Anderson, J. R., Revs. Pure Appl. Chem. (Australia)7 , 166 (1957). 184. Bond, G. C., and Addy, J. in “Chemisorption” (W. E . Garner, ed.), p. 125. Academic Press, New York, 1957. 126, Taylor, T. I., i n “Catalysis” ( P . H . Emmett, ed.), Vol. 5, p. 257. Reinhold, New York, 1957. 126. Sheridan, J., and Reid, W. D., J . Chem. SOC.,p. 2962 (1952). 187. Kemball, C., Proc. Roy. SOC.A214, 413 (1952). 188. Cremer, E., Advances in Catalysis 7, 75 (1955). 189. Aigrain, P., and Dugas, C., 2. Elektrochem. 66,363 (1952). 130. Hauffe, K . , and Engell, H. J . , 2. Elektrochem. 66, 366 (1952). 131. Weisz, P. B., J . Chem. Phys. 20, 1483 (1952); 21, 1531 (1953). 132. Wolkenstein, T., Advances i n Catalysis 9, 807,818 (1957). 133. Dowden, D. A., Mackenzie, N., and Trapnell, B. M. W., Proc. Roy. SOC.A W , 245 (1956). 134. Molinari, E., and Parravano, G., J . A m . Chem. SOC.76, 5233 (1953). 136. Parravano, G., and Boudart, M., Advances in Catalysis 7, 47 (1955). 136. Taylor, H . S., and Strother, C. O., J . A m . Chem. SOC.66,586 (1934). 137. Taylor, H . S., and Liang, S. C., J . Am. Chem. SOC.69, 1306 (1947). 138. Harrison, L. G., and McDowell, C. A., Proc. Roy. SOC.A228,66 (1955). 139. Taylor, H. S., Advances in Catalysis 1, 1 (1948). 140. Garner, W. E., Disciissions Faraday SOC.8, 211 (1950). 141. Garner, W. E., Advances in Catalysis 9, 169 (1955). 148. Morrison, S. R., Adoances i n Catalysis 7, 259 (1955). 143. Kubokawa, Y., and Toyama, O., J . Phys. Chem. 60,833 (1956). 14%~.Parravano, G., Friedrick, H. G., and Boudart, M., private communication. 1.44. Hauffe, K . , Advances in Catalysis 7, 213 (1955). 144a. Weller, S . W., and Voltz, S. E., Advances i n Catalysis 9,215 (1957). 146. Voltz, S . E., and Weller, S., J . Am. Chem. SOC.76,5227 (1953). 146. Baker, M. McD., and Jenkins, G. I., Advances i n Catalysis 7, 1 (1955). 147. Weller, S . W., and Hindin, S. G., J . Phys. Chem. 60, 1506 (1956). 148. Hindin, S . G., and Weller, S. W., J . Phys. Chem. 60, 1501 (1956). 1&a. Kohn, H . W. and Taylor, E. H., J . Phys. Chem. 63, 500 (1959). 149. Wright, L. W., Weller, S. W., and Mills, G. A,, Znd. Eng. Chem. 49, 1054 (1957). 160a. Bonhoeffer, K . F., Farkas, A., and Rummel, K. W., Z. physik. Chem. (Leipzig) B21, 225 (1933). 16Ob. Turkevich, J . , and Laroche, J . , 2. physik. Chem. (Frankfurt) [N.S.] 16, 399 (1958). 161. Green, D. E., and Strickland, L. H., Biochem. J . 26,898 (1934). 168. Gest, H., Bacteriol. Revs. 18, 43 (1954). 163. Waring, W. S., and Werkman, C. H., Arch. Biochem. 4.75 (1944). 164. Peck, H . D., Jr., San Pietro, A., and Gest, H., Proc. Natl. Acad. Sci. U . S . 42, 13 (1956).

370

J. HALPERN

166. Hoberman, H. D., and Rittenberg, D., J . Biol. Chem. 147,211 (1943). 166. Krasna, A. I., and Rittenberg, D., J . A m . Chem. SOC.76,3015 (1954).

167. Gest, H., Paper presented at International Conference on Enzyme Chemistry, Tokyo, 1957. 168. Krasna, A. I., and Rittenberg, D., Proc. Nail. Acad. Sci. U.S . 42, 180 (1956). 16s. Farkas, A., Farkas, L., and Yudkin, J., Proc. Roy. SOC.B116.373 (1934). 160. Farkas, I,., and Fischer, E., J . Biol. Chem. 147,211 (1943). 161. Rittenberg, D., and Krasna, A. I., DiSCUsSiOfl8 Faraday Soc. 20, 185 (1955). 162. Couper, A., Eley, D. D., and Hayward, A., DiSCUsSiOflS Faraday Soc. 20, 174

(1955). Farkas, A., Trans. Faraday SOC.32, 922 (1936). Beetlestone, J. G., and Couper, A., Discussiofle Faraday SOC.20, 281 (1955). Deitz, R., and Selwood, P. W., private communication. McElroy, W. D., and Glass, B., eds., “Inorganic Nitrogen Metabolism,” pp. 293374. Johns HopkinR Press, Baltimore, 1956. 167. Pliskin, W. A., and Eischens, R. P, Presented at American Chemical Society Meeting, Boston, April 1959. 168. Eley, D. D., and Inokuchi, H., Z. Eleklrochem. 69, 29 (1959).

163. 164. 166. 166.

NOTESADDEDIN PROOF (1) Recently I’liskin and Eischens (167) have observed bands a t 4.74 and 4.86 1 for hydrogen chemisorbed on alumina-supported platinum. These are associated with “weakly” and “strongly” bonded H atoms, respectively, and it is suggested that the former are adsorbed on single surface Pt atoms while the latter are adsorbed by interaction with two 1% atoms. No evidence was found for adsorbed molecular species such as H2+ and no absorption bands could be detected for hydrogen chemisorbed on silica-supported nickel. (2) Eley and Inokuchi (168) have recently reported that the deposition of a,a-diphenyl-B-picryl hydrazyl on a film of A1 or Pd raises the D.C. resistance of the latter suggesting electron transfer from the a pdiphenyl8-picryl hydrazyl to the metal. The suggestion that this may be the case also with ZnO thus appears plausible.

Author Index Numbers in parentheses are reference numbers and are included to assist in locating references when the authors' names are not mentioned in the text. Numbers in italics refer to the page on which the reference is listed.

A Abbot, H. W., 202(135), 880 Addy, J., 250(44), 251(44, 45), 252(44), 261(44), 262,348(124),369 Adkins, H., 319(44), 366 Aigrain, P.,350(129), 369 Akamatu, H.,203(139), 8.91 Aldridge, C. L., 320(49s), 366 Aleksandrova, G.I., 282(46), 298 Allen, J. A., 81(25), 188 Allison, H.W., 91(54), 101, 112(54), 114 (54), 122, 189 Alpert, D., 80(22), 188 Anderson, J. R., 226(17),228(19),229(19), 231(17), 238(19), 242(17), 243(19), 244(19), 245(19), 246(41), 247(41), 250(19, 41), 255(17), 257(46), 258(41), 260(17, 19), 261(17), 268, 348(123), 349(123), 369 Anderson, M. S., 277(26), 897 Anderson, R. B., 140(12), 817 Andrew, K.F., 155(78), 160(78), 163(78), 204(141), 219, 821 Aris, R., l68(107), 880 Armington, A. F., 155(62), 159(62), 160 (62), 161(62), 163(62), 191(62), 201(62), 205(62), 206, 819 Armstrong, W. P., 155(66), 819 Arthur, J. R., 141(28, 30), 142(30, 34), 143(36), 158(83), 176(30), 218, $19 Ashmore, P. G., 241(31), 260(31), 262 Austen, D.E. G., 140(15), 141(15), 218 Autler, S. H., 82(37), 189

B Bachran, F., 11(29), 66 Baenriger, N., 10(23), 66

Bagg, J., 125(132), 131 Baker, M. McD., 94(76), 102, 104, 130, 354(146),369 Baker, W.O . , 140(16), 141(16), 818 Bangham, D.H., 142(34), 818 Barbieri, F.H., 319(46), 366 Bardeen, J., 75(11), 98(91), 188, 130 Barsh, M. K., 306(k), 307, 316(31, 33), 366 Bartell, F. E., 267(5), 270(14), 277(25), 283(5, 14,47), 897, 898 Basolo, F.,314(27), 366 Bassi, I. W., 6(16, 19), 9(20), 10(27), 46 (27) 66 Batchelder, H.R.,155(66), 819 Bumbach, 1).O . , 207(143), 208,209(143), 881

Bayston, J., 321(53), 367 Becker, J. A., 76(13, 80(21), 83(21), 85 (21), 91(21, 51, 52, 53, 55), 92(60), 101, 108, 116(53), 127(137), 1.98, 1.99, 131, 329 (70),367 Beeck, O.,81(27, 32), 122, 125(32), 128, 199, 245(34, 35), 262, 331(76), 332, 334(86, 87, 88, 891, 349(89), 367, 368 Beetlestone, J. G., 361(164), 370 Benedict, W.S., 224(2, 3), 861 Bennett, J. E., 140(18), 218 Benson, S. W., 290(65), 898 Berghausen, P.E., 266(3), $96 Binford, J. S., Jr., 155(73), 160(73), 819 Bloyaert, F.,97(88), lOO(88), 101(88), 102, 103, 104, 107(88), llO(88), 130 Blyholder, G., 155(79), 156(79), 160(79), 161(79), 194(79), 819 Bockris, J. O'M.,322(60), 331(75), 333, 350(75), 351, 367 Bokhoven, C., 233(22), 868

371

372

AUTHOR INDEX

Bond, G. C., 250(44),251(44,45), 252(44), 261(44), 868, 348(122, 124), 369 Bonhoeffer, K. F., 119(126), 131, 322(56), 345(114), 357(150), 367, 368, 369 Bonita, E., 30(44), 66 Bonner, F., 148(50), 818, 225(10), 868 Bosworth, R. C. L., 72, 93(65, 66),99, 101, 102, 103, 105(95, 96), 112(95, 119), 113(96), 114(96, 119), 116(65), 120, 188, 129, 130, 131, 333(82), 367 Boudart, M., 107(108), 124(108), 130, 330(73), 333(85), 344(85), 352(135), 353(143a), 367, 369 Bourion, R., 103, 105(102), 130 Bowring, J. R., 142(34), 143(36),818 Boyd, G. E., 265(1), 266(l), 896 Brandes, R. G., 92(60), 189 Brattain, W. H., 91(53), 98(91), 116(53), 189, 130 Bridger, G.W., 141(29),818 Bridgman, P. W., 83(38), 189 Briggs, W. S., 250(43), 253(43), 254(43), 868

Broeder, J. J., 108(111), 109(111), 110 ( l l l ) , 130,340(103a,b), 342(103b),368 Bromley, L. A., 176(119),880 Brown, A. R. G., 174(116), 181(116), 880 Brown, F., 148(51), 818 Brown, J. F., Jr., 328(64), 367 Burger, R. M., 82(34), 189 Burwell, R. L., Jr., 234(26), 235(29), 245 (37), 246(29, 37, 42), 247(37, 42), 248 (42), 249(29), 250(26, 29, 37, 42, 43), 253(29, 37, 42, 43), 254(26, 29, 37, 42, 43), 255(42), 256(29, 42), 257(42), 868 Busche, R. M., 155(66), 819 Burshtein, R., 100(101), 130

C Calvin, M., 303(4, 5), 306(k), 307, 310 (4, 5), 316(4, 5, 32), 317(32), 327(32), 366, 366 Carman, P. C., 190(125),880 Carrington, A., 326(63a, b), 367 Carter, R. L., 140(13), 817 Cartmell, E., 9(21), 66 Castellano, S., 319(46), 368 Ceeari, M., 6(18), 66 Chalk, A. J., 306(j), 307,314(28), 316(28), 366

Chalmers, J. A., 74, 188 Chen, C. Y., 148(54), 149(54), 155(54), 160(54), 819 Chen, M. C., 155(77), 161(77), 819 Chessick, J. J., 266(2), 267(2), 270(15), 271(15, 23), 275(22), 276 20, 22, 23), 277(30), 280(20, 30, 37, 39), 281(41), 283(48), 284(15), 288(59), 289(30, 37, 61, 62,63), 292(68, 70), 896, 297, 298, 899 Christensen, C. J., 155(77), 161(77), 219 Chuckhanov, Z. F., 141(25,26), 174(113), 81 8, 880 Clark, A., 265(1), 266(1), 896 Cobb, J. W., 155(69), 819 Compere, E. L., 305(16a), 366 Conway, B. E., 322(60), 331(75), 333,350 (75), 351, 367 Copley, M. J., 100(99),103,130 Corradini, P., 2(2), 5(13, 14), 6(16, 17, 18, 19), 9(20), 10(27), 46(27), 66 Corson, B. B., 348(120), 369 Coulson, C. A., 79(17), 128 Couper, A., 343(112), 346(115a), 348 (112), 360(162), 361(162, 164), 368, 370 Craig, R. G., 270(14), 283(14), 297 Cramer,.'1 L., 203(140), 881 Cremer, E., 349(128), 369 Crisp, D. J., 292(67), 899 Csech, H., 93(69), 130 Culbertson, J. L., 283(47), 298 Culver, R. V., 95(77), 100(77), 102, 104, 125(77), 130 Cunningham, R. E., 53(50), 66 Curran, G. P., 155(71), 163(71), 819

D Dahlstrom, R., 266(1), 266(1), 896 Damon, G. H., 174(114),880 Danusso, F., 2(2), 9(22), 10(26), 39(22), 64(51), 66, 66 Davis, H., 174(115), 220 Day, J. E., 203(138), 881 Day, R. J., 141(24), 142(24), 155(24), 173 (24), 174, 175, 2oS(24), 818 Dayton, J. C., 306(h, i ) , 307, 322(57), 323(61), 367 de Boer, J. H., 78(14), 81(26, 28), 111

373

AUTHOR INDEX

(117), 114(117), 123(130), 124(14, 130), 126(133), 127(133),128, 129,131, 280 (40), 281(40), 282(40, 44), 298, 329(69), 331(74), 344(69,113), 357(74), 367,368 de Boer, N. H., 124(131), 125(131), 131, 334(93, 94a), 346(94a), 347, 349(94a), 368 Deitz, R., 364(165), 370 Deitz, V, R., 140(4), 217 Dibeler, V. H., 225(14, 15), 262 Dienes, G. J., 210(145), 211(145), 221 Dillon, J. A., 82(35), 98(35), 129 Doke, T., 96(81), 102, 103, 130 Dolch, P., 151(56), 219 D'Or, L., 97(88), lOO(SS), 101(88), 102, 103, 104, 107(88), 110(88), 130 Dorgelo, G., 245(36), 262 Dorgelo, G. J. H., 94(74), 102, 103, 130, 334(94a), 335(96b), 336(96c), 346 (94a), 347, 349(94a), 368 Dotson, J. A., 160(93), 219 Dowden, D. A., 71(2), 128, 331(78), 336, 342 (109), 348(118), 351(133), 355(133), 367, 368, 369 Drain, L. E., 282(45), 290(64), 298 Dresel, E. M., 187(123), 220 Dugas, C., 350(129), 369 Dumanskii, A. V., 279(35), 298 Dunford, H. B., 279(33), 280(33), 297 Duval, X., 141(23), 142(23), 155(64), 218, 219

E Earp, F. K., 161(99), 181(99), 203(99),220 Eckstrom, H. C., 341(108a), 368 Eischens, R. P., 107(109), 108(109, 110), 130, 340(104-108), 341, 342, 368, 370 Eisinger, J., 95(78, 79), 104, 127(78), 130 Eley, D. D., l00(100), 103, 113(120), 119 (loo), 120, 122(129), 130, 131, 303(3), 322(59), 332(3), 343(110), 346(115a), 347(116), 348(112, 119), 360(162), 361(162), 366, 367, 368, 369, 370 Ellison, A. H., 281(42), 298 Elovich, S. Y., 114(121), 131 Emmett, P. H., 140(12), 179(121), 201 (129), 217, 220, 276(21), 297, 343(111, 112), 368 Engell, H. J., 350(130), 369

Ercoli, R., 319(46), 366 Ergun, S., 146(45), 147, 159(45), 194(45), 200(45), 201(45), 218, 318(41), 366 Evans, M. G., 322(59), 367 Eyring, H., 107(106, 107), 113(107), 114 (106, 107), 121(106, 107), 122, 123 (106, 107), 125(106, 107), 130, 155 (73, 77, 791, 156(79), 160(73, 791, 161 (77, 79), 194(79), 219, 325(62a), 330 71b), 367

F Farina, M., 17(35), 32(35), 66 Farkas, A., 119(126), 131, 224(5, 6), 261, 345(114), 357(150a), 360(159,163), 368 Farkas, L., 224(5), 261, 360(159, 160), 370 Farnsworth, H. E., 82(34,35,36), 98(35), 129 Fasce, E. V., 320(49a), 366 Federova, A. J., 127(135), 131 Field, F. H., 224(7), 261 Fingas, E., 151(57), 219 Fischer, E., 360(160), 370 Flanagan, T. B., 258(47), 262 Flory, P. J., 5(15), 66 Fluornoy, J. M., 306(h), 307, 322(57), 36'7 Flynn, J. H., 321(55), 367 Foresti, R. J., Jr., 159(85), 162(85), 178 (85), 185(85), 202(85), 207(85), 819 Fowler, R. H., 84(40), 86(42), 129 Fowles, G. W., 9(21), 66 Fox, W. H., 269(13), 270(13), 297 Fraioli, A. V., 289(61), 298 Francis, S. A . , 107(109), 108(109, llO), 130, 340(104), 368 Frankenburg, W. G., 233(21), 262, 333 (81), 345(81), 367 Frank-Kamenetskii, D. A . , 165(104),218, 220 Franklin, J. L., 224(7), 234(25), 261, 262 Fraulini, F., 277(28), 297 Frechette, D., 277(27), 297 Friedel, R. A . , 319(48), 366 Friedrick, H. G., 353(143a), 369 Frumkin, A. N., 127(135), 131 Fu, Y., 283(47), 898

G Gadsby, J., 140(6), 143(6), 148(53), 149 (53), 151(53), 154(6), 155(53), 159(6), 160(53), 162(53), 217, 2f9

374

AUTHOR INDEX

Garner, W. E., 352(140, 141),569 Garten, V. A., 140(10), 817 Gentsch, H., 336(96e), 368 George, T.H., 82(34), 189 Gest, H., 359(152, 154), 360(154, 157), 362(157), 369, 570 Giachetti, E., 4(11), 13(30,31), 14(30,31, 32), 16(30, 31, 33), 19(30, 33), 20(30, 33), 23(30, 31), 24(31), 26(39, 40), 27(42), 28(42), 31(45), 32(39, 40), 33 (42), 35(46), 36(11), 39(40, 46), 51 (42),59(32), 64(52), 66,66 Giannini, U., 3(8), 66 Gilliland, E. R., 141(33), 142(33), 143(39), 154(39), 155(33), 166(So), 159(39), 161(33), 162(80), 828, 819 Giner, J., 96(82), 130 Girifalco, L.A., 274(18), 897 GiuffrB, L.,29(43), 66 Glass, B.,365(166), 370 Goldfarb, I., 319(43), 566 Golovina, E. S., l61(97), 190(128), 819, 880

f,

g, A , 307, 308(14), 309(15, 17), 310(16, 19, 19a,20,21), 311,312,313, 314(12, 28), 316(28), 366,566 Hammel, E. F., 225(8), 861 Handler, P.,98(90), 130 Hanlan, J. F., 279(33), 280(33), 897 Hansen, R. S.,292(69), 899 Hare, E.F., 269(12), 270(12),897 Harkins, W. I)., 265(1), 266(1), 268(10), 269 (lo),271,274(19),276(19), 280(36), 289(60),896, 897, 898 Harmon, C . G., 277(28), 697 Harriott, P.,156(80), 162(80), 819 Harrod, J. F., 310(19, 19a), 366 Hawif!, G.M.,229(20), 868 Harris, R.E., 267(6), 284(50), 897,898 Harrison, L. G., 352(138), 354(138), 369 Hartman, C.D.,127(137), 131 Hauffe, K.,350(130), 354(144), 369 Hayward, A., 360(162), 361(162), 570 Healey, F. H., 266(2), 267(2), 271(23), 275(22), 276(20, 22, 23), 280(20, 38, 39), 281(41), 289(61), 292(68), 296,

897, 998, 899 Gomer, R.,84(41), 92(59, 61, 63), 102, 103, 117(59,61,123), 118(63),119(63), Hedden, K.,156(81), 159(90), 210(144), 619, 881 124(41), 189, 131 Hennig, G. R., 140(17), 818 Good, R. J., 273(18), 897 Gorin, E.,151(59), 155(71), 156(59), 161 Herring, C.,73(6), 74,75(6), 76(6),80(6), 188 (59),163(59,71),819 Hickmott, T. W., 81(33), 189 Goring, G. E.,155(71), 163(71), 819 Graham, D., 140(11), 827, 275(22), 292 Higuchi, I., 107(106, 107), 113(107), 114 (106, 107), 121(106, 107), 122, 123 (69),899 (106, 107), 125(106, 107), 150 Graham, H. S.,143(41), 148(41), 155(41), Hill, G. R., 155(72), 203(72), 819 159(41), 160(41), 818 Hill, M.W., 161(99), 181(99),203(99), 280 Graham, J. A., 174(116), 181(116), 880 Hill, T.L., 273(16, 17), 276(21), 897 Green, D.E.,359(151), 569 Greenfield, H., 318(41, 48), 366 Hindin, S. G., 350(147), 357(148), 364 Greening, W.J . , 140(13), 827 (147,148), 369 Grisdale, R. O . , 210(130), 880 Hinshelwood, C. N . , 148(53), 149(53), Grodozovskii, M.K.,141(26), 828 151(53), 155(53), 160(53), 162(53), 619 Gulbransen, E. A., 155(78), 160(78), 163 Hoberman, H. D.,359, 360(155), 370 (78), 204(141), 819, 881 Holden, J. H., 160(93), 819 Gwathmey, A. T., 53(60), 66 Hollabaugh, C. H., 270(15), 271(15), 284 Gysae, B.,90(49), 189 (15),887 Hottel, H. C., 174(111), 115, 880 H Hackerman, N., 96(83), 130, 267(8), 897 Hougen, 0. A., 135(1), 827 Huffman, E.W. D.,140(9), 817 Hall, A. R., 174(116), 181(116), 810 Halpern, J.; 303(7), 304(7,8,9,10,11, 12, Hulatt, M.J., 346(115a), 368 13,14,15, M),305(11,16),306(c, d , e, Hulbert, H . M.,75(8), 118,321(55),367

375

AUTHOR INDEX

Hull, A. W., 88(45), 129 Hulm, J. R., 92(59), 103, 117(59), 129 Hunt, B. E., 160(92), 219 Hutchinson, E., 267(4, 7a), 292(4), 297

I Iguchi, M., 321(51), 367 Iler, R. K., 287(56), 298 I l k , B. V., 282(46), 286(53, 54), 298 Imperial, G., 201(132), 880 Ingles, 0. G., 151(58), 219 Ingram, D. J. E., 140(15,18), 141(15),818 Inokuchi, H., 370 Irsa, A. P., 225(10, 13), 239(13), 243(13), 250(13), 868 Ives, H. E., 92,93(64), 189

(19), 239(12, 30), 240(30), 241(31), 242(12, 30), 243(19), 244(19), 245(19, 39), 246(41), 247(41), 249(27), 250 (18, 19, 41), 252(18), 257(46), 258(41), 260 (19,31),26'2,286 (55),8Q8,349(127), 369

Key, A., 148(46),155(69),818, 21g Killman, E., 330(71a), 367 King, N. K., 321(53, 54), 367 Kingdon, K. H., 72,91(4), 92(56), 98(93), 99(93), 101,128, 129,130 Kipling, J . J., 291(66), 298 Kirch, L., 319(43, 45), 320(45), 366 Kiselev, A. V., 267(7b), 297 Kiselev, V. F., 267(76), 282(46), 286(54), 297, 298

Klibanova, T. M., 161(96), 220 Klein, R., 92(58), 103, 129 Klemm, W., 10(24), 66 J Kmetko, E. A., 202(133), 880 James, V. E., 155(70), 159(70), 163(70), Koehler, W. A., 160(93), 819 194(70), 819 Kohn, H. W., 357(148a), 369 Jenkins, G. I., 245(40), 262,354(146),369 Kolm, H. H., 98(92), 130 Johnson, A., 313, 366 Korinek, G. J., 304(13), 306(e), 307, 311, Johnson, F. B., 202(136), 280 312(25), 313, 366 Johnson, M., 112(118),131 Korswagen, A. R., 340 (103a), 368 Johnson, R. P., 80(18), 128 Kosiba, W. L., 210(145), 211(145), 221 Johnson, W. F., 267(9), 897 Kraak, J. H., 81(28), 189 Johnstone, H. F., 148(54), 149(54), 155 Krasna, A. I., 359, 360(156, 158, 161), (54), 160(54),219 361(156, 158, 161), 363(161), 370 Jolley, L. J., 148(55), 819 Kraus, G., 274(18), 297 Jonassen, H. B., 320(49a), 366 Kroger, C., 151(57), 219 Jones, G. W., 155(75), 219 Krose, E., 10(24), 66 Joyner, L. G., 276(21), 297 Krsek, G., 319(44), 366 Jura, G., 273(17), 274(19), 276(19), 280 Kubokawa, Y., 352(143), 369 (36), 297, 298 Kuchta, J. M., 174(114), 880

K Kanltgy, J. R., 279(34), 2Q7 Kant, A., 174(114),880 Kapauan, A. F., 306(1), 307, 317(36), 366 Karshavina, N. A., 155(65), 174(113),819,

L

Laby, T. H., 190(126), 880 Laidler, K. J., 243(33), 262, 312(23, 24), 329(23, 24), 362(24), 366 Lambert, J. D., 158(84), 161(95),219 220 Langmuir, I., 72, 79, 91(4, 50), 93(50), Katz, S., 160(92), 219 98(93), 99(93), 101, 108, 111(50), 114 Kauder, L. N., 225(11), 250(11), 262 (122), 116(50), 120, 188, 129, 130, 131, Kaye, G. W., C., 190(126),220 141(19),218 Keenan, A. G., 278(31), 297 Lange, E., 75(10), 96(82), 128, 130 Keler, N. P., 140(7), 217 Laporte, F., 283(47), 298 Kemball, C., 225(12), 227(18), 228(12, 19), Laroche, J., 358(150b), 369 229(19), 234(24, 27, 28), 235(12), 238 Lee, E. H., 96(83), 130

376

AUTHOR INDEX

Matsen, F. A., 234(25), 868 Maxted, E. B., 310(22), 327(22), 366 Lennard-Jones, J. E., 68(1), 188,329(71), Mayer, H., 93(67), 101, 129 Mayers, M. A., 155(67, 74), 159(67), 160 567 (91), 174(112), 219, 280 Letort, M., 161(96), 819 Mazzanti, G., 2(2), 3(8, 9 ) , 4(9), 9(22), Levine, O., 281(43), 298 lO(9, 25), 11(28), 24(9), 39(22), 46 Levy, M., 17(36), 66 Lewis, T. J., 75, 188 (281, 66 Lewis, W. K., 141(33), 142(33), 143(39), Mekryach, E. F., 279(35), 298 154(39), 155(33), 159(39), 161(33), Metlin, S., 318(41), 366 Meyer, L., 141(21), 161(21), 818 418 Mignolet, J. C. P., 79(16), 81(30), 89(47), Liang, S. C., 352(137), 569 90(30), 93(73), 97, lOO(30, 88, 89), Liggett, L. M., 202(134), 880 101(88), 102, 103, 104, 105(30), 106 Lindgren, B., 313, 566 (103), 107(16, 88), 108, 109(113), 110 Linton, H., 322(60), 367 (73, 86, 88, 103), 111(103), 119(124), Littlewood, A. B., 234(26), 250(26), 254 124(16), 125(16), 128, 129, 130, 131, (26), 862 333(83, 84), 336(83, 96d), 367, 368 Loebenstein, W. V., 140(4), 817 Long,, F. J., 140(6), 143(6), 146(43), 148 Mikos, N. N., 267(7b), 297 (43), 150(43), 154(6), 157(43), 159(6), Mikovsky, R. J., 330(73), 567 Milkovich, R., 17(36), 66 "(43, 94), 203(94), 817, 818,819 Longi, P., 3(9), 4(9), 10(9), 11(28), 24(9), Miller, S. L., 306(h), 307, 322(58), 367 Mills, G. A., 303(6), 306(k, l ) , 307, 316 46(28), 66 Los, J. M., 282(45), 298 (29), 317 (34), 320(34), 321(50), 357 (149), 366, 366, 367, 369 Lundy, R., 92(01, 03), 102, 117(61, 63), Miscenko, K. P., 75(10), 188 118(63), 119(63), 189 Miyahara, K., 242(32), 245(38), 268 M Molinari, E., 352(134), 353(134), 354(134), 369 McBride, G . T., Jr., 143(39), 154(39), Molstedt, B. V., 202(137), 820 159(39), 818 McDowell, C. A., 352(138), 354(138), 369 Mooi, J., 267(6), 284(50), 897, 898 Mooney, R. W.,278(31), 297 McDuffie, H. F., 305(16a), 566 Moore, G. E., 91(54), 101, 112(54), 114 McElroy, W. D., 305(166), 570 (54), 122, 129 Macgregor, E. R., 304(9, 161, 305(16), Moore, L. E., 110(115), 6131 306(a), 307, 310(16), 366, 566 Moraglio, G., 2(2), 21(41), 43(41), 45(41), Mackenzie, N., 351(133), 355(133), 369 McQuillan, A. I)., 127(136), 131 65, 66 McWhorter, A. L., 82(37), 189 Mori, S., 160(92), 819 Magrone, R., 161(96), 819 Morikawa, K.,224(2, 3, 4), 2661 Manchester, K. E., 267(7a), 897 Morrison, J. A., 282(45), 290(64), 298 Man'ko, N. M., 140(7), 817 Morrison, J. L., 279(33), 280(33), 897 Mantica, E., 2(2), 9(22), 17(34), 39(22), Morrison, 5.R., 352(142), 369 66,66 Moser, J. F., Jr., 202(137), 220 Markby, R., 319(48), 320(49), 366 Muller, E. W., 80(19), 92(62), 128, 189 Markham, M. C., 243(33), 868 Mulliken, R. S., 97(87), 108(112), 110 Marsh, J. D. F., 143(38), 818 (87), 130 Marshall, A. L., 140(14), 818 Muroyama, N., 121(128), 131 Martin, A. R., 319(47), 366 Muttik, G. G., 207(7b), 897 Martin, J., Jr., 190(127), 880

Lee, E. L., 337(98), 338(98), 339,345(98), 368

AUTHOR INDEX

N

377

36(11), 39(40,46), 46(48), 51(42, 49), 52(49), 53(49), 54(49), 56(49), 59(32), 64(52), 66, 66 Pate, W. N., 277(26), 897 Patrick W.A., 288(58), 298 Pauling, L., 106(104), 130, 269(49), 862, 331(77), 367 Paxton, R. R . , 141(33), 142(33), 148(49), 155(33), 158(49),161(33),218 Pearson, R. G.,314(27), 366 Peck, H.D.,Jr., 359(154), 360(154), 369 Peck, R.E., 160(92), 819 Pegoraro, M.,45(47), 66 Peraldo, M., 9(22), 39(22), 66 Peters, E.,304(8, 91, 306(c, B), 307, 310 (20,21),311,312,366 Petersen, E. E.,159(87), 167(105), 178 (87,120), 188(124), 219, 880 Pickering, H. L.,341(108a), 368 Pierce, W. C., 267(6), 275(22), 276(22), 284(50), 897, 898 Pilcher, J . M.,155(68), lsO(68), 162(68),

Nakada, I . , 96(81), 102, 103, 130 Nabors, L. G.,140(9), 817 Natta, G.,2(1-5),3(5,6,8,9),4(5,6,9,10, 11,12),5(5,13,14),6(5,16,17,18,19), 9(5, 20, 22), lO(5, 9, 25, 26, 271, 11 (5,6, 28), 13(30, 31), 14(30, 31, 32), 16(30, 31, 33), 17(34, 35), 19(30, 33), 20(30, 33), 23(30, 311, 24(9, 31)) 26 (39,40),27 (42),28(42), 29(43), 31 (451, 32(35, 39, 40), 33(42), 35(46), 36(10, ll), 39(22, 40, 46), 45(47), 46(5, 27, 28, 48), 51(42, 49), 52(49), 53(@), 54 (49), 56(49), 59(32), 64(10, 51, 52), 66, 66, 319(46), 366 Nebel, G.J., 203(140), 881 Neumann, B.,151(57), 819 Newitt, E.J., 158(83), 219 Nichols, J. R., 207(142), 208(142), ,921 Nichols, M. H.,73(6), 74, 75(6), 76(6), 80(6), 128 Nordheim, L. W., 84(40), 129 819 Norton, J. F.,140(14), 818 Pino, P., 2(2), 3(8, 9), 4(9), 9(22), lo@, 25), 11(28), 17(35), 24(9), 32(35), 39 0 (22),46(28), 66, 66 Pliskin, W. A., 107(109), 108(109, 110), Oatley, C. W., 88(44), 96(44),103,129 1.30, 340(104, 106,107), 368, 370 Ogawa, I.,96(81), 102,103,130 Polanyi, M.,302(2), 330(71b), 366, 367 O'hern, H.A., Jr., 190(127), 880 Poll, A., 148(55), 219 Olphin, A. R., 92, 93(64), 129 201(131), 202,207, 280 Orchin, M.,318(38, 39, 42), 319(43, 45), Polley, M.H., Porter, A. S.,80(20), 128 320(45), 366 Porri, L.,10(27), 46(27), 66 Orgel, L. E., 326(62b), 367 Potter, P . E., 310(19), 366 Orning, A. A . , 148(52), 218 Prater, C. D.,165(103), 166(103), 167 P (103),170(103), 188(103),220 Pratt, G.W., 98(92), 130 Parker, A. S.,174(111), 880 Pajaro, G.,14(32),27(42), 28(42), 33(42), Price, P. H.,155(61), 219 51(42, 49), 52(49), 53(49), 54(49), Pritchard, J., 95(77), 100(77), 102, 10% 125(77), 1.30 56(49), 59(32), 66, 66 Pyehev, V., 233(23), 869 Parmele, C. W., 277(27), 897 Parravano, G.,225(8), 861, 352(134, 135), R 353(134,143a),354(134), 369 Raats, E., 159(88), 167(106), 172(109), Parsons, R . , 75(9), 78(9), 128 187(106), 188(106), 219, 220 Pasquon, I . , 4(11), 6(17), 10(26), 13(30, 31), 14(30, 31, 321, 16(30, 31, 331, 19 Rabinovitch, B. S.,258(47), 262 (30, 33), 20(30, 33), 23(30, 31), 24 Raftery, M.M., 158(83), 219 (31), 26(39, 401, 27(42), 28(42), 29 Rakszawski, J. F., 202(134), 220 (43),31(45),32(39,40),33(42),35(46), Raaouk, It. I., 284(49), 298

378

AUTHOR INDEX

nee, T., 107(106, 107), 113(107), 114(106, 107), 121(106, 107, 128), 122, 123(106, 107), 125(106, 107), 130, 131 Reid, W. D., 349(126), 369 Reif, A. E., 143(37), 144(37), 145(37), 146(37,44), 151(37), 818 Reimann, A. L., 83(39), 92(57), 99(57,94), 101, 103, 199, 130 Reynolds, P. W., 348(118), 368 Rhead, T. F. E., 141(27), 818 Rice, C. W., 173(110), 880 Richardson, 0. W., 72, 188 Richardson, R. L., 290(65), 898 Rideal, E. K., 94(76), 99(95,96), l00(100), 101, 102, 103, 104, 112(95), 113(96), 114(96), 119(100, 125), 130, 131, 245 (40), 868,334(90,91), 345(90,91,115), 368 Ritchie, A. W., 245(35), 868 Rittenberg, D., 306(h), 307, 309(18), 322(58), 357(18), 359,360(18,155,168, 158, 161), 361(156, 158, lei), 362(18), 363(161), 366, 367, 370 Roberts, J. K., 80, 113(24), 122, 124(24), 188, 332(80), 345(80), 367 Roberts, L. E., 187(123), 880 Robertson, F. S., 72, 188 Roelen, O., 318(40), 366 Rossberg, M., 141(32), 142(32), 155(32), 156(32, 82), 157, 158(32), 159(32), 160(32), 165(82), 191(32), 201(32), 202(32), 204, 818, 819 Rossington, D. R., 346(115a), 368 Rowlinson, H. C., 235(29), 246(29, 42), 247(42), 248(42), 249(29), 250(29, 42), 253(29, 42), 254(29, 42), 255(42), 256 (29, 42), 257(42), 86.9 Ruch, R . J., 277(25), 897 Rummel, K. W., 357(150a), 369 Rundle, R. E., 10(23), 66 Runov, A. D., 267(7b), 897 Rusinko, F. Jr., 159(86, 89), 172(109), 202(134), 205(89), 206(89), 819, 880

S Sabatka, J. A., 337(98), 337(98), 339(98), 345(98), 368 Sachtler, W. M. H., 86(43), 93(43, 70), 94(74, 75), 98(43), 102, 103, 107(75), 108(111), l09(lll), llO(75, l l l ) , 118

(43), 126(75), 189, 130, 245(36), 868, 335(96a,b), 336(96c), 340(103b), 342 (103b), 368 San Pietro, A., 359(154), 360(154), 369 Sato, H., 203(139), 881 Savage, R. H., 140(5), 817, 276(24), 897 Scalari, F., 35(46), 39(46), 66 Schissler, D., 225(10), 268 Schlier, R. E., 82(34, 36), I89 Schneider, C. H., 294(71), 295(71), 899 Schofield, E. B., 80(23), 188 Schonland, D., 326(63b), 367 Schreiner, G. D. L., 286(55), 898 Schuit, G. C. A., 108(111), 109(111), 110 ( l l l ) , 124(131), 125(131), 130, 131, 334(93, 94a,b), 340(103b), 342(94b, 103b), 344(94b), 346(94a,b), 347, 349 (94a,b), 368 Schulmsn, F., 286(52), 898 Schuls, K., 110(114), 118(114), 126(114), 131

Schwab, G . -M., 330(71a), 448(117), 367, 368 Scott, D. S.,148(54), 149(54), 155(54), 160(54), 819 Scott, G. S., 155(75), 819 Secoy, C. H., 305(16a), 366 Selwood, P. W., 110(115), 131, 337(97102), 338(97, 98, 102), 339, 343(99), 346(98, 102), 347(100), 364(165), 368 Semechkova, A. F., 143(42), 818 Shafrin, E. G., 269(11), 270(11), 897 Shcherbakova, K. D., 267(7b), 897 Sheridan, J., 349(126), 369 Shim, B. K. S., 246(42), 247(42), 248(42), 260(42), 253(42), 254(42), 255(42), 256 (42), 257 (42), 868 Shishakov, N. A., 81(29), Id9 Shockley, W., 80(18), 188 Sihvonen, V., 141(20), 142(20), 155(76), 818, 819

Sill&, L. G., 313, 366 Simonetta, M., 17(34), 66 Skewis, J. D., 294(71), 295(71), 899 Sleightholm, P., 140(6), 143(6), 154(6), 159(6), 817 Smaller, B., 140(17), 818 Smith, A. E., 81(27), 188, 246(34), 268, 334(86), 367 Smith, H.A., 348(121), 369

379

AUTHOR INDEX

Smith, M. B., 267(7), 897 Smith, R. N., 275(22), 276(22), 897 Smith, R. P., 325(62a), 367 Smith, W. R., 201(131), 202,207(131),880 Spalaris, C. N., 187(122),211(122),880 Spence, R. W., lOO(99), 103, 130 Stiihler, A., 11(29), 66 Stephenson, M., 302(1), 359(1), 366 Sterling, E., 148(52),218 Sternberg, H. W., 318(37, 39, 41, 42), 319 (37, 48), 320(49), 366 Stellacci, V., 51(49), 52(49), 53(49), 54 (49), 56(49), 66 Stifert, A. C., 277(29), 897 Stone, H. H., 305(16a), 366 Strauss, G. R., 279(32), 897 Strickland, L. H., 302(1), 359(1, 151), 366, 369 Strickland-Constable, R. F., 141(22), 142(22), 148(47, 48), 151(47), 155(22), 48, 818 Strother, C. O., 352(136), 369 Studebaker, N. L., 140(9), 817 Suggitt, R. M., 267(5), 283(5), 297 Suhrmann, R., 86(43), 93(43, 68, 69, 70, 71,72), 94(68,72), 98(43,68),102,107 (72), llO(72, 114, 116), 118(43, 68, 114), 126(114), 129, 130, 131, 334(95, 96), 335, 336(96e), 368 Swain, G. G., 328(64), 367 Sykes, K. W., 140(6), 143(6), 146(43), 148(43, 53), 149(53), 150(43), 151(53), 154(6), 155(53), 157(43), 159(6), 160 (43, 53, 94), 162(53), 203(94), 817, 818, 819 Symons, M. C. R., 326(63a,b), 367 Szwarc, M., 17(36, 37), 66

T Takaishi, T., 342, 368 Tarbox, R. P., 155(71), 163(71), 819 Taylor, E. H., 357(148a), 369 Taylor, H. S., 223(1), 224(2, 3, 4), 225 (8,9), 86'1,330(73), 352(136,137, 139), 367, 369 Taylor, J. B., 91(50), 93(50), 101, 108, 111 (50), 120, 189 Taylor, T. I., 225(11, 14, 15), 226(16), 243(16), 250(11), 257(16), 868, 348 (125), 36'9

Teller, E., 343(111), 368 Temkin, M. I., 127(134), 131, 233(23), 868

Tester, I).A., 291(66), 898 Thiele, E. W., 165(100), 168(100), 169 (1001, 216(100), 820 Thomas, B. D., 265(1), 266(1), 296 Thomas, L. B., 80(23), 188 Thompson, S. O., 225(13), 239(13), 243 (13), 250(13), 868 Thring, M. W., 155(61), 819 Tompkins, F. C . , 80(20), 95(77), 100(77), 102, 104, 125(77, 132), 188, 130, 131 Toyama, O., 352(143), 369 Trapnell, B. M. W., 106(105), 120, 122 (127), 130, 131, 140(3), 817, 260(48), 868, 329(67, 68), 330(72), 332(68, 79), 334(72,90,91, 92), 345(79, 90,91,92), 346(79, 92), 351(133), 355(133), 367, 368, 369 Trenner, N. R., 224(4), 261 Tu, C. M., 174(115), 880 Tuddenham, W. M., 155(72),203(72), 819 Turkevich, J., 148(50), 818, 225(10, 13), 239(13), 243(13), 250(13), 262, 358 (150b), 369 Tuxworth, R. H., 235(29), 245(37), 246 (29, 37), 247(37), 249(29), 250(29, 371, 253(29, 37), 254(29, 37), 256(29), 268

v van der Knapp, W., 245(36), 868 van Heerden, C., 233(22), 868 van Reijen, L. L., 108(111), 109(111),110 ( l l l ) , 130, 334(94a,b), 340(103a,b), 342(94b, 103b), 344(94b), 346(94a,b), 347,349(94a,b), 368 Vaska, L., 337(102),338(102),345(102,368 Van Voorhis, J. J., 270(14), 283(14), 897 Vastola, F. J., 155(63), 819 Veenemans, C. F., 78(14), 124(14), 188 Vick, F. A., 112(118), 131 Vitman, L. A., 155(60), 819 Voltz, S. E., 354(144a, 145), 369 von Duhn, J. H., 96(80), 130 Vulis, L. A., 155(60),819

W Wagener, S., 90(49), 189 Walker, P. L., Jr., 143(35), 155(68), 159

380

AUTHOR INDEX

(85, 86, 88), 160(68), 162(68, 85), 167 Wolf, F. J., 234(28), 262 (l06), 172(109), 178(85), 185(85), 187 Wolfe, A. C., 140(9), 217 (106), 188(106, 124), 201 (132), 202 Wolkenstein, T., 351(132), 369 (85, 134), 207(85, 142, 143), 208(142), Woo, L. F., 305(16a), 366 Wood, L. A., 278(31), 897 200(143), 818, 819, 880, 881 Wood, R. E., 202(136), 880 Wall, M. C., 243(33), 868 Wortman, R., 92(61,63), 102, 117(61,63), Waring, W. S., 359(153), 369 118(63), 119(63), 129 Watson, K. M., 135(1), 217 Wotis, J., 319(48), 366 Watt, W., 174(116), 181(116), 680 Webster, A. H., 304(10, 11, 12, 14), 305 Wright, C. C., 143(35), 155(68), 159(85, 87), 160(68), 162(68, 85), 178(85, 87), (11), 306(d, f ) , 307, 308(14), 311, 312 185(85), 188(124), 202(85), 207(85), (12), 314(12), 366 818, 219, 280 Wedler, G., 336(96e), 368 Wright, L. W., 306(1), 307, 316(30), 317 Weiss, D. E., 140(10), 617 (34), 320(34), 357(149), 366, 369 Weissler, G. L., 88(46), 96(46), 189 Weiss, P. B., 165(103), 166(103), 167(103), Wright, M. M., 225(9), 261 170(103, 108), 188(103), 880, 350(131), Wright, P. C.,241(31), 260(31), 262 Wu, P. C., 143(40), 146(40), 154(40), 159 369 (40), 218 Weller, S. W., 303(6), 306(k, l ) , 307, 316 (29, 30), 317(34), 320(34), 321(50), Y 354(144a), 354(145), 356(147), 357(148, 149), 364(147, 148), 366, 1366, 367, Yager, W. A., 140(16), 141(16), 218 Young, C. J., 266(2), 267(2), 275(22), 276 369 (20, 22), 277(30), 280(20, 30, 38, 39), Wender, I., 318(37,39,41,42), 319(37,48), 281(41), 287(57), 288(57, 59), 289(30), 320 (49), 366 292(68, 70), 696, 297, 898, 899 Werkman, C. H., 359(153), 369 Young, T. F. 267(7), 297 Westrik, R., 233(22), 268 Yu, Y. F., 283(48), 898 Whalen, J. W., 267(9), 897 Wheeler, A., 81(27, 31), 188, 189, 165 Yudkin, J., 360(159), 370 (101, 102), 166(101), 167(101), 280, Z 245(34), 262, 321(50), 334(86), 367 Zelinski, J. J., 140(8), 217 Wheeler, R. V., 141(27), 818 Wicke, E., 141(31), 156(31, 81, 82), 158 Zettlemoyer, A. C., 266(2), 267(2), 270 (15), 271(15, 23), 275(22), 276(20, 22, (31), 159(31), 160(31), 161(31), 163 23), 277(30), 280(20, 30, 37, 39), 281 (31), 165(31,82), 167(31), 187(31), 209 (41), 283(48), 284(15, 51), 288(59), (31), 818, 819, 880 289(30, 37, 62, 63), 292(70), 294(71), Wilke, C. R., 176(118, 1191, 880 295(71), 896, 897, 298 Wilmarth, W. K., 306(h, i , k, 1 ) , 307, 316 (31, 32, 33), 317(32, 35, 36), 322(57), Zhabrova, G. M., 114(121), 151 Ziegler, K., 3(7), 17(7), 66 323(61), 327(32), 366, 367 Zielke, C. W., 151(59), 156(59), 161(59), Wilson, I. B., 329(65, 66), 367 163(59), 819 Wilson, T.N., 88(46), 96(46), 149 Zieman, W. A., 89(48), 98(48), 189, 269 Winfield, M. E., 321(52, 53, 54), 367 (11,12,13), 270 (11,12,13), 281 (42,43), Winslow, F. H., 140(16), 141(16), 818 286(52), 897, 898 Winter, L. L., 283(47), 898 Zwietering, P., 233(22), 262 Wirtz, K., 322(56), 367

Subject Index A Active centers, in polymerization, 13, 50-60 Activation energy of catalytic HZactivation, 301 desorption process, 111 gas carbon reactions, 156 Hz-DZexchange, 239, 241, 246, 252 olefin polymerization, 21 migration, 115 Activation of hydrogen, 301-368 by Agf ions, 305 by Cu++ ions, 304, 311 by Hg+ or Hg++ ions, 309, 312 by Milo4- ions, 308 in heterogeneous systems, 329-357 with ciipric and cuprous salts, 314-316 with cobalt complexes, 318 with ethyleneplatinous chloride, 321 with silver salts, 317 Adsorption and magnetic susceptibility, 337 and work function, 91 heat of, on metal surfaces, 119 ionic, on metal surfaces, 79 in solution, 291 of alkyl aluminums on a-TiClr , 50 of CO on metals, 94, 95, 100 of CZH4 on metals, 94 of gases on Ni, W, Cu, 97 on carbon, 140 of hydrogen, 69, 332, 342 on metal films, 94, 110, 105 on nickel, 93, 109, 335-340 on Pt, 93 on zinc, 352 of oxygen on nickel, 109 on metal films, 96 of xenon on nickel, 93 of hydrogen, 69, 332, 342 on semiconductors, 71 on metals, 67-128, 110

preferential, 291 thermodynamics, 271 Van der Waals, 79 Alumina, Hz-DZexchange over, 356 Aluminum, triethyl, in complex catalysts, 11, 19, 27, 30, 50 diethyl, in complex catalysts, 51 chloride, 357 Atactic polymers, 5, 24

B Biological systems, hydrogen activation in, 358 Bond formation, 106 type, in chemisorbed HZ, 342

C Catalysts, bifunctional, 327 stereospecific, 3, 10, 24 nature of coordinated anionic, 45 Carbon, adsorption of gases on, 140 dioxide, heat of reaction with carbon, 135 effect of impurities on activity of, 203 gas reactions of, 133-217 heats of reaction with 0 2 , COz , Hz , Hz0, 135 monoxide, adsorption of, 94-100 chemisorbed on metals, 104 heat of reaction with 0 2 , HzO, 135 reactivity after heat treatment, 206 Chain transfer, 26,32, 37,43 and termination, 23 Chemisorption of alkalies on metals, 101 of CO, OZ on metals, 103, 104 (see also adsorption) of HZon metals, 102,332, 352 on metal surfaces, 106 Chromic oxide, in Hz-DZexchange, 354 Cobalt carbonyls, H1 activation by, 318 Cobaltous cyanide, HZactivation by, 320

381

382

SUBJECT INDEX

Contact potential difference, 76, 78, 87 Complex, catalytic in olefin polymerization, 8 Complex formation in hydrogen activation, 310, 326 Complex metal halide-metallorganic compound catalysts, 2 Covalent bonding on Ni surface, 109 Crystal size and polymerization rate, 11 Crystallite orientation in gas-carbon reactions, 201 size in gas-carbon reactions, 205 Crystalline substrate in polymerization, 9 Cycloparaffins in a 2 - D ~exchange, 239248 (also see Hn-D$exchange)

D &Character, 333 Desorption of gases from metal surfaces, 111, 113 Deuterium exchange, 223-261 with highly symmetrical molecules, 239 hydrogen in benzene, 258 hydrogen in methylamine, 234 hydrogen in cycloparaffins, 253, 258, 257 hydrogen in paraffins, 227, 229, 239, 243,250,252,253, 254 molecules of xt low degree of symmetry, 250 Dipole moment, 78

E Electrical resistance of metal films, 334 Electrolytic hydrogen evolution, 350 Electron gas in adsorption, 128 Electron transfer from metal surfaces, 106 Equilibrium constants for gas-carbon reactions, 136 Ethane, adsorption, 243,339 Ethylene, adsorption, 94,339 Ethyleneplatinous chloride, 321 Exchange reactions, base catalyzed, 322 classification ofmultiple, 235 simple and stepwise, 233 mechanism, 236

of hydrocarbons with De , 223-261

theory for multiple, 238

F Fermi level, 70 Fermi-Dirac distribution, 75 Free energy of desorption, 113 Frequency factor in H2-D2 exchange, 241, 246, 248, 252

G Graphite, wetting of, in various liquids, 281, 283

H Halogens, effect on gas-carbon reactions, 209 Heat, immersional, 263-295 of adsorption, 119, 121 and coverage, 123,244 Heat treatment of carbons, 206 Heterogeneous catalytic systems, 329 Homogeneous catalytic systems, 303 Hydrogenase, 359 Hydrogen adsorption-see adsorption and chemisorption activation, electron configuration in, 324,330 Hydrogenation reaction, mechanism of, 348 Hydrogen-deuterium exchange, base catalyzed, 322 on alumina, 356 on AIClr , 357 on cobalt-thoria, 239, 243, 250 on chromia, 234, 254, 354 on charcoal, 358 on metal films, 243,249, 250 on molybdium, 244 on metal sulfides, 357 on nickel, 239, 244, 249, 253,254,256 on palladium, 234,239,244,249,254,256 on platinum, 239, 244, 250, 258 on rhodium, 239, 249, 251 on transition metal oxides, 355 on tungsten, 239, 250 on ZnO, 352

383

SUBJECT INDEX

I Infrared spectra of CO on Pt, Pd, 340 -of chemisorbed specie, 340-342 - o f ethylene on nickel, 341 Iron, heat of HZ adsorption on, 126 Isobutane, adsorption of on metals, 252 Isotactic polymers, 5, 24-29, 46 Irradiation, neutron, of graphite, 210 of y-AlaOs , 357

K Kinetics, of gas-carbon reactions, 153 of deuterium exchange, 230 of stereospecific polymerization, 1-66 of desorption and surface migration, 111

M Mass transport in carbon, 182 Magnetic susceptibility of nickel, 337 Metal ions, activity of HZ in aqueous solution by, 304 surface, “clean”, 79 properties of, 73 potentials, 101 work function of, 74 ~ in, 239 Methane, H z - D exchange Methylcyclopentane, Hz-Dzexchange in, 256 (+)3-methylhexane, Hz-Dzexchange in, 254

N Neopentane, Hz-L)~exchange in, 219 Nickel, covalent surface bonding, 109 gas adsorption on, 92,97,118 heat of HZadsorption on, 126 Nitric oxide, adsorption on Pt, 118

0 Oxygen, adsorption, 92, 96, 109, 140 chemisorption as negative ions, 71

P Parahydrogen conversion, mechanism of, 344 Polymerization of a-olefins, 1-66 ofgropylene, 14

Polymers, steric composition of, 25, 46 heteroblock, 64 molecular weight of, 13, 25, 61 Polar solids, 274 Polarity of solids, 284, 286 Propane, Hz-Dz exchange in, 250 Propyiene polymerization, 12, 19, 20

R Rate constants for hydrogen-deuterium exchange, 228, 231 of gas-carbon reactions, 162, 173 Reaction, diffusion in gas-carbon, 183, 188 Reaction, order of, in carbon-COz reaction, 154 in carbonsteam reaction, 155 in carbon-Hz reaction, 156 Reaction, gas, of carbon, 133-217 Rutile (TiOz), wetting of, by liquids, 281, 287

S Stereospecific catalysts, structure of, 8 polymerization, 1-66 Surface activation energy, 115 energy of solids, 273 migration, 111 modification by adsorbates, 77 nature of solid, 263-295 potential and adsorption on metals, 67-128 potentials of HZ , 0 2 , Nz , CO and other gases adsorbed on metals, 73 Syndiotactic polymers, 5

T Thermodynamics of adsorption and immersional wetting, 271 of gas carbon reactions, 135 Titanium trichloride complex as catalyst, 10, 16, 20, 50 Titanium dichloride complex as catalyst, 10

v Viscosity, intrinsic of polypropylene, 26, 41

384

SUBJECT INDEX

W Wetting of solids, 268 by organic liquids, 283 by water, 274 Wetting of, clays, 277 hydrophobic solids, 276, 283 organic fibers, 278 polar solids, 274, 280

Work function, 74, 77 and adsorption, 82, 91 measurement, 82-92

Z Zinc, diethyl, as catalyst component, 31