Modern Acetylene Chemistry
Edited by P.J. Stang F. Diederich
VCH
Modern Acetylene Chemistry Edited by P. J. Stang an...
214 downloads
1459 Views
23MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Modern Acetylene Chemistry
Edited by P.J. Stang F. Diederich
VCH
Modern Acetylene Chemistry Edited by P. J. Stang and F. Diederich
4b
VCH
Weinheim . New York . Base1 . Cambridge . Tokyo
This Page Intentionally Left Blank
Modern Acetylene Chemistry Edited by P. J. Stang and F. Diederich
Modern Acetylene Chemistry Edited by P. J. Stang and F. Diederich
4b
VCH
Weinheim . New York . Base1 . Cambridge . Tokyo
Related Titles from VCH A.Togni, T. Hayashi : Ferrocenes. VCH, 1995.
K. C. Nicolaou, E. Sorensen : Classics in Total Synthesis. VCH, 1995. J. Fuhrhop, G. Penzlin: Organic Synthesis. Second Edition. VCH, 1994.
M. N6grBdi : Stereoselective Synthesis. Second Edition. VCH, 1994.
0 VCH Verlagsgesellschaft mbH, D-69451 Weinheim (Federal Republic of Germany), 1995
Distribution: VCH, P.O. Box 10 11 61, D-69451 Weinheim (Federal Republic of Germany) Switzerland: VCH, P.O. Box, CH-4020 Base1 (Switzerland) United Kingdom and Ireland: VCH (UK) Ltd., 8 Wellington Court, Cambridge CBl 1 HZ (England) USA and Canada: VCH, 220 East 23rd Street, New York, NY 10010-4606 (USA) Japan: VCH, Eikow Building, 10-9 Hongo 1-chome, Bunkyo-ku, Tokyo 113 (Japan) ISBN 3-527-29084-2
Modern Acetylene Chemistry Edited by P. J. Stang and F. Diederich
4b
VCH
Weinheim . New York . Base1 . Cambridge . Tokyo
Prof. Dr. Peter J . Stang Department of Cheinistry University of Utah Salt Lake City, UT 84 I 12 USA
Prof. Dr. FranCois Diederich Laboratorium fur Organische Cheinie EidgenBssische Technische Hochschule ETH-Zcntrum Universitiitstrasse I6 CH-8092 Zurich Switzerland
This book was carefully produced. Ncvcrtheless, thc authors, editors and publisher do not warrant the information contained therein to be Krcc oferrors. Rcaders are adviscd to keep in mind that stalemcnts, data, illustrations, procedural details or other items may be inadvcrtcntly inaccurate.
Published jointly by VCH Verlagsgesellschaft mbH, Weinheim (Fedcral Republic of Germany) VCH Publishers, Inc., New York, NY (USA)
Editorial Director: Dr. Thomas Mager Production Manager: Dipl.-Wirt.-Ing. (FH) Bernd Riedel
Library of Congress Card No. applied for.
A catalogue record for this book is available from the British Library.
Deutsche Bibliothek Cataloguing-in-Publication Data:
Modern acetylene chemistry / ed. by P. J. Stang and F. Diederich. - Weinheim : New York ; Basel ; Cambridge; Tokyo: VCH, 1995 ISBN 3-527-29084-2 NE: Stang, Peter J. 1Hrsg.l
0VCH Verlagsgesellschaft mbH, D-6945 1 Weinheim (Federal Republic of Germany), 1995 Printed on acid-free and chlorine-free paper. All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form - by photoprinting, microfilm, or any other means - nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Composition: Filmsatz Unger & Sommer GmbH, D-69469 Weinheim Printing: Strauss Offsetdruck, D-69509 Morlenbach Bookbinding: J. Schaffer GmbH & Co. KG., D-67269 Grunstadt
Printed in the Federal Republik of Germany
Foreword
The carbon-carbon triple bond is one of the oldest and simplest functional groups in chemistry. The reactions and transformations of this humble functionality are intertwined with the history and development of organic chemistry. In the past dozen years, acetylene chemistry has experienced a major renaissance engendered by the incurrence of molecules with C = C bonds in the frontiers of modern organic chemistry - namely biochemistry and materials science. An entire family of powerful antitumor antibiotics with cis-enediynes as reactive fragments was discovered in the mid 1980s, and new members of this family continue to be found. The range and potential of these antitumor antibiotics has been greatly expanded by a family of synthetic enediynes capable, analogously to their natural counterparts, of undergoing the Bergmann cycloaromatization and efficiently nicking and cleaving DNA. On the other hand, acetylenic molecular scaffolding has been employed to prepare multinanometer-sized molecular objects with unprecedented structures, functions, and properties. Some of these materials are being developed into components for molecular electronics; others form crystals with molecular pores for separation, inclusion, and catalysis, and thereby become the organic counterparts of zeolites. Acetylenic two- and three-dimensional carbon allotropes, with structures and functions different from the natural modifications of diamond and graphite, as well as fullerenes, are under construction. New, fully conjugated, acetylenic polymer backbones complement the functional property range of polyacetylenes and polydiacetylenes. The construction of organic ferromagnets based on acetylenic backbones and scaffolds is being explored intensively. These developments, which offer plenty of fascinating perspectives at the two interfaces to materials science and biology, are efficiently fueled by the invention of powerful new synthetic methodology, based to a large extent on transition metal chemistry. The invention of new synthetic methods has particularly facilitated the cross-coupling between acetylenic sp-C atoms and alkene and arene sp2-C-atoms; reactions crucial to molecular scaffolding. Other important advances have been made in the formation of five-, six-, and higher-membered rings using alkyne transition metal chemistry. Small reactive acetylenes such as iodonium derivatives are increasingly used as reagents in organic synthesis since ways have now been found to control their reactivity and tame their previous tendency for spontaneous decomposition. The chemistry of heteroalkynes such as phosphaalkynes has emerged over the past decade. Theoretical chemistry has been challenged by the broad new developments in modern acetylene chemistry. Structures and electronic configurations of acyclic and cyclic acetylenic rr-systems have attracted the interest of both experimentalists and theoreticians, and much of the current knowledge on homoconjugation, and on through-space orbital interactions between precisely aligned chromophores, has been gained in studies of acetylenic systems. The structures and electronic properties of acetylenic all-carbon rods and rings, which are formed in the laser vaporization of graphite and occur as intermediates in fullerene production processes, have attracted much interest from theoreticians, providing attractive and challenging targets to calibrate and improve computational methods.
VI
Foreword
This multi-author monograph documents and critically analyzes these recent developments in contemporary acetylene chemistry in 13 chapters written by leading scientists in the various areas. With emphasis on the above-mentioned modern developments, the monograph does not duplicate previous treatises on alkyne chemistry such as Houben- Weyl-Miiller Vol. V/2a (Alkine, Di und Polyine, Allene, Kumulene), the volumes in the Patai series on The Chemistry of the Carbon-Carbon Triple Bond, the pioneering monograph on The Chemistry of Acetylenes by H. G. Viehe, or the book by Brandsma on Preparative Acetylenic Chemistry with a great variety of useful synthetic procedures. Rather, it builds upon these predecessors and complements them by updating the reader on the broad new developments in today’s acetylene chemistry. To enhance the practical value of the monograph, most experimental chapters include synthetic protocols which have been chosen for broad utility and application. We anticipate and hope that this monograph will further stimulate the development and application of acetylene chemistry as one of the key synthetic, structural, and functional tools of future chemistry. Salt Lake City and Zurich December 1994
Peter J. Stang Fransois Diederich
Contents
Foreword List of Contributors
1
Modern Computational and Theoretical Aspects of Acetylene Chemistry Dietmar A . Plattner, Yi Li. K . N . Houk
1.1 1.2 1.2.1 1.2.2 1.2.3 1.3 1.3.1 1.3.2 1.3.3 1.3.4 1.3.5 1.4 1.4.1 1.4.2 1.4.2.1 1.4.2.2 1.4.2.3 1.4.2.4 1.4.2.5 1.4.2.6 1.4.2.1 1.5
........................... Electronic structures of acetylene and monoacetylenes . . . . . . . . . Ground-state potential energy surfaces . . . . . . . . . . . . . . . . Excited-state potential energy surfaces . . . . . . . . . . . . . . . . Radicalions . . . . . . . . . . . . . . . . . . . . . . . . . . . Reactivities and molecular interactions of acetylenes . . . . . . . . . . Pericyclic reactions . . . . . . . . . . . . . . . . . . . . . . . . Electrophilic reactions . . . . . . . . . . . . . . . . . . . . . . . Nucleophilic additions . . . . . . . . . . . . . . . . . . . . . . Radical additions . . . . . . . . . . . . . . . . . . . . . . . . . Molecular complexes . . . . . . . . . . . . . . . . . . . . . . . Polyacetylenes . . . . . . . . . . . . . . . . . . . . . . . . . . Diacetylene . . . . . . . . . . . . . . . . . . . . . . . . . . . C, and cyclic C, . . . . . . . . . . . . . . . . . . . . . . . . . Cz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction
1
1 3 5
6
7 7 10 11 11 12
............................... c4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13 13 15 18 18 19 19 20 22 22
........................... ...........................
25 26
C3
C5. C,. andC, . C,j. C8. and Clo CI1to c,, . . .
. . . . . . . . . . . . . . . . . . . . . . . . . ......................... . . . . . . . . . . . . . . . . . . . . . . . . . c 1 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion References .
VIII 2
Contents
FunctionalizedAcetylenes in Organic Synthesis .The Case of the 1-Cyanoand the 1-Halogenoacetylenes
Henning HopJ Bernhard Witulski
. . . . . . . . . . . . . . . . . . . . . . . . . . .
33
2.1
Introduction
2.2 2.2.1 2.2.2 2.2.2.1 2.2.2.2 2.2.2.3
Synthesis and preparative use of cyanoacetylenes . . . . . . . . . . . Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preparative use of cyanoacetylenes . . . . . . . . . . . . . . . . . A short summary of the older literature . . . . . . . . . . . . . . Novel cycloadditions with cyanoacetylenes . simple and efficient methods for the construction of complex carbon frameworks . . . . . . . . . . Cyanoacetylenes as precursors for reactive and interstellar intermediates . .
39 46
2.3 2.3.1 2.3.2 2.3.2.1 2.3.2.2 2.3.2.3 2.3.3
Synthesis and preparative use of I-halogenoacetylenes . . . . . . . . . Older review of the literature on halogenoacetylenes . . . . . . . . . . Synthesis of 1-halogenoacetylenes . . . . . . . . . . . . . . . . . . The preparation of the 1-halogeno- and 1,2.dihalogenoethynes . . . . . More highly unsaturated halogenoacetylenes . . . . . . . . . . . . . Derivatives of 1-halogenoacetylenes . . . . . . . . . . . . . . . . . Novel preparative uses of 1-halogeno- and 1.2.dihalogenoacetylenes . . .
48 48 48 48 50 52 53
2.4 2.4.1 2.4.2 2.4.3 2.4.4 2.4.5 2.4.6
Experimental procedures . . . . . . . . . . . . . . . . . . . . . Cyanoacetylene (1) . . . . . . . . . . . . . . . . . . . . . . . . Dicyanoacetylene (2) . . . . . . . . . . . . . . . . . . . . . . . Dicyanodiacetylene (3) . . . . . . . . . . . . . . . . . . . . . . Chloroacetylene (93) . . . . . . . . . . . . . . . . . . . . . . . Dichloroacetylene (100) . . . . . . . . . . . . . . . . . . . . . . Diiodoacetylene (105) . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
60 60 60 61 61 62 62 63
3
Alkynyliodonium Salts: Electrophilic Acetylene Equivalents
34 34 38 38
Peter L. Stang
..........................
3.1
Introduction
3.2 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.2.6 3.2.7
Preparation and properties . . . . . . . . . . . . . . . . . . . . . Alkynyliodonium sulfonates . . . . . . . . . . . . . . . . . . . . Alkynyliodonium tetrafluoroborates . . . . . . . . . . . . . . . . . Heterocyclic alkynyliodonium species . . . . . . . . . . . . . . . . Mechanism of formation . . . . . . . . . . . . . . . . . . . . . Diynyliodonium and dialkynyliodonium triflates . . . . . . . . . . . Bis-iodonium species . . . . . . . . . . . . . . . . . . . . . . . Properties of alkynyliodonium salts . . . . . . . . . . . . . . . . .
67 68 68 69 70 71 72 72 73
Contents
IX
3.3 3.3.1 3.3.2
Characterization and structure . . . . . . . . . . . . . . . . . . . Spectroscopic properties . . . . . . . . . . . . . . . . . . . . . . X-ray and molecular structure . . . . . . . . . . . . . . . . . . .
74 74 75
3.4 3.4.1 3.4.1.1 3.4.1.2 3.4.1.3 3.4.1.4 3.4.1.5 3.4.1.6 3.4.2 3.4.3 3.4.3.1 3.4.3.2
Reactions and uses of alkynyliodonium salts . . . . . . . . . . . . . Reaction with nucleophiles . . . . . . . . . . . . . . . . . . . . . Carbon nucleophiles . . . . . . . . . . . . . . . . . . . . . . . Nitrogen nucleophiles . . . . . . . . . . . . . . . . . . . . . . . Oxygen nucleophiles . . . . . . . . . . . . . . . . . . . . . . . Sulfur nucleophiles . . . . . . . . . . . . . . . . . . . . . . . . Phosphorus nucleophiles . . . . . . . . . . . . . . . . . . . . . Halogen nucleophiles . . . . . . . . . . . . . . . . . . . . . . . Reaction with organometallic species . . . . . . . . . . . . . . . . Cycloaddition reactions . . . . . . . . . . . . . . . . . . . . . . [2 + 41-Diels-Alder cycloadditions . . . . . . . . . . . . . . . . . 1.3.Dipolar cycloadditions . . . . . . . . . . . . . . . . . . . . .
76 77 78 80 81 83 86 87 88
3.5
Conclusions
3.6 3.6.1 3.6.2
Experimental procedures . . . . . . . . . . . . . . . . . . . . . (Cyano[[(trifluoromethyl)sulfonyl]oxy)iodo)benzene, 7 . . . . . . . . . General procedure for the preparation of P-alkyl- and P-phenylethynyl(pheny1)iodonium triflates. 10 . . . . . . . . . . . . . . . . . . . . . General preparation of B-functionalized ethynyl(pheny1)iodonium triflates. 11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . General procedure for the preparation of bis-iodonium diyne bktriflates. 34 and 35 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preparation of bis(phenyl[[(trifluoromethyl)sulfonyl]oxy]iodo)ethyne, 30 . General procedure for the Diels- Alder reaction of alkynyl(pheny1)iodonium salts. 11. with 1.3.dienes. formation of cycloadducts 118-120 . . . . . . General procedure for the preparation of cyclopentenones and y-lactams . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.3 3.6.4 3.6.5 3.6.6 3.6.7
4
. . . . . . . . . . . . . . . . . . . . . . . . . . .
90 90 91 92 92 92 93
93 93 94 94 94 95
The Chemistry of Metal-Alkyne Complexes Gagik G. Melikyan. Kenneth M . Nicholas
4.1 4.2 4.2.1 4.2.2
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Bonding and structure . . . . . . . . . . . . . . . . . . . . . . . Alkyne complexes . . . . . . . . . . . . . . . . . . . . . . . . Propargylium-metal complexes . . . . . . . . . . . . . . . . . . .
4.3
Complexes of novel alkynes
4.4
Reactions of metal-alkyne complexes . . . . . . . . . . . . . . . . Reactions at the C - C triple bond . . . . . . . . . . . . . . . . .
4.4.1
....................
99 99 99 101 104 107 107
X
Contents
4.4.1.1 4.4.1.2 4.4.1.3 4.4.1.4 4.4.1.5 4.4.1.6 4.4.1.7 4.4.1.8 4.4.1.9 4.4.2 4.4.2.1 4.4.2.2 4.4.3 4.4.3.1 4.4.3.2 4.4.3.2.1 4.4.3.2.2 4.4.3.2.3 4.4.3.2.4 4.4.4 4.4.5
Nucleophilic addition . . . . . . . . . . . . . . . . . . . . . . . 107 Electrophilic addition . . . . . . . . . . . . . . . . . . . . . . . 108 M. H addition/hydrogenation . . . . . . . . . . . . . . . . . . . 109 M. C addition . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Coupling reactions with unsaturated substrates . . . . . . . . . . . . 110 Alkyne scission/metathesis/polymerization . . . . . . . . . . . . . . 114 Cluster substitution/expansion . . . . . . . . . . . . . . . . . . . 115 Demetalation . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Nucleophilic addition to mononuclear q 3-propargylium-M complexes . . 116 Reactions at the complexed acetylenic C . X bond . . . . . . . . . . . 116 Alkyne-vinylidene isomerization . . . . . . . . . . . . . . . . . . 116 Reactions of complexed terminal alkynes with base . . . . . . . . . . 117 Reactions at the propargylic (a) carbon . . . . . . . . . . . . . . . 118 Alkyne/allene isomerization . . . . . . . . . . . . . . . . . . . . 118 Reactions of dinuclear propargylium complexes with nucleophiles . . . . 118 General reaction features . . . . . . . . . . . . . . . . . . . . . 118 Proton loss/elimination . . . . . . . . . . . . . . . . . . . . . . 120 Coupling with noncarbon nucleophiles . . . . . . . . . . . . . . . . 121 Coupling with carbon nucleophiles . . . . . . . . . . . . . . . . . 122 Reactions remote from the complexed triple bond . . . . . . . . . . . 128 Reaction summary . . . . . . . . . . . . . . . . . . . . . . . . 128
4.5
Special applications of metal-alkyne complexes .
4.6 4.6.1
Selected experimental procedures . . . . . . . . . . . . . . . . . . 130 p[(q q 2-l-Methyl-2-propynylium)dicobalthexacarbonyl]tetrafluoroborate
’,
(126)
. . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.6.2 4.6.3
2-(l-Methyl-2-propynyl)cyclohexanone(127) . . . . . . . . . . . . . p-[q q 2.dl.3,4-Diphenyl.1,5.cyclooctadiyne].bi s.hexacarbony1dicob.t (128) . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
Organometallic Cycloaddition Reactions of Acetylenes
’,
128
130 131 131 132
Joseph A . Casalnuovq Neil E. Schore
. . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1
Introduction
5.2 5.2.1 5.2.2 5.2.3 5.2.4 5.2.5 5.2.6 5.2.7
Cycloadditions of acetylenes with Fischer carbenes . . . . . . . . . . Naphthols - the Dtitz reaction . . . . . . . . . . . . . . . . . . Indenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cyclobutenones . . . . . . . . . . . . . . . . . . . . . . . . . Cyclopentenones . . . . . . . . . . . . . . . . . . . . . . . . . Cycloheptadienones . . . . . . . . . . . . . . . . . . . . . . . . Cyclopropanes . . . . . . . . . . . . . . . . . . . . . . . . . . Heterocyclic ring systems . . . . . . . . . . . . . . . . . . . . .
139
. .
139 140 147 149 150 151 151 153
Contents
5.3 5.3.1 5.3.2 5.3.3
The Pauson-Khand reaction: cycloadditions of olefins. acetylenes. and CO Background and mechanism . . . . . . . . . . . . . . . . . . . . Intermolecular Pauson-Khand reaction . . . . . . . . . . . . . . . Intramolecular Pauson-Khand reaction . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
Phosphaalkynes . Starting Point for the Synthesis of Phosphorus-Carbon Cage Compounds
XI
154 155 157 161 167
Manfred Regitz. A . Hoffmann. L! BergstraJer
6.1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2
Syntheses of phosphaalkynes
. . . . . . . . . . . . . . . . . . .
6.3
Reactivity of phosphaalkynes
. . . . . . . . . . . . . . . . . . . 175
6.4
The history of phosphorus-carbon cage compounds from phosphaalkynes
176
6.5 6.5.1 6.5.1.1 6.5.1.2 6.5.1.3 6.5.1.4
177 177 177 178 180
6.5.1.5 6.5.1.6 6.5.2 6.5.2.1 6.5.2.2 6.5.2.3 6.5.3 6.5.3.1 6.5.3.2 6.5.3.3 6.5.3.4
Synthesis of phosphorus-carbon cage compounds . . . . . . . . . . . Construction by cycloaddition reactions . . . . . . . . . . . . . . . Diphosphatetracyclodecenes . . . . . . . . . . . . . . . . . . . . Phosphaprismanes and phosphabenzvalenes . . . . . . . . . . . . . Diphosphatricyclooctenes . . . . . . . . . . . . . . . . . . . . . Diphosphatetracycloundecadienones and oxadiphosphapentacyclononadecapentaenones (the tropone reaction of phosphaalkynes) . . . . . . . . Diphosphirenes as intermediates for phosphorus-carbon cage compounds Thermal cyclotetramerization . . . . . . . . . . . . . . . . . . . Construction by extrusion of Cp2Zr from phosphaalkyne dimer complexes Cp2Zr-phosphaalkyne dimer complexes . . . . . . . . . . . . . . . Tetraphosphacubanes and isomeric cage compounds . . . . . . . . . P-functionalization of the tetraphosphacubane system . . . . . . . . . Cyclooligomerization with the aid of Lewis acids . . . . . . . . . . . Spirocyclotrimerization . . . . . . . . . . . . . . . . . . . . . . Phosphaalkyne tetramers from the spirocyclotrimer 71a . . . . . . Hexaphosphapentaprismane from the spirocyclotrimer 71a . . . . . . . Phosphorus-carbon-aluminum cage compounds . . . . . . . . . . .
6.6
Outlook
6.7 6.7.1 6.7.2 6.7.3
Experimental procedures . . . . . . . . . . . . . . . . . . . . . 2,2.Dimethyl. l.(trimethylsiloxy)propy~dene(trimethylsilyl)phosphane (10a) (2.2.Dimethylpropylidyne)phosphane (9 a) . . . . . . . . . . . . . . Bis(q 5.cyclopentadienyl)(2,4.di.ter t. butyl-1,3-diphosphabicyclo[l.l.0]butan-2,4-diyl)zirconium (59a) . . . . . 2,4.6,8.Tetra.tert.butyl.l.3,5, 7.tetraphosphapentacyclo [4.2.0.02!'.O3.* .04.1' octane (53a) . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.7.4
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
173 174
182 183 184 185 186 186 188 189 189 190 192 194 195 196 196 196 197 197
XI1 6.7.5 6.7.6 6.7.7 6.7.8
7
Contents
2,4,6-Tri-tert-butyl-1,5-diphospha-3-phosphoniaspiro[3.4]hexa-1,4-diene-6-trichloroaluminate (71a) . . . . . . . . . . . . . . . . . . . . . . 197 2,5,6,8-Tetra-tert-butyl-l,3,4,7-tetraphosphatetracyclo[3.3.0.02~4.03~6]o~t-7_ene (76) 198 1,4,6.Tri.tert.butyl.2,5,7,7,8,8.hexaethyl.5,8.dialuminato.3.phosph a. 2,7-diphosphoniatetracyclo[3.3.0.~~4.03~6]~ctane (80) . . . . . . . . . . 198 2,5,7,9.Tetra.tert.butyl.3,3,4.triethyl.4.aluminato.3,6,8.triphospha.l.pho s. phoniatetracyclo[4.2.1.0'~5.0479]octane (81) . . . . . . . . . . . . . . 198 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
The Enediyne Antibiotics
K . C. Nicolaou. Adrian L. Smith
...........................
7.1
Introduction
7.2 7.2.1 7.2.2
The aromaticity era . . . . . . . . . . . . . . . . . . . . . . . . The cycloaromatization of conjugated polyenyne systems . . . . . . . . Application to the synthesis of aromatic systems . . . . . . . . . . . .
207 208 212 216 217 221 221 223 224 224 224 224 226 238
7.4.2.4 7.4.3 7.4.4
Theoretical and synthetic studies on the enediyne antibiotics . . . . . . Neocarzinostatin chromophore model systems . . . . . . . . . . . . Theoretical considerations . . . . . . . . . . . . . . . . . . . . . Synthetic studies . . . . . . . . . . . . . . . . . . . . . . . . . Calicheamicin/esperamicin theoretical and synthetic studies . . . . . . Synthetic and theoretical studies on the Bergman cycloaromatization of cyclic enediynes . . . . . . . . . . . . . . . . . . . . . . . . . Synthetic approaches to the calicheamicin aglycone . . . . . . . . . . Synthetic approaches to the calicheamicin/esperamicin carbohydrate fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total synthesis of calicheamicin y! . . . . . . . . . . . . . . . . . Dynemicin synthetic studies . . . . . . . . . . . . . . . . . . . . The chromoprotein enediyne antibiotics . . . . . . . . . . . . . . .
7.5
Medical applications of the enediyne antibiotics
7.4 7.4.1 7.4.1.1 7.4.1.2 7.4.2 7.4.2.1 7.4.2.2 7.4.2.3
7.6
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
205 205 206
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The discovery of the enediyne antibiotics 7.3 7.3.1 Neocarzinostatin . . . . . . . . . . . 7.3.2 The calicheamicins . . . . . . . . . . The esperamicins . . . . . . . . . . . 7.3.3 The dynemicins . . . . . . . . . . . 7.3.4 The chromoprotein enediyne antibiotics . 7.3.5 7.3.5.1 Kedarcidin . . . . . . . . . . . . . 7.3.5.2 C-1027 . . . . . . . . . . . . . . . 7.3.5.3 Maduropeptin . . . . . . . . . . . .
. . . . . . . . .
203
238 241 249 258 261 273
. . . . . . . . . . . . 273 Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . 274 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Contents
8
XI11
Cyclic Alkynes: Preparation and Properties
Rolf Gleiter, Roland Merger
8.1
Introduction . . . . . . . . . . . . . . . . . . .
285
8.2 8.2.1 8.2.1.1
286 286
8.2.1.2 8.2.2 8.2.2.1 8.2.2.2 8.2.2.3 8.2.3
Synthesis of cyclic acetylenes . . . . . . . . . . . . . . . . . . . Cyclic alkynes from ring-closure reactions . . . . . . . . . . . . . . Using acetylenic reactivity: nucleophilic substitution with metal acetylides and related reactions . . . . . . . . . . . . . . . . . . . . . . . Employing propargylic cations. anions. and radicals . . . . . . . . . Cyclic alkynes from elimination reactions . . . . . . . . . . . . . . 1.2.Elimination . . . . . . . . . . . . . . . . . . . . . . . . . Cycloelimination reactions . . . . . . . . . . . . . . . . . . . . . Ring contraction . . . . . . . . . . . . . . . . . . . . . . . . . Ring-enlargement reactions . . . . . . . . . . . . . . . . . . . .
8.3 8.3.1 8.3.2
Structural and spectroscopic properties . . . . . . . . . . . . . . . . Structures of cyclic mono- and dialkynes . . . . . . . . . . . . . . Photoelectron spectra of cyclic diacetylenes . . . . . . . . . . . . . .
296 296 301
8.4 8.4.1 8.4.2 8.4.3 8.4.3.1 8.4.3.2 8.4.3.3
Organic reactions of cyclic alkynes . . . . . . . . . . . . . . . . . Rearrangement of cyclic alkynes . . . . . . . . . . . . . . . . . . Transannular reactions . . . . . . . . . . . . . . . . . . . . . . Addition reactions of cyclic alkynes . . . . . . . . . . . . . . . . . Homonuclear addition reactions . . . . . . . . . . . . . . . . . . Heteronuclear addition reactions . . . . . . . . . . . . . . . . . . Cycloaddition reactions . . . . . . . . . . . . . . . . . . . . . .
303 303 305 308 308 309 309
8.5
Reactions of cyclic alkynes with metal compounds
. . . . . . . . . . .
311
8.6
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.7 8.7.1 8.7.1.1 8.7.1.2 8.7.1.3 8.7.2 8.7.2.1
314 Experimental procedures . . . . . . . . . . . . . . . . . . . . . Preparation of cyclic dialkynes of ring size Cl2.Cl. . . . . . . . . . 314 314 General procedure . . . . . . . . . . . . . . . . . . . . . . . . 314 1.7.Cyclododecadiyne (3) . . . . . . . . . . . . . . . . . . . . . l&Cyclotetradecadiyne (120) . . . . . . . . . . . . . . . . . . . 314 General procedure for Dewar benzenes 181 and 182 . . . . . . . . . 315 Dimethyl tetracyclo[l2.2.0.0'~7.08~14]hexadeca-7,15-diene-15,16-dicarboxylate 315 (182; n = 5 ) . . . . . . . . . . . . . . . . . . . . . . . . . . . Dimethyl tetracyclo[7.5.2.0.02~8]hexadeca-2,15-diene-l5,16-dicarboxylate (181; 315 n=5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cyclonon-2-ynone (91) and bicyclo[6.l.0]non-l(8)-en-9-one(92) . . . . . 315 316 References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8.7.2.2 8.7.3
286 288 292 292 293 294 295
314
XIV 9
Contents
Macrocyclic Homoconjugated Polyacetylenes
Lawrence T Scott. Mark J. Cooney 9.1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . .
321
9.2
Pericyclynes . . . . . . . . . . . . . . . . . . . . . . . . . . .
322
9.3
“Exploded” pericyclynes . . . . . . . . . . . . . . . . . . . . .
330
9.4
Homoconjugated mixed polyalkyne/diyne macrocycles . . . . . . . . .
337
9.5
Heterocyclic cognates of pericyclynes
. . . . . . . . . . . . . . . .
340
9.6 9.6.1 9.6.2
Experimental procedures . . . . . . . . . . . . . . . . . . . . . Conversion of a methyl ketone to a terminal acetylene (28-30) . . . . Conversion of a terminal acetylene to a bromoalkyne using tosyl bromide
345 345
(30-,50)
347
9.6.3 9.6.4 9.6.5 9.6.6
10
............................
Preparation of a 1.3.diyne by cross-coupling of a preformed copper acetylide with a bromoalkyne - 2 : 1 example (49 + 50 + 51) . . . . . . . Oxidative cyclization of a long-chain a.o.diyne (53 + 44) . . . . . . . Coupling of a terminal acetylene with a tertiary propargylic chloride 2: 1 example (47 + 69) . . . . . . . . . . . . . . . . . . . . . . Conversion of a 2,2.dibromovinyl compound to a bromoalkyne - two-fold example (75 + 74) . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
347 348 348 349 349
Polyacetylene
Eric J. Ginsburg. Christopher B. Gorman. Robert H . Grubbs
. . . . . . . . . . . . . . . . . . . . . . . . . . .
10.1
Introduction
10.2 10.2.1 10.2.1.1 10.2.1.2 10.2.2 10.2.2.1 10.2.2.2 10.2.3
Syntheses and properties . . . . . . . . . . . Routes from alkynes . . . . . . . . . . . . Acetylene polymerization . . . . . . . . . . Polymerization of substituted alkynes . . . . Routes from alkene precursors . . . . . . . Nonmetathetic routes . . . . . . . . . . . . Routes using olefin metathesis . . . . . . . Ring-opening of cyclooctatetraene . . . . . .
10.3 10.4 10.4.1 10.4.2 10.4.3
........... ........... ........... . . . . . . . . . . . . . . . . . . . . . . . . ........... . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental procedures . . . . . . . . . . . . . . . . . . . . . Synthesis of substituted polycyclooctatetraenes . . . . . . . . . . . . Cis/?runsisomerization of soluble polycyclooctatetraenes . . . . . . . . A precursor route to polyacetylene . . . . . . . . . . . . . . . . . .
353 358 358 358 359 363 363 366 368 376 376 376 377 377
Contents
10.4.3.1 10.4.3.2 10.4.3.3
11
XV
Synthesis of poly(diethy1 7-oxabicyclo[2.2.l]hepta-2,5-diene-2,3-dicarboxylate) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 Solid-state production of polyacetylene from poly(diethy1 7-oxabicyclo [2.2.l]hepta-2,5-diene-2,3-dicarboxylate). . . . . . . . . . . . . . . 378 Solution production of polyacetylene from poly(diethy1 7-oxabicyclo[2.2.1] hepta.2.5.diene.2,3.dicarboxylate) . . . . . . . . . . . . . . . . . . 378 379 References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acetylenic Compounds as Building Blocks for High-Spin Molecules and Molecular Assemblies
Hiizu Iwamura. Kenji Matsuda
...........................
11.1
Introduction
11.2 11.2.1 11.2.2 11.2.3
Alkynyl compounds carrying unpaired electrons . . . . . . . . . . . Alkynyl compounds carrying unpaired electrons in remote substituents . . Alkynes bonded to paramagnetic transition metals . . . . . . . . . . . 2-Propynylidenes . . . . . . . . . . . . . . . . . . . . . . . . .
385 385 387 389
11.3 11.3.1 11.3.2
Molecular crystals of organic free radicals that carry alkynyl substituents . What makes acetylenic compounds unique in assembling their molecules? Guiding principles on aligning electron spins in parallel between two neighboring molecules . . . . . . . . . . . . . . . . . . . . . . . . . Crystals of antiferromagnetic 1.3.butadiyne and ferromagnetic 1.3.5.hex a. triyne both carrying 4-chloro-3-(N-tert-butyl-N-oxyamino)phenyl as a stable free-radical substituent . . . . . . . . . . . . . . . . . . . . . .
391 391
11.3.3
11.4 11.4.1 11.4.2 11.4.3 11.4.3.1 11.4.3.2 11.5
385
392
393
Spin alignments in poly(phenylacety1enes)and poly(1. 3.butadiynes) . . . . Natural spins detected during the solid-state polymerization of 1.3.but a. diynes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Topological control of the high-spin vs. low-spin ground states of A-conjugated diradicals and dicarbenes . . . . . . . . . . . . . . . . . . Attempts at introducing stoichiometric amounts of spins in poly(pheny1acetylenes) and poly(phenyldiacety1enes) . . . . . . . . . . . . . . . . Poly(phenylacety1enes) . . . . . . . . . . . . . . . . . . . . . . Poly(phenyldiacety1enes) . . . . . . . . . . . . . . . . . . . . . .
400 402
Cyclotrimerization reaction of benzoylacetylenes in the presence of a secondary m i n e . . . . . . . . . . . . . . . . . . . . . . . . . . .
403
..........................
11.6
Conclusions
11.7 11.7.1 11.7.2 11.7.2.1
Experimental procedures . . . . . . . . . . . . . . . . . . . . Characterization of magnetic properties . . . . . . . . . . . . . Synthesis of dendritic “Starburst” dodecaketone 49 . . . . . . . 1.(3,5.Dibenzoylbenzoyl).3.(3.trimethylsilyl. 2-propynoy1)benzene (52)
395 395 398 400
409
. . .
409
409 410 410
XVI
Contents
11.7.2.2 11.7.2.3
1.(3,5.Dibenzoylbenzoy1).3.( 2.propynoyl)benzene (56) . . . . . . . . . 1,3,5.Tris[3.(3, 5.dibenzoylbenzoyl)benzoyl]benzene (49) . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
Acetylenes in Nanostructures
410 410 411
James K . Young. Jeffrey S. Moore 12.1 12.1.1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Structural parameters of phenylacetylenes . . . . . . . . . . . . . .
415 416
12.2 12.2.1 12.2.1.1 12.2.1.2 12.2.1.3
Phenylacetylene dendrimers . . . . . . . . . . . . . . . . . . . . Synthetic considerations for phenylacetylene dendrimer construction . . . The divergent and convergent synthetic approaches . . . . . . . . . . Convergent synthesis of phenylacetylene dendrimers . . . . . . . . . Effect of varying focal point functionality on the convergent synthesis of phenylacetylene dendrimers . . . . . . . . . . . . . . . . . . . . Synthesis of dendrimers by repetition of monomer enlargement (SYNDROME method) . . . . . . . . . . . . . . . . . . . . . . . . “Double exponential” dendrimer growth . . . . . . . . . . . . . .
418 419 420 421
Phenylacetylene macrocycles . . . . . . . . . . . . . . . . . . . . Phenylacetylene macrocyclic framework . . . . . . . . . . . . . . . Synthetic considerations for phenylacetylene macrocycle construction . . . The double cyclization of branched phenylacetylene oligomers . . . . . Tandem bimolecular coupling followed by intramolecular cyclization to form a foldable phenylacetylene macrotetracycle . . . . . . . . . . . Synthesis of sequence-specific phenylacetylene oligomers and dendrimers on an insoluble solid support . . . . . . . . . . . . . . . . . . .
426 428 430 430
12.2.1.4 12.2.1.5 12.3 12.3.1 12.3.2 12.3.2.1 12.3.2.2 12.4
. . . . . . . . . . . . . . . . . . . . . . . . . . .
12.5
Conclusions
12.6 12.6.1
Experimental procedures . . . . . . . . . . . . . . . . . . . . . 48-Cascade: benzene[3.1.3. 51 :(5.ethynyl.l. 3.phenylene)G. 5-ethynyl1,3.di(ter t.butyl)benzene (8) . . . . . . . . . . . . . . . . . . . . General procedure for double cyclization . . . . . . . . . . . . . . Sample preparation for mass spectrometry . . . . . . . . . . . . . . Procedures for solid-supported phenylacetylene chemistry . . . . . . . General procedure A: Pd(0)-catalyzed coupling reactions (except for trimethylsilylacetylene) . . . . . . . . . . . . . . . . . . . . . . . . General procedure B: Pd(0)-catalyzed coupling with trimethylsilylacetylene General procedure C : trimethylsilyl deprotection . . . . . . . . . . . General procedure D: liberation of the oligomeric sequence from the support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Peptide linkage to aminomethylated polystyrene (26) . . . . . . . . . Ether linkage to chloromethylated polystyrene (28) . . . . . . . . . . .
12.6.2 12.6.3 12.6.4 12.6.4.1 12.6.4.2 12.6.4.3 12.6.4.4 12.6.5 12.6.6
423 423 424
431 433 436 437 437 437 437 438 438 438 438 439 439 439
Contents
12.6.7 12.6.8
Propylaminomethylated polystyrene (29) . . . . . . . . . . . . . . Direct triazene linkage to propylaminomethylated polystyrene (31). . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
Oligoacetylenes
XVII
439 440 441
Franqois Diederich
13.1 13.2 13.2.1 13.2.2 13.2.3 13.3 13.3.1 13.3.2 13.3.3 13.3.4 13.3.5
. . . . . . . . . . . . . . . . . . . . . . . . . . . Synthetic approaches to the cyclocarbons . . . . . . . . . . . . . . . The retro-Diels-Alder route to cyclo.C, . . . . . . . . . . . . . . . Introduction
443
The 3.cyclobutene.l. 2.dione route to the cyclocarbons The transition metal complex route to cycIo-Cl8 . .
443 445 446 448
. . . . . . . . . . . . . . . . . .
Tetraethynylethenes, fully cross-conjugated n-electron chromophores. and other perethynylated molecules . . . . . . . . . . . . . . . . . . . Synthesis of tetraethynylethene (20) and geminally bisdeprotected derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synthesis of monodeprotected tetraethynylethenes . . . . . . . . . . . Synthesis of trans-bis(triisopropylsily1)-protected and trans-bisdeprotected tetraethynylethenes . . . . . . . . . . . . . . . . . . . . . . . . Synthesis of cis-bisdeprotected tetraethynylethenes . . . . . . . . . . . Other perethynylated compounds as potential monomers for carbon networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
449 449 451 451 452 453
13.4.1 13.4.2
Perethynylated dehydroannulenes and expanded radialenes: large carbon cores on the way to all-carbon sheets . . . . . . . . . . . . . . . . Perethynylated dehydroannulenes . . . . . . . . . . . . . . . . . . Perethynylated expanded radialenes . . . . . . . . . . . . . . . . .
456 456 459
13.5 13.5.1 13.5.2
Molecular wires : from polytriacetylenes to carbyne . . . . . . . . . . Linear polyynes: short oligomers of elusive carbyne . . . . . . . . . Stable soluble conjugated carbon rods with a polytriacetylene backbone .
461 461 463
13.6
Conclusions
13.7 13.7.1 13.7.2
Experimental procedures
13.4
13.7.3 13.7.4 13.7.5 13.7.6
..........................
464
464 . . . . . . . . . . . . . . . . . . . . . 2.dione (12f) . . . . . 464
3,4.Bis[triisopropylsilyl)ethynyl].3.cyclobutene.l.
Oxidative Hay coupling of 14 to the cyclobutene-fused dehydroannulenes 15-17 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.Dibromomethylene.l,5.bis(trimethylsily~).l, 4.pentadiyne (23) . . . . . (E).1,2.Diethynyl.l, 2.bis[(triisopropylsilyl)ethynyl]ethene (30a) . . . . . Eglinton-Glaser coupling of 54 to the expanded radialenes 51 and 53 . . . General procedure for solution-spray flash vacuum pyrolysis (SS-FVP) . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
465 465 466 466 466 469
This Page Intentionally Left Blank
List of Contributors
U. BergstraDe Fachbereich Chemie der Universitat Erwin-Schrodinger-Stral3e D-67663 Kaiserslautern Germany
Christopher B. Gorman Department of Chemistry North Carolina State University Raleigh NC 27695, USA
Joseph A. Casalnuovo Department of Chemistry California Polytechnic University Pomona CA 91768, USA
Robert H. Grubbs Arnold and Mabel Beckman Laboratories of Chemical Synthesis California Institute of Technology Pasadena CA 91125, USA
Mark J. Cooney Department of Chemistry Merkert Chemistry Center Boston College Chestnut Hill MA 02167, USA
A. Hoffmann Fachbereich Chemie der UniversitSlt Erwin-Schr6dinger-Stral3e D-67663 Kaiserslautern Germany
Francois Diederich Laboratorium fur Organische Chemie ETH Zentrum UniversitatsstraRe 16 CH-8092 Zurich Switzerland Eric J. Ginsburg Research Laboratories Eastman Kodak Company Rochester NY 14650, USA Rolf Gleiter Organisch-Chemisches Institut der Universitat Heidelberg Im Neuenheimer Feld 270 D-69120 Heidelberg Germany
Henning Hopf Institute of Organic Chemistry Technical University of Braunschweig Hagennng 30 D-38106 Braunschweig Germany
K. N. Houk Department of Chemistry and Biochemistry University of California Los Angeles CA 90024, USA Hiizu Iwamura Department of Chemistry Graduate School of Science The University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo 113 Japan
XX
List of Contributors
Yi Li Bristol-Myers Squibb Pharmaceutical Research Institute 5 Research Parkway P.O. Box 5100 Wallingford, CT 06492-7660 USA Kenji Matsuda Department of Chemistry Graduate School of Science The University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo 113 Japan Gagik G. Melikyan Department of Chemistry and Biochemistry University of Oklahoma Norman OK 73019, USA Roland Merger Organisch-Chemisches Institut der Universitat Heidelberg Im Neuenheimer Feld 270 D-69120 Heidelberg and Farbenlabor der BASE AG D-67056 Ludwigshafen Germany Jeffrey S. Moore Departments of Chemistry and Materials Science & Engineering Roger Adams Laboratory
K. C. Nicolaou Department of Chemistry The Scripps Research Institute La Jolla, CA 92037 and Department of Chemistry University of California San Diego, CA 92093 USA
Dietmar A. Plattner Department of Chemistry and Biochemistry University of California Los Angeles CA 90024, USA Manfred Regitz Fachbereich Chemie der Universitat Erwin-Schrddinger-Strane D-67663 Kaiserslautern Germany Neil E. Schore Department of Chemistry University of California Davis CA 95616, USA Lawrence T. Scott Department of Chemistry Merkert Chemistry Center Boston College Chestnut Hill MA 02167, USA
Box 55 600 S. Mathews Urbana IL 61801, USA
Adrian L. Smith Merck Sharpe & Dohme Research Laboratories Terlings Park Harlow Essex CM20 2QR, UK
Kenneth M. Nicholas Department of Chemistry and Biochemistry University of Oklahoma Norman OK 73019, USA
Peter J. Stang Department of Chemistry University of Utah Salt Lake City UT 84112, USA
List of Contributors
Bernhard Witulski' Institute of Organic Chemistry Technical University of Braunschweig Hagenring 30 D-38106 Braunschweig Germany
James K. Young Department of Chemistry Roger Adams Laboratory Box 55 600 S. Mathews Urbana IL 61801, USA
' Present address: Department of Chemistry, Stanford University, Stanford, CA 94305, USA
XXI
1 Modern Computational and Theoretical Aspects of Acetylene Chemistry Dietrnar A . Plattner, YiLi, K.N. Houk
1.1 Introduction Few organic molecules have been the object of more intensive physicochemical and theoretical scrutiny in recent years than acetylene. The focus of modern experimental, theoretical, and computational studies has been the characterization and elucidation of transition states and reactive intermediates, reaction potential energy surfaces and reaction dynamics of acetylene. The reason for this attention is clear. Acetylene is a simple polyatomic molecule only slightly more complex than a diatomic molecule, yet it has a variety of uses and undergoes a host of reactions like those of polyfunctional molecules. The rich and diverse chemical properties are amenable to high-level computational treatment, state-of-the-art spectroscopic measurements, and detailed theoretical interpretations of experimental data. Although acetylene is one of the most common molecules, its most fundamental properties such as bond strength are still subject to refinement, both experimentally and computationally. The rapid growing number of studies on the structures and stabilities of carbon clusters and rods have renewed interest in the bonding character of the acetylenic bond. Concepts of bonding in acetylene continue to evolve, and the understanding of complex varities of acetylenes will aid in the design of new molecules and materials. In this chapter, we review some recent developments in the theoretical and computational aspects of acetylenes. There are several detailed reviews covering various aspects of the early work [l, 21. It will become self-evident in this review that modern experimental and computational studies of acetylene constitute a paradigm for the rivalry and interplay between theory and experiment. As the theoretical treatments become increasingly sophisticated, and as the experimental design becomes more and more ingenious and precise, the better is our understanding.
1.2 Electronic Structures of Acetylene and Monoacetylenes Historically, the application of molecular orbital theory to the electronic structures of isoelectronic 14-electron molecules such as acetylene, HCN, N,, and 0, was an excellent pioneering demonstration of the value of quantum chemistry. Within the framework of molecular orbital theory, the C - C bond in acetylene is a triple bond involving one a-bond, and two orthogonal n-bonds. The a-bond is formed by two sp-hybrid orbitals from each carbon, and the two nbonds are formed from the perpendicular p-orbitals. Alternatively, the so-called “bent” or “banana” bonds have been invoked to describe the multiple C-C bonds in acetylene (Fig. 1-1) [3-51. This creates a conceptual dilemma, though one bonding model can be transformed to the other by appropriate linear combinations. It is now realized that both
2
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
models are useful for describing various aspects of bonding and reactivity [6],but neither approach is perfect in describing the electronic structure of acetylene. Due to the electron correlation effects, a multiconfiguration wavefunction is necessary to describe fully the electronic structure of a molecule. The separation of 8- and n-orbitals in the molecular orbital treatment is an approximation, and thus has limitations. A generalized valence-bond theory was developed by the introduction of Pauling’s resonance theory, which took both models into the consideration [7]. For acetylene, the descriptions of 6-71bonds or “banana” bonds comprise merely one configuration that contributes to the multiconfiguration wavefunction. n n
n
H-CGC-H
H-Cq-H
a-n bond model
bent bond model
u
Figure 1-1 The o-n and bent bond models for bonding in acetylene.
To address the question of which single-configuration bond description is a better starting point for the treatment of correlation effects, Karadakov et al. [8] examined the spin-coupled wavefunctions generated, respectively, from the Hartree-Fock molecular orbitals and the generalized valence-bond wavefunction with perfect-pairing and strong-orthogonality constraints. The results using these wavefunctions were than compared with the calculations using a multiconfiguration wavefunction consisting of a complete-active space self-consistent field. From an energetic point of view, they found that both approaches were equally good for the treatment of correlation effects beyond the one-configuration approximation. The spincoupled wavefunction from o-n orbitals recovers 63% of the CASSCF correlation energy, as compared with a 66% recovery of correlation energy using the equivalent bent orbitals. Other computational studies demonstrated the superiority of banana bonds for a variety of systems containing multiple bonds [9- 121. The superiority of one bond description over the other may depend upon the extent of conjugation of the multiple bonds [13]. In spite of the shortcomings of the single-configuration approach, the 6-71 concept has played an indispensable role in bridging theoretical understanding and chemical relevance. The HOMO-LUMO interactions in the frontier molecular orbital theory, the orbital energies either calculated by theory or measured from ionization potentials, the electron distribution and density of n-orbitals and bond orders, all have been used to understand and predict the molecular structural features, chemical stabilities, reactivities, regioselectivities, and stereoselectivities of acetylenes. Classical chemical concepts such as bond orders, the HOMO-LUMO energies, and electron densities have also been defined quantitatively and have been calculated for acetylene numerically from ab-initio calculations [14-201.
1.2 Electronic Structures of Acetylene and Monoacetylenes
3
1.2.1 Ground-state Potential Energy Surfaces On the singlet potential energy surface, acetylene (HC E CH) may undergo isomerization to vinylidene (H,C=C:). Whether singlet vinylidene exists as a bound intermediate has been the subject of extensive studies, both experimentally and theoretically [l, 21 -351. The simplest unsaturated carbene has been proposed to be involved in many chemical reactions, and is of great value in preparative organic chemistry [36-391. Because vinylidene is highly reactive, there has been limited direct experimental characterization of this species, and much debate about whether vinylidene is a minimum on the potential energy surface or a transition state for the degenerate hydrogen shift in acetylene. On the other hand, numerous computational studies have only recently provided a clear consensus on the classical barrier height for the isomerization process [26]. The lowest singlet state of vinylidene is an extremely shallow minimum on the potential energy surface. The best estimate of the classical barrier of isomerization to acetylene made by Gallo et al. is 3 kcal/mol (1 kcal = 4.184 kJ) using large basis sets and the coupled cluster method including single and double excitations [26]. The energy of isomerization of acetylene is predicted to be 43 kcal/mol at the same level of theory. Although an artifact at the MP2 level was noted, calculations at the high Msller-Plesset perturbation levels also predicted a diminishingly small barrier for the vinylidene isomerization [29, 311. Such a small barrier of 2-4 kcal/mol also led to a prediction of a lifetime of about 1 ps for the ground-state vinylidene [30, 321. The first direct observation of singlet vinylidene came from a photodetachment experiment involving the vinylidene radical anion [23]. Ervin et al. studied in detail the photoelectron spectra of the vinylidene anion and observed the vibrational structure of vinylidene [22]. The observed 2 + 0 CH, rock transition (450 cm-') indicated that the singlet vinylidene is a minimum with a barrier to rearrangement of > 1.3 kcal/mol. Its lifetime was estimated from these experiments to be 0.04-0.2 ps. Chen et al. also observed vinylidene in the high-resolution stimulated emission pumping spectrum of acetylene [24]. Although they were unable to determine the barrier height for vinylidene isomerization, a value of
-
-
/
:c=c
-46 kcali'mol
H
/
\
H
-43 kcal,'mol
Scheme 1-1 Energetics of the vinylidene-acetylene rearrangement [25, 401.
4
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
44 kcal/mol was determined for the vinylidene-acetylene isomerization energy. This is in good agreement with the results of other measurements [21], and is consistent with the prediction by calculations. One surprising feature in the transition structure obtained at various levels of theory for the acetylene-vinylidene isomerization is the extent of hydrogen migration (Scheme 1-1). For a low-barrier, highly exothermic reaction like the vinylidene rearrangement, an early transition state is expected according to the Hammond postulate. In other words, the transition structure should resemble vinylidene rather than displaying the reaction progressed halfway in terms of hydrogen transfer. This contradiction was first observed by Dykstra and Schaefer [40],and was apparently not due to the level of theory used. Petersson et al. offered a plausible explanation [25]. They considered two distinct processes that are involved in the isomerization: one corresponding to the hydrogen transfer and the other corresponding to the conversion of the carbene lone-pair electron to the R bonding electrons. The hydrogen transfer process, which breaks one C - H bond but creates another, is nearly thermoneutral, forming a species which is essentially a twisted zwitterion. The transition state for such a thermoneutral reaction should be midway according to the Hammond postulate. The second part of vinylidene isomerization is the electron reorganization from the twisted zwitterion to form acetylene, a very exothermic process. Therefore, in terms of the C - C bond length in the transition structure, the transition state closely resembles vinylidene, obeying the Hammond postulate. Petersson et al. suggested that the Hammond postulate should be applied to the energetics of individual processes, not to the total energy directly. Besides the acetylene-vinylidene isomerization, other topological regions of the lowest singlet potential energy surface have been explored in a limited number of studies to date. The stimulated-emission pumping technique has been used to probe the potential energy surface up to 28000 cm-' [24, 41, 421. These studies indicated that acetylene at energy around 26500 cm-I undergoes the transition from the regular to the chaotic regime. Sibert and Mayrhofer carried out a variational calculation on highly excited vibrational states up to 8770 cm-I 1431. Binkley reported geometries and frequencies for two additional stationary points, bridged acetylene and planar bridged acetylene [33]. Halvick et al. investigated thoroughly the singlet acetylene energy surface up to 43000 cm-' using high level ab initio calculations [34]. They located eight stationary points and characterized the minimum energy paths connecting them. This information was then used to build a topologically consistent and complete configuration space, which included all three isomerization coordinates among acetylene, vinylidene, bridged acetylene, and planar bridged acetylene (Fig. 1-2). Finally, calculations of potential energy surfaces involving bond dissociation reactions of acetylene are highly demanding on the level of theory, and have often been used as the testing ground for the development of the latest theoretical methods. Recent examples include the G2 theory by Pople and co-workers [44,451, the coupled cluster methods [46], and the density functional theory [47-511. Several authors investigated in great detail the C -H bond dissociation [52-561, and the C-C bond dissociation as well [53, 57, 581. For the C - H bond dissociation of acetylene, high-level calculations, which range from 126 to 132 kcal/mol after zero-point energy correction, are in agreement with the upper end of the experimentally measured values. The C - C bond energy of acetylene is predicted to be 206 kcal/mol by the GVB method [%I, or 226 kcal/mol by G2 theory [45], as compared with 229 kcal/mol derived indirectly from experiments [21]. Table 1-1 summarizes the C - H and C - C bond dissociation energies obtained at various levels of theory and by experimental measurements.
1.2 Electronic Structures of Acetylene and Monoacetylenes
C2h
T -
7
..
5
H
I
\
, ,
H
/
/
c=c
4
H
TS(0); 2" saddle point(@]
Figure 1-2 Stationary points on the potential energy surface of C2H2. TS = transition structure.
Table 1-1 Calculated bond dissociation energies for acetylene (Do, kcal/mol)(a)
Method
HCC - H
HC = CH
Reference
G-1 G-2
133.4 133.4 129.7 131.1 129.9 126-132
226.9 226.3 206.3
WI
GVB-CCCVDZP DFT-LDA/DN DlT-LDA/DNP
Exptl. (a)
[451 [581
[511 [511 (228.8 f 0.7)
See text
1 kcal = 4.184 kJ.
1.2.2 Excited-state Potential Energy Surfaces The lowest triplet potential energy surface of acetylene has also been studied by experiments and theory [59-621. The lowest excited state of acetylene is a cis-bent triplet state which was predicted theoretically and confirmed experimentally [62, 631. Although subsequent experimental studies by Lisy and Klemperer cast some doubts on this conclusion [64], more recent studies have resolved the apparent contradiction between the two experimental findings [59]. Theoretical work by several groups also extended to the tmns-bent triplet acetylene and its isomerization to the cis-bent triplet state [62, 651. In the case of triplet vinylidene, the energy gap between the lowest and first excited triplet state was determined to be 15 kcal/mol 1221, in good agreement with the theoretical predications [32,40]. In contrast to the singlet vinylidene, there is a significant barrier of 54 kcal/mol predicted for the isomerization of the triplet vinylidene to the triplet acetylene [60], involving a nonplanar transition structure (Scheme 1-2). This is in agreement with the experimental evidence that the lifetime of the triplet vinylidene (> 0.4 ps) is much longer than that of the singlet state [66, 671.
-
6
1 Modern Computational and Theoretical Aspects of Acetylene Chemistry
c--c 1.312A
/ \
H
H
l.081A
H
L
ii3B2 vinylidene
LHCCH = 115'
b3B, acetylene
r H
trans-bent
H b3B, acetylene
127"
184"
-
128"
C---
1
1.327A
1302acetylene
Scheme 1-2 Computed structures of triplet C,H2 and transition structures for interconversions.
In contrast to the triplet excited state, singlet excited states of acetylene are less well characterized. Recent spectroscopic studies showed there is a strong singlet- triplet coupling in the singlet excited state [68-701. This led to the speculation that the lowest singlet excited state lies close in energy to one of the transition states on the triplet potential energy surface of acetylene [60].Several computational studies have been reported on the singlet excited states of acetylene [33, 71-75].
1.2.3 Radical Ions Ionization of acetylene gives a radical cation, for which many studies have been reported in the literature. The radical cation is a Renner-Teller molecule, and is predicted to have a degenerate X'll, electronic ground state [76]. It was observed experimentally by mass spectrometry [67, 771. The structures and energies of the C,H, radical cation have been studied in detail by several groups [78-811. The isomerization barrier from vinylidene cation to the more stable acetylene cation is predicted to be 10 kcal/mol both at the UMP2/6-311G** level reported by Baker [SO] and at the CISD(+Q)/DZP level reported by Hamilton and Schaefer [79]. In addition, theoretical considerations led Ramasesha and Sinha to suggest that stacked acetylenic radical ions are prime candidates for the observation of organic ferromagnetism, because of their stable high-spin ground state [82]. The acetylene anion radical undergoes autodetachment of the electron, but the vinylidene anion can be generated easily [83]. Since the calculated isomerization barrier is 45 kcal/mol, the 'B, ground-state vinylidene anion radical is predicted to be stable with respect to the 1,2-hydrogen shift [30, 84, 851. As mentioned before, the vinylidene anion radical was used as the precursor for the generation of the singlet vinylidene in Lineberger's experimental studies.
-
-
1.3 Reactivities and Molecular Interactions of Acetylenes
7
1.3 Reactivities and Molecular Interactions of Acetylenes Theoretical and computational studies of the reactivities and molecular interactions of acetylene have exploded during recent years. Most studies are aimed at gaining theoretical understanding of the difference in the reactivities between the triple bond in alkynes and the double bond in alkenes. In the following section, we will summarize the reactivities of acetylene involved in pericyclic reactions, electrophilic reactions, and nucleophilic additions. Then, we will give a brief review of the studies probing molecular interactions of acetylene.
1.3.1 Pericyclic Reactions It has long been debated whether pericyclic reactions take place through concerted or stepwise mechanisms. Calculated reaction paths and transition structures enabled us to characterize theoretically the mechanisms of many of these reactions [86]. The concerted mechanism, as Woodward and Hoffmann predicted decades ago [87], is indeed the rule rather than the exception for most pericyclic reactions. In the cases of acetylene or alkynes, a stepwise reaction would involve intermediate vinyl radicals, in analogy with alkyl radicals in the reactions of alkenes. Furthermore, the strength of an acetylenic x-bond is different from that of an olefinic bond. Although the difference in the heat of hydrogenation of acetylene and ethylene indicates a weak n-bond in acetylene, Nicolaides and Borden recently suggested that the acetylenic n-bond is - 12 kcal/mol stronger than the x-bond in ethylene [88]. Thus, the stepwise mechanisms of additions to acetylenes involve a strong x-bond and may be less favorable than the concerted mechanism. The difference in the reactivities of acetylene in various pericyclic reactions as compared with that of ethylene was found in most cases to be due to the effects of the n-bond, which does not undergo bonding changes in the transition state. Cyclotrimerization of acetylene to form benzene is a thermally allowed and highly exothermic pericyclic reaction, but no such reaction occurs, because of its high reaction barrier. In 1979, Houk et al. performed a detailed theoretical analysis and offered an exlanation for the apparent high reaction barrier. The high activation energy was attributed to the unfavorable repulsions involving closed-shell electrons upon the approach of acetylene [89]. This study was reaffirmed years later by the calculations of Bach et al. (Scheme 1-3) [90].
r 3 HC-CH
-
H
*
-0
Scheme 1-3 The transition structure for the [2 + 2 + 21 cycloaddition of acetylene (RHF/6-31G*, Bach et
al. [90]).
Although the Diels- Alder reaction of acetylene received less theoretical attention than that of ethylene, several observations have recently been made about the reactivities of acetylene (Fig. 1-3). Coxon et al. reported an ab-initio computational study on the Diels-Alder reaction
8
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
of actylene with butadiene [91]. A concerted transition structure of C, symmetry was located at the RHF/6-31G* level, and single-point energy calculations at the MP2 level predicted an activation energy of 18 kcal/mol. For comparison, the activation energy for the Diels- Alder reaction of ethylene with butadiene was previously calculated at a similar level of theory to be 17 kcal/mol [92]. By analyzing the energies of individual molecular orbitals and the distortion energy of the reactants, these authors concluded that the major contributor to the activation energy of the Diels-Alder reaction of acetylene with butadiene was the increase in energy of the filled acetylene n-orbital that is not involved in bonding change. That is, the closed-shell repulsions are larger in the acetylene reaction than in the ethylene reaction. 1.393
1.396
[1.380]
Figure 1-3 Tkansition structures for the Diels-Alder reactions of butadiene and 2-azabutadiene with ethylene and acetylene. Bond lengths are given in A, [MP2/6-31G*], RHF/6-31G*, (RHF13-21G) [91-931.
Similarly, Gonzhlez and Houk predicted that the Diels- Alder reaction of acetylene with 2-azabutadiene is more than 2 kcal/mol higher in activation energy than the corresponding ethylene reaction [93]. In the same paper, they also investigated the substituent effects on the reactivities of alkene and alkyne multiple bonds, and the effect of Lewis acid catalysis on these reactions. Another reaction studied computationally was the Diels- Alder reaction of acetylene with a-pyrone [94]. The most extensively studied 1,3-dipolar cycloaddition reaction so far is the prototype reaction of acetylene with fulminic acid (Scheme 1-4). The early GVB calculations by Harcourt and Little attempted to resolve the controversy between Firestone’s stepwise biradical mechanism and the concerted mechanism for the 1,3-dipolar reaction [95]. Which mechanism
1.3 Reactivities and Molecular Interactions of Acetylenes
9
is lower in energy depends upon the level of theory [96-981. Multiconfiguration SCF calculations by Bernardi and co-workers led to the conclusion that the concerted mechanism was favored for the 1,3-dipolar cycloaddition of fulminic acid with acetylene [99- 1041. The HOMO-HOMO interactions of in-plane n-orbitals were also shown to contribute to the activation energy of this reaction [105].
L
. _ _2_ _ _ _ _
1.22 A
Scheme 1-4 Transition structure of the cycloaddition of acetylene with fulminic acid (MCSCF/4-31G, McDouall et al. [104]).
Another example demonstrating the difference in reactivity is the ozonolysis reactions of acetylene and ethylene. Ozonolysis of ethylene is a classical 1,3-dipolar cycloaddition reaction with an activation energy of 5 kcal/mol [106], whereas a larger activation energy of 11 kcal/mol was measured for the reaction of ozone with acetylene [107]. The 1,3-dipolar cycloaddition adduct, 1,2,3-trioxolene, has not been definitively observed as an intermediate involved in the acetylene ozonolysis. Nevertheless, according to the combined microwave and ab-initio calculation studies, the formation of similar van der Waals complexes in the course of ozonolysis has been established for both acetylene and ethylene [log]. Other pericyclic reactions of alkynes that have been studied computationally include the addition of singlet methylene to acetylene [109], the addition of carbon monosulfide to acetylene [110], the [2 + 21 dimerization [loo, 1111, and the dihydrogen transfer reaction between acetylene and ethylene [112, 1131. The Bergman cyclization of enediynes has gained a great deal of attention due to the recent isolation of a new class of antibiotics containing the enediyne moiety [114]. Various reactions of these molecules trigger the cyclization of the enediyne, and formation of reactive benzene-1,Cdiyl radicals. The reaction of the parent system, hex-3-ene-1,5-diyne, investigated experimentally for a number of simple systems by Bergman [MI, has now been studied with a number of theoretical techniques [116a-g]. Snyder used a blend of semiempirical (PRDDO - CI) and empirical calculations to study the parent reaction (Scheme 1-5) and those of a variety of substituted derivatives [116a-c]. At least four ab-initio CASSCF or CI studies of the parent reaction have been performed [116d-g]. The study by Kraka and Cremer is representative [116e]. They report the stationary point geometries shown in Scheme 1-5 for calculations performed with the CCSD(T) method - a type of CI. The calculated activation energy for cyclization is A P = 28.5 kcal/mol and the heat of reaction is AH,,, = 8.0 kcal/mol. Previous experimental estimates of these values gave 32 and 14 kcal/mol, respectively [115].
10
1 Modern Computational and Theoretical Aspects
of Acetylene Chemistry
XH1 1.220
Scheme 1-5 The Bergman reaction. Calculated geometries (CCSD(T)/6-31G(d,p)) are taken from [116e]. Bond lengths in A, bond angles in degrees.
The degenerate concerted dihydrogen exchange between acetylene and ethylene was found to have an activation barrier consistently 5 kcal/mol higher than the ethylene-ethane exchange reaction [112, 1131. The difference in the barrier heights may be attributed to a larger distortion energy and greater closed-shell repulsion in the transition structure of the acetylene-ethylene exchange reaction than in the ethylene-ethane reaction (Fig. 1-4). The lower reactivity of the enediyne than the enyne-allene in the Bergman cyclization could also be attributed in part to the four-electron repulsion between the in-plane n-bonds in the transition structure [116d].
Acety lene-Ethylene
Ethylene-Ethane
Figure 1-4 Transition structures of dihydrogen transfer reactions (MP2/6-31G*, McKee and Stanbury [112, 1131).
1.3.2 Electrophilic Reactions It is well known that alkynes are less reactive than alkenes in electrophilic reactions [117]. The theoretical explanation is not obvious [118]. Although some theoretical calculations were also reported on the reactions of acetylene with S+ [119], Si' [120] and rare metal ions (Sc+, Y+, and Ln+) [121, 1221, most studies are dealing with electrophiles such as a proton or carbonium ions. There are also computational studies on the protonation of the excited acetylene [123- 1251. There are numerous theoretical calculations on the protonation reaction of acetylene [2, 126- 1281. The focus is mostly on the protonation affinities and the structures of vinyl cations using a variety of methods and less on the dynamics of the addition. It is now known
1.3 Reactivities and Molecular Interactions of Acetylenes
11
that there is a profound effect of electron correlation in determining the structural preference of the classical versus the nonclassical, bridged, structure. The classical (unbridged) structure is favored at the HF level, whereas the nonclassical hydrogen-bridged structure is predicted to be 3 kcal/mol more stable than the acyclic form at the MP4/6-311G** level [2]. For protonation reactions, Nicolaides and Borden suggested that the smaller proton affinity of acetylene as compared with ethylene is due to a stronger n-bond that is broken in the former molecule, not to the poorer stability of the vinyl carbocation [88]. Calculations on electrophilic additions to acetylene were reported for a number of hydrocarbon cations, including C H + [129], C3H? [130, 1311, and phenylvinylium ion [132]. It was predicted that the cyclopropenylium cation, the most stable form of C3H;, forms an ion-molecule complex with acetylene but does not undergo further addition [131]. The linear propargyl cation, however, reacts with acetylene without an apparent barrier, to form many different C,H: isomers [130, 1311. Wang et al. computed the AM1 potential energy surfaces and carried out RRKM and microcanonical variational transition-state analysis for the rate of reaction of phenylvinylium ion with acetylene [132]. Ab-initio calculations on the protonation of diacetylene show that reaction takes place on a terminal carbon atom and converts a destabilizing antibonding interaction between the triple bonds in the neutral molecule into an attractive interaction, leading to a significant shortening of the C-C single bond in the protonated species [133].
1.3.3 Nucleophilic Additions In general, alkynes are more prone than alkenes to nucleophilic additions. For example, hydride addition to acetylene is predicted to be 26 kcal/mol more exothermic than the addition to ethylene [88]. This thermodynamic preference also carries over to the higher kinetic reactivity of acetylene toward nucleophiles [134]. Houk and co-workers noted the preference for the acetylenic bond to undergo a trans bending in the transition structure of hydride addition to acetylene, and suggested that the energy of such a deformation is less than trans pyramidalization of alkenes during the nucleophilic attack [134]. The trans bending of the alkyne also lowers the LUMO energy to a greater extent than the LUMO is lowered upon pyramidalization of an alkene [134]. Houk, Schleyer, and co-workers studied the nucleophilic additions of lithium hydride and methyllithium to acetylene and ethylene, respectively [135, 1361. The best estimate of the activation energy from Schleyer's study is 3 kcal/mol for the reaction of acetylene with lithium hydride. There is a negligible difference in the calculated activation energies of the acetylene-LiH and ethylene-LiH reactions, although acetylene appears to form a weaker I[complex with LiH than ethylene does. The competitive deprotonation processes were also considered by these and other workers [137].
1.3.4 Radical Additions Acetylenes are less reactive than olefins towards free radicals [138], and many theoretical studies have been made in order to understand the origin of such differences in reactivity. The additions to acetylene by a hydrogen atom [118, 139, 1401, a hydroxy radical [141], an oxygen
12
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
atom [142, 1431, and a methyl radical [144] all involve a small reaction barrier. Although some free radicals are more electrophilic than others, transition states of free-radical addition to xbonds resemble those of nucleophilic additions more than electrophilic additions. Reactions of acetylene with Li and Na atoms were also studied theoretically [30, 145, 1461 and experimentally by ESR [147] and IR in low-temperature matrices [148]. Other metal atom reactions with acetylene include B/Be/Mg [149, 1501, and Al/Si [151-1571. Among these metal-acetylene reactions, the aluminum-acetylene system has received the most attention. This is partly due to the early controversy surrounding the structural predictions by theory and the experimental conclusion, and partly due to the intriguing bonding difference betweeen Al-acetylene and Al-ethylene complexes. The Al-acetylene complex is o-bonded as observed by ESR [158]. The mbonded structure, although predicted to be a local minimum by theory, was not observed, nor was Al-vinylidene, the global energy minimum predicted at that time. It was later found that a barrier of 30 kcal/mol must have prevented the first-formed obonded complex from rearranging to the more stable Al-vinylidene complex [156]. Unlike the Al-acetylene complex, Al-ethylene forms only a x-bonded structure experimentally. In fact, calculations showed that the o-bonded Al-ethylene complex collapsed to the lower-energy xbonded structure [159].
1.3.5 Molecular Complexes There have been extensive experimental and theoretical studies devoted to the structural and bonding characterization of weakly bound van der Waals complexes of acetylene. Structures of these complexes can often be determinated experimentally by means of Fourier transform microwave and infrared spectroscopic techniques. On the theoretical side, advanced treatments are required to understand the complex nature of the weak bonding in terms of the relative contributions of polarization and dispersion interactions, interactions of multiple moments, and electrostatic interactions involved in these complexes. To determine the interaction energy in a weak complex, it is necessary to use large basis sets with the inclusion of electron correlation interactions. Theoretical calculations have been reported for van der Waals complexes of acetylene with C02 [1601, CO 1161, 1621, AlCl, 11631, NH, [164], He [165], Ar [166], H2O [167], HCN [168], HF [169-1721, HC1 [173, 1741, and acetylene itself in the forms of non-covalent dimer [175-1801, trimer [175, 1811, tetramer [175, 182, 1831, and pentamer [175]. These calculations are very useful for the determination of multiple isomeric forms of the complex. For example, calculations at the MP2/6-31G** level along with IR spectra indicate that the HCN-acetylene complex exists in a linear form in addition to the T-shaped structure observed previously by microwave studies (see Fig. 1-5) [168].
T-Shape
Linear Form
Figure 1-5 van der Waals complexes of acetylene with HCN.
1.4 Polyacetylenes
13
1.4 Polyacetylenes Studies of carbon clusters and rods during the last ten years have led to increasing interest in the structural features and properties of an extended system of conjugated triple bonds. The acetylene unit is an important unit for the construction of carbon-rich systems “4- 1861. The synthesis of polyyne carbon rods with alternating single and triple bonds (“carbyne” [187- 1891) began with the pioneering efforts of Baeyer [190] (see also ref. 49 in [191]), and was advanced by the work of Walton and coworkers [192] (see also the recent synthesis of dicyanopolyynes [193]). The study of carbon clusters dates back to 1942 [194]. Very rapid progress has been made since the discovery and large-scale preparation of the fullerenes [195, 1961. This has inspired the design of novel carbon allotropes [184, 185, 191, 197-2001, The rapidly growing body of experimental data on all-carbon compounds challenges the abilities of theorists to predict and interpret the features of such systems using quantum chemistry.
1.4.1 Diacetylene Diacetylene (1,3-butadiyne) is the first member of the polyyne series with conjugated triple bonds. It is the simplest compound with a single bond between two sp-hybridized carbon atoms and is a suitable model for the study of the influence of conjugation effects on groundand excited-state properties. Diacetylene has been the subject of numerous theoretical investigations which are discussed briefly in this section. Single-determinant ab-initio molecular orbital theory was applied to the description of the equilibrium geometry of diacetylene by Hehre and Pople in 1975 [201]. With RHF theory using an STO-3G minimal basis set, they calculated a C = C triple bond length of 1.175 A and a C - C single bond length of 1.408 A.The unusually short single bond was claimed to be the shortest such linkage in a neutral hydrocarbon. Experimental values are 1.205 and 1.376 A (rotational Raman spectra) [202], and 1.218 and 1.384 A (gas-phase electron diffraction) [203], respectively. The peculiar character of this single bond has been demonstrated by calculating the reaction energy for the isodesmic transformation, HC = C - C = CH + 2 CH4 2 C2H2+ C2H6; the reaction energy was found to be 15.3 to 17.2 kcal/mol, depending on the basis set used [204]. This very large energy is due to the strength of the sp-sp single bond in diacetylene, and to the effect of conjugation. The stabilizing effect of a conjugated triple bond in comparison with other substituents on acetylene has been demonstrated further by calculating the energy of the isodesmic reaction HC = C - C ICH + CH4+ C2H2 HCCCH, using several basis sets (e.g., 6.9 kcal/mol with the 6-31G** basis set) [2051. A thorough investigation of the basis set dependence of ab-initio molecular structures of several nonstrained hydrocarbons was made by Hafelinger et al. in 1989 [206]. The calculated C = C bond lengths in C4H2 vary between 1.1727 (STO-3G) and 1.1958 A (6-31G), and the = C -C = bond lengths vary between 1.3735 (3-21G) and 1.4082 A (STO-3G). This comparison revealed that with standard H F -SCF -MO methods the geometries closest to experiment are obtained with the 6-31G basis set (1.1958 and 1.3800 A, respectively). In a study on the proton affinity of diacetylene, Botschwina et al. reported more extended ab-initio calculations for C4H2 by allowing for effects of electron correlation, using the coupled-electron-pair approximation (CEPA) [207]. They obtained C C bond lengths of +
+
+
-
14
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
1.2114 and 1.3802 A, respectively. The proton affinity of C,H2 w a s calculated to be 177.3 kcalhol. In an earlier study, Deakyne et al. determined the proton affinity of diacetylene by ion cyclotron resonance bracketing experiments as 180 f 1 kcalhol [133]. From SCF, MP2, and MP3 calculations with the 6-31G* basis set, they obtained values of 189.9, 179.9 and 185.8 kcalhnol, respectively. An investigation of the electronic ground state and the first valence excited states of diacetylene with the aid of ab-initio methods was made by Karpfen and Lischka [208]. CI calculations based on an MCSCF reference with a 4-31G basis set led to an equilibrium geometry of C4H2 with C-C bond lengths of 1.219 and 1.374 A. In the excited states the central C-C bond distance is reduced from its ground-state value of 1.37 A to 1.29-1.30 A, whereas the outer C = C triple bonds are elongated from 1.22 to 1.28-1.31 A; all the C-C bonds in the electronicallyexcited diacetylenes have practically equal lengths. Thus, single excitation of diacetylene from the highest occupied x g orbital into the lowest unoccupied x,* orbital uniformly leads to cumulene-type structures. In all the excited states the carbon chain is quite close to linearity. A two-dimensional Hilckel molecular orbital (HMO) theory approach to acetylenic systems yielded x-bond orders of P = 0.894 for the central C - C bond and P = 1.788 for the C E C triple bonds in 1J-butadiyne [209, 2101. For comparison, P = 1 for ethylene and P = 2 for acetylene. A different criterion for determining the relative strengths of chemical bonds was used by Politzer and Ranganathan [17]. Starting from STO-3G geometries and force constants, they calculated a bond order of 1.34 for the central C-C bond in diacetylene. This corresponds to a bond dissociation energy of 150 kcal/mol [211], which compares with bond orders and bond dissociation energies of 1.14 and 88 kcalhol for ethane and 1.85 and 163 kcalhol for ethylene. Several definitions of resonance energy have been proposed in the literature, while the concept of resonance itself has been the subject of intense discussions [212-2141. Kollmar defined vertical resonance energies of conjugated hydrocarbons as the difference between the ab-initio SCF energy and the energy of a model wavefunction in which the SCF x-orbitals were replaced by appropriate nonresonating localized x MOs [ZlS]. For butadiyne, a vertical resonance energy of 22.3 kcal/mol (DZ d basis) was calculated. The short single bond of butadiyne is due to both sp hybridization and resonance in the x-system. The total adiabatic resonance energy of diacetylene is 19 kcal/mol according to this calculation. Later, Gready presented a x bond order-bond length correlation based on ab-initio STO-3G calculations and apportioned the bond shortening of the central C-C single bond in diacetylene to ca. 74% hybridization and 26% x-electron resonance contributions [216]. Theoretical investigationsof the structures and electronic properties of more extended conjugated triple bond structures in polyacetylenes have been rather rare. Approximate calculations of the extended Hitckel type for ground and excited states of cumulenes, polyenes, polyacetylenes, and C , compounds were reported by Hoffmann in a seminal paper in 1966 [217]. For polyacetylenes C,H2 (n even, up to 16), he found that charges and overlap populations vaned only slightly in the chain interior. Further, he predicted that bond alternation will persist with growing chain length, and that bond length variation will be small. Moffat studied the effect of chain lengthening on the bond distances of linear polyynes by ab-initio STO-3G calculations [218]. He found that as the length of the molecule increases, the C - C bond lengths decrease, while the C = C lengths increase; the changes are small. With chain lengthening, the HOMO energy increases, while the LUMO energy decreases.
+
1.4 Polyacetylenes
15
Fan and Pfeiffer predicted the electronic and geometric structures of a whole series of polyynes, HC,H (n = 2-10) with RHF ab-initio calculations using DZ and DZP basis sets [219]. For molecules of this type, singlet ground states (‘X;) and polyyne-type bonding structures were found for the n-even species. On the other hand, for the n-odd molecules (ethynyl carbene and its ethynylogs) the ground states were found to be triplets (3ZJ and the bonds in the interior of the n-odd chains are intermediate in length between single and double bonds. It was clearly established that the tendency for the ends of the molecules to adopt polyacetylene-like character progresses as the chain is lengthened.
1.4.2 C, and Cyclic C, Due to the development of methods for the production of cold cluster beams containing species up to CZw, spectroscopic data on all-carbon molecules are now widely available. No direct structural or energetic information is available experimentally, and this has stimulated the application of computational methods in order to evaluate the geometric and electronic structures of these molecules, as well as to predict properties such as ionization potentials and vibrational frequencies. The experimental and theoretical investigations on C, compounds up to 1989 are covered by an excellent review by Weltner and Van Zee [220]. These authors noted that “the present knowledge of C, molecules and their ions is almost a monotonically decreasing function of n”. This situation has changed somewhat now through the work of Diederich and co-workers on cyclo[n]carbons such as C,, [221]. Different synthetic routes to cyclo[n]carbons are reviewed in [I851 and [1911. In the context of this review, only the most significant theoretical contributions to the clarification of structures and properties of C, molecules are discussed. The work since 1989 is stressed. Special emphasis is placed on the structures of cyclo[n]carbons for two reasons; (i) there is still no consensus on the bonding in such species; and (ii) these molecules seem to play an important role in the formation of fullerenes [222-2241. Linear C, species may be represented simply as shown in Fig. 1-6. For an even number of carbons, the simplest electronic structure may be either a “dicarbene-cumulene” structure, or a “diradical-polyyne”. The corresponding cyclic structures will be nonlinear and strained, but formally possess closed-shell cumulene or polyyne structures. These differ by having all bond lengths equal, or alternating bond lengths, respectively. For n = odd, the linear structures may be of the “dicarbene-cumulene” or “tetraradical-polyyne” type. The cyclic isomers may be “cumulene” or “carbene-polyyne”. The application of molecular orbital theory to predict structures and properties of C, species started with the pioneering work of Pitzer and Clementi in 1959 [225]. By using semiempirical MO theory, these authors found the cumulenic linear conformation to be the most stable geometry for C,, except possibly for very large values of n. They predicted that the n-odd clusters would have closed-shell structures and lower energies than the singlet n-even species. Cyclic molecules were found to be unimportant under most conditions. Later, Hoffmann, using extended Hlickel theory, calculated that for C, molecules with n < 10, linear structures were more stable than ring structures [217]. In contrast to Pitzer and Clementi, he found considerable bond length variation for n 2 4; for example, he predicted a polyacetylenic diradical structure . C c C - C E C for C,. He also concluded that for large
-
16
1 Modern Computational and Theoretical Aspects of Acetylene Chemistry
even: dicarbene-cumulene
even: diradical-polyyne
odd: dicarbene-cumulene
odd: tetraradical-polyyne
..
Figure 1-6 Electronic structures of odd and even length (n) linear and cyclic polyynes.
n, bond alternation will not persist, but that the “end effect” would be still great, i.e., equalization of bond distances will be found only in the interior of the carbon chain. For C, with n = 4q + 2 (q = an integer), Hoffmann predicted that the cyclic isomers would become more stable than the corresponding linear chains for n 2 10. Today, through the powerful development of ab-initio MO methods, the interplay between theory and experiment has brought much insight into the nature of carbon clusters. Nevertheless, although innumerable studies applying state-of-the-art computational techniques have been published, there is still much controversy about the lowest-energy geometries and electronic states of several C, species. Among the more comprehensive theoretical studies concerned with C, clusters published to date are the following. An early MIND0/2 study by Slanina and Zahradnik [226], in which - in contrast to all the other calculations - a monotonic stability decrease with increasing n was predicted. The most stable C, clusters were found to be nonlinear. Single-determinant HF calculations by Ewing and Pfeiffer on C, (n = 2 to 6) [227]. MNDO calculations by Bernholc and Phillips for neutral and charged chains and monocyclic rings up to n = 25 [228]. For neutral clusters, linear chains were found to be the preferred structures up to n = 9 (as compared with cyclic isomers), with the odd-
1.4 Polyacetylenes
17
membered chains being more stable than the even-membered ones. At n = 10, the monocyclic rings become somewhat more stable than the chains, while the latter forms are more stable for n = 11 and 12. Particularly stable rings were predicted for n = 10, 14, 18, and 22. Graphitic structures corresponding to naphthalene and anthracene were also considered, but these were found to be more than 100 kcal/mol higher in energy than the monocyclic rings. Ab-initio calculations including the effects of polarization functions and electron correlation (fourth-order Merller-Plesset perturbation theory with the 6-31G* basis set) were reported by Raghavachari and Binkley [229, 2301. The most stable isomers calculated are shown in Fig. 1-7. Significant odd-even alternation was determined in the nature of the cluster geometries; the odd-membered species have linear structures and many of the larger even-membered clusters have cyclic structures almost equal in energy to the linear ones. In the case of neutral and positively charged clusters, the odd-membered clusters were predicted to be significantly more stable than the adjacent even-membered ones, whereas the opposite order of stability was established for the negative ions. An extensive comparative study on the performance of ab-initio and semiempirical methods when applied to carbon clusters C2 to Cl0 was presented by Martin et al. [231]. For larger systems (n = 18 to 60), Feyereisen et al. performed SCF MP2 calculations in order to answer the question whether these molecules adopt closed fullerene or cumulene/polyacetylene ring structures. For n 2 32, the fullerenes were predicted to be the most stable isomers [232]. Another semiempirical study (AM1) was devoted to the study of lowest-energy geometries of carbon clusters up to Cm [233]. Kurita et al. used an MO method based on a nonlocal density functional formalism for the calculation of binding energies of several carbon fullerenes [234]. A different concept for studying the structural properties of carbon clusters (Cn,n = 2 to 60) was applied by Xu et al. [235]. Using tight-binding molecular dynamics simulations they
c-c-c
I
c-c-c-c-c-c-c 1.270
1.264
C
\ 1.24 C
Lc;/c
1.280
/
1.38
c--c+c
1.290
/
c-c-c-c-c-c-c-c-c 1.269
1.261
1.269
1.283
I”
C
\
\C
\
119.4
c n c ’ C ‘
Figure 1-7 Structures (MP4/6-31G*) of most stable C, to C,, isomers [229].
/‘
18
-
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
found that in the range 5 In s 11 odd-membered clusters prefer a linear structure, while even-membered clusters prefer a ring structure. They observed further that monocyclic rings are energetically favorable for the clusters with 12 In 5 19, while for clusters with 20 s n 5 60 the most stable structures are cages. A density functional study of the molecular structures and vibrational frequencies of the linear and planar monocyclic isomers of the C, series (2 s n I18) was performed recently by Hutter et al. [236]. They found that for n I9 the linear chains are generally more stable than the ring structures. Their results confirmed that for n 2 10 the cyclic structures dominate.
Having described the more comprehensive studies which have been performed, we now describe the state of the art for each C,.
Many theoretical investigations on various aspects of this diatomic molecule and the C, and Cg ions were reviewed in [220]. The most throrough study is still the one by Bauschlicher and Langhoff [237]. More recent high-level calculations were performed by Scuseria et al. [238] who applied several different coupled cluster methods and compared their performance to C I S D Q and calculations by Watts and Bartlett [239]. In the latter study, various coupled cluster methods were used to calculate several electronic states of C, (ground state X ' Z l , calculated bond lengths 1.263 to 1.270 A), C; and C;.
1.4.2.2 C,
Early theoretical studies generally agreed on an electronic structure of the form :C=C=C: (singlet, four delocalized A electrons), where C3 was assumed linear. Cyclic structures were found to be noncompetitive energetically with a linear 'El ground state (2401. Many studies on this molecule were concerned with the question of whether C3 w a s slightly bent in the ground state with a small barrier to linearity, such that it becomes effectively linear by vibration ("quasi-linear"). The most complete ab-initio surface for C3 referenced in [220] was that by Kraemer et al. [241], who found a quasi-linear equilibrium structure with a barrier to linearity of 20 cm-' and an equilibrium C - C - C bond angle of 162". A more recent full valence CASSCF calculation with a very large basis set addressed the question of quasilinearity of C3 again [242]. From the ab-initio results obtained, and from a fitting to experimental data, it was concluded that the lowest-energy structure of C3 is linear. A strictly linear structure for C3 was also predicted by Kurtz and Adamowicz, who used MBPT(2) with the 6-31G* basis set to study the linear carbon chains C3 to C9 [243]. In these calculations, even-n C, molecules were assumed to be triplets, while odd-n clusters were assumed to have singlet ground states. By constraining the geometry to be planar and C2 symmetrical, all of these molecules converged to almost linear forms with nearly equal bond lengths; that is, they all have a cumulenic bonding pattern.
1.4 Polyacetylenes
19
1.4.2.3 Cd
A very thorough investigation of the bonding of the even-numbered carbon chains C4, C6, C8, and Clo was performed by Liang and Schaefer using the single and double excitation configuration interaction (CISD) method with a DZP basis set [244]. It was found that the energy difference between the lowest triplet and singlet states of the cumulenic structures and that between the cumulenic and acetylenic structures decrease monotonically as the carbon chain lengthens. For (28 and Clo, these energy differences are already very small. The lowest-energy structure of C4 has long been a matter of controversy. Fig. 1-8 shows possible structures. Several sophisticated studies indicate that the rhombic 'Ag structure (1) is somewhat lower in energy that the linear 3Z; structure (2) [US-2481. Bernholdt et al. performed an investigation of the rhombic and linear forms of C4 with various levels of coupled cluster and many-body perturbation theory [249], and found that the two isomers are essentially isoenergetic. They also argued, on thermodynamical grounds, that the linear structure would be favored at practical temperatures because of the additional degeneracy entropy. A different result was obtained by Parasuk and Almof, who used multireference configuration interaction (MRCI) methodology and large basis sets of atomic natural orbitals [250]: the linear cumulene-like 3X[ electronic state (2) was predicted to be 4.1 kcal/mol lower in energy than the rhombic structure. The polyacetylene-like :X' state (3) was found to be much higher in energy when correlation effects were included. In a recent study, Ewing addressed the question of whether triplet C4 is bent in the ground state, but found no indications for a deviation from linearity using MP2 calculations; the amount of energy required to bend linear C4 is quite small, however [251].
1
2
3
Figure 1-8 Several low-energy isomers of C,.
1.4.2.4 C5, C, and C,
There is general agreement that these molecules adopt a linear cumulenic geometry ('Z: ground state). MP2 calculations by Ewing and Pfeiffer show that a variety of alternative nonlinear structures considered for C, are much higher in energy than the linear form, which has bond lengths of 1.277 A for the outer bonds and 1.280 A for the inner bonds (2521. Calculated heats of formation for C, were reported by Martin et al. on the basis of MP4 and QCISD(T) energies [253]. For C,, Slanina et al. located a cyclic C,, symmetrical minimum structure using MP2/6-31* methodology and found it only 2.2 kcal/mol higher in energy than the linear form [254]. However, an MP4 treatment at the MP2/6-31G* geometries increases the separation to about 17.5 kcal/mol. Another cyclic C,, symmetrical isomer was calculated to be 1.12 eV
20
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
higher in energy than the linear form in an ab-initio study carried out by von Helden et al. [255]. Their results indicate that C, and C f have minima, while the lowest-energy structure of CT is cyclic.
By MP4/6-31G* calculations Raghavachari and Binkley predicted that the c6 and C,, clusters would have cyclic D3h and Dsh symmetrical geometries (bond lengths 1.316 and 1.290 A, respectively), while C, would adopt a polyacetylenic c 4 h structure (bond lengths 1.24 and 1.38 A); the linear isomers are very close in energy, however [229]. Generally, all the ring structures benefit from the additional bonding resulting from the ring closure. However, the energy gained from such bond formation has to be weighed against the angle strain energy that may be present in the more compact ring structures. This balance between two opposing factors causes the even-membered clusters, which are comparatively less stable as linear forms, to have low-lying monocyclic structures. Fig. 1-9 shows low-lying structures of c6. Raghavachari et al. predicted that the planar distorted hexagonal D3h structure (4) is the global minimum for c6, but the highly symmetrical D6h (5) form is only a few kcal/mol higher in energy [256]. The linear cumulenic 3C; structure (6) is about 10 kcal/mol higher in energy than 4, according to their calculations, and the polyacetylenic 'C,,+ structure (7)is more than 50 kcal/mol higher in energy. Contradictory results were obtained by Parasuk and Almlof, who used MCSCF and MRCI methods with large basis sets of A N 0 type [257]. They found that the linear cumulenic 3C; structure (6) was the energy minimum, the triacetylenic 3C,+ form, 7,was about 20 kcal/mol higher in energy, and the D3hand D6h symmetrical cyclic structures were about 40 kcal/mol above the global minimum. Results that partially contradict both ab-initio studies mentioned above were presented by Hutter and LUthi quite recently [258]. Using CCSD(T) and CAS-IT2 as well as density functional theory (DFT), they obtained the following energetic order: hexagonal D3hstructure < hexagonal D6h structure (saddle point) < linear cumulenic c6 chain. The authors point out that the cyclic and linear isomers are very close in energy, indicating that both structures may be observed experimentally.
4
5
Figure 1-9 Several low-energy isomers of C,.
1.4 Polyacetylenes
21
When considering such subtle differences as relative enthalpies of cyclic vs. linear carbon cluster isomers, it is important to realize that the stability of the cyclic form may be heavily outweighed by the larger relative entropy of the linear form: this causes it to become increasingly important at higher temperatures [220]. Calculations indicate that for molecules up to C8, and possibly Cl0, the vapor phase produced from pyrolysis of graphite can be considered to a good approximation to be composed of only linear (singlet for n-odd, triplet for n-even) molecules [259]. Two recent ab-initio studies are devoted to the determination of the ground-state structure of C8. Parasuk and Almldf concluded that the cyclic polyacetylenic C4hstructure ('Ag) and the linear cumulenic 3Z; state are essentially isoenergetic [260]. However, the choice of basis sets and methods (MRCI; modified coupled-pair functional, MCPF) heavily affects the difference in energy between the linear and cyclic forms of C,, and so the results are quite uncertain. On the other hand, Slanina et al. proposed a nonplanar DZdsymmetrical cyclic structure as the minimum-energy structure of C, [261]. According to their MP2/6-31G* calculations, this form is 13 kcal/mol lower in energy than the planar C4hstructure. Unexpectedly, all the C-C bond lengths in the DZdspecies are equivalent (1.339 A). DFT calculations favor the linear structure as compared with a planar cyclic one, but nonplanar species seem to have not been considered in this study [236]. CISD calculations (DZP basis set) for several cyclic and linear isomers of Clo (see Fig. 1-10, 8-12) were performed by Liang and Schaefer [262]. They concluded that (i) the monocyclic forms of Clo are considerably more stable than the linear structures; (ii) the DY symmetrical cumulenic structure 9 is the ground state, but electron correlation decreases the energy difference between 9 and the alternative ring structures 8 and 10; and (iii) the linear acetylenic structure 12 is just slightly less stable than the linear cumulene 11. Another sophisticated investigation into the nature of monocyclic Cl0, using coupled cluster methodology, was presented by Watts and Bartlett [263]. Structure 9 was found to be a local minimum, while the fully symmetrical structure 8 shows one imaginary frequency and
8
9
10
~ c ~ c - c ~ c - c ~ c - c ~ c - c ~ c ~ 12
Figure 1-10 Cyclic and linear isomers of ClW
22
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
seems to be the low-energy transition state for automerization of 9. The polyacetylenic structure 10 is always somewhat higher in energy than 9 (e. g., 5.8 kcal/mol with CCSD(T)/PVDZ), In the study by Parasuk and Almltif mentioned above [260], it was concluded that the three cyclic structures considered (8-10) were practically isoenergetic with energy differences of less than 2 kcal/mol, indicating that cyclic Clo is a highly fluxional molecule. The level of correlation treatment appeared to be decisive in determining the energy differences between these isomers.
For CI1,a cyclic C,,, symmetrical structure was predicted to be the global minimum according to MP2/6-31G* calculations [264]. The cumulenic linear form is only ca. 5 kcal/mol higher in energy, but the authors point out that a basis set extension should favor the cyclic form even more, so there is little doubt that the energy minimum of CI1 is cyclic. DFT calculations on all of the C, clusters in this series confirmed the early predictions by Hoffmann [217J, that cyclic structures are more stable than the linear chains [236]. The C,, symmetrical structures of the odd-membered clusters optimized in this study showed imaginary frequencies, which indicate that the lowest-energy structures of these molecules are three-dimensional and polycyclic systems. However, a recent infrared absorption analysis study proved the existence of a low-energy linear isomer ('El) of C13[265]. It was argued that entropy strongly favors the formation of linear over cyclic isomers in the experimental conditions applied. With the even-membered ring species, the (4n + 2) x-electron series was predicted by DFT to have cumulenic bond types, whereas the 4n x-electron series was of polyacetylenic type. However, C,, and (equal bond angles in both cases) showed imaginary frequencies, which indicate that symmetry-reducing inplane distortions would lead to lower-energy structures.
In 1989, Diederich et al. reported the synthesis of an organic precursor of cyclo[l8]carbon [221]. This molecule was designed to react further through a series of retro-Diels- Alder reactions to yield c18. Indeed, an analysis of laser flash heating experiments by time-of-flight mass spectrometry showed a fragmentation pattern according to the proposed mechanism for c 1 8 formation. In order to provide theoretical support for this interpretation, ab-initio calculations on the structures and stabilities of different cyclo[l8]carbon isomers were performed. Several low-energy isomers are shown in Fig. 1-11. Geometry optimizations at the Hartree-Fock level of theory (STO-3G and 3-21Gbasis sets) of the three different forms considered (13-15) indicated that the cyclic Dghsymmetrical polyacetylene structure 14 (alternating bond lengths) was more stable than the cumulenic Dghstructure 15 or the Dl8h structure 13, which both have equal bond lengths. According to the frequency analysis (3-21G basis set), structure 14 is a minimum on the potential energy surface, whereas structures 13 and 15 are not. It was further argued that according to thermodynamical group equivalents, the acetylenic structure 14 should be at least 60 kcal/mol more stable than the cumulenic one, so it is unlikely that the aromaticity-induced driving force could outweigh the tendency of electron-localization and equalize the bond lengths.
1.4 Polyacetylenes
23
14
13
15
Figure 1-11 Cyclic cumulenic and acetylenic isomers of CIS.
Considering the results of the theoretical investigations on smaller C, clusters summarized in this chapter, it is clear why this statement caused some controversy. Parasuk et al. found the conclusions drawn from the SCF calculations surprising, since they claimed that this molecule, with its (4n + 2) n-electron system, w a s a candidate for Huckel aromaticity and expected to adopt the delocalized structure 13 [266]. They confirmed the earlier SCF results by symmetrical form 13 to redoing these calculations with larger basis sets, but found the DIBh be most stable at the MP2 level. However, they were faced with considerable symmetry-breaking effects at the SCF level which made it impossible to treat this structure consistently at the MP2 level. In their study on the nature of monocyclic C,, mentioned above [263] Watts and Bartlett found the results obtained with MBFT of second and fourth order most likely to be artificial and questioned the applicability of perturbation theory for these systems and consequently the results of Parasuk et al. [266]. Hutter et al. concluded from their DFT calculations that the fully symmetrical cumulene structure is the most stable planar ring structure, although it turned out to be a saddle-point geometry 12361.
24
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
In view of these contradictory statements, we have recently reinvestigated whether the localized acetylenic c]8 ring structure could be favored above delocalized ring structures [267]. At the HF/6-31G* level, the energy difference between the bond-alternant structure 14 and the delocalized structures 15 and 13 is 32.1 and 68.4 kcal/mol, respectively. In order to check the validity of an MBPT approach, the stability of the single-determinant HF wavefunctions was tested [268]. M~rller-Plesset energies based on wavefunctions which are unstable with respect to UHF are highly questionable [269]. The wavefunctions for all cyclic geometries considered so far showed an RHF + UHF instability. All these structures show at least one imaginary frequency. In all cases the modes associated with these frequencies are of a ringflattening type, i. e., ring modes in which alternate atoms move toward and away from the centroid. After reducing the symmetry of the starting geometry, a polyacetylenic C9h structure with two different bond angles was found as a minimum on the potential energy surface. Density functional theory was also applied to this problem. The cumulenic Dghform 15 turned out to be a minimum structure at the B-LYP level of DFT using a 6-31G* basis set, while the fully symmetrical structure 13 is 2.4 kcal/mol higher in energy. Unfortunately, it was not possible to localize a polyacetylenic structure with the DFT methods used. The same trend holds for C,, and C,,, other candicates for Huckel aromaticity: the D,,, and D,,, symmetrical cumulenic structures are minima on the potential energy surface; polyacetylenic isomers are not stationary points using DFT methods. In order to assess the suitability of DFT-based methods for systems of this kind, additional calculations were performed on isomeric cumulenic and acetylenic structures, e. g., allene and methylacetylene. Surprisingly, while experimental geometries are reproduced very well with the B-LYP and Becke3-LYP functional, these methods favor the cumulenic structures energetically; on the other hand, simple HF/6-31G* calculations reproduce the thermodynamic energy-ordering very well. Single point energies (DFT/B-LYP) of “guessed” polyacetylenic c,8 geometries are only ca. 20 kcal/mol higher than the values for the delocalized structure 15. This is considerably less than the error inherent in the density functionals used which incorrectly favor the cumulenic structure. If one extrapolates the error inherent in the density functionals to c]8, it is quite obvious that the polyacetylene structure is the absolute minimum on the potential energy surface. The aromaticity argument for a fully delocalized structure for c18 is not in itself conclusive. The tendency for large conjugated systems to adopt electron-localized structures is well known [270-2721. Liang and Schaefer noted the strong tendency of carbon clusters to adopt the electron-localized form as compared with annulenes [262], and they suggested that a polyacetylenic c18 chain might be more stable than the cumulenic isomer [244]. Simple RHF calculations favor electron-localized structures due to the inadequate treatment of electron repulsion; nevertheless, this method correctly predicts a distorted cumulenic structure for Cl0 [262, 2631. The tendency for electron localization and thus for polyacetylenic structures in cyclo[n]carbons with growing n is clearly established from our calculations. The highest symmetrical Dnhforms are not the minimum structures for the monocyclic clusters C,,, CI4, and CIS, but are the transition states for the automerization of the less symmetrical cumulenic isomers. Another way of distorting the i 4 8 h symmetry - and according to our results also a favorable one - is the localization of electrons in c18. In view of our results it seems as if the energy gain by aromaticity cannot overrule the tendency of electron-localization in cyclo[n]carbons with growing n. Cyclic C14 appears to be the turning point in this respect, and there is clear evidence for the greater stability of the acetylenic form of c18.
References
25
1.5 Conclusion Theoretical studies of simple acetylenes and the reactions of acetylene have reached a stage of considerable accuracy, yet a vast world of acetylene chemistry remains to be studied by theoretical methods.
Abbreviations AM1 AN0 Becke3 B-LYP CAS-R2 CASSCF
CCCI CCSD CCSD(T) CEPA CI CISD CISD( Q)
+
c1sJYI-Q DFT DZP G1, G2
GVB HMO LDA MBlT(n) MCPF MCSCF MIND0/2 MP(n)
Austin Model I. One of the Dewar semi-empirical methods. atomic natural orbitals Becke3 exchange functional Becke exchange- with Lee- Yang-Parr correlation functional complete active space perturbation fheory of 2nd order complete active space self consistent field. An MCSCF technique involving all possible electron configurations among active space - i. e., partial bonding-orbitals. coupled cluster configuration interaction coupled cluster with single and double excitation coupled cluster with single and double excitation and perturbative triples coupled electron pair approximation configuration interaction. A method of correlation energy calculation. configuration interaction with single and double excitation. configuration interaction with single and double excitation and perturbative quadruples. configuration interaction with single, double, triple and quadruple excitation. density functional theory double [ with polarization functions. Description of a basis set. GAUSSIANI, GAUSSIAN2. Recipes for a series of ab initio HF, MP, and CI calculations with different basis sets, proposed by Pople and coworkers. G1 and G2 calculations give energies within “experimental error” (f2 kcal/mol) of the actual values. generalized valence bond theory. A method by W. A. Goddard to include correlation energy. Hiickel molecular orbital. local density approximation. many-body perturbation theory of n-th order. A method of calculating correlation energy. modified coupled pair functional. multi-configurational SCI? Electronic wavefunction is a linear combination of configurations. modified intermediate neglect of differential overlap. An early Dewar semi-empirical method. Merller-Plesset perturbation theory of n-th order. A specific type of MBPT.
26
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
MRCI PRDDO QCISD(T) RHF RRKM SCF STOJG, 3-21G, etc. UHF UMP(n)
multi-reference configuration interaction. CI based on an MCSCF calculation. partial retention of diatomic differential overlap. A semi-empirical method due to Lipscomb et al. quadratic configuration interaction with single and double excitation and perturbative triples. restricted Hartree-Fock theory. All MOs are doubly occupied or vacant. Rice-Ramsperger-Kassel-Marcus transition state theory. self-consistent field. Designation of basis set, which is the mathematical function used to represent the atomic orbitals used in a quantum mechanical calculation. unrestricted Hartree-Fock theory. a- and b-electrons occupy spatially different MOs. unrestricted Merller-Plesset perturbation theory of n-th order.
References [l] H. F. Schaefer 111, Ace. Chem. Res. 1979, 12, 288-296. [2]W. J. Hehre, L. Radom, P. v. R. Schleyer, J. A. Pople, A b Initio Molecular Orbital Theory, Wiley, New York, 1986. [3]J. A. Pople, Q.Rev. Chem. Soc. 1957,11, 273-290. [4]L. Pauling, The Nature of the Chemical Bond, 3rd ed., Cornell, Ithaca, 1960, pp. 136-142. [5]E. A. Walters, X Chem. Educ. 1966,43, 134-137. [a]K. N. Houk, N. G. Rondan, F. K. Brown, W. L. Jorgensen, J. D. Madura, D. C. Spellmeyer, X Am. Chem. SOC. 1983, 105, 5980-5988. [7]F. M. Bobrowicz, W. Goddard, in Modern Theoretical Chemistry, Methods in Electronic Structure Theory (Ed.: H. F. Schaefer 111), Plenum Press, New York, 1977, vol. 3, p. 79. [ 8 ] P. B. Karadakov, J. Gerratt, D. L. Cooper, M. Rairnondi, X Am. Chem. Soc. 1993,115, 6863-6869. [9]R. P. Messmer, P. A. Schultz, X Am. Chem. Soc. 1988,110, 8258-8259. [lo]R. P. Messmer, P. A. Schultz, Phys. Rev. Lett. 1986, 57, 2653-2656. [11] R. P. Messmer, P. A. Schultz, Phys. Rev. Lett. 1988,60, 860-860. [I21 P. A. Schultz, R. P. Messmer, in Molecules in Natuml Science and Medicine, (Ed.: Z . B. Maksic, M. Eckart-Maksic), Ellis-Horwood, New York, 1991, p. 309. [13] W. E. Palke, L Am. Chem. Soc. 1986, 108, 6543-6544. [14] J. Baker, Theoz Chim Acta 1985, 68, 221-229. [15] J. Cioslowski, S. T. Mixon, L Am. Chem. Soc. 1991, 113, 4142-4145. [16]A. T. Balaban, G. R. De Mark R. A. Pokier, THEOCHEM (L Mol. Struct.) 1989, 52, 103-119. [17] P. Politzer, S. Ranganathan, Chem. Phys. Left 1986, 124, 527-530. [18] 0. P. Singh, J. S. Yadav, THEOCHEM (1Mol. Struct.) 1985, 25, 287-292. [19] G. D. Paderes, W. L. Jorgensen, 1 Org. Chem. 1992,57, 1904-1916. [20]K. B. Wiberg, C. M. Hadad, T. J. Lepage, C. M. Breneman, M. J. Frisch, J. Phys. Chem. 1992,96, 671-679. [21]K. M. Ervin, S. Gronert, S. E. Barlow, M. K. Gilles, A. G. Harrison, V. M. Bierbaum, C. H. Depuy, W. C. Lineberger, X Am. Chem. Soc. 1990, 112, 5750-5759. I221 K. M. Ervin, J. Ho, W. C. Lineberger, X Chem. Phys. 1989, 91, 5974-5992. [23]S. M. Burnett, A. E. Stevens, C. S. Feigerle, W. C. Lineberger, Chem. Phys. Lett. 1983,100,124-128.
References
21
[24] Y. Chen, D. M. Jonas, J. L. Kinsey, R. W. Field, 1 Chem. Phys, 1989, 91, 3976-3987. [25] G. A. Petersson, T. G. Tensfeldt, J. A. Montgomery, J. Am. Chem. Soc. 1992, 114, 6133-6138. [26] M. M. Gallo, T.P. Hamilton, H. F. Schaefer 111, J Am. Chem. Soc. 1990, 112, 8714-8719. [27] Y. Osamura, H. F. Schaefer 111, S. K. Gray, W. H. Miller, J. Am. Chem. Soc. 1981, 103, 1904-1907. [28] L. B. Harding, J. Am. Chem. SOC. 1981, 103, 7469-7475. [29] R. Krishnan, M. J. Frisch, J. A. Pople, P. v. R. Schleyer, Chem. Phys. Lett. 1981, 79, 408-411. [30] S. Sakai, K. Morokuma, J. Phys. Chem. 1987, 91. 3661-3667. [31] B. J. Smith, R. Smernik, L. Radom, Chem. Phys. Lett. 1992, 188, 589-594. [32] T. Carrington, L. M. Hubbard, H. F. Schaefer HI, W. H. Miller, J. Chem. Phys. 1984,80,4347-4354. [33] J. S. Binkley, J. Am. Chem. Soc. 1984, 106, 603-609. [34] P.Halvick, D. Liotard, J. C. Rayez, Chem. Phys. 1993, 177, 69-78. [35] T. A. Holme, R. D. Levine, Chem. Phys. 1989, 131, 169-190. [36] P. J. Stang, Acc. Chem. Res. 1978, 11, 107-114. [37] P. J. Stang, Chem. Rev. 1978, 78, 383-405. [38] P. J. Stang, Acc. Chem. Res. 1982, 15, 348-354. [39] W. Sander, G. Bucher, S. Wierlacher, Chem. Rev, 1993, 93, 1583-1621. [40] C. E. Dykstra, H. F. Schaefer 111, J. Am. Chem. Soc. 1978, I00, 1378-1382. [41] R. L. Sundberg, J. L. Abramson, J. L. Kinsey, R. W. Field, J. Chem. Phys. 1985, 83, 466-475. [42] E. Abramson, R. W. Field, D. Imre, K. K. Innes, J. L. Kinsey, J. Chem. Phys. 1984.80, 2298-2300. [43] E. L. Sibert, R. C. Mayrhofer, J. Chem. Phys. 1993, 99, 937-944. [44] L. A. Curtiss, K. Raghavachari, G. W. Trucks, J. A. Pople, J. Chem. Phys. 1991, 94, 7221-7230. [45] L. A. Curtiss, J. E. Carpenter, K. Raghavachari, J. A. Pople, J. Chem. Phys. 1992, 96, 9030-9034. [46] J. M. L. Martin, J. Chem. Phys. 1992, 97, 5012-5018. (471 J. Andzelm, E. Wimmer, J. Chem. Phys. 1992, 96, 1280-1303. [48] B. G. Johnson, P. M. W. Gill, J. A. Pople, J. Chem. Phys. 1992, 97, 7846-7848. 1491 P. M. W. Gill, B. G. Johnson, J. A. Pople, M. J. Frisch, Chem. Phys. Lett. 1992, 197, 499-505. [50] A. D. Becke, J. Chem. Phys. 1992, 97, 9173-9177. 1511 D. Habibollahzadeh, J. S. Murray, M. Grodzicki, J. M. Seminario, P. Politzer, Znt. J. Quantum Chem. 1992, 42 267-272. [52] C. J. Wu, E. A. Carter, J. Phys. Chem. 1991, 95, 8352-8363. [53] C. J. Wu, E. A. Carter, J. Am. Chem. Soc. 1990, 112, 5893-5895. [54] J. A. Montgomery, G. A. Petersson. Chem. Phys. Latt. 1990, 168, 75-78. [55] L. A. Curtiss, J. A. Pople, J. Chem. Phys, 1989, 91, 2420-2423. [56] Y. Osamura,F. Mitsuhashi, S. Iwata, Chem Phys. b t t . 1989, 164, 205-209. [57] S. R. Langhoff, C. W. Bauschlicher, P. R. Taylor, Chem. Phys. Lett. 1991, 180, 88-94. [58] E. A. Carter, W.A. Goddard, J. Chem. Phys. 1988, 88, 3132-3140. [59] J. K. Lundberg, R. W. Field, C. D. Sherrill, E. T. Seidl, Y. Xie, H. F. Schaefer 111, J. Chem. Phys. 1993, 98, 8384-8391. [60] G. Vacek, J. R. Thomas, B. DeLeeuw, Y. Yamaguchi. H. F. Schaefer 111, J. Chem. Phys. 1993, 98, 4766-4776. [61] Y. Yamaguchi, G. Vacek, J. R. Thomas, B. J. DeLeeuw, H.F. Schaefer 111.1 Chem. Phys. 1994, 100, 4969-4980. [62] Y. Yamaguchi, G. Vacek, H. F. Schaefer 111, Theor. Chim. Actu 1993, 86, 97-113. [63] H. R. Wendt, H. Hippler, H. E. Hunziker, J. Chem. Phys. 1979, 70, 4044-4048. [64] J. M. Lisy, W. Klemperer, J. Chem. Phys. 1980, 72, 3880-3883. [65] H. Lischka, A. Karpfen, Chem. Phys. 1986, 102, 77-89. [661 A. H. Laufer, J Chem. Phys. 1980, 73, 49-52. [671 D. Sulzle, H. Schwarz, Chem. Phys. Lett. 1989, 156, 397-400. I681 M. Drabbels, J. Heinze, W.L. Meerts, J. Chem. Phys. 1994, 100, 165-174. 1691 P. Dupre, P. G. Green, Chem. Phys. Lett. 1993, 212, 555-560.
28
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
I701 P. Dupre, R. Jost, M. Lombardi, P. G. Green, E. Abramson, R. W. Field, Chem. Phys. 1991, 152, 293-318. 1711 J. Lievin, J. Mol. Spectrosc. 1992, 156, 123-146. 1721 M. Peric, R. J. Buenker, S. D. Peyerimhoff, Mol. Phys. 1984, 53, 1177-1193. [73] M. Peric, S. D. Peyerimhoff, R. J. Buenker, Mol. Phys. 1987, 62, 1339-1356. [74] S. P. So, R. W. Wetmore, H. F. Schaefer 111, J. Chem. Phys. 1980, 73, 5706-5710. [75] 0. Nomura, N. Tanaka, A Phys. B 1981, 14, L481-L485. [76) T. J. Lee, D. J. Fox, H. F. Schaefer III, R. M. Pitzer, J. Chem. Phys. 1984, 81, 356-361. [77] J. L. Holmes, J. E. Szulejko, Chem. Phys. Lett. 1984, 107, 301-303. I781 T. J. Lee, J. E. Rice, H. F. Schaefer 111, J. Chem. Phys. 1987, 86, 3051-3053. [79] T. P. Hamilton, H. F. Schaefer 111, J. Phys. Chem. 1989, 93, 7560-7562. [SO] J. Baker, Chem. Phys. Lett. 1989, 159, 447-451. [81] P. Rosmus, P. Botschwina, J. P. Maier, Chem. Phys. Lett. 1981, 84, 71-76. 1821 S. Ramasesha, B. Sinha, Chem. Phys. Lett. 1991, 179, 379-384. [83] G. C. Goode, K. R. Jennings, Adv. Muss Spectrom. 1974, 6, 797-803. [84] J. Chandrasekhar, R. A. Kahn, P. v. R. Schleyer, Chem. Phys. Lett. 1982, 85, 493-495. I851 G. Frenking, Chem. Phys. Lett. 1983, 100, 484-487. 1861 K. N. Houk, Y. Li, J. D. Evanseck, Angew. Chem., Int. Ed. Engl. 1992, 31, 682-708. [87] R. B. Woodward, R. Hoffmann, Angew. Chem., Int. Ed. Engl. 1969, 8, 781-853. 1881 A. Nicolaides, W. T. Borden, J. Am. Chem. SOC.1991, 113, 6750-6755. [89] K. N. Houk, R. W. Gandour, R. W. Strozier, N. G. Ronda, J. Am. Chem. SOC. 1979, 101, 6797-6802. 1901 R. D. Bach, G. J. Wolber, H. B. Schlegel, 1 Am. Chem. SOC.1985, 107, 2837-2841. [91] J. M. Coxon, S. T. Grice, R. G. A. R. Maclagan, D. Q. McDonald, J. Org. Chem. 1990, 55, 3804-3807, [92] R. D. Bach, J. J. W. McDouall, H. B. Schlegel, G. J. Wolber, 1 Org. Chem. 1989, 54, 2931-2935. [93] J. Gonzalez, K. N. Houk, J. OR. Chem. 1992, 57, 3031-3037. [94] E. Goldstein, A. Kallel, P. S. Beauchamp, THEOCHEM (1 Mol. Struct.) 1987, 36, 297-305. [95] R. D. Harcourt, R. D. Little, A Am. Chem. SOC. 1984, 106, 41-46. (96) P. C. Hiberty, G. Ohanessian, H. B. Schlegel, 1 Am. Chem. SOC. 1983, 105, 719-723. [97] A. Komornicki, J. D. Goddard, H. F. Schaefer I I I , J. Am. Chem. SOC. 1980, 102, 1763-1769. I981 M. Sana, G. Leroy, G. Dive, M. T. Nguyen, THEOCHEM (J. Mol. Struct.) 1982, 6, 147-163. [99] F. Bernardi, A. Bottoni, J. J. W. McDouall, M. A. Robb, H. B. Schlegel, Faraday Symp. Chem. SOC. 1984, 19, 137-147. [loo] F. Bernardi, M. Olivucci, M. A. Robb, THEOCHEM (L Mol. Struct.) 1986, 31, 97-105. [loll F. Bernardi, M. Olivucci, J. J. W. McDouall, M. A. Robb, 1 Am. Chem. SOC.1987, 109, 544-553. [lo21 F.Bernardi, A. Bottoni, M. Olivucci, J. J. W. McDouaII, M. A. Robb, G. Tonachini, THEOCHEM (1 MOI. srrucr.) 1988, 42, 341-351. 11031 F. Bernardi, M. Olivucci, J. J. W.McDouall, M. A. Robb, J. Chem. Phys. 1988, 89, 6365-6375. (1041 J. J. W. McDouall, M. A. Robb, U. Niazi, F. Bernardi, H. B. Schlegel, 1 Am. Chem. SOC.1987, 109, 642-4648. [lo51 S. M. Bachrach, A. Streitwieser, J. Am. Chem. SOC. 1986, 108, 3946-3951. [lo61 J. T. Herron, R. E. Huie, J. Phys. Chem. 1974, 78, 2085-2088. [I071 W. B. DeMore, In?. J. Chem. Kinet. 1969, I , 209-222. [I081 J. Z. Gillies, C. W. Gillies, F. J. Lovas, K. Matsumura, R. D. Suenram, E. Kraka, D. Cremer, 1 Am. Chem. SOC.1991, 113, 6408-6415. I1091 J. E. Jackson, T. A. O’Brien, Jr., 1 Phys. Chem. 1988, 92, 2686-2696. [llO] K. A. Jorgensen, E. K. Moltzen, A. Senning, J. Org. Chem. 1987, 52, 2505-2509. [111] B. A. Hess, L. J. Schaad, Int. 1 Quantum Chem. 1986, 29, 345-350. [112] M. L. McKee, D. M. Stanbury, A Am. Chem. SOC.1992, 114, 3214-3219.
References
29
(1131 M. L. McKee, M. E. Squillacote, D. M. Stanbury, J. Phys. Chem. 1992, 96, 3266-3272. (1141 Review: K. C. Nicolaou, W.-M. Dai, Angew. Chem., Int. Ed. Engl. 1991,30, 1387-1416. (1151 R. R. Jones, R. G. Bergman, A Am. Chem. SOC. 1972,94, 660-661;R. G. Bergman, Acc. Chem. Res. 1973, 6, 25-31; T.P. Lockhart, P. B. Comita, R. G. Bergman, A Am. Chem. SOC. 1981, 103, 4082-4090; T. P. Lockhart, R. G. Bergman, J. Am. Chem. SOC. 1981, 103, 4091-4096. [116] (a) J. P.Snyder, A Am. Chem SOC.1989, Ill, 7630-7632;(b) J. P. Snyder, G. E. Tipsword, J. Am. Chem. SOC. 1990, 112, 4040-4042; (c) J. P. Snyder, J. Am. Chem. SOC. 1990, 112, 5367-5369; (d) N. Koga, K. Morokuma, J. Am. Chem. SOC.1991,113,1907-1911;(e) E. Kraka, D. Cremer, d Am. Chem. Soc. 1994,116,4929-4936;(0 R. Lindh, B. J. Persson, J. Am. Chem. SOC.1994, 116, 4963-4949;(9) W. D. Allen, unpublished results. [117] G. Melloni, G. Modena, U. Tonellato, Acc. Chem. Res. 1981, 14, 227-233. [118] S. Nagase, C. W. Kern, J. Am. Chem. Soc. 1980, 102, 4513-4515. [119] C. Barrientos, A. Largo, A Phys. Chem. 1992, 96, 5808-5814. [120] A. Largo, C. Barrientos, J. Phys. Chem. 1994,98, 3978-3984. [121] C. W. Bauschlicher, S. R. Langhoff, J. Phys. Chem. 1991, 95, 2278-2282. [122] M. Rosi, C. W. Bauschlicher, Chem. Phys. Lett. 1990, 166, 189-194. [123] P. S. Martin, K. Yates, I. G. Csizmadia, Theor. Chim. Acta 1983, 64, 117-126. [I241 P. S. Martin, K. Yates, 1. G. Csizmadia, THEOCHEM (J. Mol. Stnrct.) 1988, 42, 353-363. [125] P. S. Martin, K. Yates, I. G. Csizmadia, Can. Z Chem. 1989, 67, 2178-2187. [I261 W. H.Jones, R. D. Mariani, M. L. Lively, Chem. Phys. Lett. 1984, 108, 602-608. (1271 L. A. Curtis, J. A. Pople, J. Chem. Phys. 1988, 88, 7405-7409. [128]F. Schneider, A. Palma, E. Semprini, E A. Gianturco, J. Chem. Phys. 1993,99, 1985-1989. [129]R. Lopez, J. A. Sordo, T. L. Sordo, THEOCHEM (J. Mol. Struct.) 1992,86, 443-451. [130]J. Feng, J. Leszczynski, B. Weiner, M. C. Zerner, J. Am. Chem. SOC. 1989, 111, 4648-4655. [I311 J. Leszczynski, E Wiseman, M. C. Zerner, Int. J. Quantum Chem. Quantum Chem. Symp. 1988, 22, 117-125. [132]H.Wang, B. Weiner, M. Frenklach, A Phys. Chem. 1993,97, 10364-10371. [133] C. A. Deakyne, M. Meot-Ner (Mautner), T. L. Buckley, R. Metz, A Chem. Phys. 1987, 86, 2334-2342. (1341 R. W. Strozier, P. Caramella, K. N. Houk, J. Am. Chem. SOC. 1979, 101, 1340-1343. [135]K. N. Houk, N. G. Rondan, P. v. R. Schleyer, E. Kaufmann, T. Clark, J. Am. Chem. SOC. 1985, 107, 2821-2823. [136]E. Kaufmann, S. Sieber, P. v. R. Schleyer, X Am. Chem. SOC. 1989, 111, 121-125. (1371 E. Nakamura, Y. Miyachi, N. Koga, K. Morokuma, d Am. Chem. SOC. 1992,114, 6686-6692. [I381 Y.Amiel in The Chemistry of ll-iple bonded Functional Groups, (Eds.: S . Patai, Z. Rappoport), Wiley, New York, 1983, pp. 341-382. [I391 D. C. Fang, X. Y. Fu, Int. J. Quantum Chem. 1994, 49, 3-10. [140] C. Sosa, H. B. Schlegel, Int. J. Quantum Chem. Quantum Chem. Symp. 1987, 21, 267-282. [I411 C. Sosa, H.B. Schlegel, J. Am. Chem. SOC. 1987, 109, 4193-4198. [I421 L. B. Harding, A. F. Wagner, J. Phys. Chem. 1986, 90, 2974-2987. [I431 T. H. Dunning, L. B. Harding, A. F. Wagner, G. C. Schatz, J. M. Bowman, Science 1988. 240, 453-459. [144] R. Amaud, V. Barone, S. Olivella, A. Sole, Chem. Phys. Lett. 1985, 118, 573-579. [145] M. T. Nguyen, d Phys. Chem. 1988, 92, 1426-1428. [146] C. Pouchan, Chem. Phys. 1987, 111, 87-95. [I471 P. H.Kasai, J. Am. Chem. SOC.1992, 114, 3299-3306. [148] L. Manceron. L. Andrews, J. Am. Chem. SOC. 1985, 107, 563-568. I1491 P. Chaquin, A. Sevin, A. Papakondylis, Chem. Phys. 1990, 143, 39-45. [I501 J. R. Flores, k Largo, J. Phys. Chem. 1991, 95, 9278-9288. [151] J. R. Flores, A. Largo, J. Phys. Chem. 1992, 96, 3015-3021.
30
I Modern Computational and TheoreticalAspects of Acetylene Chemistry
(1521 G. Frenking, R. B. Rernington, H. F. Schaefer 111, J. Am. Chem. SOC. 1986, 108, 2169-2173. [153]M.-D. Su, R. D. Amos, N. C . Handy, J. Am. Chem. SOC. 1990, 112, 1499-1504. [154] J. S. Tse, 1 Am. Chem. SOC. 1990,112, 5060-5065. [ M I Y. Xie, B. F. Yates, H. F. Schaefer 111, J. Am. Chem. SOC. 1990, 112, 517-523. [156]A. C. Scheiner, H. F. Schaefer 111, J. Am. Chem. SOC. 1985, 107, 4451-4453. [157] J. Miralles-Sabater, M. Merchan, I. Nebot-Gil, Chem. Phys. Lett. 1987, 142, 136-141. [l58]P. H. Kasai, J. Am. Chem. SOC. 1982, 104, 1165-1172. [159]Y. Xie, B. F. Yates, Y. Yamaguchi, H. F. Schaefer 111, J. Am. Chem. Soc. 1989,111, 6163-6167. [160]W. B. De Almeida, Chem. Phys. 1990. 141, 297-309. [I611 L. Adamowicz, Chem. Phys. Lett. 1992, 192, 199-204. [162]2. Slanina, L. Adamowicz, THEOCHEM (J. Mol. Struct.) 1992, 95, 133-140. [163]P. G. Jasien, J. Phys. Chem. 1992, 96, 9273-9278. [I641 A.-M. Sapse, D. C. Jain, J. Chem. Phys. Lett. 1986, 124, 517-521. [165]T. Slee, R. R. J. Le, C. E. Chuaqui, Mol. Phys. 1992, 77, 111-134. [166]R. G. A. Bone, J. Phys. Chem. 1994, 98, 3126-3138. [167]G. J. B. Hurst, P. W. Fowler, A. J. Stone, A. D. Buckingham, Int. J. Quantum Chem. 1986,29, 1223-1239. [168]P. A. Block, K. W. Jucks, L. G. Pedersen, R. E. Miller, Chem. Phys. 1989,139, 15-30. [169]C. R. L. Sueur, A. J. Stone, P. W. Fowler, J. Phys. Chem. 1991, 95, 3519-3522. [170]V. Magnasco, C. Costa, G. Figari, Chem. Phys. Lett. 1989, 160, 469-478. [171] V. Magnasco, G.Figari, C. Costa, THEOCHEM (J. Mol. Struct.) 1992,93, 237-253. [172] A. M. Sapse, D. C. Jain, J. Phys. Chem. 1984, 88, 4970-4973. [I731 G. B. Bacskay, D. I. Kerdraon, N. S. Hush, Chem. Phys. 1990,144, 53-69. [174]G. B. Bacskay, Mol. Phys. 1992, 77, 61-73. [175]J. Yu, S. Su, J. E. Bloor, J. Phys. Chem. 1990, 94, 5589-5592. [176]A. J. Stone, C . 3 Tong, Chem. Phys. 1989, 137, 121-135. [I771 P. Hobza, H. L. Selzle, E. W. Schlag, Collect. Czech. Chem. Commun. 1992, 57, 1186-1190. [178]W. B. De Almeida, A. Hinchliffe, Chem. Phys. 1989, 137, 143-156. [179]J. S. Craw, W. B. De Almeida, A. Hinchliffe, THEOCHEM (J Mol. Struct.) 1989,60, 69-74. [180]J. S. Craw, M. A. C. Nascimento, M. N. Ramos, Spectrochim. Acta, Part A 1991, 47a, 69-73. [181] W. B. De Almeida, A. Hinchliffe, J. S. Craw, THEOCHEM (J. Mol. Struct.) 1990, 67, 15-28. [182] W. B. De Almeida, A. Hinchliffe, J. S. Craw, THEOCHEM (1 Mol. Struct.) 1991, 74, 191-199. [183]I. L. Alberts, T. W. Rowlands, N. C. Handy, J. Chem. Phys. 1988, 88, 3811-3816. [184]U. H. F. Bunz, Angew. Chem, Int. Ed. Engl. 1994, 33, 1073-1076. [l85]F. Diederich, Nature (London) 1994, 369, 199-207. [I861 R. Gleiter, D. Kratz, Angew. Chem., Znt. Ed. Engl. 1993,32, 842-845. [187]V. V. Korshak, Y. P. Kudryavtsev, Y. V. Korshak, S. E.Evsyukov, V. V. Khvostov, V. G. Babaev, M. B. Guseva, Macromol. Chem., Rapid Commun. 1988, 9, 135-140. (1881 Y. P.Kudryavtsev, S. E. Evsyukov, V. G. Babaev, M. B. Guseva, V. V. Khvostov, L. M. Krechko, Carbon 1992, 30, 213-221. [189]A. G. Whittaker, Nature (London) 1978, 276, 695-696. [190] A. Baeyer, Ber. Dtsch. Chem. Ges. 1885, 18, 674-681. [191] F. Diederich, Y. Rubin, Angew. Chem., Int. Ed. Engl. 1992,31, 1101-1123. [192] R. Eastmond, D. R. M. Walton, Tetrahedron 1972,28, 4591-4599; R. Eastmond, T. R. Johnson, D. R. M. Walton, ibid. 1972, 28, 4601-4616; T. R. Johnson, D. R. M. Walton, ibid. 1972, 28 5221-5236. [193]T. Grosser, A. Hirsch, Angew. Chem. Znt. Ed. Engl. 1993,32, 1340-1342. [194]0. Hahn, F. Strassmann, J. Mattauch, H. Ewald, Naturwissenschaften 1942, 36, 541-542. [195]H. W. Kroto, J. R. Heath, S. C. O’Brien, R. F. Curl, R. E. Smalley, Nature (London) 1985, 318, 162-163.
References
31
[196] W. KrBtschmer, L. D. Lamb, K. Fostiropoulos, D. R. Huffman, Nature (London) 1990, 347, 354-358. [I971 R. Nesper, K. Vogel. P. E. Blochl, Angew. Chem., Int. Ed. Engl. 1993,32, 701-703. [198] H. R. Karfunkel, T. Dressler, J. Am. Chem. SOC. 1992, 114, 2285-2288. [199] R. H. Baughman, H. Eckhardt. M. Kertesz, J. Chem. Phys. 1987,87. 6687-6699. [200] R. H. Baughman, D. S. Galvgo, C. Cui, Y. Wang,D. Tomanek, Chem. Phys. Lett. 1993,204,8-14. [201] W. J. Hehre, J. A. Pople, J. Am. Chem. SOC.1975,97, 6941-6955. [202] J. H. Callomon, B. P. Stoicheff, Can. 1 Phys. 1957, 35, 373-382. [203] M. Tanimoto, K. Kuchitsu, Y. Morino, Bull. Chem. SOC. Jpn. 1971,44, 386-391. [204] J. S. Binkley, J. A. Pople, W. J. Hehre, Chem. Phys. Lett. 1975,36, 1-5. [205] P. Furet, G. Hallak, R. L. Matcha, R. Fuchs, Can. J. Chem. 1985,63,2990-2994. [206] G. HBfelinger, C. U. Regelmann, T. M. Krygowski, K. Wozniak, J. Comput. Chem. 1989, 10, 329-343. 12071 P. Botschwina, H. Schramm, P. Sebald, Chem. Phys. Lett. 1990,169, 121-126. [208] A. Karpfen, H. Lischka, Chem. Phys. 1986,102, 91-102. [209] S. J. Cyvin, Tetrahedron Lett. 1981,22, 2709-2712. [210] S. J. Cyvin, 1 Mol. Struct. 1982,86, 315-324. [211] S. W. Benson, J. Chem. Educ. 1965,42,502-518. [212] R. S. Mulliken, R. G. Parr, J. Chem. Phys. 1951, 19, 1271-1278. (2131 M. J. S. Dewar, The Molecular Orbital Theory of Organic Chemistry, McGraw-Hill, New York,
1969. [214] P. George, Chem. Rev. 1975, 75, 85-111. (2151 H. Kollmar, J. Am. Chem. SOC.1979, 101, 4832-4840. 12161 J. E. Gready, J. Comput. Chem. 1984,5, 411-426. [217] R. Hoffmann, Tetrahedron 1966,22, 521-538. [218] J. B. Moffat, J; Mol. Struct. 1977, 42, 251-259. I2191 Q. Fan, G. V. Pfeiffer, Chem. Phys. Lett. 1989,162,472-478. [220] W. Weltner, Jr., R. J. Van Zee, Chem. Rev. 1989,89, 1713-1747. [221] F. Diederich, Y. Rubin, C. B. Knobler, R. L. Whetten, K. E. Schriver. K. N. Houk. Y. Li, Science 1989,245, 1088-1090. [222] G. von Helden, P. R. Kemper, N. G. Gotts, M. T. Bowers, Science 1993,259,1300-1302; G. von Helden, N. G. Gotts, M. T. Bowers, Nature (London), 1993,363.60-63;G. von Helden, M. T. Hsu, N. G. Gotts, P. R. Kemper, M. T. Bowers, Chem. Phys. Lett. 1993,204, 15-22. [223] S. W. McElvany, M. M. Ross, N. S. Goroff, F. Diederich. Science 1993,259, 1594-1596. [224] J. M. Hunter, J. L. Fye, E. J. Roskamp, M. F. Jarrold, J. Phys. Chem. 1994,98, 1810-1818. [225] K. S. Pitzer, E. Clementi, X Am. Chem. SOC. 1959,81, 4477-4485. [226] Z. Slanina, R. Zahradnik, J. Phys. Chem 1977,81, 2252-2257. (2271 D. W. Ewing, G. V. Pfeiffer, Chem. Phys. Lett. 1982,86, 365-368. [228] J. Bernholc, J. C. Phillips, J. Chem. Phys. 1986, 85, 3258-3267. I2291 K. Raghavachari, J. S. Binkley, 1 Chem. Phys. 1987,87, 2191-2197. [230] K. Raghavachari, Z. Phys. D 1989, 12, 61-64. [231] J. M. L. Martin, J. P. Francois, R. Gijbels, J. Comput. Chem. 1991, 12,52-70. [232] M. Feyereisen, M. Gutowski, J. Simons, J. Almlof, J. Chem. Phys. 1992,96,2926-2932. [233] J. J. Novoa, M.-H. Whangbo, J. M. Williams, Inorg, Chim. Acta 1992, 198-200,133-138. [234] N. Kurita, K. Kobayashi. H. Kumahora, K. Tago, K. Ozawa, Chem. Phys. Lett. 1992.188, 181-186. [235] C. H. Xu, C. Z. Wang, C. T. Chan, K. M. Ho, Phys. Rev. B 1993, 47, 9878-9880. [236] J. Hutter, H. P. LUthi, F. Diederich, 1 Am. Chem. SOC. 1994, 116, 750-756. [237] C. W. Bauschlicher, Jr., S. R. Langhoff, J. Chem. Phys. 1987,87, 2919-2924. [238] G. E. Scuseria, T. P. Hamilton, H. F. Schaefer 111, J Chem. Phys. 1990,92,568-573. 12391 J, D. Watts, R. J. Bartlett, J. Chem. Phys. 1992, 96, 6073-6084.
32
I Modern Computational and Theoretical Aspects of Acetylene Chemistry
[UO] R. A. Whiteside, R. Krishnan, M. J. Frisch, J. A. Pople, P. v. R. Schleyer, Chem. Phys. Lett. 1981, 80, 547-551. [241] W. P. Kraemer, P. R. Bunker, M. Yoshimine, J. Mol. Spectrosc. 1984, 107, 191-207. [242] P. Jensen, C. McMichael Rohlfing, J. Almlof, J. Chem. Phys. 1992, 97,3399-3411. [243] J. Kurtz, L. Adamowicz, Astrophys. J. 1991, 370,784-790. [244] C. Liang, H. F. Schaefer 111, Chem. Phys. Lett. 1990, 169, 150-160. [245] R. A. Whiteside, R. Krishnan, D. J. Defrees, J. A. Pople, P. v. R. Schleyer, Chem. Phys. Lett. 1981, 78,538-540. [246] D. H. Magers, R. J. Harrison, R. J. Bartlett, J. Chem. Phys. 1986, 84, 3284-3290. [247] J. P. Ritchie, H. F. King, W. S . Young, L Chem. Phys. 1986, 85, 5175-5182. [248] J. M. L. Martin, J. P. Francois, R. Gijbels, J. Chem. Phys. 1991, 94,3753-3761. [249] D. E. Bernholdt, D. H. Magers, R. J. Bartlett, J. Chem. Phys. 1988, 89, 3612-3617. [250] V. Parasuk, J. Almlof, 1 Chem. Phys. 1991, 94,8172-8177. [251] D. W. Ewing, Z. Phys. D 1991, 19,419-422. [252] D. W. Ewing, G. V. Pfeiffer, Chem. Phys. Lett. 1987, 134,413-417. [253] J. M. L. Martin, J. P.FranGois, R. Gijbels, J. Chem. Phys. 1991, 95, 9420-9421. [254] Z. Slanina, J. Kurtz, L. Adamowicz, Chem. Phys. Lett. 1992, 196,208-212. [255] G. von Helden, W. E. Palke, M. T. Bowers, Chem. Phys. Lett. 1993, 212, 247-252. [256] K. Raghavachari, R. A. Whiteside, J. A. Pople, J. Chem. Phys. 1986, 85, 6623-6628. [257] V. Parasuk, J. Almlof, J. Chem. Phys. 1989, 91, 1137-1141. [258] J. Hutter, H. P. LOthi, J. Chem. Phys. 1994, 101,2213-2216. [259] R. J. Van Zee, R. F. Ferrante, K. J. Zeringue, W. Weltner, Jr., D. W. Ewing, J. Chem. Phys. 1988, 88, 3465-3474. [260] V. Parasuk, J. Almlof, Theor. Chim. Acta 1992, 83, 227-237. [261] Z. Slanina, J. Kurtz, L. Adamowicz, Mol. Phys. 1992, 76,387-393. [262] C. Liang, H. F. Schaefer 111, J. Chem. Phys. 1990, 93, 8844-8849. [263] J. D. Watts, R. J. Bartlett, Chem. Phys. Lett. 1992, 190, 19-24. [264] J. M. L. Martin, J. P. Francois, R. Gijbels, J. Almlof, Chem. Phys. Lett. 1991, 187, 367-374. [265] T. F. Giesen, A. Van Orden, H. J. Hwang, R. S. Fellers, R. A. Provencal, R. J. Saykally, Science 1994, 265, 756-159. [266] V. Parasuk, J. Almlof, M. W. Feyereisen, J Am. Chem. Soc. 1991, 113, 1049-1050. [267] D. A. Plattner, K. N. Houk, J Am. Chem. SOC. 1995, 117, 4405-4406. [268] R. Seeger, J. A. Pople, J. Chem. Phys. 1977, 66,3045-3050. [269] P. Carsky, 1. Hubac, Theor. Chim. Acta 1991, 80, 407-425. [270] R. E. Peierls, Quantum Theory of Solids, Oxford University Press, London, 1955. [271] R. Hoffmann, Solids and Surfaces: A Chemist’s View of Bonding in Extended Structures, VCH, New York, 1988, p. 92. [272] L. Salem, The Molecular Orbital Theory of Conjugated Systems, W. A. Benjamin, Reading, MA, 1974, p. 466.
2 Functionalized Acetylenes in Organic Synthesis - The Case of the 1-Cyano- and the 1-Halogenoacetylenes Henning Hopi Bernhard Witulski
2.1 Introduction There is no other functional group in organic chemistry that can compete with the carbon-carbon triple bond in richness and diversity of chemical reactivity. Whether elements of the left-hand side of the periodic table or of the right are involved, whether electrophilic or nucleophilic additions, stepwise or concerted reaction modes, transformations initiated by heat, light, or catalysts - the triple bond can always participate. That triple bonds come so close to an “ideal functional group” is not really surprising: with organic chemistry being the chemistry of carbon compounds, what would be better qualified as the ideal substrate than the naked, highly reactive carbon atom to which any partner could be added, any bond be attached? This being the case, it appears as a hopeless enterprise to write an invited chapter on “Functionalized Acetylenes” in a multi-author volume - which in addition to the reactive diversity of the triple bond demonstrates impressively the use of alkynes for constructive i. e., molecular-engineering work in organic chemistry - and keep it of reasonable and comprehensible length. Rather than even attempting complete coverage of functionalized alkynes we have therefore decided to select triple bond systems carrying two types of functional groups: the cyano group and the halogens. Both of these acetylene derivatives are anything but new. We nevertheless focus on them at the present time for several reasons: the cyano acetylenes have been discovered as important contributors to the growing number of organic compounds in interstellar space, and nearer home, “on earth”, they have been shown to be very useful starting materials in organic synthesis with a synthetic potential which still leaves a lot of room to be developed. That this should be so comes as no surprise, since - after all - the parent system, cyanoacetylene is a composite of two of the most important industrial chemicals, namely acetylene and hydrogen cyanide. Although the halogenoacetylenes so far are mostly known and feared for their explosive nature, we believe that they - with appropriate care and the necessary experimental technique - could be developed into very useful synthetic intermediates, especially since they are so readily available from inexpensive substrates. We would therefore like to convince our colleagues that these interesting compounds should be used more often in their future work, be it as starting materials in complex syntheses, as building blocks for acetylene scaffolding, or to generate highly reactive intermediates from them. It is our opinion that, after lying dormant for several decades, these highly reactive acetylenes are about to experience a renaissance. In toto, functionalized acetylenes have not been summarized recently [l]. However, there are some excellent reviews which may be consulted in order to become familiar with this class of unsaturated compounds. Monographs which have become classics in the field of acetylene chemistry have been compiled by Raphael [2], Rutledge [3], and Viehe [4]. There are furthermore excellent volumes on acetylenes in the Houben-Weyl-Muller [ 5 ] and in the Patai series
34
2 Functionalized Acetylenes in Organic Synthesis
[6].In the last two cases it is recommended to search the volumes on halogens [7] and nitriles [8,91 as well if one is interested in the title compounds. For all practical purposes of acetylene chemistry, the experimental procedures developed or collected by Brandsma [lo] are unsurpassed: $ a Brandsma procedure cannot be reproduced it is clearly the experimentalist’s own fault. To our knowledge no comprehensive review on the spectroscopic and structural properties of functionalized alkynes has appeared. Many of these, often rod-like and electronically unusual, compounds have been of great interest to spectroscopists for a long time.
2.2 Synthesis and Preparative Use of Cyanoacetylenes 2.2.1 Synthesis Cyanoacetylene (l),dicyanoacetylene (2) and dicyanodiacetylene (3) may be called the three “basic systems”, since 1 and 2 are the simplest conceivable derivatives possessing one and two cyano functions, respectively, and 3 is the simplest cyanocarbon that may be produced by oxidative coupling, a technique often used in this area of organic synthesis. H-C=C-CN
NC-CGC-CN
I
NC-CGC-CEC-CN
2
3
These three compounds have been known for more than 85 years, and during the intervening decades a sizeable number of methods to prepare them have been developed [ll]. This is particularly true for 1; however, in our hands its original synthesis [12], which involves the dehydration of the amide 5 of propiolic acid 4 with phosphorous pentoxide [Eq. (l)], is still the most convenient and reliable one, especially when the variant of this sequence introduced by Franck-Neumann (dehydration by a suspension of phosphorous pentoxide in sulfolane, and removal of 1 as it is formed by vacuum transfer [13]) is employed.
I
6
4
Cyanoacetylene (1) may be prepared by this approach in ca. 90% yield in multigram quantities [14]. The compound can be stored for extended periods of time in the deep freeze, and we never experienced an uncontrolled reaction during handling. The toxicity of 1 is unknown; because of its structural resemblance to hydrogen cyanide (of which it is a “stretched” version) we recommend working with 1 only under a well-ventilated hood. Several methods have also been described for the preparation of dicyanoacetylene (2) [ll]. In our experience the original one [15], which involves dehydration of the diamide 7 of acetylenedicarboxylic acid 6 with phosphorous pentoxide [Eq. (2)], is the method of choice if gram quantities of 2 are needed. HOZC-CEC-CO~H
6
-
-H20
HzNOC-C=C-CONH2
7
NC-C-C-CN
2
(2)
2.2 Synthesis and Prepamtive Use of Cyanoacetylenes
35
For larger amounts we recommend the gas-phase pyrolysis of 4,5-dicyano-1,3-dithiol-2-one (8) [Eq. (3)], which is readily available from sodium cyanide, carbon disulfide, and phosgene [16]. Both the starting materials and most of the pyrolysis products are extremely toxic, and the price to be paid for large-scale production is hence high.
Ncn,>o 6oo-aoo0c
NC-C=C-CN
+
CO
+
COS
+
CS2
+
S
(3)
NC
2
8
The diacetylene 3 is best obtained by oxidative coupling of the copper salt of cyanoacetylene 9 [12, 14, 15, 171 [Eq. (4)]. Although the raw yield of this simple reaction is high (>8O%),
much of the product is lost during its purification [14].
101
NC-C=C-Cu
NC-CEC-CEC-CN
(4)
3
9
Higher ethynylogs of both 1 and 2/3 are known. Thus cyanodiacetylene 12 has been obtained by reacting the doubly protected butadiyne 10 with cyanogen chloride to form the trimethylsilyl derivative 11, from which the parent system 12 can be liberated by deprotection over alumina [18] [Eq. (5)]. CI-CN
Et3Sn -CEC-C?C-SiMe3
-
NC-CEC-CGC-SiMe3
AIC13. CH2Clz
A1203
I1
10 N C -C
-
C-C
C -H
(5)
12
1-Cyanohexatriyne (13) has been prepared analogously [19]. Both 12 and 13 are already very unstable cyanoalkynes, and that they may be used for synthetic purposes appears very unlikely at present. They are of interest for spectroscopic reasons [18, 191 since they have been detected - as has 1 - in interstellar space [20]. NC-CEC-CEC-CEC-H
13
The dicyanopolyacetylenes 14-18, which are also of importance in connection with interstellar chemistry and novel forms of carbon, have been obtained very recently by vaporizing graphite in the presence of cyanogen under Kratschmer-Huffman conditions [21]. The polyyne fraction obtained consisted of 55% 14, 35% 15 and 10% of the higher homologs 16-18 as analyzed by HPLC. Cyanocarbon 14 has been obtained in analytically pure form; as expected, it is only stable in dilute solution under exclusion of light and moisture at low temperatures [21].
2 Functionalized Acetylenes in Organic Synthesis
36
NC-C+C+CEC-CN n
2 3
15
16 17 5 18 4
The preparation of alkyl, alkenyl, and aryl derivatives of cyanoacetylene (1) poses no particular problems. In most cases a terminal acetylene is metalated or converted into an alkynyl Grignard reagent, and these intermediates are subsequently intercepted by a “cyan0 source”, which in most cases is either cyanogen chloride or bromide and sometimes cuprous cyanide or phenyl cyanate. Of course, dehydration of an acetylenic amide as described for the parent molecules is also possible and has occasionally been employed. Table 2-1 - without attempting to be comprehensive - gives a selection of these preparations from the literature. Table 2-1 A selection of methods for the preparation of substituted I-cyanoacetylenes R in R - C = C - H
Metalated intermediate
CN source
R-C=C-CN yield (70)
Ref.
Me, Et, Pr, n-Bu R-CSHI~,CyClohexyl, cyclohexenyl, tert-Bu, Ph Ph tert-Bu Ph
R - C = C -Li R - C = C -Li
CI - CN Ph-OCN
85-92 70-80
P21 ~ 3 1
Ph - C r C -CU Me$ - C CMgBr Ph - C = C -CU
Br - CN CONH, --* CN CuCN
60 66
[241 t251
70
WI
Both the triethylstannyl (19) and the trimethylsilyl (20) derivatives of cyanoacetylene (1) were prepared several years ago. Derivative 19 is available by reacting bis(triethylstanny1) ether with 1 in the presence of calcium hydride in benzene [27], and 20 can be obtained from 19 by trimethylsilyl bromide treatment in hexamethylphosphoramide (HMPA) [28] [Eq. (6)]. In a more recent method for the preparation of 20, the commercially available 21 is first converted into the doubly protected 22 which, on treatment with cyanogen chloride, loses its triethylstannyl group preferentially to yield the desired 20 [29]. benzene. CoH+
H-CEC-CN
(E
-EC-CN
I
19
-
19
MeSSiBr
HMPA.
Me3Si-CEC-CN
A
20
MeSSi - C E C - H
21
-
1
CI-CN
1
Me3Si -C3C-SnEt3
22
2.2 Synthesis and Preparative Use of Cyanoacetylenes
37
The halocyanoacetylenes 23-25 have been prepared by standard methodology [Eq. (7); see below] from cyanoacetylene (1)[30-321. Although they have been investigated in detail from the structural and spectroscopic viewpoint, their chemistry remains largely unexplored. These compounds are not particularly unstable, and can thus be handled by more or less routine laboratory techniques. Of particular interest should be their behavior in high-temperature pyrolysis as well as their photochemical behavior. Metal-catalyzed oligomerization could turn out to be another interesting field of application of these functionalized cyanocarbons.
[Hal’]
H-CEC-CN
Hal-CEC-CN
(7)
I
For the preparation of fluorocyanoacetylene (27) a different approach has been used: decomposition of various fluorinated heteroaromatics by electrical discharge [33]. The best yields - which still did not exceed a few percent - were realized when perfluorobenzonitrile (26) was subjected to plasma conditions [Eq. (8)]. CN
26
The hydroxy derivatives 29, a preparatively most valuable (see Section 2.2.2) class of functionalized cyanoacetylenes, are obtained when the propargylic alcohols 28 are first subjected to the Strauss reaction and the resulting bromoacetylenes are subsequently treated with cuprous cyanide in dimethylformamide [34, 351 [Eq. (911. HO *i.)-CEC-H
1. Br2/OH-
-
2. CUCN, DMF, 50°C
R2
HO R i . ) - ~ ~ (9)~ - ~ ~ ~2
29
28
The cyanoacetylenes30 and 31,which are derivatives of 12 and 13,have been described [36], as has the carboxylic acid 32, a natural product (“diatetryne”) isolated from Clytocybe diutretu and which possesses antibiotic activity [37].
H~C-c~C-c~C-ciEC-cN
31 HOzC-CH =CH-C G C-C E C - C N
32
38
2 Functionalized Acetylenes in Organic Synthesis
2.2.2 Preparative Use of Cyanoacetylenes 2.2.2.1 A Short Summary of the Older Literature In their review on cyanocarbon and polycyano compounds Ciganek et al. [ll a] wrote in 1970: “Although their [i. e., the cyanoacetylenes] physical properties have received considerable attention, reports on the chemistry of these compounds are not as abundant as might be expected in view of their high reactivity”. For nearly a quarter of a century this situation has not changed very much; in fact, it was one of the reasons why we began our own work in this area. Among the classical (summarized in [ll]) reactions of cyanoacetylenes (salt and n-complex formation, nucleophilic addition of, inter afia, amines and alcohols, Diels- Alder additions with 2 (see Section 2.2.2.2),addition of halogens and hydrogen halides, etc.) their polymerization might deserve a second look since the products formed - polyacetylenes - have attracted much attention during the last two decades. Thus 1, several of its derivatives, and 2 have been polymerized with anionic initiators (triethylamine, sodium cyanide, butyllithium) to give black, low-molecular-weight polymers claimed to have structures like 33 138-401, which is obtained from 2 by treatment with butyllithium.
33
A structurally related compound, the stable alkylidene-l,6-diphosphome 35, has been obtained by treating 2 with triphenylphosphine, with the betaine 34 postulated as an intermediate in this process. The “trimerization” is accompanied by the formation of “polymerized acetylene” of unknown structure [41, 421 M.(lo)]. Ph3PI
NC-CZC-CN
+
- [“.$q
__Ic
NC
2
34
+
PhsPO\
’acetylenic polymer’
(10)
NC
PPh30 NC
36
That nitriles derived from propargyl alcohols are versatile substrates in fine organic chemistry has already been mentioned. The actual richness of this chemistry may be deduced from Scheme 2-1, which has been adapted from a recent review article [35]. Whether four-, five- or six-membered, polyfunctional heterocycles all are readily obtained from 29-derived starting materials..
2.2 Synthesis and Prepamtive Use of Cyanoacetylenes
39
R’ = R2 = olkyl. cycloolkyl
Scheme 2-1 Preparative uses of cyanoacetylenes derived from propargylalcohol according to [35].
- Simple and Efficient Methods for the Construction of Complex Carbon Frameworks
2.2.2.2 Novel Cycloadditions with Cyanoacetylenes
Like the esters of acetylenedicarboxylic acid (6),dicyanoacetylene (2) has often been employed in cycloaddition reactions; it participates in Diels-Alder and 1,3-dipolar cycloadditions, and ene reactions as well as homo additions are also known [9, 111. Although kinetic measurements with 2 seem not to have been performed so far, from “working knowledge” it appears that 2 is a more reactive dienophile than any other activated acetylene including hexa-
40
2 Functionalized Acetyiena in Organic Synthesis
fluoro-Zbutyne. It is thermally more stable than the (also less reactive) acetylenedicarboxylates, which are normally more difficult to remove during work-up than 2 when they have been used in excess. The reactivity of 2 in [2 + 41 cycloadditions is emphasized by its ready addition to aromatic “dienes”, although benzene itself requires further activation by either heat or Lewis acid catalysis [43]. When the strain-activated [44] [2.2]paracyclophane (36)is reacted with 2 at 12OoC, the 1 :1 adduct 37 is formed as the major product (Scheme 2-2). Increasing the addition temperature to 170“C causes formation of the 2: 1 Diels-Alder adduct 39. In fact, originally the product isolated was thought to possess the isomeric structure 40 [43]. However, recent X-ray structural evidence shows 39 to be correct [45]. The adducts 37 and 39 are derivatives of barrelene and double barrelene [46], respectively, and as such might be expected to undergo interesting isomerization reactions. A first experiment involving 37 indeed shows that photoisomerization to the “semibullvalenophane” 38 may be accomplished [45, 471.
& 6 NC
-
120 “C
hw
i + 0 benzene
CN
0
I
+ If C I
36
CN
2
0
0
toluene
37
38
CN
1
170 OC
benzene
-
CN
NC
CN
39
NC
40
Scheme 2-2 The thermal addition of dicyanoacetylene 2 to [2.2]paracyclophane 36.
Compared with 2, cyanoacetylene (1) has hardly been used so far as a dienophile. Since from the reactivity viewpoint it should not differ very much from 2, but since its adducts should be easier to convert into the corresponding hydrocarbons, we have started a comprehensive program aimed at elucidating the scope and limitations of the use of 1 and its derivatives as a partner in thermal, photochemical and catalyzed addition reactions. To our surprise, our very first experiment [46], the reaction of 1 with [2,2]paracyclophane (36), did not give the expected product(s), i. e., mono and dinitriles corresponding to 37 and 39,but adducts of very different structure, as shown by NMR spectroscopy and X-ray structural analysis: the 2: 1-adducts 41-44 [Eq.(ll)]. In fact “normal” Diels-Alder products were not obtained at all in this experiment. Rather, the formation of the four-membered ring compounds - which are all derivatives of the Nenitcescu hydrocarbon, the formal 1P-adduct of cyclobutadiene to benzene - was accompanied by formation of varying amounts of 1,2,4- (51)
2.2 Synthesis and Prepamtive Use of Cyanoacetylenes
41
A
8+ w + NC
0
1
+
36
41 ( 1 5 % )
42 (6%)
43 (6%)
44 (0.5%)
L
22 h, 160 OC benzene
+
benzonitriles
and 1,2,3-tricyanobenzene (52), with the 1,3,5-isomer (53; see Scheme 2-3) conspicously lacking. To rationalize this outcome we propose (Scheme 2-3) that, on heating, 1 first undergoes a [2 + 21 cycloadditionleading to a cyclobutadiene dicarbonitrile which can exist in the form of H-EC-EN
I x 21A
47
40
49
50
I I no:
CN
1
CN
A
CN
I
52 Scheme 2-3 Thermal di- and trimerization of cyanoacetylene 1.
42
2 Functionalized Acetylenes in Organic Synthesis
two valence tautomers, 45 and 46. These highly reactive intermediates may either be trapped by 36 to provide adducts 41-44 or react with another molecule of 1 to lead to the isomeric Dewar benzenes 47-50. Under the reaction conditions these thermally ring-open to 51 and 52, a process well known to occur on heating Dewar benzene derivatives. As can be seen from Scheme 2-3, there is no pathway to the (not observed) 53; interestingly, 51 and 52 are the only adducts produced if the reaction is carried out in the absence of a trapping reagent. The mechanism is further supported by the observation that 47 is produced in excellent 62% yield when 1 is subjected to a pressure of 13 kbar at 40°C [48]. The above sequence of events also takes place with other cyclophanes [14, 47, 491, as illustrated in Scheme 2-4 with the benzofuranophane 54 as just one additional example. Here the primary adduct 55 may either stabilize itself to the ortho,para-cyclophane 56 or ring-open to the tropylidenophanes 57 and 59. We assume that the latter interconvert thermally by 1,5-hydrogen shifts since the necessary intermediate, 58 (not isolated), can be trapped via its norcaradiene valence isomer 61 to the 3 : 1 adduct 62, a process which is accompanied by the formation of an isomer of 56, the para-dinitrile 60:
0
H;Zl-CN
benzene -I
L
54
66
66
NC
O
57S
N
*
58
It
CN
L
60
61
62
Scheme 2 4 Cycloaddition of cyanoacetylene 1 to benzofuranophane 54.
That strain activation is no prerequisite for a successful addition is shown by offering classical aromatics to cyanoacetylene (1). Thus anthracene (63) provides the expected adducts 65-67 as well as the secondary dinitriles 68 and 69 (Scheme 2-5). As shown by control experiments, 68 is produced by the addition of the diene 63 to 65, and 69 is produced from 67 by electrocyclic ring-opening. The “classical” [2 + 41 cycloaddition product 64 is observed in this case for the first time, and the trimers 51 and 52 are produced in up to 30% yield [14, 471.
43
2.2 Synthesis and Preparative Use of Cyanoacetylenes
8
&+
H-CSC-
64 (50%)
CN
1
22 h. 160 OC
\ /
benzene
65(22%)
63
I
67 (3%)
‘CN
+65
J
+
A
69 (3%) +benzonitriles
Scheme 2-5 Cycloaddition of cyanoacetylene 1 to anthracene 63.
That anthracene participates in Diels-Alder addition has, of course, been known for a long time. Still, even monocyclic aromatics like hexamethylbenzene (70)and p-xylene (73) react with 1 with formation of the cyclobutadiene adducts 71/72 and 74/76, respectively, although the yields have now become very low indeed - a clear reflection on the decreased “diene character” of 70 and particularly 73 [Eqs. (12), (13)].In the latter cases the trimerization of 1 has become by far the dominating process (combined yield of 51 and 52: 42%) [14,471.
$
H-CIDCN
- %+
N * \
22 h, 160 OC
\
benzene
70
$ 73
71 (29%)
H- CEC-CN
x
+
N
72 (9%)
c
p
+
F
C CN
1 22 h, 160 ‘C
N
(1 3) \
\
benzene
74
75
(combined yield 3%)
76 (0.03%)
Many of the above adducts have turned out to be interesting substrates for further transformations, as illustrated by the following representative reactions. A mixture of the adducts 41/42 is thermally isomerized to the dihydrobenzonapthalenophane 77,from which the fully
44
2 Functionalized Acetylenes in Organic Synthesis
N
41/42
C
DDQ, A
e
180 OC benzene-
Chloro benzene
(30%)
(40%)
N &
00
77 78 Scheme 2-6 Preparation of naphthalenophanes 78 from cyanoacetylene adducts 41/42.
aromatized system 78 can be obained by dehydrogenation with (DDQ) [14,471 (Scheme 2-6). Not surprisingly, adducts 41 and 71 undergo photochemical ring-closure between their most proximate double bonds to yield the basketene derivatives 79 and 80, respectively [14,471 m.(141,( ~ 1 . NC
c
A >300 n m toluene
41
79 CN
A >300 ' . ' - n&m( 1 5 ) toluene
The most varied and interesting chemical behavior of any of the adducts between 1and'an aromatic diene is displayed by the anthracene-derived dinitriles 65 and 67. On heating, these isomers ring-open to 81 and 69, which on either direct or sensitized irradiation are partly reconverted to their cyclobutene isomers 65 and 67, partly to the dibenzoisobullvalene 82 and the dibenzobullvalene 83, respectively, thus allowing a completely new entry into this class of polycyclic hydrocarbons [14,471 (Scheme 2-7). Furthermore, 82 with its vinylcyclopropane subunit is a born candidate for a thermal ring-enlargement reaction. This takes place on heating it to 230°C and the dibenzotriquinacene 84 is formed in excellent yield. Small amounts of the lumibullvalene derivative 85 are also produced [14,471 (Scheme 2-8). Considering the two structurally very simple starting materials, 1and 63, it is indeed-astonishing what level of molecular complexity can be reached from them by a sequence of straightforward thermal and photochemical addition and isornerization processes. Thrning again to the addition of dicyanoacetylene (2) to cyclophanes, novel polycyclic structures with interesting electronic properties are also obtained with the heterophane 54 (Scheme 2-9). When these two components are heated in benzene at 160°C the substituted phane system 87 is produced. That this Diels-Alderhetro-Diels- Alder process involves the initial formation of 86 is made likely by the production of this primary adduct in a highpressure experiment at room temperature and 8 kbar [48].If, however, 54 is treated with 2 in
2.2 Synthesis and Prepamtive Use of Cyanoacetylenes
65 (BOX)
1
-
67
45
60 X
A
CN
hu
h >300 nrn CDC13
/‘
hu
h >300 nrn acetone (65 X )
hu
h >300 nrn acetone
(36 2)
a3 Scheme 2-7 Thermal and photochemical isomerizations of the anthracene adducts 65 and 67.
82 230 O C I d toluene
CN
Scheme 2-8 Ring enlargement of 82 by vinylcyclopropane rearrangement.
a 5 M lithium perchlorate/diethyl ether solution at room temperature, the [2 + 21 adduct 88 is formed as the primary product in 50% yield. Heating the latter in toluene results in the formation of 89 (68%), which was the first oxepinoparacyclophane to be reported and which owes its beautiful deep-red color to a charge-transfer interaction between its “aromatic” donor and its “antiaromatic” acceptor subunits [14, 47, 491. Lithium perchlorate accelerated
46
2 Functionalized Acetylenes in Organic Synthesis
a I&] NC-CYC-CN
d
0
I
54
c
160 'C
benzene
-
*
0
C2H2
87
86
N
.
-
NC-CZ C-CN
160 'C 5 M LiCIO, / Et20 r. t.
C
0
*
toluene
88
0 89
Scheme 2-9 Cycloadditions of heterophane 54 with dicyanoacetylene 2.
Diels-Alder reactions with 1 and 2 and their derivatives will very likely be of growing importance in the future. The photochemical behavior of 1 and 2 as well as photoadditions of these nitriles to other unsaturated systems have been studied recently. Thus the photolysis of 1 with 185- or 206-nm light yields 1,3,5-tricyanobenzene (53; see above), while 254-nm radiation leads to a mixture of tetracyanocyclooctatetraenes, 1,2,4-tricyanobenzene(51), and 53. Photolysis of mixtures of 2 and acetylene with either 185- or 206-nm light yields 1,Zdicyanobenzene and (E/Z)-1-buten3-yne-l,4-dicarbonitrile.These products are also obtained with 245-nm light, along with a mixture of tetracyanocyclooctatetraenes. Finally, the photolysis of 1 and 2 in the presence of ethylene with 185-nm light provides I-cyanocyclobutene and 1,2-dicyanocyclobutene, respectively. When light of longer wavelength is employed in this latter case (254 nm), only the ringopened products 2-cyano-1,3-butadiene and 2,3-dicyano-l,3-butadieneare formed. As in the thermal processes (see above), cyclobutadiene intermediates are postulated to rationalize some of these findings [50].
2.2.2.3 Cyanoacetylenes as Precursors for Reactive and Interstellar Intermediates As already mentioned, cyanoacetylenes are important components of interstellar matter [20, 511, and it appears likely that besides 1 and its higher ethynylogs, polyyne dinitriles derived from 2 are also present in cosmic space. It is also conceivable that other highly reactive species derived from these cyanocarbons are produced in interstellar space, and actually the cyanoethynyl radical CCCN' has been detected as an interstellar species [52]. To find experimental evidence for these assumptions, neutralization-reionization mass spectrometry is often the method of choice. In fact, this technique has been of particular importance for the generation of functionalized acetylenes which have been sought and discussed in acetylene chemistry for a long time: HO - C = C -OH, H2N- C =C - NH,, H2N - C = C -OH, and some of their derivatives are the most prominent representatives in this context 1531. In solution these compounds are inaccessible because of solvent-induced rearrangements.
2.2 Synthesis and Preparative Use of Cyanoacetylenes
NC-CZC-CN
2
-E -
CCN'
+
CCCN'
+ CN*
CCCCN'
+
47
CCN*
(16)
N-
70 eV
NC-(C=C)*-CN
CCCCCN'
+ CN.
(17)
3
As described already some time ago [54], 70-eV electron impact ionization of 2 [Eq. (16)j and 3 [Eq. (17)] affords by direct cleavage processes the complete series of C,N+ ions. On the other hand, the polycarbon nitride radicals C,N' with n = 2-5, thought to be of prime importance in the genesis of interstellar organic molecules, are obtained in the gas phase by neutralization of the corresponding C,N ions (n = 2 - 5 ) using neutralization-reionization mass spectrometry employing tandem mass spectrometric methods. Furthermore, collision-induced dissociation reactions of mass-selected C,N+ ions support the notion that these species are "carbon rods" bearing a nitrogen atom at one terminus [55]. In contrast to neutral dicyanoacetylene (2) and its highly energized radical cation, the slow dissociations (metastable ion) of 2 ' are dominated by the extrusion of molecular nitrogen and concomitant generation of a C:* cluster, 91, of unknown structure [56] [Eq. (18)]. For this result to occur a substantial geometry change must take place on ionization of 2, such that the terminal nitrogen atoms are eventually permitted to form a nitrogen-nitrogen bond. It is not unreasonable to speculate that nitrogen extrusion from the metastable 2'' intermediate is another example of the operation of an electrocyclic process, involving the intermediate 90 in this case, which seems to be greatly facilitated by the open-shell nature of the precursor species. Interestingly, in the metastable ion spectrum of ionized dicyanodiacetylene 3 + * there is no signal due to the elimination of N,. Rather, this mass spectrum is dominated (> 95 9'0)by the thermodynamically favored loss of CN' to generate an ion C5N+, a species whose gas-phase chemistry has been studied recently in great detail [57]. +
ii
1+=
C
I il C I
-e'
N
2
2
+.
90
91
As these few examples demonstrate, cyanoacetylenes are indeed useful and versatile precursor molecules for the generation of highly reactive molecules and clusters, and it can safely by assumed that novel insights into the mechanisms of formation of these species and their electronic structures will come forth, with more - and structurally new - cyanocarbons becoming available.
48
2 Functionalized Acetylenes in Organic Synthesis
2.3 Synthesis and Preparative Use of 1-Halogenoacetylenes 2.3.1 Older Review of the Literature on Halogenoacetylenes As mentioned in the Introduction (Section 2.1), I-halogenoacetylenes are no newcomers to organic chemistry; this may be illustrated by chloroacetylene, which was described first in 1908 [58] and dibromoacetylene, which is yet three years “older” [59]. Over the years all the “basic systems” (see below) have been prepared, and their chemistry has been developed. The review literature already mentioned contains summaries of varying coverage depth [2-7, 101; a review dedicated to the haloacetylenes exclusively and considering the literature up to 1966 has also appeared (601. In our opinion the chapter by Viehe and Delavarenne in [4] is outstanding both in organization and wealth of information. Inter alia, it contains extended tables which present all 1-halogenoacetylenesprepared up to 1969. The Houben- Weyl-Miiller volume [7] covers the literature up to 1976/1977. Hence there exists a 20- years period in which - as far as we know - the haloacetylenes have not been reviewed comprehensively. Again, as stated for the 1-cyanoacetylenes (see above) we cannot fill this gap in a monograph of the present type. Rather, we intend to show that these functionalized alkynes are of importance in their own right, for structural reasons (spectroscopic model compounds), and for preparative chemistry in general.
2.3.2 Synthesis of 1-Halogenoacetylenes 2.3.2.1 The Preparation of the 1-Halogeno- and lJ-Dihalogenoethynes
The 1-halogenoethynes 92-95 and the 1,2-dihalogenoethynes 96-105 are the parent systems of the halogenoacetylenes. All these derivatives have been prepared, most of them decades ago (see above).
H-CGC-Hal
Hal F
92
CI
93
Br
94
I
95
Ha11-C=C-Ha12
Hal‘ Hal2
F
C1
Br
I
F
96
97
98
99
CI Br
I
100
101 102 103 104 105
2.3 Synthesb and Preparative Use of I-Halogenoacetylenes
49
As a rule, the simple halogenoacetylenes are strong lachrymators, they are poisonous (although a systematic investigation of their toxicity has not been reported), and they are unstable, sometimes explosive, compounds. Among the dihalogenoacetylenes only diiodoacetylene is comparatively stable [61]. Since their synthesis has been reviewed elsewhere [4,601 in detail, only a general scheme due to Kloster-Jensen, the leading worker in this area, providing a variety of dihalogenoethynes and illustrating the typical preparative procedures employed, is presented here (Scheme 2-10). Although the reactions look simple L‘onpaper” - with structurally simple and readily available starting materials such as 1,2-dichloroethene (106), its dibromo analog 107, and acetylene (108), and straightforward reaction conditions - their practical realization represents a major synthetic achievement [61].
-
-
PhLi. ether CIHC=CHCI
ooc
PhLi
[CI-CEC-H]
CI-CEC-Li
106 1. PhLi 2. Clz, heptane, -50°C
I
CI - C E C - B r
CI -C_C-CI
BrHC=CHBr
107
CI - C f C - I
I02
101
I00
above
-
[Br-CEC-H
]
-
2 PhLi
[Li-cGc-Li
94
1
1
1. LINH,. Nz 2. 12. -50°C I -CGC-I
Br -C=C-I
104 H-C
EC-H I08
9
LLi-CGC-Li]
105
Br2
ether, -7O’C
Br-CeC-Er
103
Scheme 2-10 The preparation of 1,2-dihalogenoethynesfrom 1,2-dihalogenoethenes according to [61].
The last two derivatives to yield to synthetic efforts were 98 and 99. For the former, several approaches have been described. A “classical” one starts with the commercial product 109, which was converted to 98 via the intermediate halogenoethenes 110 and 111 [62], and a “modern” one starts from l,l-dibromo-2,2-difluoroethene(112),which was either metalated to the alkali derivative 113 or the tin compound 114 (Scheme 2-11). Loss of alkali fluoride or trimethyltin fluoride (by flash vacuum pyrolysis) then leads to the desired 98 [63]. Finally, iodofluoroacetylene (99) was obtained by applying the Bieri method (electrical discharge; see above) to an appropriate aromatic precursor, 115 in this case [64], as represented in Eq. (19). With few exceptions (see below) the simple halogenoacetylenes 92-105 have not been used for synthetic purposes. They have, however, played an important role in structural and theoretical chemistry as well as in spectroscopy. From a large number of studies in this area only a few can be cited here, by means of leading references: 1-chloroacetylene (93) and
50
2 Functionalized Acetylenes in Organic Synthesis
I-bromoacetylene (94) [65], 1,2-difluoroacetylene (96) [66], I-chloro-2-fluoroacetylene (97) and 1-bromo-2-fluoroacetylene(98) [67], 1-fluoro-2-iodoacetylene(99) [64],and 1,2-dichloroacetylene (100) [68, 691. BryF
KOH/H20
Br
-
100°C
1. Br2
2. base
Li(Na,K)
F+(r
[ x:]-Li(Na,K)F
Br
H
Ill
I10
I09
F
’@:
KOH 4I solid 11ooc Br-C=C-F
r
98
112
113
114 Scheme 2-11 Various approaches to bromofluoroacetylene 98. 1
electrical
* discharge
F*
I-CEC-F
(19)
F
116
99
2.3.2.2 More Highly Unsaturated Halogenoacetylenes As in the case of the cyanoacetylenes (see above), we intend to discuss in this section the preparation of the ethynologs of the parent systems, i. e., the “extended versions” of the simple halogeno- (92-95) and dihalogenoacetylenes (96- 10.5). Although a sizeable number of these molecules have been prepared, their chemistry - because of their very high reactivity and the difficulties involved in handling them (see below) - remains largely unexplored. AU four monohalogenated butadiynes 116- 119 are known. H-C=C-CEC-Hal
118 I19
For the preparation of the fluoro derivative 116, electrical discharge is again the method of choice, and (in fact) the only reported one so far [70, 711. Thus both 1,3,5-trifluorobenzene
2.3 Synthesis and Preparative Use of I-Halogenoacetylenes
51
(120) and 1,2,3,4,5-pentafluorostyrene(121) fragment, as shown in Eq. (20), to give - among other products - this diacetylene, whose spectroscopic data have been reported (photoelectron spectrum [70] ; microwave spectrum [71]).
bF electrical
F
electrical
-
F-CZC-C=C-H
discharg;
discharge
FW (20)
F
I4
121
116
120
For the preparation of the other three halogenobutadiynes, standard methodology could be employed [Eq. (21)] : 1,Cbutadiyne (122) is first treated with one equivalent of phenyllithium to provide the lithium salt 123, which is subsequently trapped by the appropriate halogen [72]. Routine as they may appear, the reactions are preparatively a veritable tour de force, since all reactions and the complete work-up must be carried out in the -50°C temperature region. Chemical applications of 116-119 have not been reported, to our knowledge; however, their spectroscopic properties are well known (microwave spectra of 116 [71] and 117 [73] ; IR spectra of 118 and 119 [74]; 'H-NMR spectra of 117-119 [75]).
H-CGC-CEC-H
-
PhLi, e t h e r -5O'C
H-CGC-CEC-Li
Cl2
H-CEC-CGC-CI(Br,I)
(21)
(Br2.12)
122
123
117-118
The symmetrical dihalogenodiacetylenes 124-127 have all been prepared; the methods employed are basically those already used for 116-119, i. e., electrical discharge for the generation of 124 (with hexafluorobenzene as the precursor) [70,761, and reaction of the dilithium salt of 122 for the preparation of 125-127 [72], whose infrared and Raman spectra have been measured [77]. F-C3C-C=C-F
124 Br - C 3 C - C E C - B r
126
CI - C Z C - C ~ C - C I
125
I -C_C-C_C-I 127
Again, as in the case of 1,2-diiodoacetylene(105), the diidobutadiyne derivative 127 appears to be the most stable, and has hence been used for synthetic work. It reacts with tungsten hexachloride (128) to form the tungsten complex 129, which on treatment with diethyl ether yields the etherate 130, and with iodomethylphosphonium chloride the complex 131 (Scheme 2-12). In the latter, halogen exchange may be performed with silver chloride providing the dichlorobutadiyne complex 132 [78], which - like the other complexes - is thermally and mechanically stable.
52
2 Functionalized Acetylenes in Organic Synthesis
2 WCl*
+
-
2 I-cGc-C-c-1
rw~cl~(I-=-=-I)l~ +
l/x
C4C1412
I29
128 + EtzO
[W*CI,(
+ [P~sPCH~IICI
I-=-=-I) (Et2O) 1
( P P ~ ~ C H Z I ) ~ I WI-=-=-I)I ~CI~(
131
130 ~
(PP h3CHzI)2[Wz Cle( CI -E-=-Cl) J
+ 2 AgCl
-2 AgI
I32
Scheme 2-12 Selected transformations of 1,Cdiiodobutadiyne127.
2.3.2.3 Derivatives of LHalogenoacetylenes Although fragmentation reactions, rearrangements, and other “special” processes are by no means unimportant in preparative acetylene chemistry, these methods cannot compete with the classical routes to alkynes: eliminations from suitable halogenoalkanes or halogenoalkenes, and substitutions on acetylenic triple bonds (“acetylenes from acetylenes”). What is true in general, is even more true in halogenoacetylene chemistry, since the elimination reactions - because of lack of starting materials and the frequent occurrence of side reactions - are of limited significance in this instance. Cases like the conversion of 111, 112 or 114 into 98 (see Scheme 2-11) are hence rare. Thus substitutions provide the most general access to derivatives of 1-halogeno- and 1,2-dihalogenoacetylenes. In principle, two routes are possible: one beginning with a triple-bonded compound which already contains the halogen, the derivative 133, the other starting with a triple-bond-containing substrate 134 which already carries the other desired substituent and introducing the halogen subsequently, often in “positive” form (Scheme 2-13). The first route is the preferred one for the preparation of chloroacetylenes, whereas the second one is general (except for fluoroacetylenes, which often require a special methodology [4] and are discussed below).
t MetX
Hal-CEC-Met
+
R-X
133 [R-CEC-]
+
[Hal’]
Hal-C3C-R
134 Scheme 2-13 Preparation of I-halogenoacetylene derivatives.
The easily accessible bromoacetylenes 135 are normally obtained by the Strauss reaction, the treatment of terminal acetylenes with alkaline solutions of sodium or potasium hypobromite [Eq. (22)], which has turned out to be of great value in this area [79,SO]. R-CGC-H
+
NaOBr
-
R-CEC-Br
135
+ NaOH (22)
2.3 Synthesis and Preparative Use of I-Halogenoacetylenes
53
A recent convenient method for the preparation of 1-bromoalkynes in excellent yields ( > 90%) consists in the treatment of terminal acetylenes with the triphenylphosphine/tetra-
bromomethane bromination reagent [81]. In a further new development, 1-bromo- and 1-iodoacetylenes 137 are obtained by reacting trimethylsilyl-protected alkynes 136 with either N-bromo- (NBS) or N-iodosuccinimide (NIS) in the presence of silver nitrate [82] [Eq. (23)]. NIS or NBS R - C 3 C- Si Me3
+
136
(23)
R-CEC-I(Br)
ASNO3
137
Because the preparation and chemistry of 1-halogenoacetylenes has been reviewed several times [4-7, 601 and Brandsma's laboratory manual [lo] contains several detailed descriptions of the preparation of I-bromo- and 1-chloroacetylenes,the reader is referred to these valuable sources. Additional information on these compounds is also given in Section 2.4 below. As already mentioned, the preparation or generation of 1-fluoroacetylenes often necessitates methods differing from those used to prepare the other 1-halogenoacetylenes. In many cases these involve lengthy elimination/addition sequences [83] or fragmentation or rearrangement processes [4]. A remarkable example has recently been described in the flash vacuum pyrolysis of the perfluorodialkyl-l,2,3-triazine138 which leads to perfluoro-3-methyl-l-butyne,139 [Eq. (24)]. The 1-fluoroacetylene which was obtained in quantitative yield is a stable compound which does not undergo thermal oligomerization on heating up to 200 "C - in striking contrast to the explosive 1-fluoroacetylene (92) [84] ! CF3
F
600°C
F3cF H
f
C
NbNA
13a
F
3
F3c\
F-C-CGC-F F3
d
(24)
139
Several derivatives of the halogenodiacetylenes 118 and 119 have been prepared from the corresponding terminal diynes by the metalation/halogenation approach [85, 861, and their preparative use will be described below. Finally, calculations have been published on 1-cyano-4-iodobutadiynesto assess the use of these systems as model compounds for push-pull polyynes, which in turn are of interest in connection with organic nonlinear optical materials [87].
2.3.3 Novel Preparative Uses of 1-Halogeno- and 1,2-Dihalogenoacetylenes Rather than fearing and consequently avoiding the high reactivity and often explosive character of 1-halogeno- and 1,2-dihalogenoacetylenes,there are more and more examples in the modern literature which deliberately set out to use this reactivity imaginatively for synthetic purposes. 1,2-Dichloroacetylene (100) is a good example to illustrate this point. Originally reported in 1930 [88], its dangerous nature has frequently been noted: it ignites spontaneously and may explode violently, especially on contact with air. However, it was also noted relatively early
54
2 Functionalized Acetylenes in Organic Synthesis
that, in the presence of ether, the autoxidation of 100 is retarded [89].Still, until very recently, no really safe and convenient route to this potentially very useful building block was available. This situation has now changed and at least four routes, claimed to be secure, have been reported, all of them starting from trichloroethylene (140) and submitting it to base treatment under various conditions [90-931. In one of the more recent approaches [92] the dichloroacetylene-diethyl ether complex 141 is obtained from a mixture of 140 and diethyl ether in the presence of a phase-transfer catalyst in an aqueous solution of sodium hydroxide at 70 “C [Eq.(25)]. This procedure has been claimed to avoid the problems of the previously described methods (anhydrous solvents, expensive reagents and equipment). CI
cI
ether, 50X aq. NaOH
Ph-CHzN’Et3
EtzO
CI - C E C - C I
(25)
Cl-, 70%-
140
141
Once in hand, 100 may be used in multitudinous ways. Treatment with a thiol (both aliphatic and aromatic ones having been employed) in the presence of potassium hydride in tetrahydrofuran (THF) solution affords the corresponding bis(thioethers) 142 in high yield [94](Eq.(26)). Dichloroacetylene behaves as a very electrophilic compound which is readily attacked even by bulky nucleophiles (with tert-butylthiol the product 142 is obtained in 98% yield) [95J :
-
2 RSH, KH CI - C E C - C I TH F
I42
100 R-S-C=C-S-R R
-
(26)
R-S-CGC-S-R
MCPBA
(H3C)3C-SO2-C3C-S02-C(CH3)3
(27)
CHC13
tert-Bu 142
143
When 142 (R = tert-Bu) is oxidized with m-chloroperbenzoic acid (MCPBA), the disulfone 143 is obtained in quantitative yield [96] (Eq.(27)). As expected, the latter shows high dienophilic activity in Diels-Alder additions, and since the sulfonyl moiety can be easily removed from the Diels-Alder adducts (also obtained in good to excellent yields) by a variety of methods, 143 serves as another equivalent for acetylene in [2 41 cycloadditions. The high propensity for addition to 1-halogeno- and 1,2-dihalogenoacetyleneshas often been noted [4,601,and 100 is no exception in this respect [93].With secondary amines, the
+
HNRz CI-CEC-CI
+ HNR2
CI
I00
145
NRz
A
144
146
Scheme 2-14 Preparation of ynediarnines 146 from 1,Z-dichloroacetylene100.
2.3 Synthesis and Preparative Use of 1-Halogenoacetylenes
55
adducts 144 are obtained which undergo substitution with a second equivalent of the amine to provide the 1,l-bis(dialky1amino)ethenes 145, useful precursors for the ynediamines 146 [97] (Scheme 2-14). An addition which completely “destroys” the triple bond - but at the same time illustrates the synthetic usefulness of 100 - is demonstrated by the trapping of its etherate complex (see above) with dimethylamine under in-situ conditions; the glycinamide 147 is isolated in 50% yield [92] according to Eq. (28). Formally, 100 has behaved in this transformation as an equivalent of the synthon 148 in the same sense as disodium acetylide (149) is synthetically equivalent to the dianion 150.
147
100
Tbrning to structurally more complex applications of 100, it has been shown that it can function as a Michael acceptor. For example, when the enolate of 2,Cdimethyl-cyclohexen-3-one (152) is treated with 100 in the presence of lithium hexamethyldisilazane (LiHMDS), dichlorovinylation takes place and 153 is formed. On the other hand, with lithium diisopropylamide (LDA) as base, the I-chloroacetylene derivative 151 is produced 198-1001 (Scheme 2-15). The reaction, which also takes place with other 1-chloroacetylenes,most likely involves the “Michael intermediate” 154 which - depending on reaction conditions - is either protonated or loses a chloride ion. On treatment with copper powder in tetrahydrofuradacetic acid, 151 is dechlorinated; the resulting terminal acetylene has been used for further transformations.
-
LDA
LiHMDS
c -
+ 100
+
162
151
I+
H
I00
IS3
base,
100
L 164 Scheme 2-15 1,2-Dichloroacetylene 100 as a Michael acceptor.
56
2 Functionalized Acetylenes in Organic Synthesis
Among the other dihalogenoacetylenes listed in Section 2.3.2.1 1,Zdiodoacetylene (105) has been used most often in preparative and mechanistic work, which is not surprising considering the relative stability of this compound and its ease of preparation. In a remarkable displacement reaction, 105 reacts with Mn(C0)S to afford the bis-metal(carbony1 complex) 155; whereas with other metal carbonyls such as rhenium pentacarbonyl, a formal 1’ abstraction leading to 156 and 157 is observed [loll (Scheme 2-16). M n (CO)5-
7
I-CGC-
1
(OC)5Mn-C~C-Mn(C0)5
4
165
156
157
Scheme 2-16 Transformations of 1,2-diiodoacetylene 105.
The photoiodination of 105 in h e m e has been studied [102], as has its thermal decomposition, a free-radical reaction initiated by homolysis of a C -I bond and leading, inter alia, to triiodoethylene, a product not detected in previous investigations [103]. 1-Halogenoacetylenes have so far rarely been used as starting materials for the generation of highly reactive compounds and intermediates (see the discussion on the cyanoalkynes in Sect. 2.2.3). In view of the developments in carbon chemistry during recent years (cf. Chapter 13 in this volume), we believe that 1-halogenoacetylenes and their derivatives and 1,2-dihaloacetylenes - including the higher ethynologs - could serve as useful and interesting starting materials for novel forms of carbon. That these hopes are not unfounded is indicated by an experiment which we carried out in 1980 in which lithium chloroacetylene (158) was thermally decomposed [Eq. (29)] in the hope of finding a selective way to C2 (159). Although we found no experimental proof for its formation (no trapping products in the presence of cyclohexene), the formation of deep-black “polymeric material” was noted. Clearly, these observations deserve closer scrutiny [la].
A
Li-C-C-CI
158
[IC=CI]
+ LiCl + ‘black polymer‘
(29)
169
The last decade has witnessed the application of 1-halogenoacetylenes as crucial intermediates for the synthesis of increasingly complex structures, especially in natural product chemistry. In pheromone synthesis it is essential to create double-bond systems diastereoselectively, and a route often taken consists in the preparation of a suitable alkyne precursor which is then converted into the final olefin by various addition reactions (catalytic hydrogenation, metalation, etc.). For the construction of the alkyne precursor to the pheromone, l-bromo(94) and 1-iodoalkynes (95) have been particularly valuable since they can easily be subjected to metal-catalyzed coupling reactions [105]. For example, the unsaturated ester 163, which is a sex attractant of Lepidoptera (moths and butterflies), has been prepared by first converting the terminal acetylene 160 into its 1-iodo derivative 161. This is subsequently hydrogenated
2.3 Synthesis and Preparative Use of 1-Halogenoacetylenes
57
with diimine to the diastereoisomerically pure iodoolefin 162 which, by Grignard coupling, deprotection, and esterification, yields the desired 163 in an isomeric purity of better than 99% [lo61 (Scheme 2-17). 12
-
morpholine
H-CEC-(CH,),OTHP
I-C~~-(CH~),OTHP
benzene, 100 X
160
161
I62
163
Scheme 2-17 Use of I-iodoacetylenes in pheromone synthesis.
The decisive step in the stereoselective synthesis of (E)-1-bromo-and (Z)-1-iodo-1-alkenes, which were required for the preparation of various pheromones, consists in the hydroboration of different 1-bromo- and 1-iodoacetylenes [107- 1091; conjugated alkenynes could be readily prepared by metal-catalyzed cross-coupling of 1-alkenylboranes with 1-brornoalkynes [110]. In an application from the area of prostaglandin synthesis, the protected 1-iodopropargyl alcohol 164 was reduced to the (Z)-iodoolefin 165, which was subsequently converted via 166 into the functionalized organocuprate reagent 167. The latter then provided the 13-cis-prostaglandin 168 'by a highly stereoselective conjugate addition to the appropriate cyclopentenone derivative [lll] (Scheme 2-18).
/ O 0
/
C5H11
H
H
O Q
x
4
LiCu
1. BuLi
2. (Me3P)2CuI*
x CSH11
H
--
167
166
OH
168 Scheme 2-18 Use of 1-iodoacetylenes in prostaglandin synthesis.
2 Functionalized Acetylenes in Organic Synthesis
58
A coupling reaction long used in acetylene chemistry is the Cadiot-Chodkiewicz coupling [112]; its use in retinoid synthesis is demonstrated by the transformations depicted in Scheme 2-19 (1131. The enyne precursor 169, on Cadiot-Chodkiewicz coupling with 3-bromo-2-propyn-1-01(170)yields the diyne 171 which, by methodology long established in retinoid chemistry [1141, may either be chain-elongated to dehydroretinal 172 or - via the diynal 173 - to the bis-acetylenic retinal 174.
169
171
170
173
172
174
Scheme 2-19 Use of 1-bromoacetylenes in retinoid synthesis.
Whereas the acetylene function in most of the examples discussed so far is a means to achieve certain preparative ends, triple bonds are the conditio sine qua non in the enediyne antibiotics, which because of their DNA-cleaving properties belong to the presently most studied natural products [115] (cf. Chapter 7 in this volume). In a model reaction for the preparation of the esperamicin/calicheamicin aglycones (Scheme 2-20) the enediyne alcohol 175 is converted into the terminal iodo derivative 176 by
dH 176
176
Scheme 2-20 Use of I-iodoacetylenes in enediyne synthesis.
177
2.3 Synthesis and Preparative Use of I-Halogenoacetylenes
59
iodine/morpholine treatment followed by (PCC) oxidation [116].Intramolecular Nozaki reaction [117]then leads to the cyclized products 177 with yields depending on the ring size (n = 1, 34%; n = 2, 76%). How valuable this process is for the preparation of medium-sized, strained-ring systems is underscored by the cyclization of the furanoside 178 to the oxabicyclo[7.2.1]enediyne system 179 [118][Eq. (3O)J. The corresponding substrates carrying hydrogen or trimethylsilyl in place of the iodine substituent in 178 could not be made to cyclize under a variety of conditions. Further variations upon this theme have recently appeared in the literature [119, 1201, and we predict that, with the numerous metal-mediated coupling reactions now known, the use of 1-halogenoacetylenes will be of rapidly increasing importance in preparative organic chemistry.
CrCI2/NiCl2
Ik$ocH3
OCH3
THF
*
OCH3
179
I78
Finally, in the area of naturally occuring polyacetylenes, derivatives of the halogenobutadiynes 118 and 119 are turning out to be useful synthetic intermediates. Thus coupling of either the copper 180 or the zinc organic allene 181 with iodide 182 or bromide 183, respectively (R in both cases being either methyl or trimethylsilyl), yields the highly unsaturated products 184 from which, for R = (CH,),Si, the first naturally occuring allene to be isolated, marasin (185), could be liberated (Scheme 2-21); 185 is an antibiotic effective against Staphylococcusaureus [85]. Polyacetylenes which occur in Basidiomycete fungi have been synthesized analogously [86]. [EtOCH(CH3)0CH2CH2C=C=CHCu1
[EtOCH(CH3)0CH2CH2C=CICHZnCll
180
181
+ R-C=C-C=C-I 182
R-C=C-CpC-CH-C=CH-CH2CHOCH(
CH3)OEt
184
t H-C=C-C=C-CH=C=CH-CHzCHzOH 185 Scheme 2-21 Use of I-halogenodiacetylenes in natural products chemistry.
60
2 Functionalized Acetylenes in Organic Synthesis
2.4 Experimental Procedures In this brief experimental section we wish to present some procedures for the preparation of several simple 1-cyano- and 1-halogenoacetylenes which we have been using for several years. None of these procedures is original; rather, these descriptions represent optimizations of methods previously published with the appropriate references already given in the main sections of this review. For the procedures for preparing 1-chloroacetylene (93)and 1,2-diiodoacetylene (105) we are indebted to Professor Dr. E. Kloster-Jensen (University of Oslo).
2.4.1 Cyanoacetylene (1) To 500 mL of liquid ammonia kept at -45"C, 50.0 g (0.595 mol) of methyl propiolate is slowly added. After stirring at this temperature for 24 h, 200 mL of anhydrous diethyl ether is added and the ammonia is evaporated at room temperature. The ether is removed in VCICUO, and the solid residue recrystallized from dichloromethane/pentane at - 20 "C: 34.6 g (0.501 mol, 84%) of propiolic amide (5) as colorless needles. 'H-NMR (200.1 MHz, DMSOd,, int. TMS): 6 = 8.09 (s), 7.63 (s), 4.06 (s). I3C-NMR (DMSO-d,, int. TMS): 6 = 153.4 (s), 78.7 (s), 75.6 (s). In a 250-mL three-necked flask equipped with a dropping funnel and a mechanical stirrer and connected to a cold trap via a U-tube filled with glass wool, 80 mL of sulfolane is placed which has freshly been distilled from P,Olo. To the reaction mixture is added 3-5% of 3-methylsulfolane to keep the reaction medium liquid at room temperature, thus allowing easier dispersion and redistillation, respectively. Under stirring 20.0 g (72 mmol) of P4010is added, and the reaction flask is purged thoroughly with N,. The apparatus is connected to a vacuum pump and under vigorous stirring and heating to 110°C (oil bath) 5.0 g (7.2 mmol) of 5 in 30 mL of sulfolane (with the 3-methylsulfolane added as described above) is slowly added while a pressure of ca. 20 Torr is maintained. The addition is accompanied by strong frothing, and the dehydration agent turns increasingly black. The cyanoacetylene (1) condenses in the trap (-78 "C) in the form of colorless needles (yield: 3.26 g, 64 -01, 87%). 'H-NMR (400.1 MHz, CDCl,/int. TMS): 6 = 2.59 (s). I3C-Nh4R (100.6 MHz, CDCI,, int. TMS): 6 = 104.46 (s), 73.36 (d), 57.35 (s). After completion of the dehydration the solvent may be recovered by distillation.
2.4.2 Dicyanoacetylene (2) A solution of 23 g (0.162 mol) of dimethyl acetylenedicarboxylate in 30 mL of diethyl ether is slowly added to 200 mL of liquid ammonia kept at -40 "C. After a short time, acetylenedicarboxamide (7) begins to precipitate. To complete the reaction, the mixture is stirred for 8 h at -40"C. The excess ammonia is evaporated at room temperature, and the residue is extracted with 50 mL of boiling ethanol followed by the same amount of methanol. After cooling these extracts to room temperature, the precipitate formed is isolated by filtration and dried under high vacuum: 11-16 g (0.10-0.14 mol, 62-88%) of 7. Into a 500-mL three-necked flask equipped with dropping funnel and mechanical stirrer and connected via a U-tube filled with glass wool to a cold trap cooled to - 78 "C, 100 mL of freshly distilled sulfolane is placed and 25 g (88 mmol) of P40,0 is added. The apparatus
2.4 Experimental Procedures
61
is purged thoroughly with N, under stirring, and 8 g (71 mmol) of bisamide 7 suspended in 100 mL of sulfolane is added in portions under vigorous stirring while a pressure of 20 Torr and an external temperature of 110'C are maintained. Since 7 is poorly soluble in sulfolane, it is recommended to pulverize it thoroughly to guarantee a homogeneous suspension. Dicyanoacetylene (2) is pumped off as it is formed and condenses as colorless needles in the cold trap (1.7 g, 22 mmol, 40%). 13C-NMR(100.6 MHz, int. TMS): 6 = 103.3 (s), 55.2 (s).
2.4.3 Dicyanodiacetylene (3) Cuprous chloride (9.9 g) is dissolved in 30 mL of a 14% solution of ammonia in water, the solution is filtered and water is added until 900 mL of a clear solution is obtained. The solution is cooled with an ice bath, and a stream of 1.4 g (0.027 mol) of cyanoacetylene (1)is swept into the reaction flask by a stream of nitrogen. Copper cyanoacetylide (9) soon begins to precipitate in the form of yellow needles. These are removed by filtration under N, cover, washed twice with a 1.5% solution of aqueous ammonia, and finally with water. Since dry 9 is an explosive compound, it is used in moist form for the oxidative coupling. The copper acetylide is suspended in a mixture of 30 mL of CCl, and 20 mL of water, and, at O"C, ca. 18 g of K3[Fe(CN)6Jin 20 mL of water is added under stirring within 5 min. After additional stirring for 10 min, the reaction mixture is distributed to several precooled centrifuge tubes, and after adding a few milliliters of CCl,, centrifugation is performed as quickly as possible. The organic phases (bottom layer) are removed with a pipette, combined, and dried with MgSO,. The solution is concentrated to ca. 4 mL by distillation with a Vigreux column, and the remainder of the solvent is removed by passing a stream of dry N, through the solution, which is cooled continuously by an ice bath. The still-wet raw product is subjected to sublimation at aspirator pressure with the cold finger of the sublimation apparatus kept at -10°C. Yield: 100-300 mg (1-3 mmol, 8-24070) of 3 in the form of colorless needles. The yield of the reaction strongly depends on the speed of the work-up. Yields of the raw product (immediately after the coupling step) can be as high as 80%. I3C-NMR (100.6 MHz, CDCl,, int. TMS): 6 = 103.9, 64.5, 53.4 (all s).
2.4.4 Chloroacetylene (93) [I221 The preparation of pure 93 from Hg(I1) chloroacetylide has been described in a very detailed procedure [123] which has been used sucessfully by later workers [122]. Toward purification, a sample of crude 93 prepared from 4.5 g of the mercuric salt under helium is trapped in a cold trap cooled by liquid nitrogen. After evacuation to lo-' Torr, the temperature is raised to - 80 "C and the chloroacetylene is expanded into a volume of 1 L. This freeze-pump-thaw cycle is repeated two more times. Final purification is achieved by low-temperature distillation at - 115"C such that the product is first passed through a trap cooled to - 140 "C and subsequently into a receiving trap held at liquid nitrogen temperature yielding 1.5 mL of 93, which is a water-clear liquid at - 80 "C.
62
2 Functionalized Acetylenes in Organic Synthesis
2.4.5 Dichloroacetylene (100) For the preparation of 100, we have used repeatedly and without safety problems the method described in [90]. The amount of the dichloroacetylene in the ethereal solutions obtained can be determined either by refractometry [121] or by iodometric titration [90],and typical yields are in the 50% range. As mentioned above, in the meantime several other procedures for preparing 100 [91-931 have been published, which apparently provide a further increase in the ease of access to 100. Furthermore, p. 145 of [lo] mentions unpublished work describing the preparation of dichloroacetylene by treating trichloethylene (140) in ether at - 70 "C with LDA. Dibromoacetylene (103) has been obtained analogously [lo].
2.4.6 Diidoacetylene (105) [61] Compound 105 can be prepared on a 10 g scale by bubbling acetylene through a solution of iodine in liquid ammonia [124]. The white solid which precipitates is recrystallized from pentane and sublimed twice over P,Olo at 50°C/10-2 Torr, yield ca. 50%; colorless crystals, m.p. 76.0-76.5 "C. The compound has an iodoform-like odor.
Acknowledgements We thank Joachim Steckelberg and Martin Vogtherr for their help in preparing the manuscript for this chapter, and A.Hirsch, E. Kloster-Jensen, and E. Winterfeldt for literature references and critical comments on the manuscript.
Abbreviations DDQ
2,3-dichloro-5,6-dicyano-1,4-benzoquinone dimethyl sulfoxide HMDS hexamethyldisilazane HMPA hexamethylphosphoramide LDA lithium diisopropylamide MCPBA m-chloroperbenzoic acid NBS N-bromosuccinimide NIS N-iodosuccinimide PCC pyridinium chlorochromate room temperature r. t. THF tetrahydrofuran THP tetrahydropyranyl TMS trimethylsilyl
DMSO
References [l] A recent article by E. Winterfeldt, Acetylenes in synthesis, in the Modern Synthetic Methods Series, Vol. 6, (Ed.: R. Scheffold), Verlag Helvetica Chimica Acta, Basel, and VCH, Weinheim, 1992, pp. 103-226, assembles a wealth of interesting information on the use of functionalized acetylenes in modem organic chemistry, with, again, no attempt to cover this group of compounds totally.
References
63
[2] R. A. Raphael, Acetylenic Compounds in Organic Synthesis, Butterworth, London, 1955 [3] T. F. Rutledge, Acetylenic Compounds, Reinhold, New York, 1968, cf. T. F. Rutledge, Acetylenes and Allenes, Reinhold, New York, 1969. [4] H. G. Viehe (Ed.), Chemistry of Acetylenes, Marcel Dekker, New York, 1969. [5] Methoden Org. Chem. (Houben- Weyl-Miiller) Alkine. Di- und Polyine Allene, Kumulene, 1977, Vol. V/2a, pp. 33. [6] S. Patai (Ed.), The Chemistry of the Carbon-Carbon Thiple Bond, Parts 1 and 2, John Wiley, Chichester, 1978. [7] Methoden Org. Chem. (Houben- Weyl-Miiller), Halogene, Fluor und Chlor, 1960, Vol. V/4, pp. 719. ibid, Brom und Iod, 1962, Vol. V/3, p. 592. [8] Methoden 0%.Chem. (Houben- Wql-Miiller), Carbonsguren und Carbonsaure-Derivate, 1985, Vol. E5, Teil 2, p. 1531. [9] 2. Rappoport (Ed.), The Chemistry of the Cyuno Group, Interscience, London, 1970; cf. S. Patai, Z. Rappoport (Eds.) Supplement C, The Chemistry of Thiple-bonded Functional Groups, Parts 1 and 2, John Wiley, Chichester, 1983. [lo] L. Brandsma, Preparative Acetylenic Chemistry, 2nd ed., Elsevier, Amsterdam, 1988. 1111 (a) For a comprehensive review on the work published until 1970 see E. Ciganek, W. J. Linn, 0. W. Webster in [9]. (b) For a more recent (though very brief) summary of cyanocarbons see W. H. Hertler, W. Mahler, L. R. Melby, J. S. Miller, R. E. Putscher, 0. W. Webster, Mol. Cryst. Liq. Ciyst. 1989, 171, 205-216. [12] C. Moureu, J. C. Bongrand, Comp. Rend. 1910, Z51, 946. [13] M. Franck-Neumann in Methoden Org. Chem. (Houben- Weyl-Miiller) 1977, Vol. V/2a, p. 677. [14] B. Witulski, Ph. D. Dissertation, Braunschweig, 1992. [15] C. Moureu, J. C. Bongrand, Ann. Chim. 1920, 14, 5; cf. C. Moureu, J. C. Bongrand, Bull. Soc. Chim. Fr. 1909, 846-847. [16] E. Ciganek, C. G. Krespan, J. Org.Chem. 1968, 33, 541-544. [17] F. J. Brockman, Can. J Chem. 1955, 33, 507-510; cf. A. J. Saggiomo, 1 Org. Chem. 1957, 22, 1171-1175. [18] A. J. Alexander, H. W. Kroto, D. R. M. Walton, J Mol. Spectrosc. 1976, 62, 175-180. [19] C. Kirby, H. W. Kroto, D. R. M. Walton, ibid. 1980,83,261-265; cf. T. Oka, ibid. 1978, 72,172-174. [20] J. P. Hare, H. W. Kroto, Acc. Chem. Res. 1992,25, 106-112; cf. H. W. Kroto, Angew. Chem. 1992, 104. 113-133; Angew. Chem., Znt. Ed. Engl. 1992, 31, 111-131. [21] Th. Grosser, A. Hirsch, Angew. Chem. 1993, 105, 1390-1392; Angew. Chem. Int. Ed. Engl. 1993, 31, 1340-1342. In the meantime C,,N2 has also been prepared and all polyynes (C& to C&) have been isolated by HPLC methods and have been fully characterized: A. Hirsch, private communication. [22] R. A. van der Welle, L. Brandsma, Recl. Tmv. Chim. Pays-Bas 1973, 92, 667-672. [23] R. E. Murray, G. Zweifel, Synthesis 1980, 150-151. [24]P. L. Compagnon, B. Grosjean, ibid 1976, 448-449. [25] B. Fisher, C. A. Grob, Helv. Chim. Acta 1956, 39, 417-422. [26] A. Casarini, P. Denbeck, G. Reginato, A. Ricci, G. Secondi, Tetrahedron Lett. 1991, 32, 2169-2170. In a recent method for the preparation of cyanoalkynes, terminal acetylenes are coupled with cuprous cyanide in the presence of trimethylsilyl chloride, water, catalytic amounts of sodium iodide, and acetonitrile in DMSO at 50°C: F.-T.Luo, R.-T. Wang, Tetrahedron Lett. 1993, 34, 5911-5914. [27] F. G. Kleiner, W. P. Neumann, Liebigs. Ann. Chem. 1968, 716, 19-28. [28] E. T. Bogoradovskii, V. S. Zavgorodnii, B. V. Polozov, A. A. Petrov, Zh. Obshch. Khim. 1982, 52, 455-456; J. Gen. Chem. USSR 1982, 52, 399-400. [29] A. J. Alexander, S. Firth, H. W. Kroto, D. R. M. Walton, J. Chem. SOC.Furaday Thans. 1992, 88, 531-533. [30] E. Kloster-Jensen, Actu Chem. Scand. 1963, 17, 1859-1861.
64
2 Functionalized Acetylenes in Organic Synthesis
[31] E. Kloster-Jensen, Acta Chem. Scand. 1963, 17, 1862-1865. [32] E. Kloster-Jensen, Acta Chem. Scand. 1964, 18, 1629-1934. [33] G. Bieri, J. P. Stadelmann, F. Thomnen, J. Vogt, Helv. Chim. Acta 1978, 61, 357-360. [34] S. R. Landor, P. F. Asobo, 2. F. Fonum, R. Roberts, J. Chem. Soc. Perkin nuns. I, 1991, 1201-1204 and references quoted therein. [35] For a very informative recent summary on the synthetic use of these compounds, see B. A. Trofmov, A. G. Mal’kina, Yu. M. Skvortsov, Zh. Org. Khim. 1993,29, 1268-1291; J. Org. Chem. USSR 1994, 30, 1053-1072. [36] F. BoNmann, H. J. Mannhardt, Chem. Eer. 1956, 89, 2268-2272. [37] M. Anchel, Science 1955, 121, 607. 1381 M. Benes, J. Peska, 0. Wichterle, Chem. Ind. (London) 1962, 562-563. [39] M. Benes, J. Peska, 0. Wichterle, J. Polym. Sci. (C) 1964, 1377. [40] B. J. MacNulty, Polymer 1966, 7, 275. [41] M. A. Shaw, J. C. Tebby, R. S. Ward, D. H. Williams, J. Chem. Soc. (C) 1968, 1609-1612. [42] P. J. Butterfield, D. C. Tebby, J. Chem. SOC, Perkin Trans. I, 1979, 1189-1191. 143) E. Ciganek, ntmhedron Lett. 1967, 3321-3325. [44] H. Hopf, C. Marquard, in ChemicalZmplicationsof Strain, (Ed.: A. de Meijere, S. Blechert) Reidel, Nato AS1 Series, Series C, Vol. 273, 1989, pp. 297-332. (451 H. Hopf, B. Witulski, P. G. Jones, D. Schomburg, Liebigs Ann. 1995, 609-612. [46] H. Hopf, B. Witulski, L. Ernst, P. G. Jones, Chem. Be,: 1990, 123, 2015-2022; cf. H. Hopf, B. Witulski, Nachr. Chem. Tech. Lab. 1991, 39, 286-290. [47] H. Hopf, B. Witulski, unpublished results. [48] V. Breitkopf, H. Hopf, F.-G. Klaner, B. Witulski, B. Zimny, Liebigs Ann. 1995, 613-617; cf. B. Zimny, Ph. D. Dissertation, Universitat Essen, 1994. [49] H. Hopf, B. Witulski, Pure Appl. Chem. 1993, 65, 47-56. [50] J. P. Ferries, J. C. Guillemin, J. Org. Chem. 1990, 55, 5601-5606. [51] M. Morris, B. E. lkrner, P. Palmer, B. Zuckerman, Astrophys. 1 1976, 205, 82; cf. A. P. C. Mann, D. A. Williams, Nature (London) 1980,283, 721-725. For the occurrence of 1in the atmosphere of Titan, see V. G. Kunde, A. C. Aikin, R. A. Hanel, D. E. Jennings, W. C. Maguire, R. E. Samuelson, Nature (London) 1981, 292, 686-688. [52] V. Bujarrabal, J. Gomez-Gonzalez, R. Bachmiller, M. Martin-Pintado, Astron. Astrophys. 1988,204, 242. [53] For a summary, see: J. K. Terlouw, P. C. Burgers, B. L. M. van Baar, Th. Weiske, H. Schwarz, Chimia 1986, 40, 357-359; cf. B. van Baar, W. Koch, C. Lebrilla, J. K. Terlouw, Th. Weiske, H. Schwarz, Angew. Chem. 1986, 98, 834-835; Angew. Chem. Znt. Ed. Engl. 1986, 25, 827. [54] V. H. Dibeler, R. M. Reese, J. L. Franklin, J. Am. Chem. Soc. 1961, 83, 1813. (551 D. Siilzle, K. Seemayer, B. Witulski, H. Hopf, H. Schwarz, Chem. Ber. 1991, 124, 1481-1483. [56] D. Siilzle, K. Seemayer, H. Schwarz, B. Witulski, H. Hopf, Znt. J. Mass. Spectrom. Zon Proc. 1991, 105, Rl-R4. [57] D. C. Parent, J. Am. Chem. Soc. 1990, 112, 5966-5973. [58] K. A. Hofmann, H. Kirmreuther, Ber. Dtsch. Chem. Ges. 1908, 41, 314-317. [59] J. I. Iotschich, Bull. Soc. Chim. F,: 1905, 34, 180-184. [60] K. M. Smirnov, A. P. Tomilov, A. 1. Shchekotikhin, Usp. Khim. 1%7,36,777-803; Russ. Chem. Rev. 1%7,36, 326-338. [61] E. Kloster-Jensen, Tetrahedron 1971, 27, 33-49. 1621 M. Andolfatto, B. Kleibomer, A. Guarnieri, Z. Naturfosch. 1982, 37A, 1449-1453. [63] A. Runge, W. W. Sander, Tetrahedron Lett. 1990, 5453-5456. [64]T. Hirao, T. Okabayashi, M. lknimoto, .l Mol. Spectrosc. 1993, 162, 358-365. [65] N. Heineking, H. Harder, H. Dreizler, J. Mol. Spectrosc. 1990, 141, 299-308; cf. M. L. H. Jeng, B. S. Ault, J. Phys. Chem. 1989, 93, 5426-5431.
References
65
[66]H. Burger, S. Sommer, J. Chem. Soc. Chem. Commun. 1991,456-458.In this communication a new method for preparing 96 by gas-phase pyrolysis of perfluoro-1,2,3-triazineis described; cf. H. Burger, W. Schneider, S. Sommer, W. Thiel, J. Chem. Phys. 1991,95,5660-5669 for matrix and highresolution infrared studies and ab-initio calculations. Photochemically % has been prepared from difluoromaleic anhydride (J. C. Brahms, W. P. Dailey, J. Am. Chem. SOC. 1989,111,8940-8941)and difluoropropadienone (J. C. Brahms, W. P. Dailey, J. Am. Chem. SOC. 1990,112,4046-4047), inter alia. [67]T. Okabayashi, M. Tanirnoto, J. Mol. Spectrosc. 1992, 154, 201-206. [68] D. McNaughton, Struct. Chem. 1992, 3, 245-252. [69] D. McNaughton, D. McGilvery, F. Shanks, J. Mol. Spectrosc. 1991, 149,458-473. [70] G. Bieri, A. Schmelzer, L. Aasbrink, M. Jonsson, Chem. Phys. 1980,49, 213-224;cf. T. Hayashi, M. Kikuchi, T. Fujioka, S. Komiya, Proc. Ion. Eng. Congr. 1983,3, 1611-1616. [71] T. Okabayashi, K. Tanaka, T. Tanak, J. Mol. Spectrosc. 1989, 137, 9-12. [72]E. Heilbronner, V. Hornung, J. P. Maier, E. Kloster-Jensen, J. Am. Chem. SOC.1974,96,4252-4262. [73]A.Bjoerseth, E. Kloster-Jensen, K.-M. Marstokk, H. Moellendahl, J. Mol. Struct. 1970,6,181-204. [74] P. Klaeboe, E. Kloster-Jensen, S. D. Cyvin, Spectrochim. Acta. Purt A 1967,23, 2733-2748 and D. H. Christensen, I. Johnsen, P. Klaeboe, E. Kloster-Jensen, ibid. 1969, 25, 1569-1576;cf. M. K. Phibbs, ibid. 1973,29, 599-602;P. Klaeboe, D. H. Christensen, ibid. 1974, 30, 1167-1168;M. K. Phibbs, L. Mannik, ibid. 1975,31, 1103-1104. [75]E. Kloster-Jensen, R. Tabacchi, Tetrahedron Lett. 1972, 4023-4026. [76]G. Bieri, E. Heilbronner, J.-P. Stadelmann, J. Vogt, W. von Niessen, J. Am. Chem. SOC. 1977, 99, 6832-6838;cf. G. Bieri, L. Aasbrink, W. von Niessen, J. Electron Spectrosc. Relat. Phenom. 1981, 23, 281-322;K. Kamienska-Tsela, P. Gluzinski, Croat. Chem. Acta 1986,59, 883-890. [77]P. Klaeboe, E.Kloster-Jensen, E. Bjarnov, D. H. Christensen, 0. F. Nielsen, Spectrosc. Acta. Part A, 1975, 31, 931-943. [78]K. Stahl, K. Dehnicke, J. Organomet. Chem. 1986,316, 85-93. [79]F. Strauss, L. Kollek, W. Heyn, Ber. Dtsch. Chem. Ges. 1930, 63, 1868-1885. [80] F. Strauss, L. Kollek, H. Hauptmann, Ber. Deutsch. Chem. Ges. 1930, 63, 1886-1899. [81] A. Wagner, M. P. Heitz, C. Mioskowkski, Tetrahedron Left. 1990,31, 3141-3144. [82]T.Nishikawa, S. Shibuya, S. Hosokawa, M. Isobe, Synlett 1994,485-486;cf. P. Bovonsombat, E. McNelis, Tetrahedron Lett. 1992,33, 4123-4126 for the preparation of a,a-dihaloacetophenones by the reaction of phenylacetylene with N-iodosuccinimide in the presence of catalytic amounts of ptoluenesulfonic acid. [83]R. E. Banks, M. G. Baslow, W. D. Davies, R. N. Hasezeldine, D. R. Taylor, J. Chem. SOC.(C) 1969, 1104-1107. [84] R. D. Chambers, T. Shepherd, M. Tamusa, M. R. Bryce J. Chem. SOC. Chem. Commun. 1989, 1657-1658. [85]W. de Graf, A. Smits, J. Boersma, G. van Koten, W. P. M. Hoekstra, Zktruhedron 1988, 44, 6699-6704. [86]A. Ahmed,J. W. Keeping, T. A. Macrides, V. Thaller, L Chem. SOC.Perkin I , 1978, 1487-1489,and previous papers in this series. [87]M. Jain, J. Chandrasekhar, J. Phys. Chern. 1993, 97, 4044-4049. [88] E. Ott, W. Ottemeyer, K. Packendorff, Ber. Dtsch. Chem. Ges. 1930, 63, 1941-1944. [89] E. Ott, G. Dittus, H. Wissenburger, Ber. Dtsch. Chem. Ges. 1943, 76, 88-91 and references quoted. The method as reported by Ott has nevertheless been found to be extremely dangerous: R. Riemenschneider, K. Brendel. Justus Liebigs Ann. Chem. 1961, 640, 5-13. The toxicology of 100 has been investigated: D. Reichert, G. Liebaldt, D. Henschler, Arch. Toxicol. 1976, 37, 23. [90] J. Siege], R. A. Jones, L. Kurlansik, 1 0%.Chem. 1970, 35, 3199. [91] A. S. Kende, P. Fludzmski, Synthesis 1982, 455-456. [92]J. Pielichowski, R. Popielarz, Synthesis 1984, 433-434.
66
2 Functionalized Acetylenes in Organic Synthesis
(931 J.-N. Danis, A. Moyano, A. E. Greene, J. Org. Chem. 1987, 52, 3461-3462. [94] A. Riesa, F. Cabre, A. Moyano, M. A. Pericas, J. Santamaria, Tetrahedron Lett. 1990, 2169-2172. 1-Chloroacetylene (93) behaves analogously: J. Flynn, V. V. Badiger, W. E. Truce, J. Org.Chem. 1963,28,2298-2302. For the preparation of alkynyl sulfides and selenides from 1-bromoacetylenes, see: A. L. Braga, A. Reckziegel, P. H. Menezes, H. A. Stefani, Tetrahedron Lett. 1993,34,393-394. 195) The question of the mechanism(s) of nucleophilic substitution at the acetylenic carbon atom is an important one and has been investigated and discussed for a long time. For leading references, see [4] and A. Fujii, J. 1. Dickstein, S. I. Miller, Tetrahedron Lett. 1970, 39, 3435-3438 and references cited therein; R. Tanak, M. Rodgers, R. Simonaitis, S. I. Miller, Tetrahedron 1971, 27, 2651-2659; A. Commercon, J. F. Normant, J. Villiers, Tetrahedron 1980, 36, 1215-1221. [96] A. Riera, M. Mosti, A. Moyano, M. A. Pericas, J. Santamaria, Tetrahedron Lett. 1990, 2173-2176. [97] R. van der Heiden, L. Brandsma, Synthesis 1987, 76-77; cf. J. Pielichowski, D. Bogdal, Bull. SOC. Chim. Belg. 1993, 102, 393-395. 1981 A. S. Kende, P. Fludzinski, Tetrahedron Lett. 1982, 23, 2369-2372. [99] A. S. Kende, P. Fludzinski, Tetrahedron Lett. 1982, 23, 2373-2376. [loo] A. S. Kende, P. Fludinski, J. H. Hill, W. Swenson, J. Clardy, J. Am. Chem. SOC. 1984, 106, 3551-3562. [loll J. A. Davies, M. El-Ghanam, A. A. Pinkerton, D. A. Smith, J. Organomet. Chem. 1991, 409, 367-376. [lo21 J. W. Tamblyn, G. S. Forbes, J. Am. Chem. SOC.1940, 62, 99-104. [lo31 S. E. Krikonian, W. R. Moore, J. Org. Chem. 1994, 59, 3742-3743. [lo41 H. Hopf, S. Ehrhardt, Diplomarbeit, Braunschweig, 1981. [lo51 For a summary on work employing palladium- and nickel-catalyzed cross-coupling processes, see E. Negishi, Acc. Chem. Res. 1982, IS, 340-348. 1106) D. Michelot, Synthesis 1983, 130-134. [lo71 H. C. Brown, V. Somayaji, Synthesis 1984, 919-920. [lo81 H. C. Brown, D. Basavaiah, S. M. Singh. N. G. Bhat, J. 0%.Chem. 1988, 53, 246-250. [lo91 H. C. Brown, D. Basavaiah, S. M. Singh, Synthesis 1984, 920-922. [110] N. Miyaansa, K. Ymada, H. Suginome, A. Suzuki, J. Am. Chem. SOC. 1985, 107, 972-980. [Ill] A. F. Kluge, K. G. Untch, J. H. Fried, J. Am. Chem. SOC. 1972, 94, 9256-9258. 11121 P. Cadiot, W. Chodkiewicz in [4], Chapter 9, pp. 597-647. [113] H. Hopf, N. Krause, Tetrahedron Lett. 1985,26, 3323-3326. [114] H. Hopf, N. Krause in W. H. Okamura, M. Dawson (Eds.), Chemistry and Biology of Retinoids, CRC Press, Boca Raton, FL, 1990, pp. 177-199. [115] K. C. Nicolaou, Angew. Chem. 1993, 105, 1462-1471; Angew. Chem. Int. Ed. Engl. 1993, 32, 1377-1386 and references cited therein. 11161 G. Crevisy, J.-M. Boan, Tetrahedron Lett. 1991, 32, 3171-3174. [117] K. Takai, M. Yagashira, T. Kuroda, T. Oshima, K. Uchimoto, H. Nozaki, J. Am. Chem. SOC.1986, 108, 6048-6050; cf. T. D. Aicher, K. R. Buszek, F. G. Fong, C. J. Forsyth, S. H. Jung, Y. Kishi, M. C. Maletich, P. M. Scola, D. M. Spero, S. K. Yoon, 1 Am. Chem. SOC. 1992, JJ4, 3162-3164. [118] M. E. Maier, T. Brandstetter, Etrahedron Lett. 1992, 33, 7511-7514. [I191 Y.-F. Lu, C. W. Harwig, A. G. Fallis, J. Org. Chem. 1993, 58, 4202-4204. 11201 T. Nishikawa, S. Shibuya, S. Hosakawa, M. Tsobe, Synlett 1994, 466-485. [121] E. Ott, Ber. Dtsch. Chem. Ges. 1942, 75, 1517-1522. [122] H. J. Haink, E. Heilbronner, V. Hornung, E. Kloster-Jensen, Helv. Chim. Actu 1970,53,1073-1083. [123] L. A. Bashford, H. J. Emeleus, H. V. A. Briscoe, J. Chem. SOC. 1938, 1358-1364. [124] T. H. Vaughn, J. A. Nieuwland, J. Am. Chem. SOC. 1932, 54, 787-789; cf. G. N. Taylor, Chem. Br. 1981, 17, 107.
3 Alkynyliodonium Salts: Electrophilic Acetylene Equivalents Peter .l Stang
3.1 Introduction The unsubstituted carbon-carbon triple bond, by virtue of its n-bonds, is electron-rich and hence generally not disposed toward interaction with other electron-rich species. Therefore, even acetylenes bearing a leaving group, such as haloalkynes, do not undergo the direct SN-1 or SN-2type of nucleophilic displacement reactions. In fact the parent alkynyl cation, HC; , is estimated to be some 60 kcal/mol less stable than the methyl cation [l]. As a consequence the vast majority, if not all, nucleophilic substitutions at an acetylenic carbon occur via some type of addition-elimination process [2, 31. Since acetylenic esters of any type, but especially ones with a good leaving group, like sulfonates, were unknown until recently [4], early substrates for nucleophilic acetylenic substitutions [S,-A] were primarily the haloacetylenes 1 [2, 3, 51. Unfortunately, product yields in these reactions tend to be moderate at best, usually because of competing reactions including displacement of RC=C- via direct attack on the halogen itself. Undoubtedly the most powerful leaving group in organic chemistry is neutral nitrogen from a diazonium ion. Regrettably, ethynyldiazonium ions, 2, are unknown to date [6]. However, loss of neutral iodobenzene from alkynyl(pheny1)iodoniium salts, 3, should be comparable to loss of N2 from 2. Recent data [7] indicate that the leaving ability of PhI from iodonium species is about lo6 times greater than that of a triflate which, as one of the “super” leaving groups, is known to be some lo8 times better than chloride [8]. Unlike the hitherto unknown alkynyldiazonium salts, 2, stable alkynyl(pheny1)iodoniumspecies 3 have been known [9] for nearly a dozen years. RC-CX 1: X=I,Br,CI,F
RC-&% 2
RC=CiPN< 3
Alkynyl(pheny1)iodonium salts, 3, are members of the family of polyvalent iodine [lo] species where two organic ligands, an alkyne and an aryl group, are bound to a positively charged iodine(II1) atom. Since the preparation of the first stable examples of 3 as tosylate salts [ll-131, well over 100 different alkynyliodonium salts have become known. In this chapter we shall describe the preparation, properties, characterization, and chemistry of these novel functionalized alkynes that readily serve as electrophilic acetylene equivalents. Besides their reaction with a wide variety of nucleophiles, including organometallic species, their cycloaddition reactions will be discussed along with relevant mechanistic considerations. Emphasis is placed on our own contributions to the field and key experimental procedures are given at the end.
68
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
3.2 Preparation and Properties The overwhelming majoritiy of currently known alkynyliodonium species are prepared by interaction of a terminal, sila- or tin-acetylene with an electrophilic L3-iodane, also referred to as a 10-1-3 hypervalent species [14]. Key reagents are iodosobenzene (4), [hydroxy(tosyloxy)iodo]benzene (5, HTIB) [151 the p-oxo-h3-iodane 6 1161, and cyano(pheny1)iodonium triflate 7 [17].
-
.. .. ..:
Ph--I=O
Ph-i*
:
5
4
"-1 '
0 I'-Ph
ROSO
&OH 6Ts
I
OSO2R
6
3.2.1 Alkynyliodonium Sulfonates The first stable class of alkynyl(ary1)iodonium salts were the tosylates, 9, prepared by the interaction of HTIB (3, with terminal alkynes in refluxing chloroform [Eq. (I)] 111-131. Unfortunately, this method suffers from lack of generality, separation problems from the concomitantly formed alkene salt, 8, and low product yields of 9.
Improvements [18] and modifications [19] in this procedure have provided product yields of the alkynyliodonium tosylates of 60-90% as well as broader applicability to a greater variety of P-alkyl groups R. These modified procedures are also applicable to the formation of methanesulfonates, CH3S0F, as well as p-NO2C6H4SO, salts. A more general, simpler procedure I201 takes advantage of the in situ formation of the p0x0-bis-triflate (6: R=CF3) and its interaction with a sila- or tin-acetylene [Eq.(2)]. This methodology affords a wide variety of stable, alkynyl(pheny1)iodonium triflates 10 in good to excellent yields and is applicable to the synthesis of the parent [21] ethynyl(pheny1)iodonium triflate (10: R=H) from n-Bu3SnCeCH.
PhIO
+ (CF3S02$20
OW
RC!&Ph6S02CF3 + @&A) 2O
10 ( 4 5 9 3 %
RC=C!MRi :M=SiSn cH2C12. go to 20 W
3.2 Preparation and Properties
69
The best and most versatile contemporary method [22] of preparation of alkynyl(ary1)iodonium salts employs readily available alkynylstannanes [23], and the easily prepared [24] cyano(pheny1)iodoniurn triflate 7 as an iodonium transfer agent in dichloromethane at low temperatures [Eq.(3)]. This procedure provides excellent yields of iodonium triflates 10 and is applicable to a very broad range of alkynylstannanes, including those with strongly electronwithdrawing groups as summarized in Scheme 3-1. Particularly noteworthy and valuable are the P-keto- and P-amido-substituted species 11 and the cyano-functionalized molecule 11 (Y=CN) [25, 261.
RC=CSnRi
+
A C N OTf
CHZCIZ -42 oc to 2o oc w
R C E C b h OTf
(3)
10
7
YC=CSnBu3
+ RjSnCN
+ PhiCN GSO2CF3
YC=CbhGSOzCF3
CH2C1z
-420 to 20 OC
+
Bu3SnCN
11
11 Y(iso1ated yield): CH,(85 %), 1-cyclohexenyl(73%), MeOCH2(77 %), C1CH2(54 %), BrCH2(76 %), CN(72
%), Cl(72 %) ArS0,(85 %), MeC(OWPh(80 %), t-BuC(0)(82 %), Me,(Et)C(0)(75 %), c-C3H,C(0)(59 %), c-
C$V30)(58
%), ~-C&,C(0)(47 %), WC(0)07 %), 1-adamantylC(0)(52 %), 2-furyl-C(0)(75
a),Z-thienyl-
C(O)(88 %), MeOC(0)(42 %), Me2NC(0)(89 %)
o - C ( O ) (45%) C N - C ( 0 ) (79 %)
C
n N-C(O) (82 %) w
N-C(O) (55 %) 0
Scheme 3-1 Preparation and yield of various P-substituted ethynyl(pheny1)iodonium triflates, 11, via PhI(CN)OSO,CF,.
3.2.2 Alkynyliodonium Tetrafluoroborates Interaction of sila-alkynes with excess iodosobenzene 4 in the presence of excess triethylox-
onium tetrafluoroborate in dichloromethane at room temperature leads [27] to the formation of alkynyliodonium tetrafluoroborates 12 [Q. (4)]. A variation of this procedure employs EtzO BF3 followed by treatment with aqueous NaBF, [Eq. (S)] [27, 281. To date only alkyl-
-
and aryl-substituted homologs, along with the silyl ethynyliodonium tetrafluoroborates, 12, have been reported via these procedures. RC=CSiMe3 + 1.6 PhIO 4
1.6 Et30+BF4 + RC&'hBF4 CHZC12, RT. 12
(4)
1%R(Yie1d) :Ph(65 %), PhC% (56 %), PhC&CH2 (75 %), n-CgH17 (70%), c-Wl1 (64 %)
70
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
12 :Rwield): Ph (79 %) P h C b (54 %), n-qH,7 (85 %), Me& (83 5%)
Interestingly, treatment of Me3SiC= CH under these conditions does not yield the parent iodonium tetrafluoroborate (12, R=H). The parent system can be obtained by reaction of the silyl system with 48 Vo H F [Eq. (6)] [28]. Direct conversion of a terminal alkyne to 12, via treatment with the poxo-bis-BF4 13, has also been reported [29] [Eq. (7)]. However, the generality and scope of this interesting and simple reaction was not reported.
12 R=H
CH3(CH2),C=CH
+
+ P&O-bh
2gF4 CHzaz' R'T;
13
CH3(CH&CECIPh BF4
(7)
12: (42%)
Similarly, treatment of t-BuC = CH with a 1 : 1 mixture of iodosobenzene and CF,SO,H (PhIO - TfOH) is reported to give the t-butylethynyl(pheny1)iodonium triflate [30]. Benzilic oxidation and formation of the acylalkyne 15, rather than alkynyliodonium salt, was observed in the reaction of 14 with PhIO and BF, OEt, in dioxane [Eq. (8)] [27].
ArCHzC=CSiMe3
+
-
EtzOSBF,, R.T. PhIO Dioxane
0 I1 ArC-CECSih4e3
15
14
3.2.3 Heterocyclic Alkynyliodonium Species
-
Reaction of trimethylsilylalkynes with 16 in dichloromethane in the presence of BF, OEtz at room temperature followed by heating in methanol at 60 "C results in the stable heterocyclic alkynyliodonium salts 17 [Eq. (9)] [31]. These species represent intramolecular iodonium salts where the counterion is a carboxylate. Unlike the acyclic alkynyl(pheny1)iodonium carboxylates 18, that are unstable to isolation and decompose to the corresponding alkynyl benzoates 19 [Eq. (lo)] [32], the cyclic analogs 17 are readily isolable.
16 17:R= c-C6Hll (34%); R= n-GHl7 (22 %), R= t-Bu (35 %)
17
3.2 Preparation and Properties
71
Likewise, interaction of terminal alkynes with 20 in refluxing acetonitrile in the presence of toluenesulfonic acid gives the intramolecular heterocyclic alkynyliodonium salts 21 [Eq.(ll)] [33]. A wide range of the alkyl- and phenyl-substituted congeners 21 of these intramolecular analogs of the acyclic iodonium tosylates, 9, are obtained in 26-70% isolated yields.
Treatment of alkynylsilanes with 22, prepared from iodosobenzene and two molar equivalents of CF3S03H,in acetonitrile results in the bis-iodonium (p-phenylene) bistriflates 23 in yields of 49-83% [Eq. (12)l [34].
22
23
3.2.4 Mechanism of Formation Few, if any, direct mechanistic investigations have been reported in the literature on the formation of alkynyliodonium salts. However, all available evidence suggests that the reaction involves initial electrophilic addition of a highly polar or ionic h3-iodane to the triple bond and formation of a vinyl cation 24 (or vinyl cation-like intermediate). The reaction of the cyano species, 7, is illustrative, as summarized in Scheme 3-2.
RCmMR;
+
-R;MCN*
RC-CiPh
PhiCN6S02CF3
-
Ph,,/CN Re=C
/
24
GSO2CF3
Scheme 3-2 Mechanism of formation of alkynyliodonium salts.
Among the evidence for this mechanism is the fact that PhIO alone does not react with alkynes and that activation of the iodosobenzene by Lewis acids such as BF, . OEtz is re-
72
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
quired. Similarly, the more ionic triflate species, i. e., 6 or 7, seem to react better than the less ionic tosylate such as HTIB (5). Relatively polar solvents like CH2C12 and CH3CN are required for reaction. Most importantly, the p-substituted silicon or tin plays a key role in the stabilization of the incipient vinyl cation 24. The stabilization of carbocations by p-Si substitution, and even more by B-Sn substitution, is well established and understood [35]. Finally, all the best iodine(II1) reagents carry a ligand that has a high affinity for Si or Sn, facilitating the elimination step leading to the desired final product. In other words, the mechanism of formation of these alkynyliodonium species involves a simple addition-elimination process.
3.2.5 Diynyliodonium and Dialkynyliodonium Triflates Reaction of butadiynyltributylstannanes25 with 7 in dichloromethane at -40 "Cresults in the corresponding 1,3-diynyl(phenyl)iodonium triflates 26 [Eq. (13)] [36]. These novel iodonium salts represent a new type of functionalized, conjugated diyne as well as an interesting iodine(II1) species. Iodonium salts bearing two alkynes as organic ligands are also known [37]. These species, 28, are prepared by the reaction of sila-alkynes with iodosyl triflate, 27 1%. (W11381. R C ~ - C = C S n F l u 3 + d C N OTf 25
cH2"2. -40 o c
Rm-CebhGTf
7
(13)
26
26: Rvield): Me (77 %), n-Bu (80 %), t-Bu (84 %), MgSi (96 %), Ph (80%)
27
28
28: Rvield): t- Bu (46 %), MqSi (43 %), i-Pr3Si (83 %)
3.2.6 Bis-iodonium species A number of interesting bis-iodonium acetylenes and bis-iodonium diynes were reported in the early 1990s. Reaction of bis-tin-acetylene 29 and bis-tin-diyne 31 with two equivalents of 7 in cold dichloromethane results in the formation of novel bis-iodoniurn ethyne 30 [24] and bis-iodonium diyne 32 [39], respectively [Eqs. (15), (16)]. Bu3SnC=CSnl3t~ + 2ph;CN 6Tf 29
7
t
cH2c12-
N28
-30 0 to - 20
t
PhIC=CIPh 20Tf
30 (80%)
3.2 Preparation and Properties
73
Similarly, tethered conjugated as well as nonconjugated bis-iodonium diyne bistriflates 34, 35 [40], and ditosylates 36 [41], have been reported [Eqs. (17), (lS)]. Reaction of the respective bis-tin-alkynes with two equivalents of 7 results in good to excellent yield of the corresponding bis-iodonium diynes 34-36. B u 3 S n C E C e nC s S n B u 3
-780 to R.T. + 2PhI+CN&€ CH2C12, -2Bu3SnCN w
34: n=l(82 %); n=2 (92 %)
Bu3SnC=C fcHz)nC=CSnBg
+ 2&CN 6S02R
CHzCl2, -780 to R.T. -2Bu3SnCN
P h b@ 2 ) $ & P h
20SOzR
35: n=2, R e 3 (90%), n=4, R=C& (93 %); n=5, R = Q (90%) n=6, R=Clj (89 %) n=8, R=CF3 (90 %) 36: n=4, R=p-MeC& (61 %) n=6,R = p - M e w (67%) n=8, R=p-Me=
(74 %)
Likewise, the reaction [42] of three equivalents of 7 with the tris-tin-alkyne 37 in cold dichloromethane gives the symmetrical tris-iodonium salt 38 [Eq. (19)].
+
C S P h
+ 3PhkN6Tf *CH&Iz, -78 OC
30Tf
- 3 Bu3SnCN
C Bu3Sd
37
c*
(19)
+
CPh
38 (78 %)
3.2.7 Properties of Akynyliodonium Salts All pure alkynyliodonium species are microcrystalline solids. Their solubility is limited to nonnucleophilic polar solvents. They are insoluble in hydrocarbon, aromatic, ether, and other nonpolar solvents. They tend to decompose in methanol and ethanol as well as in dimethyl sulfoxide (DMSO), dimethylformamide (DMF), etc. With the exception of the parent ethynyliodonium salt they are also insoluble in water. The best solvent seems to be acetonitrile. Reactions may also be carried out under initially heterogeneous conditions in solvents like CH2C12, CHCI, where the mixture becomes homogeneous as the reaction proceeds. The stability of alkynyliodonium species is highly dependent upon both the counter-anion and the P-substituent on the alkyne. The more nucleophilic the counter-ion, the less stable the iodonium salt. Hence, the order of stability as a function of counterion is approximately:
74
3 Alkynyliodonium Salts Elecirophilic Aceiylene Equivalents
CF,SO, = BF; a CIO, >ArSO, = CH3S0, > CF3C02 > (RO),PO, >C1- > C6H5C0, > CH,CO,. Hence, the most stable and therefore the most widely used species are the alkynyliodonium triflates, tetrafluoroborates and tosylates. In general, the parent ethynyliodonium salts and simple alkyl-substituted compounds are the most stable and may be indefinitely stored as pure solids in a refrigerator. The p-functionalized (11) and bis-iodonium (34-36) compounds are somewhat less stable but may be isolated, characterized, and stored cold for several days in pure form. The stability of both the diynyliodoniums 26 and the dialkynyl compounds 28 are greatly dependent upon the substituents R. The greater the steric bulk of R, the more stable the compound. Perhaps, the least stable, fully characterized species is the bis-iodonium diyne 32, which decomposes rapidly above -20°C and hence is best made fresh and used in situ. The large majority of alkynyliodonium salts, and all alkyl-substituted ones are white or offwhite in color. Some P-functionalized (11) and conjugated bis-iodonium (34, 38) compounds are yellow. Although a few explosions have been reported [43]with Phi -0-iPh2BF4 and also with perchlorates, we have not experienced any problems to date with any of the alkynyliodonium triflates or tosylates. Nevertheless, it is prudent to exercise due caution in the handling of all iodonium species.
3.3 Characterization and Structure Alkynyliodonium species are readily characterized by modern spectroscopic techniques. Furthermore, several single-crystal X-ray determinations have provided unambiguous structural information.
3.3.1 Spectroscopic Properties The Fast Atom Bombardment (FAB) mass spec+ra of alkynyliodonium salts generally show reasonable peaks for the intact cations RC = CIPh [(M -Anion)+] that are very useful for the identification of the individual compounds. Subsequent fragmentation patterns can also be valuable in identification. The infrared spectra have a weak, but clearly discernible, C = C absorption between 2120 and 2190 cm-'. Strong absorption, highly characteristic of the anions is also present: intense, broad bands between 1OOO and 1100 cm-' for BF; and two intense signals around 1000 and 1270 cm-' for CF3S0C. Moreover, the P-functionalized ethynyliodonium salts 11 display useful, characteristic absorptions for the specific functionality such as C =N, carbonyl, etc. Equally characteristic and very useful are the NMR spectral data. The '% spectrum has a sharp singlet at around -78 to -79 ppm for the ionic CF,SO, for all alkynyliodonium triflates and at about - 150 ppm for BF; . The aromatic region of the 'H NMR spectra for all alkynyl(pheny1)iodonium species is highly characteristic, with three distinct multiplets between 1.4 and 8.3 ppm in a 2 : 1 :2 ratio. The ortho protons of the phenyl group resonate between 8.1 and 8.3 ppm, the para proton at about 7.7 ppm, and the metu protons around 7.5 ppm; these may be compared with -7.6 ppm and -7.0 ppm for the ortho protons, and
3.3 Characterization and Structure
75
the meta and para protons, of iodobenzene itself. The deshielding of these aromatic protons is in accord with the strong electron-wit\drawing effectpf the iodonium moiety. Whereas oI is only 0.45 for iodine, it is 1.24 for PhIBF; (with PhI (YTf- similar), 1.17 for -1C1, and 0.85 for -I(OAc), [44]. Likewise, the I3C NMR spectra are highly characteristic and informative. Most distinctive are the resonances of the C, and Cp acetylenic carbons, with the former generally between 20 and 40 ppm and the latter at 110- 120 ppm. The shielding and concomitant upfield shift of the C, signal compared with the common acetylenic carbon signals of 60-90 ppm are attributed to the spin-orbital effects [45] of the heavy iodine atom, whereas the considerable downfield shift of the+Cp resonancz is due to the resonance-induced electron deficiency of RC =C=IArX-. For example, for the parent ethynyl(phethis carbon: RC = C - IArXny1)iodonium triflate HCp=C,IPh OTf-, the C, is at 27.3 ppm and the Cp at 98.3 ppm compared with 71.9 ppm for acetylene itself. Interestingly, the 13C NMR spectrum for 30 shows the acetylene resonance as a singlet at 51.8 ppm, in between those of the a- and l3-carbons of ethynyliodonium triflate, but upfield from the signal of acetylene itself at 71.9 ppm. The aromatic carbon signals of the aromatic group of alkynyl(pheny1)iodonium salts are also shifted downfield except for the ipso-carbon that is actually shifted upfield in the region of 110 ppm. In fact, a deshielding of 20 to 30 ppm is observed upon going from iodine(1) of iodobenzene to iodine(II1) of the iodonium species [46]. For iodonium triflate salts there is I320 Hz), a further signal in the I3C NMR spectra, usually centered at 121 ppm (9, JCPF arising from the fluorine-coupled carbon of the CF3SOj- group.
7
3.3.2 X-ray and Molecular Structure To date, six single-crystal X-ray molecular structures of alkynyliodonium compounds have been reported: four alkynyl(pheny1)iodonium salts including the substituted cyanoethynyl salt, one heterocyclic and one dialkynyliodonium salt, all with oxyanions. Key structural data for these compounds are summarized in Table 3-1. The data in Table 3-1 indicate that alkynyliodonium salts, like the related 10 - I - 3 species PhICl, , PhI(OAc), , and HTIB (5), can be considered as distorted pseudo-trigonal-bipyramidal (39), or approximately T-shaped in the solid state.
39a
39b
In other words, there are two apical ligands, an equatorial ligand and two lone-pair electrons also in the equatorial position around the central iodine(II1) atom. As expected by simple electrostatic considerations, the most electronegative groups occupy apical positions and the least electronegative group is in the equatorial position. Hence, in all known alkynyl(pheny1)iodonium species to date the alkyne and counter-anion occupy the apical sites whereas the less electronegative phenyl resides in the equatorial position.
16
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
Table 3-1 Key X-Ray Molecular Structural Data for Alkynyliodonium Salt Bond distances (A) Compound HCEC-I-OTf
1 Ph
Pli€ZC--I--OTs
I
Bond angles (deg) o-I-c,pl
Refs.
Csp-1
I-C,2
1-0
2.017
2.108
2.620
93.2
170.9
78.3
[20b, 211
1.969
2.120
2.556
95.0
170.9
76.8
[18bl
2.01 2.00
2.10 2.1 1
2.62 2.56
93.6 92.1
171.7 175.7
80.5 83.9
[22c, 251
2.006
2.124
2.618
93.1
172.4
81.1
[471
2.03
2.14
2.34
90.9
166.7
75.8
13 11
2.69
92.6
-180
c,p-I-cspz
c,-r-o
Ph
i-Pr,Si Cm-t-OTf
6111
C i-Pr$i
-89 2.02 2.01
[371
@)Twocrystallographically distinct species
The C,, -I - CSp2bond angle is between 90 and 95" and the C,, -I - 0 angle is in most cases around 171". The apical C,,-I bond length is just a bit over 2.0 A whereas the equatorial I - CSp2bond length is slightly over 2.1 A. The I - 0 bond length varies between 2.3 and 2.7 A and is well outside the s u m of the theoretical covalent radii of iodine and oxygen and the I - 0 single-bond lengths of 1.99 br, indicating the ionic, salt-like, nature of these compounds, even in the solid state. Theoretical calculations [48] are in accord with these structural data and further indicate the importance of ionic bonding and negative hyperconjugation over d-orbital participation in the bonding nature of hypervalent molecules.
3.4 Reactions and Uses of Alkynyliodonium Salts Polycoordinated iodine(II1) chemistry has experienced a renaissance in the last decade, largely due to the ready availability of alkynyliodonium and the related alkenyliodonium species. Moreover, the carbon-carbon triple bond is one of the oldest, simplest and most useful functional groups in organic chemistry. Besides the common hydrocarbon acetylenes, a large variety of functionalized alkynes are known and play an important role in numerous organic
3.4 Reactions and Uses of Alkynyliodonium SaIts
77
transformations. Iodonium-substituted alkynes add a new dimension to acetylenic transformations and in particular to the reaction of alkynes with nucleophiles and organometallic complexes.
3.4.1 Reaction with Nucleophiles As indicated in Section 3.1, SN-1 or direct S,-2 type displacements on acetylenes are unknown and all reactions of alkynes with nucleophiles proceed by alternative pathways. In the case of alkynyliodonium salts, all available evidence indicates that the first step in their reaction with nucleophiles is a Michael addition to the electron-deficient P-carbon to form an ylide, 40, as the initial intermediate (Scheme 3-3). This ylide intermediate may be protonated to give a stable vinyliodonium salt, 41, as the product, in the presence of a ready proton source. More likely, the ylide undergoes loss of iodobenzene (analogous to the loss of N, from a diazonium ion) to give the well-known 1491 unsaturated carbene 42 as the next intermediate. If either of the two P-substituents of this carbene is a group or atom with a high migratory aptitude, the carbene undergoes rearrangement to the alkyne 43. The end result of this process is a nucleophilic acetylenic substitution reaction via an addition-elimination-rearrangement pathway, as outlined in Scheme 3-3. Alternatively, if both the precursor nucleophile and R of 42 are groups or atoms with a poor migratory aptitude, and in the absence of viable external traps, the carbene inserts into any available 1,5-carbon-hydrogen bond, resulting in a substituted cyclopentene 44 as the product. All three products 41, 43, and 44 have been observed, depending upon both the exact reaction conditions as well as the nucleophile employed. NU\
C=CHbh
d
44
41
42
Scheme 3-3 Mechanism of reaction of alkynyliodonium salts with nucleophiles (note Nu- stands for both a neutral and charged nucleophile).
For reasons not yet completely understood, only soft nucleophiles react well with alkynyliodonium salts. Hard nucleophiles such as alkyllithium, alkoxides, simple enolates, etc., give only decomposition products. A possible explanation for this observation might be that hard nucleophiles either attack directly on the iodine or undergo electron transfer pro-
78
3 A Ncynyliodoniurn Salts Electrophilic Acetylene Equivalents
cesses (rather than Michael addition) and subsequent decomposition. A wide variety of nucleophiles interact with alkynyliodonium salts. Here we shall discuss, in order, the reaction of carbon, nitrogen, oxygen, sulfur, and phosphorus nucleophiles, as well as halogens, with alkynyliodonium salts.
3.4.1.1 Carbon Nucleophiles
The most investigated carbon nucleophiles are 0-dicarbonyl or related enolates, generally cyclic in structure and fully substituted (i. e., tertiary) at the nucleophilic carbon. The products of reaction of these enolates with alkynyliodonium salts are 0-dicarbonyl compounds, and their formation is mostly dependent upon the migratory aptitude of the substituent on the ethynyliodonium salt. As a consequence of the superior migratory aptitude of hydrogen, the parent ethynyliodonium salt gives exclusively rearrangement (alkynylation) products with diverse P-dicarbonyl nucleophiles in 63-78% yield (Scheme 3-4) [28]. This reaction represents a very convenient way of introducing the HC = C - functionality into keto and ester P-dicarbonyl compounds.
Me 0
ONO'
EtOK C=CH -XPh 48
CECH
50
49
+
Scheme 3-4 Products of reaction of &&carbony1enolates with HC = CIPh or THF [28].
BF; , t-BuOK in t-BuOH
A similar interaction of (2-oxoazetidiny1)malonates 51 with sila-ethynyliodonium triflate - 78 "C affords the corresponding ethynylmalonates, 53, in 92% yield [50] [Eq. (2011.
(52) in THF at
53
Analogously, malonate 54 gave exclusively alkynylation products 55 in 33-95 Yo yields with a variety of alkynyliodonium triflates [Eq. (21)] [20a].
3.4 Reactions and Uses of Alkynyliodoniurn Salts
79
CECR &C=N-C(COZJ~)~
+
RCEChh OTf THF, OC.
54
I
Pl&=N--C(C02&)2 55: R=Me3Si, Ph, n-Bu, t-Bu
(21)
In contrast, the reaction of a range of P-dicarbonyl nucleophiles with a variety of alkynyliodonium salts substituted with an alkyl group possessing a y-CH bond results in the cyclopentene products derived by carbene insertion [51] as illustrated in Scheme 3-5. In fact, as shown in Eq. (22), the alkyl chain need not be restricted to the alkynyliodonium salt but may instead be part of the enolate nucleophile [51].
R2
+
R,CH$H,CH$=Cbh
iF,
t-BuOK, t-BuOH or THF
0
Scheme 3-5 Cyclopentene annulations via a tandem Michael addition-carbene insertion reaction [51].
CH3
With bis-iodonium salts 56,bis-insertion products 57-59 are observed [52], as summarized in Scheme 3-6.
mNu
Nu
PhiC-f
CH2jnCm&h 20Tf + 2Nu-
56: n=S,6,7
57: (73-77 8)
-
f
-2PhI
\ n=7\
Nu 58: (72-748)
Nu&CH~Q 59: (66-73%)
Nu
Scheme 3-6 Bis-cyclopentene formation via double Michael addition and carbene insertions in the reaction of nucleophiles with bis-iodonium diynes.
80
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
Furans 61 are obtained if activated carbonyl compounds with acidic methylene protons 60 are employed as nucleophiles [51]. In these cases, insertion of the carbene into the enolic 0 - H bond occurs [Eq. (23)].
__
0
II PhCCH2Y
+
RCdPhkF4
kBuoK t-BuOH or THF
Y
H PhC=C-C-R
I
(23)
Y
R
61 R=Me, n-CgHI7
Reaction of vinylcopper reagents, 62, with alkynyliodonium tosylates results in conjugated enyne 63, [53] [Eq. (24)l. The reaction is stereospecific with retention of olefin geometry. By appropriate order of addition, either of the two possible isomeric trisubstituted olefin isomers, 63, can be obtained in good isolated yields and excellent (> 99%) stereoselectivity. Likewise, conjugated diynes, 65, are obtained [54] in the reaction of dialkynylcuprates, 64, with alkynyliodonium tosylates [Eq. (25)]. This method may be used for the preparation of unsymmetrical diynes. The mechanism of these coupling reactions is not understood at present.
( RCe%Cu(CNr&i +
+ R'C=CPhOTs
TIE * -70 O C to R.T.
RCW-CWR
(25)
65
64
Finally, alkynyliodonium tosylates are subject to alkoxycarbonylation and formation of alkynoic esters, 66, via Pd-catalyzed CO insertion under very mild conditions [55] [Eq. (26)]. + RCECPh OTs
+ CO + R'OH
pd(OAc)2 R.T.
Et3N
*
0
II RC=CCOR'
(26)
66
3.4.1.2 Nitrogen Nucleophiles Reaction of fi-functionalized alkynyliodonium triflates, 11, with LiNPh, results in various push-pull ynamines, 67, in 43-66% isolated yields [56] [Eq. (27)]. Treatment of alkynyliodonium tetrafluoroborates with Me,SiN, in wet CH,Cl, results in the stereoselective formation of (2)-B-azidovinyl iodonium salts 68 [Eq. (28)] in 50-91 070 isolated yields [57].
3.4 Reactions and Uses of Alkynyliodonium Salts
+ -
YC=CIPhOTf + LiNPh2
Et20
b
-78 OC to R.T.
11
YCECNPh2
81
(27)
67: Y = A r S 0 2 - , PhC(0)-, etc.
68
These latter reactions are postulated to involve the in-situ formation of HN3 where the azide ion adds in a Michael fashion and the intermediate ylide is subsequently protonated (see Scheme 3-3) to give the observed vinyliodonium salt, 68. Likfwise, the P-azidovinyliodonium tosylate is isolated in 68 070 yield in the reaction of PhC = CIPhTsO- with NaN3 in CH2C12 in the presence of 18-crown-6 [58]. In contrast, reaction of the t-butyl compound in the presence of Et3SiH results in a 61 Vo yield of the P-azidovinylsilane, 69 [Eq. (29)], presumably via insertion of the intermediate unsaturated carbene (Scheme 3-3) into the Si - H bond [%I.
3.4.1.3 Oxygen Nucleophiles The reaction of alkoxide ions with alkynyliodonium salts is unproductive, leading to only decomposition products rather than the desired alkoxyacetylenes. Similarly, reaction of R3SiO- does not lead to any siloxyalkynes. In contrast the softer sulfonate, carboxylate, and phosphate nucleophiles all readily react with alkynyliodonium salts leading to the corresponding alkynyl sulfonate, carboxylate and phosphate esters [4]. Reaction of alkynyliodonium sulfonates, 9, in dry acetonitrile in the presence of catalytic amounts of AgOTs or CuOTf leads to the formation of alkynyl sulfonates, 70, in reasonable yields [Eq. (30)] [MI. In a similar manner bis(alkyny1iodonium) tosylates, 36, give modest yields of bisalkynyl tosylates, 71, accompanied by some monotosylates, 72 [Eq. (31)] [41].
82
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
Alkynyl carboxylate esters, 75, are obtained in the reaction of lithium acetylides with bis(acyloxy)iodobenzene, 73 [Eq. (32)] [59]. These reactions are likely to proceed through the intermediacy of the respective alkynyliodonium carboxylates (74), although no such salts have been isolated to date as they spontaneously decompose, via loss of iodobenzene, to the alkynyl carboxylates, 75. Only benzoate esters (75: R = C,H,) are sufficiently stable to isolate and store pure for longer periods. Simple alkylcarboxylates such as acetates are not stable although the hindered pivaloate ester (65: R = t-Bu, R = t-Bu) has been isolated in low yield [59]. Among the reasons for the instability of these esters is their sensitivity to moisture; they both readily add water and undergo subsequent hydrolyses [60]. Because of the sensitivity to moisture, the isolated yield [41] of bisalkynyl benzoates, 76, from the bisalkynyliodonium triflates, 35, is only 6-15% [Eq. (33)]. 0
*[
PhI(OCOR)2 + R'CECLi -78 oc THF to R.T. 73
I1
RC&h
6 C O R ] s R'C=COCR 75
14
35
(32)
R
0
(33)
ArCOCWfCH2~CECOCAr 11 n
76: n = 6.8
In contrast, alkynyl dialkyl phosphate esters, 78, are formed in good isolated yields by either the treatment of alkynyliodonium triflates with (RO),PO,Na or the reaction of terminal alkynes with [hydroxy(phosphoryloxy)iodo]benzene, 77 [Eq. (34)], or the sequential treatment of alkynylsilanes with PhIO Et20BF, followed by aqueous (R0)2P0,Na [Eq. (35)] [61]. These new, alkynyliodonium-derived, acetylenic esters have potent biological activity [4] : in particular, the alkynyl benzoates are protease inhibitors [62], whereas the alkynyl dialkylphosphates, 78, are inhibitors of a bacterial phosphotriesterase [63].
-
RCECH
0
0 II
+ PhI(OH)(OP(OR')2) I7
cH2a2
II m RCECOP(0R')z 78
(34)
78
Phenoxide ion, unlike alkoxides, interacts favorably with alkynyliodonium species. Reaction of two equivalents of lithium phenoxide with the [bis(phenyliodonium)ethyne, 30, results in diphenoxyacetylene, 79 [N] [Eq. (36)]. Benzofurans, 81, are obtained in the reaction of 23 with PhONa in methanol [Eq. (37)] [MI. As indicated, these products arise via insertion of the intermediate carbene, 80, into the ortho-C - H bond.
3.4 Reactions and Uses of Alkynyliodoniurn Salts
+ f phICsCIPh 2OTf
+
2PhOLi
CH2Q2
b
-78 O C to R.T.
83
(36)
PhOCECOPh 19
30
80
23
6Tf
1
81
3.4.1.4 Sulfur Nucleophiles
A wide variety of sulfur nucleophiles react readily with alkynyliodonium salts. Reaction with sodium thiocyanate in aqueous CHzC12 afforts alkynyl thiocyanates, 82, in 70-94 Yo yields [Eq. (38)] [65]. Similarly, diyne dithiocyanates, 83 and 84, are obtained in 69430% yield from reaction of 34 and 35 with NaSCN [Eqs. (39) and (40)], respectively [41]. Likewise, alkynyl thiocyanates, 82, are obtained from 23 and KSCN in DMF [Eq. (41)] [66].
83: n = 1,2
+
P&iCfCH2~C=CIPh n 35
-
20Tf
+ 2NaSCN
HzO, CH3CN
NCSC=C+H~+SCN n 84: n = 2,4,6,8
The reaction of ArS(0)2SK [67] and (R'O),PS,K [68] with alkynyliodonium salts results in alkynyl thiotosylates, 85, and alkynyl phosphorodithioates, 86, respectively [Eqs. (42), (43)] in good isolated yields. Interaction of thiocarboxylates, 87, with alkynyliodonium triflates gives the hitherto unknown alkynyl thiocarboxylates, 88 [Eq.(44)][69].
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
84
P
~
~
l
C
~
R
2
+OKSCN T
-
f DMF
RCSSCN
23
R C E C b h 6Tf
+
82
0 ArS(0)zSK
CH2Q2w
20 0
RCEC-S-SAr 0 85
Interaction of alkynyliodonium compounds with arylsulfinate salts is particularly interesting. When the R group of the alkynyl moiety lacks a y-CH bond, alkynyl sulfones (89, 90) are formed in excellent isolated yields [70,711 [Eqs. (43,(46)J.When y-CHbonds are available, the intermediate unsaturated carbene (Scheme 3-3) prefers insertion over rearrangement and hence cyclopentenyl sulfones, 91, predominate, although some alkynyl sulfone formation is also observed as illustrated in [Eq. (47)l [72].
RCECbh6Tf
+
ArS02Na
-
0 RCZCSAr
(45)
0
89
n-Bu, h0; SPh, 0 OC THF,H20
& P fh
+ PhfCH*%WSPh 0
(47)
0
PhS 0
(minor) 91 (major)
Because of the poor migratory aptitudes of both sulfones and the keto as well as the amido moieties, exclusive cyclopentene formation is observed in the reaction of P-ketoethynyl- and p-amidoethynyl-iodonium triflates, 11, with sodium p-toluenesulfinate in anhydrous dichloro-
85
3.4 Reactions and Uses of Alkynyliodonium Salts
methane at 20°C [23]. The full synthetic potential of this reaction is summarized in Table 3-2, which shows that this new methodology readily affords not only simple cyclopentenones and y-lactams but also fused bicyclic systems, and hence nicely complements the Nazarov [73] and related cationic cyclizations and the Pauson-Khand [74] Co-mediated cyclizations for cyclopentenone construction, of importance in numerous natural products. Table 3-2 Cyclopentenones and y-Lactams via the Reaction of Na02SC,H,CH, with P-Keto- and
P-
Amidoethyliodonium Triflates Starting Iodonium
Product, Yield
>(k,
Starting Iodonium
Product, Yield
IPh 6Tf
(72%)
0
(75%)
In contrast to the behavior of sulfinate salts toward alkynyliodonium salts, phenylsulfinic acid, 92, in methanol, trap [72] the initially formed ylide, resulting in high yields of P-sulfonylvinyliodonium salts, 93, [Eq. (48)l.
(a-
86
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
RC&hiF4
+
R\ c=c/H
(3330H* PhSOzH
p ~ ’
92
0
‘bhBF4 93
Sodium thiophenoxide reacts with the bisiodonium species 30 and 32 affording the bis(phenylthio)alkynes, 94, in 66-67% yields [24, 391 [Eq. (49)l. cH3cN3 0 n = 1,32:n = 2
PM-f C S C i S P h
(49)
%n=1,2
3.4.1.5 Phosphorus Nucleophiles
All varieties of alkynyliodonium salts readily react with triphenylphosphine resulting in the corresponding alkynylphosphonium salts in excellent yields. For example, reaction of alkynyliodonium triflates with Ph,P in cold dichloromethane gives alkynylphosphonium salts, 95, in nearly quantitative yields 1751 [Eq. (5011. + RCECIPh OTf
+m
P
m2Q2 -78 oc to Ref
+
-
RCECPPb OTf 95
Likewise, the bis-iodonium diyne triflates 34 and 35 give the bisphosphonium diynes 96 and 97 in high isolated yields [40][Eqs. (51), (52)l.
34
% n = 1,2
Advantage has been taken of the ready interaction of phosphines with alkynyliodonium salts in the alkynylation of tetra(t-butyl)tetraphosphacubane, 98, to give the phosphacubane salt, 99 [76] [Eq. (53)].
3.4 Reactions and Uses of Alkynyliodonium Salts
98:R = t-Bu
WR=t-Bu R'=H,CH3
87
111
C
R'
Similarly, reaction of t-butylethynyliodonium tosylate with bis(diphenylphosphino)methane, 100,in benzene gives the novel 1h5,3h5-diphospholium ion, 101,in 77% yield [77][Eq.(54)].
100
101
Trialkyl phosphites undergo reaction with alkynyliodonium tosylates, resulting in dialkyl alkynylphosphonates, 102, via an Arbuzov-type process 1781 [Eq.(55)l.
@'0)3p+ R C & ' h 6 T s
-
0
II
(55)
RC=C-P(OR')r
102
3.4.1.6 Halogen Nucleophiles
Reaction of alkynyliodonium tetrafluoroborates with either LiX in acetic acid or HX in methanol results in the stereoselective formation of (9-P-halovinyliodonium halides, 103 [79] [Eq.(56)]. Once again, the ylide resulting from Michael addition of the halide to the P-carbon of the alkynyliodonium salt is protonated, prior to loss of PhI and carbene formation (Scheme 3-3), to give the observed (3-halovinyliodonium species in high isolated yields. The P-halovinyliodonium species 103 serve as precursors to P-haloalkylidene carbene 104, via base-initiated a-elimination of PhI; subsequent rearrangement gives the corresponding haloalkyne, 105, and/or halocyclopentenes [80] [Eq. (57)].
+
-
RC=PhBF4
WAcOH
R\
/H
x/c=c\+
P h X-
103:X = Br, C1
103
104
105
88
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
3.4.2 Reaction with Organometallic Species The great majority of o-acetylide transition metal complexes are prepared by interaction of a metal halide with acetylide, RC=C-, or the formal oxidative addition of terminal alkynes or alkynyl stannanes to the metal center. As amply demonstrated in the previous section, alkynyliodonium salts may serve as electrophilic acetylene equivalents. In other words, transition metal complexes may act as nucleophiles in reactions with alkynyliodonium species. Indeed, the reaction [81] of the square planar Vaska’s complex, 106, and its Rh analog, 107, with a variety of alkynyliodonium triflates in toluene results in 89-96% isolated yield of the hexacoordinate o-acetylide complexes, 108 and 109 [Eq. (58)J. Reaction is essentially instantaneous and occurs with retention of stereochemistry around the metal center.
R
OTf
106:M=Ir 107: M = Rh
Conjugated transition metal complexes are a promising class of molecules for use as advanced materials in areas of nonlinear optics, organic conductors, and liquid crystals [82, 831. This is a consequenc of the ability of the metal to participate in n-delocalization, as well as the interaction of the metal d-orbitals with the conjugated n-orbitals of the organic moiety [83, 841. Moreover, the ability of organometallic complexes to participate in metal-to-ligand and ligand-to-metal charge transfers allows significant reordering of the n-electron distribution [85]. Recent studies have shown that a-acetylide metal complexes exhibit very encouraging thirdorder nonlinear optical (NLO) properties [85, 861. Hence, there is considerable current interest in a-acetylide complexes and conjugated, bridging, bimetallic systems. We have employed our conjugated bis[phenyl(iodonium)] diyne triflates, 34 and 110, and the tris-triflate, 38, along with Vaska’s complex, 106, and its Rh analog, 107, to give the novel conjugated bimetallic systems 111-114 (Scheme 3-7) [87]. Complexes 111-114 form in a matter of minutes and are isolated as stable, yellow, microcrystalline, solids in yields ranging from 65 to 96% with the majority isolated in greater than 85% yield. These complexes are formed under very mild conditions and are remarkably thermally stable and insensitive to air and moisture, allowing for facile handling and storage. This new methodology complements existing oxidative addition techniques and amounts to a reversal of classic metal acetylide chemistry by using the organometallic species as the nucleophile and the iodonium salts as the “alkynylating” agents. Moreover, the presence of the acetonitrile ligand afford an opportunity for possible further derivatization of these rigid linear complexes, particularly via the use of bidentate ligands and organometallic polymer formation. Interaction of alkynyliodonium triflates with bis(tripheny1phosphine)ethylene Pt(0) complex, 115, may lead to either the o-alkynylplatinum(I1) complex, 116, or the novel q3-propargyl/allenyl Pt complex, 117 (Scheme 3-8), depending both upon the group R and the exact
3.4 Reactions and Uses of Alkynyliodonium Salts
89
106M=lr lWM=Rh
34a:n = 1; R = H 34b:n = 2; R = H 110a:n= 1; R = CH,
-
2+
IoTF12
I06
llOb
Ph$’ J
L
113 +
106
114
Scheme 3-7 Formation of conjugated bridging bimetallic complexes via iodonium chemistry.
reaction conditions employed [88].Bulky substituents such as t-Bu and Me3Si favor formation of complex 117 whereas smaller substituents such as CH, favor the o-complex 116. Bubbling added ethylene through the solution during progress of the reaction also favors formation of complex 117, whereas bubbling argon (removal of ethylene) through the reaction mixture favors formation of o-complex 116. These observations clearly indicate that o-complex formation occurs via the disassociated (Ph3PhPt fragment whereas the q 3-complex, 117, is formed from the undissociated Pt(0) complex 115.
115
116
Scheme 3-8 Reaction of alkynyliodonium triflates with a Pt(0) complex.
117
90
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
3.4.3 Cycloaddition Reactions Acetylenes with electron-withdrawing substituents such as Me0,C = CC02Me or NCC = CCN have a rich cycloaddition chemistry. As alkynyliodonium salts are highly electron-deficient acetylenes they are expected to undergo a variety of electrocyclic processes.
3.4.3.1 [2
+ 41-Diels-Alder
Cycloadditions
Diels-Alder cycloadditions are among the most useful and valuable synthetic reactions. A large variety of electron-deficient olefins as well as acetylenes interact with diverse dienes resulting in cyclic products. Although no Diels- Alder cycloadditions have been reported for the simple alkylethynyliodonium salts, the p-substituted systems 11 readily react with diverse dienes [25]. The full scope of this reaction is summarized in Scheme 3-9, from which it may be seen that cyclic, acyclic, and endocyclic dienes react with 11 under unusually mild conditions to form cycloadducts, 118-121, in good to excellent isolated yields. All the cycloadducts 118-121 are stable, microcrystalline solids. These adducts have the additional advantage of carrying two different functionalities, Y and iPh OTf, that may be employed for further synthetic elaboration [89].
Y'
1%: Y = CN, 79% b Y =p-CH3C&SO2, 75%
118a:Y=cN.81% b: Y=p€H3C&SI&.
55%
d: Y e: Y = i-BUC(O), 73%
. .-. -".-\-,,
%
&,W,
CH3CN.200. 1.5-3 hr.
*.
20 min-3 In.
0
Scheme 3-9 Summary of Diels-Alder cycloadditions of 0-functionalized ethynyliodonium triflates, 11, with various dienes [25].
91
3.4 Reactions and Uses of Alkynyliodonium Salts
The bisiodonium ethyne 30 is even more reactive than 11 and undergoes Diels-Alder cycloaddition [24] with cyclopentadiene and furans in a matter of minutes at low temperatures (Scheme 3-10). The structure of adduct 122 was unambiguously established by X-ray analysis [ a ] . Cycloadducts 122 and 123 can be reacted with RC = CLi to give enediynes, 125, or with nucleophiles [90] to give 126 (Scheme 3-10).
C=CR
CH-CN. -350 to 20 OC/
P h b C b h 26Tf 30
122:Z = CH, (69%) 123:2 = 0 (73%)
7
126: 2 CH?,0 Nu:CN,Br, I
CH$N, -350 to 20 0
fph 2aTf
124 (47%) Scheme 3-10 Diels-Alder cycloaddition of bis-iodonium ethyne, 30,with cyclopentadiene and furans and subsequent reaction of the adducts.
3.4.3.2 1,3-Dipolar Cycloadditions The strongly polarized C I C bond of alkynyliodonium salts, along with their propensity for Michael additions, predicts that they should be good 1,3-dipolarophiles. Indeed, reaction of arylethynyliodonium tosylates with arenenitrile oxides, 127, gives a mix:ure of cycloadducts, 128 and 129, in 62-80% yields (911 [Eq.(59)]. Similarly, Me,SiC=CIPh m f and various diazocarbonyl compounds, 130, result [92] in cycloadducts 131 [Eq. (60)l. Likewise, alkynyliodonium salts react with methyl and phenyl azide to give low yields of triazines, 132, as adducts [Eq. (61)].
121
128
129
92
*
3 Alkynyliodoniurn Salts Electrophilic Acetylene Equivalents
H
0 M e $ i C & P h 6Tf
+
II
R-CCHN:!
Me3Si CHZCI,, R.T.
I
mPhI
130
C(0)R
131: R = MeO, EtO, Ph, t-Bu
'y) R' I
c
-
RzCIPhOTf +
R'--N3
THF or CH3CN,
n
+ TR) PhI
N
132
3.5 Conclusions It is evident that alkynyliodonium salts represent a highly versatile, new class of valuable, functionalized acetylenes. Although they have only been available for a dozen years, they provide an+added dimension to acetylene transformations. Due to the superb leaving ability of the PhI moiety, alkynyliodonium salts serve as electrophilic acetylene equivalents par excellence. They react with diverse nucleophiles, including organometallic species, thereby facilitating the preparation of hitherto unknown or not readily available functionalized acetylenes such as the new alkynyl esters. They are superb cycloaddition partners in a variety of electrocyclic reactions. Most recently, some alkynyl and other iodonium salts have shown biological activity as potent inhibitors [94] of PQQ, an organic cofactor in biological redox processes, particularly in microorganisms. However, this is just the start of the many possible applications and uses of these novel, easily prepared, functionalized acetylenes. Their ready availability from commercial precursors, reasonable stability, versatility and ease of handling should stimulate imaginative uses and thereby greatly enhance the continued development of acetylene chemistry.
3.6 Experimental Procedures 3.6. (Cyano { [(trifluoromethyl)suIfonyl]oxy iodo)benzene, 7 To a stirred suspension of PhIO (8.8 g, 40 mmol) in dry CH2CI2(100mL) at -20°C under N2 was added trimethylsilyl triflate (7.75 mL, 40 mmol). The mixture was warmed to 10°C and stirred for 10 min until the formation of a bright yellow precipitate. The reaction mixture was recooled to -30°C and trimethylsilyl cyanide (5.33 mL, 40 mmol) was added via a syringe: a white precipitate formed instantaneously upon additon of the TMSCN. The mixture was warmed to 0°C and stirred for an additional 15 min. The precipitate was filtered, washed several times with cold ether (5 mL) and dried in vacuo yielding 13.5 g (89%) of 7: mp 111-112°C (dec).
3.6 Experimental Procedures
93
3.6.2 General Procedure for the Preparation of P-Alkyl- and P-Phenylethynyl(pheny1)iodonium 'Ikiflates, 10 To a stirred suspension of PhIO (8.8 g, 40 mmol) in dry CH,Cl, (60 mL) at -20°C under N2 was added, dropwise, 7.75 mL (40 mmol) of Me3SiOTf. A bright yellow suspension of the p-0x0-bis-triflate, 6, formed immediately after completion of the addition of the Me3SiOTf. With the reaction mixture at - 20 "C, the appropriate sila-acetylene or tin-acetylene (RC = CSiMe, or RC = CSnR;, 40 mmol) was added dropwise over approximately 30 min. The mixture was allowed to warm to room temperature and concentrated to about 25 mL, and cold dry ether (30 mL) was added. The principitate was filtered and washed several times with cold, dry ether (5 mL) and then dried in vacuo. Use of n-Bu3Sn= CH gave the pa+rent HC = Ci in 50-60% yield, mp 100-101 "C (dec); CH3C=CSnBu3 gave CH3C=CIPh OTf in 74% yield, mp 115-117°C (dec); t-BuC=CSiMe, gave t-BuC=Ci m f in 86% yield, mp 132-133°C (dec); Me,SiC=CSiMe? gave Me,SiC=Cimf in 89% yield, mp 138-139°C (dec); PhCECSiMe, gave P h C r q I P h OTf in 83% yield, mp 85-96°C (dec); nin 55 Yo yield, mp 67-68 "C (dec). C6H& = CSnBu, gave n-C6HI3C=CIPh
mf
mf
3.6.3 General Preparation of P-Functionalized Ethynyl(phenyl)iodonium Triflates, 11 A solution of the appropriate functionalized alkynylstannane (YC ICSnR,, 1.03-30.9 mmol, a 3-5% molar excess) in dry CH,Cl, (10 mL) was added dropwise to a stirred 0.08 M suspension of PhICN OTf, 7 (1.00-30.0 mmol), in CH2Cl, at -42°C (CH,CN/dry-ice slush bath) under nitrogen. The initial suspension became a clear, homogeneous solution after completion of the addition. Stirring was continued at -42°C for an additional 45 min, then twice the volume of cold pentane was slowly added to precipitate the product. The microcrystalline solid was filtered cold under a nitrogen atmosphere and washed with cold, dry ether (3 x 30 mL). The resulting product was immediately recrystallized from CHzC12/EtzO/pentane, filtered, and dried in vacuo. It is important for the success of this procedure and in order to obtain good product yields that both the starting reagent (7) and the alkynylstannane be pure, as even small amounts of impurities result in diminished yields and impure products.
3.6.4 General Procedure for the Preparation of Bis-iodonium Diyne Bis-triflates, 34 and 35 A solution of the appropriate bis(tributy1stannyl)diacetylene (5.0 mmol) in dry CHzClz (20 mL) was added to a stirred suspension of PhICN OTf, 7 (3.79 g, 10.0 mmol), in CH,Cl, (100 mL) at - 78 "C. The mixture was allowed to warm to room temperature and stirred for 30 min. Dry hexane was added to precipitate the product and the resulting microcrystalline solid was filtered under nitrogen and washed with dry hexane (100 mL), then dried in vacuo. Once again it is important to use pure starting materials. Analytically pure products can be obtained by recrystallization from a concentrated solution of CH3CN and addition of CH2C1, and ether.
94
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
3.6.5 Preparation of Bis(pheny1[ [(trifluoromethyl)sulfonyl]oxy) iodo)ethyne, 30 A solution of bis(tributylstanny1)acetylene (3.02 g, 5 mmol) in dry CH2C12 (20 mL) w a s added to a stirred suspension of PhICN OTF, 7 (3.79 g, 10 mmol), in dry CH2C12(100 mL) at -78°C under nitrogen. The mixture was allowed to warm slowly to 0°C and stirred for about 10 min at 0°C until a white precipitate formed. The microcrystalline precipitate was filtered cold under nitrogen and washed with cold CH2C12(100mL), then dried in vacuo, yielding 2.96 g (81 Vo) of 30, mp 127-128°C (dec). It is important to use pure reagents and dry solvents, and to do all operations in the cold. The pure product, 30, is a white, microcrystalline solid that can be stored in a refrigerator for several weeks.
3.6.6 General Procedure for the Diels-Alder Reaction of Alkynyl(pheny1)iodonium Salts, 11, with 1,3-Dienes: Formation of Cycloadducts 118-120 The appropriate diene (1.2-4.5 mol equiv.) was added dropwise to a degassed, stirred solution of the appropriate iodonium salt, 11 (0.40-1.2 mmol), in dry CHJCN (10 mL) at 20°C under nitrogen. Stirring was maintained at room temperature for 20 min to 3 h, at the end of which the excess diene and solvent were removed using a rotary evaporator. The crude product was taken up in dry CH2C12(5 mL) and recrystallized by the addition of dry ether (10 mL) and pentane ( 5 mL). The microcrystalline solid was further purified by recrystallization from dry CH2C12/Et20/pentane, isolated by filtration, washed with dry ether (2 x 10 mL) and dried in vacuo.
3.6.7 General Procedure for the Preparation of Cyclopentenones and y-Lactams The appropriate iodonium salt, 11 (1.00 mmol), was added as a solid to anhydrous sodium p-toluenesulfinate (1.01 mmol) in CH,C12 (15 mL) at 20°C under nitrogen and stirred for 15 min, then 10 mL of water was added. The organic layer was separated. The aqueous layer was extracted with additional CH2C12 (2 x 5 mL) and the combined organic extracts were dried over anhydrous MgSO,. The solution was filtered, hexane (30 mL) was added and the majority of the solvent was removed by rotary evaporation, precipitating the product. The crude product was collected by filtration, washed with pentane (3 x 10 mL) and dried in vacuo. Further purification of the products was effected by radial chromatography (silica gel) using CH2C12/hexane (1 : 1) as eluent.
Acknowledgements I am grateful for the dedication and experimental skills of my able co-workers, as cited, and for financial support by the National Cancer Institute of the NIH (CA 16903) for our own work described in this chapter.
References
95
Abbreviations DMF DMSO FAB HTIB NLO PQQ R.T. SN-A TEBA TMSCN 61
dimethylformamide dimethyl sulfoxide fast atom bombardment [hydroxy(tosyloxy)iodo]benzene
nonlinear optical methoxatin room temperature nucleophilic acetylenic substitution triethyl benzyl ammonium chloride trimethylsilyl cyanide inductive substituent constant
References [l] P. J. Stang. Z. Rappoport, M. Hanack, L. R. Subramanian, Vinyl Cations, Academic Press, New York, 1979. [2]S. I. Miller, J. I. Dickstein, Acc. Chem. Res. 1976,9, 358-363. [3]J. I. Dickstein, S. I. Miller, The Carbon-Carbon niple Bond (Ed.: S. Patai), Wiley, London, 1964, Chapter 19, pp. 813-955. [4]P. J. Stang, Acc. Chem. Res. 1991,24, 304-310. [5]A. S . Kende, P. Fludzinski, Tetrahedron Lett. 1982,23, 2373-2376. [6]R. Glaser, 1 Am. Chem. SOC. 1987, 109, 4237-4243. [7]T. Okuyama, T. Takino, T. Suedo, M. Ochiai, 1 Am. Chem. SOC. 1995, 117, 3360-3367. [8]P. J. Stang, M. Hanack, L. R. Subramanian, Synthesis 1982, 85-126. [9]Reviews: (a) P. J. Stang, Angew. Chem., Znt. Ed. Engl. 1992, 31, 274-285;(b) M. Ochiai, Reviews on Heteroatom Chemistrx 1989,2, 92-11 1. [lo]A. Varvoglis, The Organic Chemistry of Polycoonlinated Zodine, VCH Publishers, New York, 1992. [ll] G. F. Koser, L. Rebrovic, R. H. Wettach, 1 Org.Chem. 1981, 46, 4324-4326. [12) L. Rebrovic, G. F. Koser, 1 Org.Chem. 1984, 49, 4700-4702. [13]A. J. Magida, G. F. Koser, I Org. Chem. 1984, 49, 4703-4706. [14] J. C. Martin, Science 1983, 221, 509-514. [15]For a review of 5 and its chemistry see: (a) R. M. Moriarty, K. V. Vaid, G. F. Koser, Synlett 1990, 365-383; (b) 0.Prakash, N. Saini, P. K. Sharma, Heterocycles 1994, 38, 409-431. I161 (a) N.S. Zefirov, V. V. Zhdankin, Y. V. Dan’kov, A. S. Koz’min, Zh. Org. Khim. 1984,20, 446-447; (b) J. Gallos, A. Varvoglis, N. W. Alcock. 1 Chem. Soc. Perkin lFam. 1 1985, 757-763; (c) T. R. Hernbre, C. P. Scott, J. R. Norton, I 0%.Chem. 1987, 52, 3650-3654. [17]V. V. Zhdankin, C. M. Crittell, P. J. Stang, N. S. Zefirov, Bfmhedron Lett. 1990,31, 4821-4824. [18] (a) P. J. Stang, B. W. Surber, 1 Am. Chem. SOC. 1985,107, 1452-1453;(b) P. J. Stang, B. W. Surber, Z. C. Chen., K. A. Roberts, A. G. Anderson, ibid. 1987, 109, 228-235. [19] T. Kitamura, P. J. Stang, 1 Org. Chem. 1988,53, 4105-4106. [20](a) M.D. Bachi, N. Bar-Ner, C. M. Crittell, P. J. Stang, B. L. Williamson, I Org. Chem. 1991, 56, 3912-3915;(b) C. M. Crittell, Ph. D. Dissertation, University of Utah, 1993. [21]P. J. Stang, A. M. Arif, C. M. Crittell, Angew. Chem., Znt. Ed. Engl. 1990,29, 287-288.
96
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
[22] (a) P. J. Stang, V. V. Zhdankin, B. L. Williamson, J. Am. Chem. SOC. 1991, 113, 5870-5871; (b) V. V. Zhdankin, M. C. Scheuller, P. J. Stang, Tetrahedron Lett. 1993, 34, 6853-6856; (c) B. L. Williamson, Ph. D. Dissertation, University of Utah, 1994. I231 (a) W. E. Davidsohn, M. C. Henry, Chem. Rev. 1967, 67, 73-106; (b) C. Cauletti, C. Furlani, A. Sebald, Gau. Chim. Ztal. 1988, 118, 1-35; (c) I. Omai, Organotin Chemistry, Elsevier, New York, 1989; (d) B. L. Williamson, P. J. Stang, Synlett 1992, 199-200. [24] P. J. Stang, V. V. Zhdankin, J. Am. Chem. SOC. 199l, 113, 4571-4576. [25] B. L. Williamson, P. J. Stang, A. Arif, J. Am. Chem. SOC. 1993, 115, 2590-2597. [26] B. L. Williamson, R. R. ’ifrkwinski, P. J. Stang, J. Am. Chem. SOC. 1994, 116, 93-98. [27] M. Ochiai, M. Kunishima, K. Sumi, Y. Nagao, E. Fujita, Tetrahedron Lett. 1985, 26,4501-4504 1281 M. Ochiai, T. Ito, Y. Takaoka, Y. Masaki, M. Kunishima, S. Tani, Y. Nagao, J. Chem. SOC, Chem. Commun. 1990, 118- 119. [29] V. V. Zhdankin, R. nkwinski, R. Caple, B. Berglund, A. S. Koz’min, N. S. Zefirov, TetrahedronLert. 1988, 29, 3717-3720. [30] T. Kitamura, R. Furuki, H. Taniguchi, Tetrahedron 1992, 48, 7149-7156. (311 M. Ochiai, Y. Masaki, M. Shiro, J. Org. Chem. 1991, 56, 5511-5513. [32] P. J. Stang, M. Boehshar, H. Wingert, T. Kitamura, J. Am. Chem. SOC. 1986, 108, 7832-7834. [33] G. F. Koser, G. Sun, C. W. Porter, W. J. Youngs, J. 0%.Chem. 1993, 58, 7310-7312. [34] T. Kitamura, R. Furuki, L. Zheng, T. Fujimoto, H. Taniguchi, Chem. Lett. 1992, 2241-2244. [35] Review: J. Lambert, Tetrahedron 1990, 46, 2677-2689. [36] P. J. Stang, J. Ullmann, Synthesis 1991, 1073-1075. [37] P. J. Stang, V. V. Zhdankin, A. M. Arif, J. Am. Chem. SOC. 1991, 113, 8997-8998. [38] J. R. Dalziel, H. A. Carter, F. Aubke, Znorg. Chem. 1976, I5, 1247-1251. [39] P. J. Stang, V. V. Zhdankin, J. Am. Chem. SOC. 1990, 112, 6437-6438. [40](a) P. J. Stang, R. Tykwinski, V. V. Zhdankin, J. Org. Chem. 1992, 57, 1861- 1864; (b) R. Tykwinski, Ph. D. Dissertation, University of Utah, 1994. [41] R. Tykwinski, P. J. Stang. Tetrahedron 1993, 31, 3043-3052. [42] P. J. Stang, R. nkwinski, .lAm. Chem. SOC. 1992, 114, 4411-4412. [43] (a) G. van Look, Chem. Eng. News, 1989, 67 (30), 2; (b) P. J. Stang, ibid. 1989, 67 (34), 4. [44] A. A. Mironova, I. I. Maletina, S. V. Iksanova, V. V. Orda, L. M. Yagupolskii, Zh. 0%.Chim. 1989, 25, 306-308. [45] (a) A. A. Cheremisin, P. Schastnev, J. Magn. Reson. 1980, 40, 459-462; (b) P. Pyykk6, A. GBrling, N. Rdsch, Mol. Php. 1987, 61, 195-205. [46] M. Schmeisser, K. Dahmen, P. Sartori, Chem. Ber. 1970, 103, 307-319. [47] M. Ochiai, M. Kunishima, K. Fuji, Y. Nagao, M. Chiro, Chem. Pharm. Bull. 1989, 37, 1948-1950. [48] A. E. Reed, P. v. R. Schleyer, J Am. Chem. SOC. 1990, 112, 1434-1445. [49] (a) P. J. Stang, Methoden Org. Chem. (Houben-Weyl)4th ed. 1988, Vol. E19, pp. 84-165; (b) P. J. Stang, Acc. Chem. Res. 1982, 15, 348-354; (c) P. J. Stang, Chem. Rev. 1978, 78, 383-405. [SO] M. D. Bachi, N. Bar-Ner, P. J. Stang, B. L. Williamson, .l Org. Chem. 1993, 58, 7923-7924. [51] M. Ochiai, M. Kunishima, Y. Nagao, K. Fuji, M. Shiro, E. Fujita, J. Am. Chem. SOC. 1986, 108, 8281-8283. [52] R. R. Skwinski, P. J. Stang, N. E. Persky, Tetrahedron Lett. 1994, 35, 23-26. [53] (a) P. J. Stang, T. Kitamura, J. Am. Chem. SOC.1987, 109, 7561-7563; (b) P. J. Stang, T. Kitamura, Org. Synth. 1991 70, 215-225. [54] T. Kitamura, T. Tanaka, H. Taniguchi, P. J. Stang, J. Chem. SOC, Perkin Trans. 1 1991, 2892-2893. [55] T. Kitamura, I. Mihara, H. Taniguchi, P. J. Stang, J. Chem. SOC, Chem. Commun. 1990, 614-615. [56] P. Murch, B. L. Williamson, P. J. Stang, Synthesis 1994, 1255-1256. [57] M. Ochiai, M. Kunishima, K. Fuji, Y. Nagao, J. Org. Chem. 1988, 53, 6144-6145. [58] T. Kitamura, P. J. Stang, Dtrahedron Lett. 1988, 29, 1887-1890. [59] P. J. Stang, M. Boehshar, H. Wingert, T. Kitamura, J Am. Chem. SOC. 1988, 110, 3272-3278.
References
97
[60] V. K. Brel, A. A. Gakh, V. V. Zhdankin, N. S. Zefirov, A. S. Koz’min, A. K. Korkin, T. G. Kutateladze, R. Caple, S. A. Lermontov, I.G. Plokhikh, S. 0. Safronov, P. J. Stang, N. G. Chovnikova, Dokl. Akad. Sci. SSSR 1990, 313, 1131. [61] (a) P. J. Stang, M. Boehshar, J. Lin, J. Am. Chem. SOC. 1986, 108, 7832-7834; (b) P. J. Stang, T. Kitamura, M. Boehshar, H. Wingert, J. Am. Chem. Soc 1989, I l l , 2225-2230. [62] D. Segal, Y. Shalitin, H. Wingert, T. Kitamura, P. J. Stang, FEBS Lett. 1989, 247, 217-220. [63] J. N. Blenkenship, H. Abu-Soud, W. A. Francisco, F. M. Raushel, D. R. Fischer, P. J. Stang, J. Am. Chem. SOC. 1991, 113, 8560-8561; J. A. Banson, J. M. Kuo, B. Miles, D. R. Fischer, P. J. Stang, F. M. Raushel, Biochemistry 1995, 34, 743-749. [64] T. Kitamura, L. Zheng, H. Taniguchi, M. Sakurai, R. Tanaka, TetrahedronLett. 1993,34,4055-4058. [65] D. R. Fischer, B. L. Williamson, P. J. Stang, Synlett 1992, 535-536. [66] T. Kitamura, R. Furuki, L. Zheng, T. Fujimoto, H. Taniguchi, Chem. Lett. 1992, 2241-2244. [67] B. L. Williamson, P. Murch, D. R. Fischer, P. J. Stang, Synlett 1993,858-860. [68] 2. D. Liu, Z. C. Chen, J. Org. Chem. 1993, 58, 1924-1925. [69] P. Murch, P. J. Stang, unpublished results. 1701 R. R. Tykwinski, B. L. Williamson, D. R. Fischer, P.J. Stang, A. M. Arif, J. Org. Chem. 1993, 58, 5235-5237. [71] Z. D. Liu, Z. C. Chen, Synth. Commun. 1992, 22, 1997-1999. [72] M. Ochiai, M. Kunishima, S. Tani, Y. Nagao, 1 Am. Chem. SOC. 1991 113, 3135-3142. [73] (a) S. E. Denmark, in Comprehensive Organic Synthesis (Eds. : B. M. Trost, I. Fleming), Pergamon, New York, 1991,Vol. 5, pp. 751 -784; (b) C. Santelli-Roovier, M. Santelli, Synthesis 1983,429-442. [74] (a) P. L. Pauson, Tetrahedron 1985, 41, 5655-5860; (b) N.E. Shore in Comprehensive Organic Synthesis (Eds.: B. M. Trost, I. Fleming), Pergamon, New York, 1991, Vol. 5, pp. 1037-1064; (c) N. E. Shore, Chem. Rev. 1988, 88, 1081-1119; (d) P. L. Pauson in Organometallics in Organic Synthesis. Aspects of A Modern Interdisciplinary Field (Eds. : A. de Meijere, H.tom Dieck), Springer, Berlin, 1988, p. 233; (e) M. E. Krafft, I. L. Scott, R. H. Rornero, S. Feibelmann, C. E. Van Pelt, J Am. Chem. SOC. 1993, 115, 7199-7207. [75] P. J. Stang, C. M. Crittell, J. Org. Chem. 1992, 57, 4305-4306. [76] K. K. Laali, M. Regitz, M. Birkel, P. J. Stang, C. M. Crittell, J. Org. Chem. 1993, 58, 4105-4109. [77] A. Schmidpeter, P. Mayer, J. Stocker, K.A. Roberts, P. J. Stang, Heteroatom Chem. 1991,2, 569-573. [78] J. S. Lodoya, G. F. Koser, 1 Org. Chem. 1990, 55, 1513-1516. [79] M. Ochiai, K. Uemura, K. Oshima, Y. Masaki, M. Kunishima, S. Tani, Tetrahedron Lett. 1991, 32, 4753-4756. [80] M. Ochiai, K. Uemura, Y. Masaki, J. Am. Chem. SOC. 1993, 115, 2528-2529. [81] P. J. Stang, C. M. Crittell, Organometallics 1990, 9, 3191-3193. [82] (a) D. Kanis, M. A. Ratner, T. J. Marks, Chem. Rev. 1994, 94, 195-242; @) A. Hirsch, M. Hanack in Conjugated Polymeric Materials: Opportunities in Electronics, Optoelectronics and Molecular Electronics (Eds.: J. L. Bredas, R. R. Chance), Kluwer, New York, 1990. [83] (a) T. P. Pollagi, T. C. Stoner, R. F. Dallinger, T. M. Gilbert, M. D. Hopkins, 1 Am. Chem. SOC. 1991, 113, 703-704; (b) J. C. Calabrese, L.-T. Cheng, J. C. Green, S. R. Marder, W. Tam, J. Am. Chem. SOC.1991. 113, 7227-7232; (c) L. K. Meyers, C. Langhoff, M. E. Thompson, J. Am. Chem. SOC. 1992, 114, 7560-7561 and references therein. [84] (a) D. L. Lichtenberger, S. K. Renshaw, R. M. Bullock, J Am. Chem. SOC. 1993, 115, 3276-3285; (b) D.L. Lichtenberger, S. K. Renshaw, A. Wong, C. D. Tagge, Organornetallics 1993, 12, 3522-3526. [85] C. C. Frazier, S. Guha, W. P.Chen, M. P. Cockerham, P. L. Porter, E. A. Chauchard, C. H. Lee, Polymer 1987, 28, 553-555 and references therein. [86] (a) S. Guha, C. C. Frazier, P. L. Porter, K. Kang, S. E. Finberg, Opt. Lett. 1988, 14, 952-954; (b) W. J. Blau, H. J. Byme, D. J. Cardin, A. P. Davey, J. MateK Chem. 1991, I, 245-249. [87] R. Tykwinski, P. J. Stang, Organometullics 1994, 13, 3203-3208. [88] P. J. Stang, C. M. Crittell, A. M. Arif, Organornetallics 1993, 12, 4799-4804.
98
3 Alkynyliodonium Salts Electrophilic Acetylene Equivalents
P. J. Stang, T. Blume, V. V. Zhdankin, Synthesis 1993, I, 35-36. P. J. Stang, A. Schwarz, T. Blume, V. V. Zhdankin, Tetrahedron Lett 1992, 33, 6759-6762. E. Kotali, A. Varvoglis, A. Bozopoulos, 1 Chem. Soc., Perkin Trans. I 1989, 827-831. G. Maas, M. Regitz, U.Moll, R. Rahm, F. Krebs, R. Hector, P. J. Stang, C. M. Crittell, B. L. Williamson, Tetrahedron 1992, 48, 3527-3540. [93] P. M. Gallop, M. A. Paz, R. Fliickiger, P. J. Stang, V. V. Zhdankin, R. Tykwinski, J. Am. Chem. SOC.1993, 115, 11702-11704.
[89] [90] [91] 1921
4 The Chemistry of Metal-Alkyne Complexes Gagik G. Melikyan, Kenneth M. Nicholas
4.1 Introduction The transition metal organometallic chemistry of alkynes had its beginnings in the pioneering and belatedly published studies of Reppe [l], who uncovered the diverse nickel-catalyzed oligomerization chemistry of alkynes, and in the work of Hubel [2] on stoichiometric alkyne-metal carbonyl reactions, which can yield a bewildering array of organic and organometallic compounds. Since these seminal studies the organotransition metal chemistry of alkynes has blossomed and proved to be a fertile testing ground for the development of structure/bonding correlations and for the discovery of interesting and useful metal-promoted reactions. The resulting expansiveness of this field prevents a comprehensive review. It is our intention, therefore, to highlight some of the most important and recent developments in the chemistry of metal-alkyne complexes. We will largely limit our coverage to that chemistry which clearly involves the intervention of metal-x-bonded alkyne complexes. We thus exclude the chemistry of metal-acetylide derivatives and mention only briefly the burgeoning number of metalcatalyzed reactions for which alkyne complexes are only presumed intermediates. Prior reviews of metal-alkyne chemistry may be consulted for more complete coverage of the older literature [3].
4.2 Bonding and Structure 4.2.1 Alkyne Complexes Alkynes can coordinate to transition metals in a variety of ways (Scheme 4-1, 1-5) depending on the number of metals present, their electronic nature, and that of the alkyne. Coordination typically results in substantial changes in the structure of the alkyne unit (1) including: (1) a distortion from linearity (a < 180"); and (2) an increase in the coordinated C - C bond length (1). These two effects can be understood in terms of the now-classical Dewar-Chatt-Duncan-
?
1
Scheme 4-1
2
3
4
5
100
4 The Chemistry of Metal-Alkyne Complexes
son model [4] which features two primary metal-alkyne bonding components (Scheme 4-2): (1) ligand-to-metal electron donation via interaction of the filled x-bonding alkyne orbital with an appropriate acceptor orbital on the metal (6) and (2) electron back-donation from a filled metal d-orbital into the x* of the alkyne (causing weakening of C = C; 7). The relative importance of these synergistic contributions depends on the energy match and overlap of the contributing orbitals as determined by the metal, its oxidation state and auxiliary ligands, and the electronic character (releasing or withdrawing) of the alkyne substituents. Contributor 6 tends to predominate with early or higher-oxidation-state metals and electron-rich alkynes whereas contribut0.r 7 becomes increasingly important with later, lower-oxidation-state metals and electron-poor alkynes. The net effect of these two components reduces the C - C bond order and confers multiple bond character on the metal-alkyne linkage, leading to its occasional representation as a metallacyclopropene 10 (Scheme 4-3). For electron-deficient metals the metal-alkyne linkage can be further strengthened by secondary x (a1kyne)-to-(M)interactions 8, 9 which result in the alkyne serving formally as a four-electron donor, represented in valence bond terms by the biscarbene contributor 11.
Scheme 4-2
(-M
c--)
M<
M<
10
11
R Scheme 4-3
The magnitude of coordinative distortions can be taken as an indication of the importance of the above-mentioned bonding components 151. We can illustrate three extreme situations with complexes 12-14 (Scheme 4-4). Thus, the alkyne unit in 12, the first lanthanide-alkyne x-complex [6],shows very little distortion from the free ligand (a = 177.4", 1 = 1.15 A vs. 1.21 A in free butyne), consistent with nearly exclusive contribution from M and minimal backbonding (the metal is do); the rare main group-alkyne n(a1kyne) +
12 Scheme 4-4
13
14
4.2 Bonding and Structure
101
complex, Cp?Ca(Me,SiC = CC = CSiMe,) [7], shows similar features. On the other hand the formally Pt(0) complex 13 exhibits a strongly bent R - C - C angle and a C - C bond length similar to a C = C bond [8].A continuum of intermediate structures lies between these two extremes with the IR v(C - C) of alkyne complexes, diminished ca. 100-700 cm-', providing a convenient, semiquantitative indicator of the reduced bond order resulting from the two contributors. The geometrical changes attendant with coordination may also be understood in terms of rehybridization of the metal-bound C-atoms with increased p-character. The alkyne ligand in 14 is considered to be a four-electron donor based on the metal's tendency to accommodate a total of 18 electrons and supporting X-ray diffraction (e. g., short W - C distance) [9] and spectroscopic data, e.g., I3C NMR chemical shifts of ca. 200 ppm [lo]. Binuclear complexes are found most commonly with the alkyne perpendicular to the M - M axis (2) and serving as a four-electron donor (two to each metal). Less common are examples with the alkyne parallel to M - M (3), essentially a o-bonded dimetallacyclobutene. The differing frontier molecular orbital requirements and the interconversion barriers for these two bonding modes have been analyzed by Hoffmann [ll]. Rare examples with a skew geometry (twist from perpendicular = 35", 31") [12, 131 have been analyzed using the X a method by Cotton, who found this situation to be favored when there is a small HOMO/LUMO gap, giving rise to a second-order Jahn-Teller distortion. Other calculational methods and X-ray diffraction-derived electron density mapping of Co2(CO),(BuC2Bu) and Cp2Ni2(HC2H)have been used to conclude that the M - M bent axis shows electron deficiency and that A backdonation amounts to ca. 0.88 e-, agreeing with calculations from the IR C-C stretching frequency [14]. Among trimetallic alkyne derivatives two orientations of the alkyne relative to the cluster are found (4 and 5), the preferred arrangement being accounted for on the basis of the cluster electron count [IS].
4.2.2 Propargylium-Metal Complexes Bimetallic p-alkyne derivatives, especially the - C O ~ ( C Oclass, ) ~ have probably been the most extensively investigated class of alkyne complexes [16]. Of these the compounds 15, 16 derived from a-ionization of dinuclear propargyl complexes have received particular attention because of their novel structural and useful reactivity features [17]. Propargylic cations were first shown to be stabilized (pKR+ca. -7, comparable with Ph3C+) by coordination to the - C O ~ ( C O ) ~ group, i. e., 15 [18,191. These compounds (L=CO, PR,) can be generated in situ or isolated as red, moisture-sensitivesolids upon Lewis acid treatment/protonation of the corresponding pro)~ [20-231 pargyl alcohols, ethers, acetates, acetals, aldehydes, or ene-yne - C O ~ ( C Ocomplexes which, in turn, are derived from alkyne complexation by C O ~ ( C O(Scheme )~ 4-5) [24]. The corresponding (propargylium)M2(C0),Cp$ (M = Mo, W; 16, 17) derivatives have been obtained similarly [25]. Heterobimetallic (propargyli~m)Co(CO)~MCp(CO)$ (M =Mo, W) complexes of propargyl alcohols have been obtained with moderate diastereoselectivity via metal vertex substitution of the Co2-complexes by (C,H,R)M(CO)F (M=Mo, W) [26]; these serve as precursors to the corresponding heteronuclear cationic complexes 18 [27]. The powerful cation) ~ - CpzM2(CO), units is further demonstrated by the stabilizing ability of the - C O ~ ( C Oand generation of novel dicutionic complexes, [(H2CC=CCH2)M2L,][BF4]2, (M=Mo, Co), e. g., 19, 20 (Scheme 4-6), by protonation of butynediol precursors [28-301.
102
4 The Chemistry of Metal-Alkyne Complexes
Lewis or Bronsted acid ____)
15 M'=Co, L=L'=CO 16 M'=Mo, L=L'=Cp; 17 M'=W, L=L'=Cp; 18 M'=Co, L=CO, M'=Mo, L'=Cp
15
- 18
Scheme 4-5
19 Scheme 4-6
Several -Cp2M2(CO), derivatives 16, 17 (M=Mo, W) have proven amenable to X-ray studies [31-351 which demonstrate - q ',q 3-coordination of the propargylic ligand as shown in Scheme 4-5. The metal - C(3) bond distances indicate a stronger bond for primary vs. tertiary cations, with the secondary species intermediate in value. Variable-temperature + 401 NMR studies of the (HC=CCRlR2)-Cp2M~2(CO),f [36-381 and - C O ~ ( C O ) ~ L[39, derivatives (for which no X-ray structures exist) are also consistent with unsymmetrical q2,q3-bonding present in solution with a degree of fluxionality which depends on the particular metals involved and the substituents at the propargylic carbon. %o dynamic processes involving the propargylic ligand have been proposed : a lower-energy antarafacial migration of the -CRIRz unit from one metal to the other and a higher-energy (sydanti) rotation about the C(2) - C(3) bond (Scheme 4-7). For both processes AG * increases from tertiary to primary cations (ca. 11 to >15 kcal/mol for syn/untQ. NMR studies of the -Coz(CO)5(PPh,) derivatives [21] have revealed a significantly higher syn/anti interconversion barrier relative to the -Co,(CO), complexes (17-20 vs. 13 kcal/mol for secondary cations). Interestingly, variable-temperature NMR studies of the heterobimetallic Mo - Co complexes 18 showed no fluxional behavior and it was suggested that the propargylic carbon is localized on the Mo center [27]. These results were supplemented by EHMO calculations indicating that the positive charge is better accommodated at the CpM(C0)' vertex and that the transition state for antarafacial migration is strongly disfavored. Further experimental evidence for charge delocalization onto the - C O ~ ( C Ogroup ) ~ in 15 includes IR data which show an increase in v(C0) (40-60 cm-') relative to the neutral complexes [19] and dramatically shielded 13C NMR resonances relative to free propargyl carbocations [41, 421. Recently a number of mononuclear q 3-propargyl/allenyl complexes have been reported, e. g., 21-24 (Scheme 4-8). These have been produced in a variety of ways including protona-
6 LnM-
-M*Ln
4.2 Bonding and Structure
-
antarafacial migration
103
R.p$
h
L n -~ -M'Ln
0
0 R suprafacial migration
supra migration facial1 rotation (syn/ant i isomerization)
82
w
L
antarafacial
F Z p
l
hM+\*Ln
migration R
R
Scheme 4-7
21
22
23
24
Scheme 4-8
tion of q2-propargyl alcohols/ethers [43], hydride abstraction from q 2-alkynes [44],halide abstraction from 0-propargyl or d e n y 1 complexes [45], and reactions of metal halide derivatives with propargyl Grignards [46]. The bonding in these compounds may be considered in terms of two primary resonance contributors, propargyl structure 25 and allenyl structure 26 (Scheme 4-9), leaving one of the C = C x-bonds not interacting with the metal. X-ray crystallographic studies reveal that the C, skeleton is bent (C - C - C angle ca. l5Oo), the metal center is almost coplanar with the C3 fragment, and the two C - C bond lengths range from considerably different (favoring contributor 25) to nearly equal (favoring contributor 26) 1461.
25 Scheme 4-9
26
104
4 The Chemistry of Metal-Alkyne Complexes
4.3 Complexes of Novel Alkynes The distinctly bent geometry of coordinated alkynes has allowed the preparation of complexes of cycloalkynes which are highly reactive or not isolable as the free ligand. These have been accessed by trapping the strained cycloalkyne with a reactive complexing agent, by complexing a cycloalkyne precursor followed by C = C generation, or by complexation of an acyclic alkyne and subsequent cyclization. The interested reader is referred to reviews by Bennett [47] for a more complete, earlier review of this topic; some highlights and recent developments follow. Adducts of the marginally stable cyclooctyne and cyclooctenyne with the Pt[PPh,], [48] and - CO,(CO)~[49, 501 fragments have been prepared by direct complexation. Reaction of 1,2-dibromocycloalkeneswith sodium amalgam in the presence of Pt(PPh,), has produced complexes 27-29 of the unisolable cycloheptyne, cyclohexyne [51], and even cyclopentyne [47] (Scheme 4-10); the free alkynes apparently are not intermediates in these reactions.
27
- 29
Scheme 4-10
Alternative indirect methods have been used in the synthesis of (3,3-dimethylcyclopen[53]. The smallest cycloalkyne tyne)Cp2Zr(PMe3) [52] and (cyclohe~yne)[CpMo(CO)~]~ stabilized by complexation is cyclobutyne, incorporated by Adams into tri- and tetrametallic ruthenium and osmium clusters as a b3-ligand 30 using 1-Br and -SPh substituted cyclobutene precursors (Scheme 4-11) [54]. Liberation of these strained alkynes from their metallic bondage has not been reported but some have been shown to be reactive towards insertion of unsaturated substrates, as illustrated in Scheme 4-12 with the Zr-cyclohexyne derivative 31 [ 5 5 ] . Several transition metal complexes of the highly reactive benzyne ligand have been reported in recent years, some of which are sufficiently stable to permit complete (including X-ray) characterization, e.g., Ta [56], Ni [57], Zr [58] and Ru derivatives [59] 32-35 (Scheme 4-13). In other cases transient benzyne complexes have been implicated as key intermediates in the thermolysis of a-aryl-metal complexes, e. g., Cp2ZrPh2 -, Cp2Zr(q2-benzyne). The latter, like the corresponding cyclohexyne complex (Scheme 4-12), undergoes insertion with a variety of unsaturated substrates to form metallacycles, some of which have been converted to organic
30 Scheme 4-11
4.3 Complexes of Novel Alkynes
105
R = H, I-CH-CHMe Scheme 4-12
32
- 35 36
Scheme 4-13
products by protonolysis [60]. Additionally, various polynuclear complexes in which benzyne serves as a multiply bridging ligand of the type 4, e.g., in O S ~ H ~ ( C O ) ~ ( ~ [61], ~ - Chave ~H~) been characterized. Complexation to Cp2Zr(PMe3)- [62] and (CyzPCHzCHzPCy2)Ni- 1631 fragments has been employed to stabilize even more exotic tetradehydrobenzenoid molecules, e. g., 36. Finally, we note the conspicuous absence of methods for the release of free benzyne from these metal complexes. The bending associated with alkyne complexation has been increasingly employed to facilitate the cyclization of acyclic alkyne complexes. Schreiber's group studied intramolecular nucleophilic trapping of propargylium-cobalt complexes by an ally1 silane (see below [22], using an endocyclic variant to produce medium-sized cycloalkyne (7,8-membered) derivatives (Scheme 4-14). Highlighted later are intramolecular nucleophilic trapping reactions of the cobalt-propargylium complexes by enol derivatives, elegantly used to generate the strained bicyclic ene-diyne cores characteristic of the remarkable ene-diyne antibiotics [ a ] . Melikyan and Nicholas recently developed the reduction (by Zn or NaBPK; BPK = benzophenone ketyl) of dicationic propargylium complexes 37-39 as a general regio- and stereoselective entry to 1,s-cycloalkadiynes, including the previously poorly accessible C, [65], C9
106
4 The Chernisrry of Metal-Alkyne Complexes
Scheme 4-14
and Clo [66] derivatives 40-42 (Scheme 4-15), presumably via the corresponding diradical species. The heretofore-unknown novel cyclooctene-diynederivative 43 has also been prepared by this method [65] and is of interest in terms of the ability/facility of the free ligand to undergo Bergman cyclization [67]. Initial studies indicate that the free diyne derivatives can be liberated by “red-ox” demetalation with NaBPK-molecular oxygen [65, 681.
2 e-
M 1.3
M ICO&O),;
11.3
R = H, Ak. Ar
1.3
-
-
37 39
40 42
43
Scheme 4-15
The interesting antiaromatic triynes tribenzocyclyne (TB) and trithienocyclotriyne (TTC) exhibit unusual coordination chemistry. The small cavity of these compounds allows the incorporation of first-row metals including Co (691, Cu [70], and Ni[71] in novel arrangements as illustrated by structures 44-46 in Scheme 4-16. When partially reduced with alkali metals, Ni(TBC) shows a lo4-fold increase in conductivity.
do(CO)3 44
45
46
Scheme 4-16
Diederich and co-workers have synthesized a triscluster complex 47 of the unusual CI8 molecule via oxidative cyclotrimerization of the hexatriyne complex 48 (Scheme 4-17) [72]. The cyclocoupling efficiency benefits from the bent geometry of the coordinated triyne (a = 138”). Other noteworthy features include: (1) the regioselective complexation of the silylated triyne; (2) efficient desilylation of the complexed triyne.
4.4 Reactions of Metal-Alkyne Complexes
107
Si(iPr),
I
CU(OAC)~
I
Si(iPr),
A
___)
pyridine
48
L = Ph2PCH2PPh2
Scheme 4-17
4.4 Reactions of Metal- Alkyne Complexes 4.4.1 Reactions at the C - C Triple Bond 4.4.1.1 Nucleophilic Addition Although typical alkynes are rather unreactive towards nucleophiles, coordination to electrophilic metal fragments can activate them towards nucleophilic attack, typically resulting in the formation of vinyl-metal derivatives (Scheme 4-18, X = Nu). The formation of tmns-adducts 49 (M relative to Nu) is generally taken to indicate direct attack of the nucleophile at the alkyne and is typically observed with coordinatively saturated, nonlabile complexes. The formation of cis adducts 50 probably proceeds via initial attack at the metal (51, X = Nu) or at an auxiliary ligand, e. g., Cp, followed by migration of Nu to the alkyne. The former type of process is illustrated by the addition of a variety of soft carbon nucleophiles to CpFe(CO)L(alkyne)+ complexes producing trans-adducts 52 which undergo oxidative carbonylation (Scheme 4-19) [73]. The latter process operates in the addition of BH; to CpM~[P(OMe)~]~(alkyne)+ in the presence of P(OMe), which initially gives the cis-vinyl derivative (74). Highly electrophilic L,PtMe(alkyne)+ cations display carbocationic character in their reaction with alcohols, giving both trans-vinyl adducts and alkoxycarbene complexes [75]. Bianchini has described a (tripod)Rh-catalyzed synthesis of enol esters via carboxylic
51
Scheme 4-18
-
cis 50
108
4 The Chemistry of Metal-Alkyne Complexes
L
A J
Nu = Alk,CuLi, Ph,CuLi, etc.
52
Scheme 4-19
acid addition to alkynes (Scheme 4-20) which proceeds with high Markovnikov selectivity via the coordinated alkyne [76]. Another noteworthy metal-promoted addition to alkynes is Pt(I1)-catalyzed hydration which proceeds with moderate regioselectivity in aqueous THF V71.
Scheme 4-20
In contrast, some cationic and neutral 4e- alkyne complexes 53 add nucleophiles to give qz-vinyl derivatives, 54 (Scheme 4-21) [78].This outcome allows the metal to retain its preferred 18-electron count by virtue of the four electrons donated from the q2-vinyl ligand. R
+
M-(
NU M<
R1 Nu
R1 53
54
Cp[(MeO)3P]2Mo+, CpMoCl(a1kyne). CpWCl(alkyne) Nu = H-, PRg R2CuLi Scheme 4-21
4.4.1.2 Electrophilic Addition
Reactions of alkyne complexes with electrophilic agents lead to the isolation of metal-vinyl complexes or ultimately to substituted olefins (Scheme 4-18, X = E). Since for most alkyne complexes there appears to be a net electron withdrawal by the organometallic fragment, they are typically less reactive than free alkynes towards electrophiles. A striking exception to this generalization is provided by the powerful nucleophilic character of NbC13(THF)2(alkyne) complexes which react with 1,2-aryldialdehydes regioselectively to form 2,3-disubstituted-1napthols (Scheme 4-22) [79].
4.4 Reactions of Metal-Alkyne Complexes
R = Ak, Ar, etc.
109
0
Scheme 4-22
In the addition of protic acids to alkyne complexes the stereochemistry varies from system to system, apparently reflecting a number of available mechanistic pathways. For several mononuclear complexes in which 0-vinyl complexes have been isolated, e. g., (Ph,P),R(RC =CR) [80], the metal and proton are added in a cis fashion suggesting that initial protonation occurs at the metal with subsequent olefin insertion into the M - H bond. Further protonation of the metal-vinyl complexes may lead to the formation of cis- and/or trans-olefins as found in the protonation of (Ph,P),Pt(RC = CR), where 30- 100% of the trans-isomer was found depending on the nature of R and the acid employed [81, 821. These observations may be the result of protonation/isomerization of the vinyl complex or isomerization of the liberated olefin. A number of classes of alkyne complexes afford alkenes as protonation products [83, 841, usually (but not always) [85] with the cis-isomer formed selectively. This is probably the result of initial protonation at the metal, cis-M - H addition (alkyne insertion), and stereoretentive cleavage of the M-C o-bond via initial metal protonation. In contrast, the reaction of (RC iCH)Co,(CO), complexes with mineral acids in refluxing methanol gives p,-alkylidyne derivatives, (RCH,C =))CO,(CO)C, [86].
4.4.1.3
M - H Addition/Hydrogenation
In addition to the classical heterogeneous Lindlar catalysts for cis-semihydrogenation of alkynes, homogeneous semihydrogenation has also been reported. Bianchini et al. have described Fe and Ru catalysts of the type [(tetraphos)MH(q2-H2)]BPh4which catalyze the cis-semihydrogenation of terminal alkynes to alkenes [87]. In contrast, Muetterties described a short-lived binuclear catalyst system for the trans-hydrogenation of alkynes which proceeds via the bridging alkyne complex 55 and an intermediate p-vinyl species 56 (Scheme 4-23) [88]. Mononuclear RhH,(02COH)[P(iPr)3]3 also has been shown to catalyze the trans-hydrogenation of PhC2Ph and R02CC2C02Rvia an intermediate alkyne complex [89].
55
Scheme 4-23
56
110
4 The Chemistry of Metal-Alkyne Complexes
Stoichiometric studies of M - H additions to alkynes also show mixed stereochemical results. The more common cis-addition is typified by the CO-promoted transformation of Cp2Nb(H)(RC= CR) to cis-Cp,Nb(CO)(q '-CR= CHR) [90]. In mononuclear systems where tmns-additions have been found, radical-type mechanisms have been implicated [91] or ciskrans isomerization of the intermediate vinyl species 1921 has been found. Although the intermediacy of alkyne complexes has not been established, Schwartz's hydrozirconation of alkynes [93] by Cp2ZrHC1 represents a general entry to vinyl-metal species which can be transformed stereoselectively to alkenes, vinyl halides, and/or carboxylic acids.
4.4.1.4 M
- C Addition
The addition of transition metal alkyls to alkynes is less common and often less facile than the corresponding metal-hydride additions, although this reaction is probably key to many metal-catalyzed alkyne polymerizations (see below). In one case where an alkyne adduct has been established, i. e., the reaction of L,PtClMe with electrophilic alkynes, cis-M - C addition is observed (Scheme 4-24) [94].
Scheme 4-24
4.4.1.5 Coupling Reactions with Unsaturated Substrates
Alkynes enter into a remarkable variety of metal-promoted coupling reactions with olefins, alkynes, and other unsaturated species leading to a diversity of cyclization, oligomerization, and polymerization products of synthetic value. In many instances alkyne complexes are presumed intermediates in these reactions but often this has not been proven. The reader is referred to other reviews 195-971 for more complete coverage of this topic. We briefly summarize here the most useful of these processes, highlighting those systems in which metal-alkyne complexes have been demonstrated as intermediates. Scheme 4-25 summarizes the most common metal-promoted intermolecular cyclocoupling reactions of various species with alkynes. The most prominent organic products include arenes, cyclooctatetraenes, cyclohexadienes (with olefins), pyridines (with nitriles), cyclopentenones (with olefin + CO; the Pauson-Khand reaction [98]), pyrones (with CO,), and fivemembered heterocycles (with X = S, Se); common organometallic products include cyclobutadiene complexes, cyclopentadienone complexes, and metallacyclopentadienes. With alkynes alone, typical products are those of cyclotrimerization, i. e., arenes, although cyclotetramerization to cyclooctatetraenes has been directed selectively by Ni catalysts [l], providing an early commercial process for cyclooctatetraene. Numerous catalysts are known for the former process and, depending on the catalyst and substrate, moderate to high regioselectivity (1,2,4- or 1,3,5-) can be obtained with unsymmetrical alkynes [95, 961. Selective mixed
4.4 Reactions of Metal-Alkyne Complexes
111
Scheme 4-25
alkyne coupling has also been achieved through the use of the sterically hindered bis(trimethylsily1)acetylene [99]. Some mechanistic insight and control over these intermolecular processes have been gained through the studies by Yamazaki of the CpC~(PPh~)~(alkyne) system [loo], by Bianchini with (tetraphos)M(alkyne)+ (M= Co, Rh) [loll, and by Wigley with (ArO),Ta(alkyne) [102]. In general it appears that steric effects dominate over electronic in controlling the regioselectivity of metallacycle formation with unsymmetrical or different alkynes; for example, in the CpCoLpromoted reactions (Scheme 4-26) the cobaltacyclopentadiene 57 which places the bulkier substituents a- to the metal is favored. Formation of arenes may proceed via metallacycloheptatriene or Diels-Alder [4 + 21 processes but these alternatives have rarely been differentiated (Scheme 4-27). B o ex ceptions are provided by the previously mentioned Rh 11031 and Ta [lo21 systems in which q4-benzene (for RH) and metallanorbornadiene (for Ta, e.g., 58 in Scheme 4-28) intermediates, have been isolated and shown to be arene precursors and catalysts. In the latter system, metallacyclopentadiene (59) and alkyne (60) intermediates also can be isolated depending upon the reacting alkyne. In contrast, cyclotrimerization with Pd(I1) catalysts proceeds by an alternative route involving sequential insertion reactions via a cyclopentadienylmethyl complex [104]. 1
c 9 $
+
m3p
Fb4-
R2 R1, R2 = Ph. C@Me, CN; R3, R4 = H, Me, Ph, C02Me, Fc Scheme 4-26
+
cp;y R3
PhaP
b 57
112
4 The Chemistry of Metal-Alkyne Complexes
R
0+ 11
k
J?\
F. R
R
R
Scheme 4-21
SiMe,
Ta(DIPP)2C13 + 2NaMg
Me3SiC+SiMe3 _____+
(DIPP),CITa \
SiMe,
60
Ph
58
59
Scheme 4-28
Probably the synthetically most useful versions of cyclotrimerization have been developed in the intramolecular sense through the pioneering work of Vollhardt. Since this topic has been thoroughly reviewed [95], we point out here only that the most powerful cyclocouplings have been two-component (e. g., diyne + mono-yne) or single component (e. g., ene-diyne) ones. The former strategy is elegantly illustrated in the CpCo(CO)2-catalyzed synthesis of estrone (61) depicted in Scheme 4-29 [105].
II
61
Scheme 4-29
Among the most useful hetero-cyclocouplings (Scheme 4-25) are the CpCoL,-promoted reactions with nitriles giving pyridines [lo61 and the Ni(0)-catalyzed reactions with CO, to give pyrones [107]. In both cases considerable regioselectivity can be achieved and intermediate metallacycles have been implicated. Another interesting diversion in the cyclo-
4.4 Reactions of Metal-Alkyne Complexes
113
coupling of alkynes is provided by the formation of novel metallapyran derivative 62 by PPh3-induced deoxygenation of Cp*Re03 in the presence of alkynes via an isolable Cp*Re(O)(alkyne) intermediate; oxidation of 62 with O2or I, affords substituted furan 63 (Scheme 4-30) [log]. A synthetically attractive and novel route to substituted furans is the regioselective Ta-promoted three-component cyclocoupling of alkynes, aldehydes, and isonitriles (Scheme 4-31), presumably via low-valent Ta-alkyne complexes [log].
R = R = Me, Et R = ph, R' = H
62
63
Scheme 4-30
R', R* = n-C5Hll, R3 = n-Pi, n-Oct Scheme 4-31
Finally, a few other recent developments in the area of intmmolecular alkyne cyclocoupling should be mentioned because of their considerable synthetic value, although the involvement of alkyne complexes is uncertain. These include Livinghouse's Rh-catalyzed Diels- Alder reactions [110], Negishi's stoichiometric Zr-promoted bicyclization of ene-ynes [97] (Scheme 4-32), and Rost's Pd-catalyzed cycloisomerization of ene-ynes [lll] (Scheme 4-33).
114
4 The Chemistry of Metal-Alkyne Complexes
4.4.1.6 Alkyne Scission/Metathesis/Polymerization Perhaps the most remarkable illustration of the ability of metals to activate alkynes comes from reactions in which complete scission of the carbon-carbon triple bond occurs. On the stoichiometric level these include examples in which carbyne complexes are produced from alkyne complexes as in the melt-thermolysis of CpCo(PPh,)(RC = CR) [I121 or from reactions of alkynes with unsaturated metal species (Scheme 4-34) [113]. The remarkable alkyne metathesis reaction (Scheme 4-35), which involves overall cleavage and regeneration of two oand four x-bonds, is conceptually related. A variety of functionalized alkynes can be tolerated as metathesis substrates [114] and especially effective catalysts for these reactions are Mo(V1)and W(V1)-carbyne complexes. Metallacyclobutadienes 64, formed by the reaction of the alkyne with a metal-carbyne complex, appear to be central intermediates in these reactions and the equilibrium between metallacycle and alkyne/metal-carbyne is observable in some cases [115]. w2(o!-BU)6 + RC-R
+2 (!-BUO),WMR
R = Me, Et, Pr Scheme 4-34
Scheme 4-35
In addition to the cyclooligomerization processes summarized previously, linear oligomerizations/polymerizations are also prominent metal-catalyzed reactions of alkynes. These reactions are catalyzed by a wide variety of metal complexes, including the commercial ZieglerNatta olefin polymerization catalysts [116], and provide entry to polyacetylene and polyacetylene derivatives which have attracted great attention as organic conductors. Mechanisms which have been proposed for these reactions have as alternative propagating steps: (a) alkyne insertion into vinyl-metal complexes; or (b) alkyne metathesis via metal-carbene complexes. The potential intermediacy of metal- alkyne complexes in the latter pathway is supported by Geoffroy’s generation of the metastable carbene-alkyne complex 66 which leads to alkyne polymerization upon warming, presumably via the metallacycle 65 (Scheme 4-36) [117]. Also
Ph
65
4.4 Reactions of Metal-Alkyne Complexes
115
relevant is formation of the isolable cobaltacyclobutene 67 from the reaction of CpCo(alkyne)(PPh3) with ethyl diazoacetate, leading ultimately to the formation of diene complexes 68 (Scheme 4-37) [118].
NzCH(C02)Et
TMS
Ph3P S02Ph
-
SOpPh
67
68
Scheme 4-37
4.4.1.7 Cluster Substitution/Expansion
Although polynuclear alkyne complexes are often prepared by reaction of the alkyne with a suitable metal cluster fragment, heteropolynuclear complexes 69 (Scheme 4-38) have been obtained also by isolobal metal fragment substitution, as noted previously [26]. Highernuclearity alkyne complexes also can be produced by the addition of various metal carbonyl fragments to a lower-nuclearity alkyne complex [119]. A novel entry to heterobi- (and tri-)metallic neutml p-propargyl complexes (e. g., Fe/Mo) via protonation of trinuclear p-q 2,q 2-o-propargyl derivatives 70 was recently described by Wojcicki and coworkers [120, 1211.
Scheme 4-38
4.4.1.8 Demetalation
To exploit the utility of metal-mediated organic reactions, it may be necessary to remove the metal fragment from stable alkyne complexes after a desired transformation has been effected. This has been variously accomplished for alkyne complexes, most commonly by oxidative decomplexation; less commonly Gsubstitution or reductive processes have been employed. )~ are typically demetalated by mild oxidizing The widely investigated - C O ~ ( C Oderivatives agents, including Ce(1V) [122], Fe(II1) [123], I, [124], and arnine oxides [125]. Oxidative demetalation (NMO or 13 of the cyclic ene-diyne complexes, e.g., 71, has been useful for producing the free ene-diynes 72 or the Bergman-cyclized aromatics 73 (Scheme 4-39) [126]. Demetalation under reducing conditions is possible using NaBPK [65]. An early claim of alkyne displacement from (alkyne)Co2(CO), by more electrophilic alkynes [127] has proven irreproducible [128].
*om
4 The Chemistry of Metal-Alkyne Complexes
116
B"B"*]
OH 71
73
72
Scheme 4-39
4.4.1.9 Nucleophilic Addition to Mononuclear q3-Propargylium-M Complexes
The recently reported mononuclear rl 3-pr~pargyliumcomplexes 74-76 have been found generally to add nucleophiles specifically at C2 with formation of q b l l y l or metallacyclobutene complexes (Scheme 4-40] 143-461. With aprotic nucleophiles, e.g., PR, and LiC = CCMe3, reaction with 74 produces the rhenacycles 77 (M = CP*R~(CO)~) exclusively [44]. Protic nucleophiles such as HzO, ROH, or R2NH react readily with 75 and 76 to give the corresponding q3-allyl derivatives 78 [43. 451, a net addition with the heteroatom attaching to C2 and the proton to c1. Some strongly coordinating nucleophiles add to the platinum complexes at the metal Center to give o-propargyl or -allenyl derivatives [45 a].
78
-
74 76
77
Scheme 4-40
-
4.4.2 Reactions at the Complexed Acetylenic C X Bond 4.4.2.1 Alkyne-Vinylidene Isomerization
Some reactions of terminal alkynes with monometallic species do not result in isolation of the n-complexed alkyne 79 but rather afford a vinylidene complex 80 wherein the acetylenic hydrogen has been transferred to the P-carbon (Scheme 4-41)[129].This is particularly the case if the organometafic Species is 16-electron or labile. In some cases isolable n-complexes
Scheme 4-41
4.4 Reactions of Metal-Alkyne Complexes
117
of terminal alkynes thermally rearrange to the vinylidene isomers [130- 1321. Hoffmann used extended Huckel calculations to analyze the energetics and potential mechanisms of the isomerization for mono-, di-, and trinuclear complexes [133] and found that a "concerted" metal-slip/l,2-H migration pathway is favored for monometallic systems but that stepwise processes are favored in the higher-nuclearity cases. Despite the above theoretical prediction, a number of mononuclear alkyne complexes have been found to rearrange via observable/iso83 conversion lable M(H)(alkynyl) intermediates 81 [131, 1321 as illustrated by the 82 (Scheme 4-42) [134]. Werner's group has shown that alkynylsilanes also undergo 1,Zsilyl migration in the same system via an intermediate x-complex [135]. Alternatively, the rearrangement may be catalyzed by base [136] or induced by sequential deprotonation/protonation [137]. The relative stability of the alkyne and vinylidene complexes is dependent on the electron density and the d-electron count of the metal, as illustrated by the behavior of the d4-Mo complexes 84 in which the alkyne is a four-electron donor; addition of CO causes the 84 4 85 conversion whereas tautomer 84 is favored with the phosphite ligand (Scheme 4-43) [137]. -+
H
OHL = P(iPr),
83
82
Scheme 4-42
co - 78O-r.t. P t f
MO--
L
4 L
%
84
L =PMe2Ph
d 40
O0 L = P(OMe),
85
Scheme 4-43
4.4.2.2 Reactions of Complexed Terminal Alkynes with Base
Besides promoting the above alkynehinylidene rearrangement, reactions of n-bonded terminal alkynes with base have received limited attention. In some cases, e. g., with the cationic complex 86, such reactions produce a-alkynyl derivatives 87, which can be further converted into vinylidene complex 88 (Scheme 4-44) [9]. In contrast, few successful attempts to substitute the alkynyl C - X unit (X = H, SiR,, SnR,) of binuclear alkyne complexes have been reported and it appears that this bond is deactivated relative to that in the free alkyne, consistent with its decreased s-character in the complex. Treatment of (Me3SiC= CH)Co,(CO), with LiN(SiMe,), has been proposed to form (MesSiC=CLi)Co2(CO), which reacts with various electrophiles primarily to form complexed 1,3-diynes [138] - products, however, which could be derived from the corresponding radical, (Me3SiC= C .)Co,(CO),. The efficient protodesilylation of com-
118
4 The Chemistry of Metal-Alkyne Complexes
plexed silylalkynes of the type (Me,SiC = CZ)Co,(CO), (Z = R, OR) has been achieved using methanol/K,CO, 11391 and THF/H20/Bu,NF [72]. This transformation has been used in the preparation of complexes of chiral, nonracemic alkoxyacetylenes (1391for potential use in asymmetric Pauson-Khand reactions and in the synthesis of the novel all-carbon molecules, CISand c 2 4 [721.
86
88
87
Scheme 4-44
4.4.3 Reactions at the Propargylic (a) Carbon 4.4.3.1 AlkyneiAllene Isomerization
In contrast to the behavior of free alkynes, I,3-isomerization of alkyne complexes to produce complexed allenes [140, 1411 is rather rare. A synthetically useful example [140] is provided by the CpMn(CO), complexes of electrophilic alkynes 89 which undergo alumina-promoted isomerization to the corresponding allene derivatives 90;these, in turn, can be oxidatively demetalated to the free allenes 91 (Scheme 4-45). H,
EWG C’
basic A120,
Cp(CO),Mn-Il
CecI
s
s FC
___)
R,/ 89 R’, R2 = H, Alk ; E = C02Et,COMe, CHO
,EWG
90
h 2
91
Scheme 4-45
4.4.3.2 Reactions of Dinuclear Propargylium Complexes with Nucleophiles 4.4.3.2.1 General Reaction Features
The (propargyliurn)C~~(CO)~L (L=CO, PR,, P(OR),) complexes 92 serve as electrophilic propargyl synthons in their reactions with a wide variety of hetero- and carbon-centered nucleophiles (Scheme 4-46). Attack occurs exclusively at the propargylic carbon (C3), thus avoiding the allenic by-products which plague reactions of classical propargyl electrophiles [17]. Acetylenic aldehydes and acekals have served as useful precursors for oxygen-substituted cations 92 (R2 = OH, OR). Nucleophilic reactions of the less reactive C~,MO,(CO)~
4.4 Reactions of Metal-Alkyne Complexes
119
5
Scheme 4-46
derivatives 16 (Scheme 4-5)also lead to regioselectively substituted alkyne complexes [32,331 (with one exception [142]). The acid-promoted opening of a$-epoxyacetylene complexes proceeds with a variety of carbon and heteronucleophiles, affording exclusively the products derived from cleavage of the C - 0 bond a- to the electron-releasing (q2-alkynyl)Co2(CO), unit [143]. A recently reported intmmolecular version of this reaction provides a regioselective and stereospecific route to 2-ethynyl-3-hydroxytetrahydrofurans 93 [I441 (Scheme 4-47).a-Vinyl-substituted propargylium complexes, produced from vinylethynyl carbinol derivatives 94, react with carbon nucleophiles exclusively at the terminus remote from the organometallic substituent and with complete Estereoselectivity (Scheme 4-48)11451. Cobalt-complexed alkynyl cyclopropyl carbinols react with HBr/ZnBr, to give homoallylic bromides with much greater E-stereoselectivity than the free ligands [146].Interestingly, the cations 92 (R, = cyclopropyl) undergo attack by carbon nucleophiles exclusively at the a-carbon without ring opening [147].
93 Scheme 4-47
120
4 The Chemistry of Metal-Alkyne Complexes
,OH
i) BF3 ' Et20 ii) Nu
94 Scheme 4-48
One approach to stereocontrol in these reactions has involved the generation of complexes possessing a chiral cluster core, e. g., 95,96. Thus, complexes of chiral propargyl alcohols react with triphenylphosphine highly stereoselectively, giving configurationally stable, separable diastereomers of 95 (Scheme 4-49, L=PPh3; * indicates stereocenter) [148] but with little or no diastereoselectivity in the reactions of phosphines with complexed chiral propargyl ethers [149]. The chiral cluster cations 96, which are more configurationally rigid than the parent hexacarbonyl complexes (vis a vis Scheme 4-7) [21], are diastereoselectively quenched with oxygen-centered nucleophiles. Their diminished electrophilicity compared with the Co2(C0), complexes, however, is reflected in their failure to react with mild, synthetically useful carbon nucleophiles (e. g., silylenol ethers, ally1 silanes). On the other hand, incorporation of the relatively bulky, weakly donating P[OCH(CF,),], ligand in the cationic complex 96 allows facile, diastereoselectivecoupling with mild carbon nucleophiles and, when derived from enantioenriched propargyl alcohols, chirality transfer occurs with considerable diastereoselectivity and virtually complete enantioselectivity (Scheme 4-49) [150]; carbonylation followed by oxidative decomplexation gives the propargylation product 97, e. g., H C = CCHPhCH,C(O)Ph, enantiomerically pure.
96
95
Nu R,-Ad+12 H
101
R+,
~
4
97
1
NU-
\ \
L(co),co-co(co),
Scheme 4-49
4.4.3.2.2 Proton Loss/Elimination
The facile acid-promoted dehydration of cobalt-complexed propargyl alcohols [18], via P-proton loss from the derived cations (Scheme 4-50) offers improved chemo-, regio- and stereoselectivity v i s a vis the free propargylic alcohols, with a strong preference for the more substituted (E)-ene-yne complex [151, 1521. Such Co-mediated dehydration has afforded routes to enantiomerically pure manicone and normanicone (4,6-dimethyl-4-octen-3-one,
4.4 Reactions of Metal-Alkyne Complexes
121
3,5-dimethyl-3-hepten-2-one) [151], acetylenic analogs of leukotriene-E, [152], 16a, 17a-epoxycorticosterone [153], and the side chain of isolaurepinnacin [154].
98 Scheme 4-50
4.4.3.2.3 Coupling with Noncarbon Nucleophiles
Protonation of 98 followed by reduction with hydride (CF3CO2H-NaBH4 [155] or -BH3 SMe, [156]) provides a one-pot sequence for converting tertiary propargyl alcohols to the corresponding sec-alkylacetylenes, an attractive alternative to the often-inefficient, direct acetylide/sec-alkyl halide coupling. The latter reagent combination has been employed in a highly stereoselective route to 113-methylcarbapenem precursors 99 [157] (Scheme 4-51). TBDMS3
'.
HO
0
Me (c016
i) CF,CO,H, BH,. Me2S ii) ce4+
H
TBDMSD
* H
0 99
Scheme 4-51
Oxygen-centered nucleophiles (OH-, OMe-) have been used to intercept the cations 100 generated by &addition of electrophiles to 1,3-enyne cobalt complexes 101 (E' = RCO', Nu- =OH-, OR-), constituting a method for the selective a-hydroxy- and a-methoxy-0-acylation of ene-ynes (Scheme 4-52) (1581. The Co-mediated, highly stereoselective epimerization of 1-alkynyl pyranose derivatives 102 103 (Scheme 4-53) takes advantage of both the cation-stabilizing ability and the sterically demanding nature of the - (alkynyl)Co,(CO), unit [159]. The Co- and Mo-propargylium complexes react with primary and secondary amines in the presence of a hindered base to give, respectively, bis- and monopropargylated tertiary -+
101 Scheme 4-52
100
4 The Chemistry of Metal-Alkyne Complexes
122
102
SiMe3
SiMe3 103
Scheme 4-53
amines in good yield (Scheme 4-46)[160,1611. Based on this reactivity, Magnus utilized the - (propargyl)Co,(CO), unit as a novel N-protecting group in a nonoxidative methodology for the synthesis of vinblastine and vincristine model compounds [162].An alternative method for N-propargylation by the cobalt complexes comes from their use in the Ritter reaction with acetonitrile [163] or their reaction with sulfonamides [164]. Jaouen’s group synthesized [(HC = CCH2Nu)Co,(C0)6]BF4 from reaction of 92 with sulfides, phosphines, and pyridine, and found that the sulfide derivatives serve as “time release” precursors of the propargylium complexes with attenuated reactivity toward various nucleophiles [l65]. Recently, these workers have also reported the DBU induced deprotonation of the Mo2-phosphonium salt 104 which produces the novel ylide 105 (Scheme 4-54)whose X-ray structure suggests significant stabilization of the negative charge by delocalization onto a metal center [166].
104
105
Scheme 4-54
4.4.3.2.4 Coupling with Carbon Nucleophiles
Aromatics
The Co-complexed cations 92 undergo electrophilic substitution with electron-rich aromatic nucleophiles such as anisole, phenol and NN-dimethylaniline; after demetalation, good to excellent yields of C-propargylated aromatic derivatives result (Scheme 4-46) [167].Selective substitution of - (propargyl)Co2(C0)6 groups on polysubstituted arene rings has been studied in the zeranol, p-resorcyclic acid and 6-methyl-~-resorcyclicacid series [168].Grove has demonstrated that the cobalt-complexed cations can be used in intramolecular aromatic alkylations to generate useful cis-fused tricyclic ring systems 106 (Scheme 4-55)[169].“Labeling” of the A-ring of aromatic steroids by electrophilic substitution with 92 has been investigated with a view to using Fourier-transform infrared spectroscopy (FTIR) as a tool in receptor binding studies (see Sect. 4.5) [170].
4.4 Reactions of Metal-Alkyne Complexes
123
106
Scheme 4-55
Heteroaromatic substrates including furans and thiophenes can be alkylated efficiently at the 2-position, leading to the synthesis of prostaglandin analogs [171], while indole and tryptamine derivatives react with high C3 regioselectivity [172, 1731. The cobalt complexes of propargylium and a-vinylpropargylium cations have been found to couple efficiently with 2- and 3-siloxyfurans producing propargylated furanones in good yields [174]. Enol Derivatives Ketones with a-hydrogens react with the (propargylium)Co,(CO),+ complexes [175], undergoing regioselective alkylation at the more highly substituted a-position; classical direct (e. g., enolate) and indirect (enamine, acetoacetic ester) methods for ketone propargylation are often complicated by the formation of allenic by-products. &Diketones and ketoesters are also readily C-alkylated by 92 (Scheme 4-46) [176]. More generally useful is the coupling of silylenol ethers with -CO,(CO)~ complexes of propargyl alcohols 11751 and ethers, and acetylenic acetals and aldehydes, 107, 108 (Scheme 4-56). Lewis-acid-promoted reactions of complexed propargyl ethers give alkylated ketones 109 with moderate to excellent syn diastereoselectivity [22], which increases with the size of the remote acetylenic substituent, probably reflecting the bent geometry of the coordinated propargyl unit [177]. The reactivity of 92 is modified by hexamethyldisilazane, presumably via reversible adduct formation, allowing selective reaction at enol sites in the presence of activated aromatic rings as illustrated in a synthesis of 16a-substituted-17P-estradiols [178]. Alternatively, the Co-complexed cations 92, formed by electrophilic addition to 1,3-enyne complexes, have been trapped by trimethylsilylenol ethers (and ally1 silanes) [179], providing entry to a$-C-functionalized acetylenes.
108 or
+
A4
HBF, or
BF,-EI,O
~
R3 R R, = R3 = R, = H, Alk R2 = Alk, OAlk R5 = OH, OAlk, Alk; M = SIRS, SnR3, BR2
107 Scheme 4-56
1 09
124
4 The Chemistry of Metal-Alkyne Complexes
A consequence of the stereochemical nonrigidity of the -CO~(CO)~ complexes is the racemization which accompanies ionizationkoupling of homochiral propargyl ether complexes, indicating that antarafacial migration of the secondary cations (Scheme 4-7) is fast relative to the rate of alkylation [40]. Seeking an enantioselective approach to the propargyl-cobalt coupling reactions, Schreiber’s group found that Evans-type homochiral boron enolates, e. g., 110 (Scheme 4-57) [40], react highly stereoselectively (12: 1-35 : l), allowing the production of optically active alkylation products 111. The selectivity was rationalized by a double stereodifferentiating process wherein the rapidly equilibrating cation enantiomers react at different rates, leading to net kinetic resolution.
12 (R = H, R1 = Me) : 1 (R1 = H, R = Me)
110
111
Scheme 4-57
Early examples of the use of cobalt-mediated propargylation in organic synthesis included routes to dihydrojasmone [NO], the guaiane sesquiterpene cyclocolorenone [181] and its relative isocyclocolorenone [177], all of which take advantage of regioselective hydration of the intermediate propargylated ketones [182]. High stereoselectivitieswere observed for the key alkylation step in the latter two syntheses (6: 1, > 20: 1). Enol silane propargylation (followed by acetylene hydration and aldolization) has also been used in the stereoselective construction of bicyclic acetylenic ketone 112, which was converted to a demonstrated precursor 113 of damsin (Scheme 4-58)“3- 1851. Another useful example of enol silane propargylation comes from a recent formal synthesis of stemodin by the Vollhardt group “61.
112
113
Scheme 4-58
Jacobi and co-workers have applied the above SchreibedEvans chiral boron enolate methodology to afford stereoselective routes to precursors of biologically important tetrapyrroles [187], pyrromethanenones (114) (Scheme 4-59) [188], phycocyanin and phytochrome precursors, and p-amino acids “91, versatile intermediates for p-lactams of the carbapenem class. Generally, reaction of achiral or “matched” enolates with racemic cobalt complexes gave excellent syn selectivity. With a careful choice of “mis-matched” chiral enolate, moderate to good anti selectivity could also be achieved, leading to a formal total synthesis of thienamycin [190].
125
4.4 Reactions of Metal-Alkyne Complexes i)B B 9 0 l l
I
A
R = SCHOMeCH3, R' = CH2CH2C02Me
114
Scheme 4-59
Intramolecular alkylations of enol derivatives by the Co-complexed propargyl cations have found wide application in approaches to the synthesis of the ene-diyne anticancer antibiotics, especially by the Magnus group. For example, synthesis of the esperamicinkalicheamicin core unit employed ether 115,which undergoes efficient closure to 116 (Scheme 4-60) [126]. Similar alkylative cyclizationsleading to the ene-diyne core skeletons of dynemicin (117) [191], neocarzinostatin chromophore (72) [192], and related structures [193, 1941 have also been effected.
q-oq% ''6
RoMe0
R=TBDMS
(co)~co4o(co)~ 115
'I,
-
','z-
(co)3co-co(co)3 116
(co)3 0*0(co)3 117
Scheme 4-60
Lewis-acid-promoted alkylations of silylenol ethers and silyl ketene acetals [195] with Cocomplexed acetylenic acetals [196] and acetylenic aldehydes [197, 1981 (Scheme 4-56) also proceed with fair to excellent syn diastereoselectivity, in contrast to the low selectivity reactions of the free acetylenic derivatives [199, 2001. Reactions of the complexed aldehydes with lithium enolates are stereospecific, with (Z)-enolates giving syn selectivity and (E)-enolates anti selectivity [201]. The complementary stereoselectivity of the crossed aldol reactions of free and cobalt-complexed propynals with silyl ketene O,S-acetals has been elaborated by Hanoaka; exclusive syn selectivity is exhibited by the complexes and high anti selectivity is found with pro-
R=SCH2CH2NHAc (k) -6-epi-PS-5
Scheme 4-61
-
126
4 The Chemistry of Metal-Alkyne Complexes
pynal itself [202]. Applications of these stereoselective reactions to the synthesis of p-lactam antibiotics (*)-PS-5 (118) and (f)d-epi-PS-5 (119) and all four stereoisomers of (*)blastmycinone [203] have been described. Lastly, Roth has described the reaction of cobalt-complexed propargyl cations with enamines leading to intermediate iminium ion salts which were treated in situ with carbon nucleophiles; decomplexation gives precursors for the synthesis of five- and six-membered nitrogen heterocycles [204]. Ally1 Metals
Allylsilanes also couple readily with the propargylium-cobalt complexes, providing a regiocontrolled route to 1,5-enynes(Scheme 4-46) [205]. As noted earlier, Schreiber’s group studied intramolecular variants of the allylsilane coupling reactions [22], an endocyclic version producing medium-sized cycloalkyne (seven- and eight-membered) complexes (Scheme 4-14), and an exocyclic version proceeding with complete tmns-1,2-stereocontrol. In the course of studies directed toward the synthesis of macrocyclic and medium-ring natural products, Marshall showed that a cobalt-complexed alkynyl aldehyde 120 undergoes intmmolecular coupling with an allylstannane unit (Scheme 4-62) [206] stereoselectively producing 12-membered alcohol 121 in much higher yield than for the uncomplexed substrate. Roush has found that asymmetric crotylboration of (2- and 3-de~ynal)Co,(CO)~ proceeds with high syn diastereoselectivity and significantly higher enantioselectivities (> 90 Vo) than the corresponding free aldehydes (Scheme 4-63) [207, 2081. Similarly, Ganesh and Nicholas investigated the coupling of homochiral y-alkoxyallylboranes with acetylenic aldehyde complexes [209] which proceed efficiently to produce 3,4-dioxy-1,5-enynes diastereospecifically (2 --t syn, E --t anto and with high enantioselectivity ( 2 95 To). The metal-free acetylenic derivatives are promising intermediates for the synthesis of natural products possessing multiple adjacent stereocenters.
L 120
Scheme 4-62
Scheme 4-63
Co2(co),
J
121
4.4 Reactions of Metal-Alkyne Complexes
127
Alkyls/Cyanide With regard to the development of efficient methods for propargyl/hydrocarbyl coupling, the combination of organoaluminum reagents with complexes of propargyl acetates has proved the most effective (Scheme 4-46) [210]. A variation on this theme involves the use of (RC= C)3Al to produce 1,Cdiynes [211]. Stuart and Nicholas found that complexes of propargyl acetates and acetylenic acetals efficiently couple with Et2A1CN [212], providing a convenient route to propargyl nitriles and cyanohydrin derivatives (Scheme 4-46). 4.4.3.2.5 Tandem Nucleophilic Coupling/Pauson-Khand Reaction
In concluding this section on the reactions of propargylium complexes with nucleophiles, we note that use of appropriate unsaturated nucleophiles as reaction partners with the cobalt cations or unsaturated electrophiles in the Ad, reaction of Co-complexed 1,3-enynes [213-2151 offers an efficient route to variously substituted a,o-enynes, valuable precursors for intramolecular Pauson-Khand cyclizations (Scheme 4-64) [96, 981. This tandem methodology has been employed to produce, among others, bicyclo[3.3.0]octenones and their 3-oxa analogs [22, 216, 2171, the fusicoccin sesquiterpenoid skeleton [218], linear and angular fused tri- and tetracyclics [219], and fenestrane derivatives [220].
Scheme 4-64
4.4.3.3 Reduction and a-Radical Reactions
The formation of 1J-diyne complexes as side-products in some reactions of cations 92 with metal alkyls [41] and the facile, regioselective reduction of a propargylic chloride complex by Zn/HOAc, used in the synthesis of the insect pheromone 5-(Z)-tetradecenyl acetate [221], provided early evidence for the intermediacy of ( p ~ o p a r ~ ~ ) c o ~ (mdicals. c o ) 6 In a similar vein, the binuclear molybdenum-complexed propargyl cations react with Na/Hg to produce 1J-diyne derivatives [32, 331. More recent studies by Melikyan and Nicholas have focused on the intentional generation and chemistry of these novel organometallic radicals. Thus, reaction of primary and secondary Co,cation complexes with Zn regioselectively produces 1J-diyne derivatives in fair to excellent yield complementing the aforementioned intmmolecular version (Scheme 4-15) [66]. (Pr~pargyl)Co~(CO)~ radicals presumably are also involved in the Mn(II1)-mediated addition of P-dicarbonyl compounds to complexed 1,3-enynes, which produces highly functionalized dihydrofuran derivatives 122 (Scheme 4-65) [222, 223). The chemo-, regio-, and stereoselectivity of these reactions stands in contrast to the variable selectivity associated with the corresponding reactions of free enynes [224]. The formation of ethers 123 in methanol (Scheme 4-65) suggests that the cobalt -propargyl radicals initially produced are rapidly oxidized by Mn(II1) to the stabilized carbocations.
128
4 The Chemistry of Metal-Alkyne Complexes
123
122
Scheme 4-65
4.4.4 Reactions Remote from the Complexed Triple Bond Early work by Pettit [123] and Seyferth [225] demonstrated the concept of using the - C O ~ ( C O )unit ~ as a protecting group for the C-C triple bond. Since these alkyne complexes are reasonably resistant to attack by moderate electrophilic, nucleophilic and reducing agents, remote reductions, dehydrations, hydroborations, and Freidel-Crafts reactions can thus be effected selectively, leaving the complexed alkyne unit unchanged. Acetylenic analogs of tetrathiafulvalene have been prepared through Wittig reactions involving cobalt-complexed acetylenedicarboxaldehyde;complexation helped overcome problems of low yields and sideproducts often encountered with the uncomplexed acetylenic aldehydes [226]. The protecting group concept has also been used to direct chemoselective radical additions 1222, 2231 and dipolar cycloadditions at the double bond of 1,3-enyne complexes [227]. The unprotected counterparts of these reactions generally proceed with attack at both the double and triple bonds [224, 2281.
4.4.5 Reaction Summary We conclude our reaction survey with Scheme 4-66,which summarizes the known reactivity characteristics of metal-alkyne complexes.
4.5 Special Applications of Metal- Alkyne Complexes Jaouen and co-workers have promoted the use of metal carbonyl complexes as FTIR markers in biological studies [229, 2301. The Co2(CO),-labe1ed female hormonal steroid, 178estradiol, proved superior for quantitative analysis purposes than corresponding - Cr(CO)3 or - Cp,Mo,(CO), labels [231]. The alkylating ability of (propargylium)MzL, complexes may play a key role in the efficacy of complexed 17-alkynyl estradiols 124 (Scheme 4-67) as suicide substrates for the study of receptor proteins [232). These results correlate with the p K 2 values of model propargyl alcohol complexes, i.e., the least stable (most reactive) cations show the strongest inactivation, suggesting that receptor inactivation is derived from irreversible alkylation (of a cysteine - SH?) by the intervening electrophilic complex.
4.5 Special Applications of Metal-Alkyne Complexes
Triple bond protectlon & deprotectlon slte (RzTMS)
€lectrophlllc M a c k sltes In neutral complexes, produdng a-vlnyl end alkylldyne complexes
129
Carbocatlon center generated by lonlrstlon , AdE or ,H-abstraction reactlon
I
Radlcal center generated by Adl, OT cerbocatlon reductlon of acetylldes
Electrophlllc center In nucleophlllc substltutlon reaction ( X I LG)
1.2-H Shlfi (R=H),
produclng vlnylldene
Nucleophlllc attack alte In $ propargy//um complexes
-
Nucleophlllc attack sltee In cdonlc alhyne complexes
M-C Insertlona (coupllng with unsaturated substrates, cycloollgomerlzatlon. polymerlzatlon, metathesis)
Metel cluster chemlstry (Ilgmd exchange, cluster substltutlon, expanslon)
Scheme 4-66
It is perhaps fitting to conclude this review with an “ashes to ashes, dust to dust” illustration of the use of alkyne-metal complexes in materials science. For this we cite the recent production of nanoscale platinum clusters in glassy (sp2) carbon with unusual electrocatalytic properties by pyrolysis (600 “C) of (Ph3)2Pt-complexed poly(phenylenediacety1ene) (125) (Scheme 4-68) [233].
\ %(PPh3)2WWd
A
_____)
125
Scheme 468
Ptxonglassy cahn
130
4 The Chemistry of Metal-Alkyne Complexes
4.6 Selected Experimental Procedures We include here a few representative procedures involving work of the authors. The reaction sequences are shown in Schemes 4-69 and 4-70.
127
Ph
GOH
129
i)HBFq ii)Zn
pxh 128
Scheme 4-70
4.6.1 ~-[(~2,~2-l-Methyl-2-propynylium)dicobalthexa~rbonyll Tetrafluoroborate (126) POI In a well-ventilated fume hood a 2-L, two-necked, round-bottom flask fitted with a magnetic stirring bar, stopper, and a gas inlet T-tube attached to a mineral oil bubbler is flame-dried under a flow of nitrogen. The flask is charged with 200 mL of dry dichloromethane and 13.0 g (0.185 mol) of 3-butyn-2-01. After the mixture is stirred for 15 min, 65.0 g (0.19 mol) of dicobalt octacarbonyl is added in portions over a few minutes while maintaining a slow stream of nitrogen. Vigorous gas evolution (carbon monoxide!) is observed. The mixture is stirred for 4-5 h, and the solvent is then removed under reduced pressure (20-25 mm). The residual solid (alkyne)Co,(C0)6 complex is dissolved in 40 mL of propionic anhydride under nitrogen and cooled to - 45 "C in a dry-ice/acetonitrile bath. Tetrafluoroboric acid-dimethyl etherate (37.3 g, 0.28 mol) is added with stirring. After 30 min, 600-800 mL of anhydrous diethyl ether is added with continuous stirring. The burgundy-red salt which precipitates is isolated by filtration under a flow of nitrogen and is thoroughly washed with anhydrous diethyl ether to give 60-61 g (76-77070) of 126 as a dark-red, moisture-sensitive solid. This material is used immediately in the following step.
4.6 Selected Experimental Procedures
131
4.6.2 2-(l-Methyl-2-propynyl)cyclohexanone(127)[20] In a well-ventilated hood a 2-L, two-necked, round-bottom flask is equipped with a magnetic stirring bar, stopper, and a pressure-equalizing dropping funnel fitted with a gas inlet T-tube connected to a mineral oil bubbler. The flask is flushed with nitrogen and charged with 150 mL of dry dichloromethane and 60.0 g (0.141 mol) of 126. The mixture is stirred and cooled to -78 "C in a dry-ice/2-propanol bath, and 23.9 g (0.141 mol) of l-trimethylsiloxycyclohexene is added dropwise over a few minutes. The mixture is stirred at - 78 "C for 4 h. After the solution is warmed to room temperature, dichloromethane is removed under reduced pressure and replaced with 400 mL of acetone. The dark-red solution of the alkyne complex is cooled to - 78 "C and 175 g (0.32 mol) of ceric ammonium nitrate is added in portions. The mixture is stirred until the gas evolution (carbon monoxide!) ceases (ca. 4 h). The reaction mixture is warmed to room temperature, poured into 1 L of saturated brine solution, and extracted with four 250-mL portions of diethyl ether. The combined ether extracts are dried over magnesium sulfate, filtered, and concentrated on a rotary evaporator. The residual red oil is distilled at reduced pressure to afford 15.0-15.2 g (71-72%) of 127 as a pale-yellow liquid, bp 57-60°C (10 mm).
4.6.3 ~1,[(~~,~~-dl-3,4-Diphenyl-1,5-cyclooctadiyne]-bis-hexacarbonyldicobalt (128)[651 A flame-dried round-bottom flask was charged with 517 mg (0.6 mmol) of 129 and 0.5 mL propionic anhydride at 20°C under a N, atmosphere. The reaction flask was cooled to - 5 "C, 214 mg (1.32 mmol) HBF, EtzO was added dropwise via syringe (3 min), and the mixture was stirred for 30 min. The dark-red dication salt thus formed was thoroughly washed with dry ether (3 x 30 mL) at -1O"C, warmed to 20°C and dissolved (partially) in dry CH2C12(120 mL). Zn dust (2.74 g, 42 mmol, 325-mesh) was added and the reaction mixture was stirred for 19 h (TLC control). The dark-brown solution was cannulated and chromatographed after evaporation (SiO,, 15 g, petroleum ether, 35-60 "C). Thus obtained were 237 mg (47.7%) dl-isomer 128 and 3 mg (0.6%) of the meso-isomer, de 97.6%.
-
Acknowledgements We thank Dr. A. J. M. Caffyn for assistance in coverage of the chemistry of propargylium complexes [16]. We are grateful for financial support of our research in this area provided by the National Institutes of Health and the Petroleum Research Fund (A. C. S.).
Abbreviations Ad, AdR BPK
electrophilic addition radical addition benzophenone ketyl
132
4 The Chemistry of Metal-Alkyne Complexes
CP CP* DABCO DBU
DIPP EHMO
EWG Fc HOMO/LUMO
L TB TBDMS tetraphos Tf
THF TMS
TTC
cyclopentadienyl
pentamethylcyclopentadienyl 1,4-diazobicyclo [2.2.2]octane 1,5-diazobicyclo[5.4.O]undec-5-ene 2,6-diisopropylphenoxide extended Huckel molecular orbital electron-withdrawing group ferrocene highest occupied MO/lowest unoccupied M O ligand tribenzocyclyne t-but yldirnethylsilyl
tris-(diphenylphosphinornethy1)phosphine trifyl (trifluoromethanesulfonyl) tetrahydro furan trimethylsilyl trithienocyclotriyne
References [I] W. Reppe, 0. Schlichting, K. Klager, T. Toepel, Liebigs Ann. Chem. 1948, 560, 1-92. [2] W. Hubel in Organic Synthesis via Metal Carbonyk, (Eds.: I. Wender, P. Pino, Vol. 1, Wiley, New York, 1%8, p. 273. (31 (a) K. M. Nicholas, M. 0. Nestle, D. Seyferth in Transition Metal Organornetallics in Organic Synthesis, Vol. 2 (Ed. :H. Alper, Academic Press, New York, 1978, Chapter 1, (b) J. L. Davidson in Reactions of Coordinated Ligands, Vol. 1 (Ed. : P. S. Braterman), Plenum Press, London, 1986, (c) J. P. Collman, L. S. Hegedus, J. R. Norton, R. G. Finke in Principles and Applications of Organotransition Metal Chemistry, University Science Books, Mill Valley, CA, 1987, Chapter 18. [4] M. J. S. Dewar, Bull. SOC.Chim. Fr. 1951, 18, C71-79; J. Chatt, L. A. Duncanson, J. Chem. SOC. 1953, 2939-2947. [5] S. D. Ittel, J. A. Ibers, Adv. Organomet. Chem. 1976, 14, 33-61. [6] C. J. Bums, R. A. Andersen, J. Am. Chem. SOC. 1987, 109, 941-942. [7] R. A. Williams, T. P. Hanusa, J. C. Huffman, J. Am. Chem. SOC. 1990, 112, 2454-2455. [8] J. 0. Glanville, J. M. Stewart, S. 0. Grim, J. Organomet. Chem. 1967, 7, P9-PIO. 191 S . G. Feng, C. C. Phillipp, A. S. Gamble, P.S. White, J. L. Templeton, Organometallics 1991, 10, 3504-3512. [lo] J. L. Templeton, B. C. Ward, J. Am. Chem. SOC. 1980, 102, 3288-3290. [11] D. M. Hoffman, R. Hoffmann, C. R. Fisel, J. Am. Chem. SOC.1982, 104, 3858-3875. [12] K. J. Ahmed, M. H. Chisolm, K. Folting, J. C. Huffman, Organometallics 1986, 5, 2171-2181. [13] F. A. Cotton, X. Feng, Znorg. Chem. 1990, 29, 3187-3192. [14] F. Baert, A. Guelzim, J. M. Pobiet, R. Wiest, J. Demuynck, M. Bernard, Znorg. Chem. 1986, 25, 1830- 1841. [I51 B. E. R. Schilling, R. Hoffmann, Acta Chem. Scand. B 1979,33,213-218; J.-F. Halet, J.-Y. Saillard, R. Lissillour, M. J. McGlinchey, G. Jaouen, Znorg. Chem. 1985, 24, 218-224. [16] Review: A. J. M. Caffyn, K. M. Nicholas in Comprehensive Organometallic Chemistry ZZ, Vol. 12, (Eds.: E. W. Abel, F. G. A. Stone, G. Wilkinson), Pergamon, Oxford (in press).
References
133
[17] Review: K. M. Nicholas, Acc. Chem. Res. 1987, 20, 207-214. [18] K. M. Nicholas, R. Pettit, J. Orgunornet. Chem. 1972, 44, C21-C24. [19] R. E. Connor, K. M. Nicholas, J. Organomet. Chem. 1977, 125, C45-C48. [20] V. Varghese, M. Saha, K. M. Nicholas, Org.Synth. 1989, 67, 141-148. [21] D. H.Bradley, M. A. Khan, K. M. Nicholas, Organometallics 1992, 11, 2598-2607. [22] S. L. Schreiber, T. Sammakia, W. E. Crowe, J. Am. Chem. SOC. 1986, 108, 3128-3130. [23] A. A. Schegolev, W. A. Smit, Y. B. Kalyan, M. Z. Krimer, R. Caple, Tetrahedron Lett. 1982, 23, 4419-4422. [24] Review: R. S. Dickson, P. J. Fraser, Adv. Organomet. Chem. 1974, 12, 323-377. [25] V. I. Sokolov, I. V. Barinov, 0. A. Reutov, Izv. Akad. Nauk SSSR, Ser. Khim. 1982, 1922. [26] M. F. D'Agostino, C. S. Frampton, M. J. McGlinchey, Organometullics1990, 9, 2972-2984. [27] M. F. D'Agostino, C. S. Frampton, M. J. McGlinchey, J. Organomet. Chem. 1990, 394, 145-166. [28] 0. A. Reutov, I. V. Barinov, V. A. Chertkov, V. I. Sokolov, J. Orgnnomet. Chem. 1985,297, C25-C29. [29] S. C. Bennett, M. A. Phipps, M. J. Went, J. Chem. SOC, Chem. Commun. 1994, 225-226. [30] M. Buchmeiser, H. Schottenberger, Organometallics 1993, 12, 2472-2477. [31] I. V. Barinov, 0. A. Reutov, A. V. Polyakov, A. I. Yanovsky, Y. T. Struchkov, V. I. Sokolov, X Organomet. Chem. 1991, 418, C24-C27. [32] N. Le Berre-Cosquer, R. Kergoat, P. L'Haridon, Organometallics 1992, 11, 721-728. [33] A. Meyer, D. J. McCabe, M. D. Curtis, Organometaflics 1987, 6, 1491-1498. [34] H. El Amouri, J. Vaissermann, Y. Besace, K. P.C. Vollhardt, G. E. Ball, Organometallics1993, 12, 605-609. [35] M. Gruselle, C. Cordier, M. Salmain, H. El Amouri, C. Gukrin, J. Vaissermann, G. Jaouen, Organornetallics 1990, 9, 2993-2997. [36] M. V. Galakhov, V. I. Bakhmutov, I. V. Barinov, 0. A. Reutov, 1 Organomet. Chem. 1991, 421, 65-73. [37] C. Cordier, M. Gruselle, G. Jaouen, V. I. Bakmutov, M. V. Galakhov, L. L. Troitskaya, V. I. Sokolov, Organometallics 1991, 10, 2303-2309. [38] L. L. Tkoitskaya, V. I. Sokolov, V. I. Bakhmutov, 0. A. Reutov, M. Gruselle, C. Cordier, G. Jaouen, J. Organomet. Chem. 1989, 364, 195-206. [39] S. Padmanabhan, K. M. Nicholas, J. Organomet. Chem. 1984, 268, C23-C27. [40]S. L. Schreiber, M. T. Klimas, T. Sammakia, J. Am. Chem. SOC. 1987, 109, 5749-5759. [41] S. Padmanabhan, K. M. Nicholas, J. Organomet. Chem. 1981, 212, 107-113. [42] G. A. Olah, C. Pittman, J. Am. Chem. SOC. 1965, 87, 5632-5637. [43] V. V. Krivykh, E. S. Taits, P. V. Petrovskii, Y. T. Struchkov, A. I. Yanovski, Mendeleev Commun. 1991, 103-104. [44] C. P. Casey, C. S. Yi, J. Am. Chem. SOC. 1992, 114, 6597-6598. [45] (a) P. W. Blosser, D. G. Schimff, J. C. Gallucci, A. Wojcicki, Organornetallics 1993, 12, 1993-1995. (b) T.-M. Huang, J.-T. Chen. G.-H. Lee, Y. Wang, J. Am. Chem. SOC. 1993, 115, 1170-1171. [46] Review: A. Wojcicki, New J. Chem. 1994, 18, 61-68. [47] Reviews: M. A. Bennett, H. P. Schwemlein, Angew. Chem., Int. Ed. Engl. 1989,28, 1296-1320; M. A. Bennett, Pure Appl. Chem. 1989, 61, 1695-1700. [48] T. L. Gilchrist, F. J. Graveling, C. W. Rees, J Chem. Soc, Chem. Commun. 1968, 821-822. [49] M. A. Bennett, P. B. Donalson, Znorg. Chem. 1978, 17, 1995-2000. [50] H. Petersen, H.Meier, Nouv. J. Chim. 1980, 4, 687-689. [51] M. A. Bennett, T. Yoshida, J. Am. Chem. SOC. 1978, 100, 1750-1759. (521 S. L. Buchwald, R. T. Lum,R. A. Fischer, W. M. Davis, J. Am. Chem. SOC.1989, Ill, 9113-9114. [53] A. J. Mayr, B. Carrasco-Flores, L. Parkanyi, K. H. Pannell, J Am. Chem. SOC. 1992,114,5467-5469. [54] R. D. Adams, L. Chen, X . Qu, Organometallics 1994, 13, 1992-1998 and references therein. [55] S. L. Buchwald, R. T. Lum, J. C. Dewan, J. Am. Chem. Soc. 1986, 108, 7441-7442.
134
4 The Chemistry of Metal-Alkyne Complexes
[56] S. J. McLain, R. R. Schrock, P. R. Sharp, M. R. Churchill, W. J. Youngs, J. Am. Chem. SOC. 1979, 101, 263-265. [57] M. A. Bennett, T. W. Harnbley, N. K. Roberts, G. B. Robertson, Organometallics1985,4, 1992-2000. [58] S. L. Buchwald, B. T. Watson, J. C. Huffrnan, J. Am. Chem. SOC. 1986,108, 7411-7413. [59] J. F. Hartwig, R. A. Andersen, R. G. Bergman, J. Am. Chem. SOC. 1989, 111, 2717-2719. [60] S. L. Buchwald, R. B. Nielsen, Chem. Rev. 1988, 88, 1047-1058. [61] A. J. Deeming, M. Underhill, J. Chem. SOC., Dalton Tram. 1974, 1415-1419. [62] S. L. Buchwald, E. S. Lucas, J. C. Dewan, J. Am. Chem. SOC. 1987,109, 4396-4397. [63] M. A. Bennett, J. S. Drage, K. D. Griffiths, N. K. Roberts, G. B. Wickramasinghe, Angew. Chem. 1988, 100, 1002-1004; Angew. Chem, Int. Ed. Engl. 1988,27, 941-942. [64] K. C. Nicolaou and W.-M. Dai, Angew. Chem. 1991,103,1453-1481; Angew. Chem., Int. Ed. Engl., 1991, 30, 1387-1416. [65] G. G. Melikyan, M. Khan, K. M. Nicholas, Organometallics 1995, 14, 2170-2173. [66] G. G. Melikyan, R. C. Combs, J. Lamirand, K. M. Nicholas, Tetrahedron Lett. 1994,35, 363-366. [67] R. G. Bergman, Acc. Chem. Res. 1973,6, 25-31. [68] G. G. Melikyan, K. M. Nicholas, unpublished results, 1994. [69] D. Solooki, J. D. Bradshaw, C. A. Tessier, W. J. Youngs, Organometallics 1994, 13, 451-455. [70] J. D. Ferrara, C. Tessier-Youngs, W. J. Youngs, Orgunometullics 1987, 6, 676-678. [71] J. D. Ferrara, A. A. Tanaka, C. Fierro, C. A. Tessier-Youngs, W. J. Youngs, Organometallics 1989, 8, 2089-2098. [72] Y. Rubin, C. B. Knobler, F. Diederich, J. Am. Chem. SOC. 1990,112, 4966-4968. [73] D. L. Reger, K. A. Belmore, E. Mintz, P. J. McElligot, Orgunometullics 1984,3, 134-140. [74] M. Green, M. Botrill, J. Am. Chem. SOC. 1977, 99, 5795-5796. [75] H. C. Clark, M. H. Chisolm, Acc. Chem. Res. 1973, 6, 202-209. [76] C. Bianchini, A. Meli, M. Peruzzini, F. Zanobini, C. Bruneau, P. H. Dixneuf, Orgunometallics1990, 9,115-122. [77] W. Hiscox, P. W. Jennings, Organometallics 1990, 9, 1997-1999. [78] J. L. Davidson, I. E. P. Murray, P. R. Preston, M. V. Russo, L. Manojlovic-Muir, K. W. Muir, J. Chem. Soc., Chem. Cornmun. 1981,1059-1061; J. L. Davidson, W. F. Wilson, L. Manojlovic-Muir, K. W. Muir, J. Orgunomet. Chem. 1983,254,C6-C10; S. R. Allen, R. G. Beevor, M. Green, N. C. Norman, A. G. Orpen, I. D. Williams, X Chem. Soc, Dalton Ifans. 1985,435-450;J. R. Morrow, T. L. Tonker, J. L. Templeton, J. Am. Chem. SOC. 1985, 107, 6956-6963. [79] J. B. Hartung, Jr., S. F. Pedersen, J. Am. Chem. SOC. 1989, I l l , 5468-5469. [80] B. E. Mann, B. L. Shaw, N. I. lbcker, J. Chem. SOC. A, 1971,2667-2673. [81] P.B. Tripathy, B. W. Renoe, K. Adzamli, D. M. Roundhill, J. Am. Chem. SOC.1971,93,4406-4410. [82] D. M. Barlex, R. D. W. Kemmitt, G. W. Littlecott, J. Chem. Soc, Chem. Commun. 1969,613-614. [83] J. A. Labinger, J. Schwartz, J. Am. Chem. SOC. 1975, 97, 1596-1597. [84] J. L. Thomas, Inorg. Chem. 1978, 17, 1507-1511. [85] P. B. Tripathy, D. M. Roundhill, X Am. Chem. SOC. 1970, 92, 3825-3826. [86] Review: D. Seyferth, Adv. Organomet. Chem. 1976, 14, 97-144. [87] C. Bianchini, A. Mali, M. Peruzzini, F. Vizza, P. Frediani, Organometallics 1989,8, 2080-2082. [88] R. R. Burch, E. L. Muetterties, R. G. Teller, J. M. Williams, J. Am. Chem. SOC. 1982, 104, 4257-4258. [89] T. Yoshida, W. J. Youngs, T. Sekaeda, T. Ueda, S. Otsuka, J. A. Ibers, J. Am. Chem. SOC.1983,105, 6273-6278. [go] J. A. Labinger, J. Schwartz, J. M. Townsend, J. Am. Chem. SOC. 1974, 96, 4009-4011. [91] H. C. Clark, C. S. Wong, J. Am. Chem. SOC.1977, 99,7073-7074. [92] D. W. Hart, J. Schwartz, J. Organomet. Chem. 1975,87, Cll-C14. [93] Ref. [3c]; D. W. Hart, T. F. Blackburn, J. Schwartz, X Am. Chem. SOC. 1975, 97, 679-680. [94] H. C. Clark, R. J. Puddephatt. Inorg. Chem. 1970,9, 2670-2675.
References
135
(951 K. P. C. Vollhardt, Angew. Chem. 1984, 96, 525-541; Angew. Chem. Int. Ed. Engl. 1984, 23, 539-556. (961 Review: N. E. Schore in Chapter 9.4, Vol. 5 , Comprehensive Organic Synthesis, Pergamon, Oxford, 1991, pp. 1129-1162; Chapter 5 of the present monograph. (971 Review: E.4. Negishi in Comprehensive Organic Synthesis, Vol. 5, Pergamon, Oxford, 1991, Chapter 9.5, pp. 1163-1184. [98) Review: P. L. Pauson, Tetrahedron, 1985, 41, 5855-5860. (991 E. R. F. Gesing, J. A. Sinclair, K. P. C. Vollhardt, J. Chem. SOC.. Chem. Commun. 1980, 286-287. [lo01 Y. Wakatsuki, 0. Nomura, K. Kitaura, K. Morokuma, H. Yamazaki, J. Am. Chem. SOC. 1983,105, 1907-1912. [loll Review: C. Bianchini, Pure Appl. Chem. 1991, 63, 829-834. (1021 J. R. Strickler, P. A. Wexler, D. E. Wigley, Organometallics 1988, 7, 2067-2069. 11031 C. Bianchini, K. G. Caulton, C. Chardon, 0. Eisenstein, K. Folting, T. J. Johnson, A. Meli, M. Perutzini, D. J. Rauscher, W. E. Streib, F, Vizza, Z Am. Chem. SOC. 1991, 113, 5127-5129. [lo41 Review: P. M. Maitlis, Acc. Chem. Res. 1976, 9, 93-99. [lo51 R. L. Funk, K. P. C. Vollhardt, J. Am. Chem. SOC. 1977, 99, 5483-5484. [lo61 H. Bonnemann, Angew. Chem. 1978,90,517-526; Angew. Chem. Int. Ed. Engl. 1978,17,505-515 and references therein. [lo71 Y. Inoue, Y. Itoh, H. Hashimoto, Chem. Lett. 1978, 633-634; H. Hoberg, D. Schaefer, G. Burkhardt, J. Organomet. Chem. 1982, 228, C21-C24. [lo81 E. J. M. de Boer, J. de With, A. G. Orpen, J. Am. Chem. SOC. 1986, 108, 8271-8273. [lo91 K. Takai, M. Tezuka, Y. Kataoka, K. Utimoto, J. Org. Chem. 1990, 55, 5310-5312. [110] R. S. Jolly, G. Leudthe, D. Sheehan, T. Livinghouse, J. Am. Chem. SOC. 1990, 112, 4965-4966. [lll] Review: B. M. Trost, Acc. Chem. Res. 1990, 23, 34-42. [I121 H. Yamazaki, Y. Wakatsuki, K. Aoki, Chem. Lett. 1979, 1041-1044. (1131 R. R. Schrock, M. L. Listemann, L. G. Sturgeoff, J. Am. Chem. SOC. 1982, 104, 4291-4293. [I141 J. H. Wengrovius, J. Sancho, R. R. Schrock, 1 Am. Chem. SOC. 1981, 103, 3932-3934. [115] L. G. McCullough, R. R. Schrock, J. C. Dewan, J. C. Murdzek, X Am. Chem. SOC. 1985, 107, 5987-5998. [116] Review: W. Keim, A. Behr, M. Roper, Alkene and alkyne oligomerization, cyclooligomerization and telomerization reactions, in Comprehensive Organometallic Chemistry, (Eds. : G. Wilkinson, F. G. A. Stone, E. W. Abel), Pergamon, Oxford, 1984. (1171 H. Foley, L.M. Strubinger, T. S. Targos, G. L. Geoffroy, J. Am. Chem. SOC.1983,105, 3064-3073. [I181 J. M. O’Connor, H. Ji, M. Iranpour, A. L. Reingold, J. Am. Chem. SOC. 1993, 5 , 1586-1588. [119] Review: H. Vahrenkamp, Adv. Organomet. Chem. 1976, 22, 169-208. [120] G. H. Young, A. Wojcicki, M. Calligaris, G. Nardin, N. Bresciani-Pahor, J. Am. Chem. SOC. 1989, 111, 6890-6891. [121] Review: A. Wojcicki, C. E. Schuchart, Coord. Chem. Rev. 1990, 105, 35-60. [122] D. Seyferth, M. 0.Nestle, A. T. Wehman, J. Am. Chem. Soc. 1975, 97, 7417-7426. [123] K. M. Nicholas, R. Pettit, ntrahedron Lett. 1971, 3475-3478. [124] P. Magnus, R. T. Lewis, J. C. Huffman, J. Am. Chem. SOC. 1988, 110, 6921-6923. [125] Y. Shvo, E. Nazum, J. Chem. Soc, Chem. Comrnun.1974, 336-337; P. Magnus, D. P. Becker, ibid. 1974, 640-642. [126] P. Magnus, T. Pitterna, J. Chem. Soc, Chem. Commun.1991, 541-543. [127] G. Cetini, 0. Gambino, R. Rossetti, E. Sappa, J. Organomet. Chem. 1%7, 8, 149-154. [128] G. G. Melikyan, K. M. Nicholas, unpublished results, 1993. [129] Reviews: M. I. Bruce, Chem. Rev. 1991, 91, 197-257; A. J. L. Pombiero, R. L. Richards, Cooml. Chem. Rev. 1990,104, 13-38; H. Werner, Angew. Chem. 1990,102, 1109-1121; Angew. Chem. Int. Ed. Engl. 1990, 29, 1077-1176. [130] J. R. Lomprey, J. P. Selegue, J. Am. Chem. SOC. 1992, 114, 5518-5523.
136
4 The Chemistry of Metal-Alkyne Complexes
[131] C. Bianchini, M. Puruzzini, A. Vacca, E Zanobini, Organometallics 1991, 10, 3697-3707. [132] C. Bianchini, D. Masi, A. Meli, M. Peruzzini, J. A. Ramirez, A. Vacca, F. Zanobini, Organometallies 1989,8, 2179-2189. I1331 J. Silvestre, R. Hoffmann, Helv. Chim. Acta 1985, 68, 1461-1506. [134] H. Werner, T. Rappert, Chem. Ber. 1993, 126, 669-678. [135] D. Schneider, H. Werner, Angew. Chem., Int. Ed. Engl. 1991, 30, 700-702. [136] N. E. Kolobova, L. L. Ivanov, 0. S. Zvanko, Izv. Akad. Nauk SSSR, Ser. Khim. 1980, 478-479, 2646-2649. [137] P. Nicklas, J. P. Selegue, B. A. Young, Organometallics 1988, 7, 2248-2250. [I381 P. Magnus, D. P. Becker, X Chem. SOC, Chem. Comrnun. 1985, 640-642. [139] V. Bernardes, X. Verdaguer, A. Moyano, M. A. Percias, A. Riera, A. E. Greene, J. Organomet. Chem. 1994, 470, C12-Cl4. [140] M. Franck-Neumann, F. Brion, Angew. Chem. 1979, 91, 736-737; Angew. Chern., Int. Ed. Engl. 1979, 18, 688-689. [141] D. L. Hughes, A. J. L. Pombiero, C. J. Picket, R. L. Richards, X Chem. SOC, Chem. Commun. 1984, 992-993. [142] S. F. T. Froom, M. Green, K. R. Nagle, D. J. Williams, J. Chem. Soc., Chem. Commun. 1987, 1305-1307. [143] M. Saha, K. M. Nicholas, X Org. Chem. 1984, 49, 417-422. [144] C. Mukai, Y.-I. Sugimoto, Y.Ikeda, M. Hanaoka, J. Chem. SOC,Chem. Commun. 1994,1161-1162. [145] S. Padmanabhan, K. M. Nicholas, Tetrahedron Lett. 1982,23, 2555-2558. [146] C. Descoins, D. Samain, Tetrahedron Lett. 1976,745-748. [147] M. Saha, Ph. D. Dissertation, Boston College, 1985. [148] D. H. Bradley, M. A. Khan, K. M. Nicholas, Organometallics 1989,8, 554-556. [149] J. A. Dunn, P. L. Pauson, X Organomet. Chem. 1991,419, 383-389. [150] A. J. M. Caffyn, K. M. Nicholas, J. Am. Chem. SOC. 1993, 115, 6438-6439. [MI G. G. Melikyan, A. Mineif, 0. Vostrowsky, H. J. Bestmann, Synthesis 1991, 633-636. [I521 A. K. Saksena, M. J. Green, P. Mangiaracina, J. K. Wong, W. Kreutner, A. R. Gulbenkian, Tetrahedron Lett. 1985,26, 6423-6426. [153] A. M. Tbruta, A. V. Kamernitsky, T. M. Fadeeva, L. D. Huy, Mendeleev Commun. 1992, 47-48. [154] D. Berger, L. E. Overman, P. A. Renhowe, X Am. Chem. SOC. 1993,115, 9305-9306. [155] K. M. Nicholas, J. Siegel, X Am. Chem. SOC. 1985, 107, 4999-5001. [156] D. F. McComsey, A. B. Reitz, C. A. Maryanoff, B. E. Maryanoff, Synth. Commun. 1986, 16, 1535-1549. [157] J. S. Prasad, L. S. Liebeskind, Tetrahedron Lett. 1987,28, 1857-1860. [I581 W. A. Smit, A. A. Schegolev, A. S. Gybin, G. S. Mikaelian, R. Caple, Synfhesis 1984, 887-890. [I591 S. Tanaka, T. Tsukiyama, M. Isobe, Tetrahedron Left. 1993,34. 5757-5760. [I601 M. Gruselle, V. Philomin, F. Chaminant, G. Jaouen, K. M. Nicholas, 1 Ofgunomet. Chem. 1990, 399, 317-326. [I611 K.-D. Roth, U. Miiller, Tetrahedron Lett. 1993, 34, 2919-2922. [I621 P. Magnus, M. Ladlow, J. Elliott, C. S. Kim, J. Chem. SOC, Chem. Commun. 1989, 518-520. [I631 S. Top, G. Jaouen, X Org.Chem. 1981, 46, 78-82. [I641 N. Jeong, S. Yoo, S. J. Lee, S. H. Lee, Y. K. Chung, Tetrahedron Lett. 1991,32, 2137-2140. [165] H. El-Amouri, M. Gruselle, G. Jaouen, J. C. Daran, J. Vaissermann, Inofg. Chem. 1990, 29, 3238-3242. [166] H. El-Amouri, M. Gruselle, Y. Besace, J. Vaissermann, G. Jaouen, Ofganometallics 1994, 13, 2244-2251. [167] R. F. Lockwood, K. M. Nicholas, Tetrahedron Lett. 1977, 4163-4166. [168] M. Gruselle, J.-L. Rossignol, A. Vessitres, G. Jaouen, J. Organomet. Chem. 1987,328, C12-Cl5. [I691 D. G. Grove, F. Miskevich, C. C. Smith, J. R. Corte, Tetrahedron Lett. 1990, 31, 6277-6280.
References
131
[170] S. Greenfield, M. Gruselle, G. Jaouen, V. Varghese, K. M. Nicholas, A&. Organomet. Chem. 1987, I , 441-447. (1711 H. J. Jaffer, P. L. Pauson, J. Chem. Res. (S), 1983, 244. [I721 M. Nakagawa, J. Ma, T. Hino, Heteroqcles 1990, 30, 451-462. [173] K.-D. Roth, Synlett 1993, 529-533. [I741 J. G. Stuart, K. M. Nicholas, Heterocycles 1991, 32, 949-963. [I751 K. M. Nicholas, M. Mulvaney, M. Bayer, J. Am. Chem. SOC. 1980, 102, 2508-2510. [176] H. D. Hodes, X. M. Nicholas, Tetrahedron Lett. 1978, 19, 4349-4352. [177] M. Saha, S. Muchmore, D. van der Helm, K. M. Nicholas, J. Org. Chem. 1986,51, 1960-1966 and references therein. [178] M. Gruselle, S. Greenfield, G. Jaouen, J. Chem. SOC, Chem. Commun. 1987, 1353-1355. [179] G. S. Mikaelian, A. S. Gybin, W. A. Smit, R. Caple, Tetrahedron Lett. 1985, 26, 1269-1272. [180] S. Padmanabhan, K. M. Nicholas, Synth. Commun. 1980, 10, 503-507. [I811 M. Saha, B. Bagby, K. M. Nicholas, Tetrahedron Lett. 1986, 27, 915-918. [182] R. Borch, G. Stork, J. Am. Chem. SOC. 1964, 84, 935-940. [183] A. M. Montana, K. M. Nicholas, M. A. Khan, J. 0%.Chem. 1988, 53, 5193-5201. [184] A. M. Montana, K. M. Nicholas, J. Org. Chem. 1990, 55, 1569-1578. [185] A. M. Montana, K. M. Nicholas, Mag. Res. Chem. 1990, 28, 486-495. [186] J. Germanas, C. Aubert, K. P. C. Vollhardt, J. Am. Chem. SOC. 1991, 113, 4006-4008. [I871 P. A. Jacobi, S. Rajeswari, Tetrahedron Lett. 1992, 33, 6231-6234. [188] P. A. Jacobi, R. W. DeSimone, Tetrahedron Lett. 1992, 33, 6239-6242. [189] P. A. Jacobi, W. Zheng, Tetrahedron Lett. 1993, 34, 2581-2584. [I901 P. A. Jacobi, W. Zheng, Tetrahedron Lett. 1993, 34, 2585-2588. [191] P. Magnus, S. M. Fortt, J. Chem. Soc., Chem. Commun. 1991, 544-546. [192] P. Magnus, P. Carter, J. Elliott, R. Lewis, J. Harling, T. Pitterna, W. E. Bauta, S. Fortt, J. Am. Chem. Soc. 1992, 114, 2544-2559. [I931 K. Tomioka, H. Fujita, K. Koga, Tetrahedron Lett. 1989, 30, 851-854. [I941 M. E. Maier, B. Greiner, Liebigs Ann. Chem. 1992, 855-861. [195] C. Mukai, K. Sukuki, K. Nagami, M. Hanaoka, J. Chem. SOC, Perkin Dam. 1 1992, 141-145. [I961 R. Tester, V. Varghese, A. M. Montana, M. Khan, K. M. Nicholas, J. Org. Chem. 1990,55, 186-192. [I971 J. Ju, B. R. Reddy, M. Khan, K. M. Nicholas, J. Org. Chem. 1989, 54, 5426-5428. I1981 C. Mukai, K. Nagumi, M. Hanaoka, Tetrahedron Lett. 1989,30, 5623-5626. [I991 A. G. Angoh, L. J. Clive, J. Chem. Soc, Chem. Commun.1984, 534-536. [200] S. Kobayashi, S. Matsui, T.Mukaiyama, Chem. Lett. 1988, 1491-1494. [201] C. Mukai, K. Nagami, M . Hanaoka, Tetrahedron Lett. 1989, 30, 5627-5630. [202] C. Mukai, 0. Kataoka, M. Hanaoka, Tetrahedron Lett. 1991, 32, 7553-7556. [203] C . Mukai, 0. Kataoka, M. Hanaoka, J. Org. Chem. 1993, 58, 2946-2952. [m]K.-D. Roth, Synlett 1992, 435-438. [205] J. E. O’Boyle, K. M. Nicholas, Tetrahedron Lett. 1980, 21, 1595-1598. [206] A. Marshall, W. Y. Gung, Btrahedron Lett. 1989, 30, 309-312. [207] W. R. Roush, J. C. Park, J. Org. Chem. 1990, 55, 1143-1144. [208] W. R. Roush, J. C. Park, Tetrahedron Lett. 1991, 44, 6285-6288. [209] P. Ganesh, K. M. Nicholas, 1 Org. Chem. 1993, 58, 5587-5588. [210] S. Padmanabhan, K. M. Nicholas, J. Organornet. Chem. 1981, 212, 115-124. [211] S. Padmanabhan, K. M. Nicholas, 2trahedron Lett. 1983, 24, 2239-2242. [212] J. G. Stuart, K. M. Nicholas, Tetrahedron Lett. 1983, 24, 2239-2242. [213] W. A. h i t , S. 0. Simonyan, G. S. Mikaelian, S. S. Mamyan, A. S. Gybin, A. S. Shashkov, R. Caple, Izv. Akad. Nauk SSSR. SeK Khim. 1989, 334-344. [214] R. Caple, in Organic Synthesis: Modern %nu3 (Ed. : 0. Chizhov), Blackwell, Oxford, 1987, p. 119.
138
4 The Chemistry of Metal-Alkyne Complexes
[215] W. A. Smit, A. S. Gybin, A. S. Shashkov, Y. T. Struchkov, L. G. Kyz’mina, G. S. Mikaelian, R. Caple, E. D. Swanson, Tetrahedron Lett. 1986,27, 1241-1244. (2161 A. S. Gybin, W. A. Smit, A. L. Veretenov, S. 0. Simonian, A. S. Shaskov, Y. T. Struchkov, L. G. Kyz’mina, R. Caple, Izv. Akad. Nauk SSSR, Ser. Khim. 1989, 2756-2769. [217] A. L. Veretenov, A. S. Gybin, W. A. Smit, V. A. Chertov, A. S. Shashkov, Izv. Akad. Nauk. SSSR, Ser: Khim. 1989, 1879-1888. [218] T. F. Jamison, S. Shambayati, W. E. Crowe, S. L. Schreiber, 1 Am. Chem. SOC. 1994, 116, 5505-5506. [219] A. S. Gybin, W. A. Smit, R. Caple, A. L. Veretenov, A. S. Shashkov, L. G. Vorontsova, M. G. Kurella, V. S. Chertkov, A. A. Karapetyan, A. Y. Kosnikov. M. S. Alexanyan, S. V. Lindeman, V. N. Panov, A. V. Maleev, Y. T. Struchkov, S. M. Sharpe, J. Am. Chem. SOC. 1992,114, 5555-5566. [220] W. A. Smit, S. M. Buhanjuk, S. 0. Simonyan, A. S. Shashkov, Y. T. Struchkov, A. I. Yanovsky, R. Caple, A. S. Gybin, L. G. Anderson, J. A. Whiteford, Btrahedron Lett. 1991,32, 2105-2108. [221] G. G. Melikyan, V. M. Mkrtchyan, K. A. Atanesyan, G. K. Asaryan, S. 0. Badanyan, Bioorg. Khim. 1990, 16, IOOO-1001. [222] G. G. Melikyan, 0. Vostrowsky, W. Bauer, H. J. Bestmann, J. Organomet. Chem. 1992, 423, C24-C27. [223] G. G. Melikyan, 0. Vostrowsky, W. Bauer, H. J. Bestmann, M. A. Khan, K. M. Nicholas, J. Org. Chem. 1994, 59, 222-229. [224] Review: G. G. Melikyan, Synthesis 1993, 833-850. [225] D. Seyferth, M. 0. Nestle, A. I. Wehman, J. Am. Chem. SOC. 1975,97, 7417-7426. [226] A. Khanous, A. Gorgues, M. Jubault, Tetrahedron Lett. 1990,31, 7311-7314. I2271 B. DuCroix, S. Bernard, K. M. Nicholas, unpublished results, 1994. [228] C. Grundmann, P. Grunanger, The Nitrile Oxides, Springer, Berlin, 1971, p. 222. [229] G. Jaouen, A. Vessihres, Pure Appl. Chem. 1989, 61, 565-572. [230] Review: G. Jaouen, A. Vessihres, I. S. Butler, Acc. Chem. Res. 1993, 26, 361-369. [231] M. Salmain, A. Vessihres, G. Jaouen, I. S. Butler, Anal. Chem. 1991, 63, 2323-2329. I2321 A. Vessikres, S. Top, C. Vaillant, D. Osella, J.-P. Mornon, G. Jaouen, Angew. Chem. 1992, 104, 790-792; Angew. Chem, Int. Ed. Engl. 1992,31, 753-755. [233] N. L. Pocard, D. C. Alsmeyer, R. L. McCreery, T. X. Neenan, M. R. Callstron, J. Am. Chem. Soc. 1992, 114, 769-771.
5 Organometallic Cycloaddition Reactions of Acetylenes Joseph A . Casalnuovo, Neil E. Schore
5.1 Introduction Cycloaddition reactions have long been cornerstones of synthetic organic methodology. The ability to construct multiple carbon-carbon bonds, especially with well-defined regio- and stereochemical characteristics, has been an essential tool in synthesis for decades. Transitionmetal-mediated organic chemistry greatly expands the range of substrates which can participate in such processes and, correspondingly, increases the variety of ring systems that can be accessed therefrom. While a great many cycloaddition reactions of alkenes are known and are very useful, such processes reduce the oxidation level of the olefinic C-atoms to that of an alkane. In contrast, cycloadditons of acetylenes preserve functionality at these C-atoms in the form of the remaining unsaturation, which may be the target of further transformation. This feature has been widely exploited in synthetic applications and is an important stimulus for ongoing research into acetylene cycloaddition chemistry. While the number of transition-metal-promoted cycloadditions of acetylenes is quite large [l],we have chosen to focus on two in particular. The first, reaction with late transition metal (so-called “Fischer”) carbenes, is noteworthy in the variety of ring systems to which it gives access and the richness of functionality that its products possess. The second, co-cycloaddition with alkenes and carbon monoxide to give cyclopentenones, has become a valuable supplement in the organic chemist’s arsenal of five-membered ring syntheses. These processes were both first reported in the early 1970s: metal carbene cycloadditions by Karl Heinz DBtz [2], and the cyclopentenone synthesis by Peter Pauson and Ihsan Khand [3]. Significant mechanistic information is available in both cases, permitting the rational evaluation of the utility of either in a wide variety of situations. Finally, both have attracted the attention of organic chemists worldwide and have been employed in numerous sophisticated synthetic applications.
5.2 Cycloadditions of Acetylenes with Fischer Carbenes In his initial paper in 1975, DBtz reported that the thermal cycloaddition of pentacarbonyl(methoxypheny1carbene)chromium with diphenylacetylene in di-n-butyl ether yielded a chromium-complexed 4-methoxy-I-naphthol [Z]. Soon thereafter, he related that the same reactants in n-heptane produced not only naphthol product, but also indene, furan, and cyclobutenone products [4]. As it turned out, these results foreshadowed the extraordinary richness of organic structural types that may be derived from cycloadditions of alkynes with Fischer carbenes, as well as very recent contributions to reaction chemoselectivity through control of reaction conditions. Indeed, in the years since, the field has seen the introduction of a number of newly discovered cycloaddition types and, maybe more importantly, has
140
5 Organornetallic Cycloaddition Reactions of Acetylenes
undergone a degree of taming, the tailoring of reaction chemoselectivity according to predictable reactivity patterns. Thus, the types of ring systems and functionality available through rational syntheses has truly exploded into a formidable array 11, 51. This is, of course, a boon to the synthetic chemist confronted with the task of reproducing the dazzling variety of rings and functionality present in natural products. Even so, the Ddtz reaction to form naphthols has easily remained the most intesively studied and exploited coupling between Fischer carbene and alkyne reaction partners. As a result, it will be treated first, while other cycloaddition reactions will be introduced under separate headings according to the general ring system produced. The intent in these sections will be to provide the reader with a general background of the patterns of reactivity that have become apparent since the 1970s, with particular attention to the discoveries and methodologies reported in the last few years.
-
5.2.1 Naphthols the Dotz Reaction The Dotz reaction, in its most common incarnation, produces a chromium-complexed naphthol through the reaction of an alkyne with a chromium arylalkoxycarbene under mild thermal conditions. An uncomplexed organic product can easily be isolated in almost every instance, and thus equations will be depicted in this fashion. The reaction is general for internal or terminal alkynes, with either aryl or alkyl substituents, giving 45-75 Yo yields. Terminal alkynes react with total regioselectivity, while internal alkynes allow good to excellent regiocontrol depending on the substituents [Eq. (l)]. The regioselectivity displayed in a given reaction is almost entirely dependent on the relative steric bulk of the alkyne substituents, with examples of contrasteric regiocontrol due to electronic effects extremely rare [a]. The products in Eq. (1) can be viewed as a joining of the three fragments in Fig. 5-1, drawn to display the prevailing product regiochemistry, with the alkyne carbon bearing the smaller substituent connected to the carbene carbon.
qoMe ReCMe, (*Bu)pO, 45°C.2 h
Cr(C015 R = Et, 56% R = &BU.59%
OMe 2.5 14
OMe 1 1
Figure 5-1 Schematic of connectivity in Dbtz naphthol formation.
Historically, the proposals for the mechanism of the Dotz reaction have come about largely on the basis of reactivity studies [7]. The most widely cited mechanisms for the Dotz reaction are presented in Scheme 5-1. The loss of carbon monoxide has been determined to be the in-
5.2 Cycloadditions of Acetylenes with Fkcher Carbenes
141
itial and rate-determining step [7c], followed by alkyne coordination to form 1. Subsequent formal [2 + 21 cycloaddition generates the metallacyclobutene 2, which undergoes electrocyclic ring opening to give a vinylcarbene complex 3. Carbon monoxide insertion may then lead to vinylketene 4, which undergoes electrocyclic ring closure to the cyclohexadienone 5. Finally, aromatization yields the isolated chromium-complexed 4-alkoxynaphthol. An alternative route has also been proposed in which 3 instead undergoes electrocyclic ring closure to the chromacyclohexadiene 6, which after CO insertion also leads to 5.
Scheme 5-1 Mechanistic pathways for the formation of naphthol, indene, and cyclobutenone products.
More recently, a number of experimental and theoretical results which have direct bearing on the reaction mechanism have been reported. Although no genuine intermediates have been isolated from the Ddtz reaction, a molybdenum derivative of the ultimate intermediate 5, and a chromium q 4-vinylketene complex, have recently been obtained [S]. The latter discovery and additional circumstantial evidence significantly strengthen the case for a vinylketene, compared with a chromacyclohexadiene, pathway [9]. In addition, the intermediacy of a chromacyclobutene in the Ddtz reaction has been called into question by the theoretical work of Hofmann; he concludes that the metallacyclobutene would be significantly higher in energy than its complexed vinylcarbene tautomer, and, thus, the alkyne insertion should take place directly (1 -+ 3) [lo]. The implications of this proposal are far-reaching, as insertion (1 + 2) from the preferred alkyne conformation has been suggested as the origin of regioselectivity in the Dotz reaction [7], and metallacyclobutene intermediates have been widely proposed in mechanistically related reactions [ll]. According to the modified mechanism, the regioselectivity arises as a steric preference for the larger alkyne substituent to reside at C1 versus C2 on the vinylcarbene complex, thus avoiding the steric congestion due to an axial carbon
142
5 Organometallic Cycloaddition Reactions of Acetylenes
monoxide ligand, as shown for the disfavored regioisomer in Fig. 5-2 [lob]. In general, the remaining mechanistic transformations present in Scheme 5-1 have so far survived scrutiny. However, recent detailed studies on the effect of reaction conditions on product distributions have suggested that solvent molecules, and the acetylenes themselves, play vital roles as simple ligands for the mechanistic intermediates, directing the observed chemoselectivity [9].
Figure 5-2 Steric congestion in the disfavored vinylcarbene regioisomer in a modified D6tz mechanism.
Although a few other late transition metal carbenes are capable of effecting the Dotz transformation, chromium has remained the metal of choice as it generally produces conveniently stable carbene complexes which provide good chemoselectivity for phenol formation. The other Group VI metal carbenes, however, have become important agents in the related reaction to form indenes (see Section 5.2.2). As the chromium-complexed arene product is generally air-sensitive, work-up procedures usually include decomplexation agents which may produce the naphthol (CO or Fe(III)), the quinone (Ce(IV)/H20), or the monoa c e d (Ce(IV)/CH,OH) as desired [4, 121. The Ddtz reaction was traditionally carried out in ethereal solvents for many years; more recently solvents of low coordinating ability have been found to be superior in most situations, with yields in excess of 80% available from annulation of simple alkynes to arylcarbenes in hexane [13]. In fact, it has become increasingly clear that manipulation of reaction conditions, including solvent, temperature, and concentration, can be used in a synthetically useful way to drastically alter product distributions (see Sections 5.2.2 and 5.2.3). Additionally, dry-state adsorption techniques, which have expanded the utility of the Pauson-Khand reaction (see Section 5.3.3), show promise for increased reaction rates and yields in the DOtz reaction [14]. Annulation can take place with virtually any x-system attached to a carbene C-atom, including a variety of aromatic, heteroaromatic, and simple vinyl systems [Eqs. (2)-(4)l [15].
9 Me r(C0)5
PhCSPh, (f?-Bu)pO,6O"C,1 h 19%
HO
5.2 Cycloadditions of Acetylenes with Fischer Carbenes
143
The reaction is exceptionally tolerant for functionality, which has permitted its exploitation in quite a number of natural product syntheses [5]. The cycloaddition of a cyclohexenylcarbene with an alkynyl lactone in Eq. (4) was the key step in the demonstration of a formal synthesis of daunomycinone, a tetracyclic anthracyclinone [16]. The functionally may even redirect product regiochemistry as in the annulation of 2-naphthylcarbenes, which is ordinarily driven toward C1 to give phenanthrenes [Eq. (2)], but can be diverted toward C3 if C1 alkoxy substitution is present. This allowed a high yield of the anthracene derivative and its subsequent conversion to a known tetracyclic precursor to daunomycin [Eq. (5)] [17]. The chemoselectivity of the reaction can also be exploited, as in the penultimate step in a synthesis of 1-0-methyldefucogilvocarvin V, in which the carbene preferentially reacts with the alkyne group, leaving alkene functionality intact. Lactonization of the cycloaddition product completes the preparation of the gilvocarcin V aglycone [Eq. (6)] [18]. Propargyl oxygen substituents, including methoxy groups, are a noteworthy exception to this functional group tolerance, diverting the reaction toward indene and furan formation, possibly through internal metal coordination of the oxygen [19]; this limitation can be circumvented by using bulky protecting groups, as exemplified in the synthesis of a fredericamycin A precursor [Eq.(7)] [20].
*
(1) a
O M e MeO--Cr(C0)4
t-BuOMe, 50°C,10 min (2) CO, 75 bar, Et20, 70"c, 76%
ppMe +
Me0
Cr(CO),
OMe
MeO2/
Me02C Me0
/
(5)
OMe
heptane, ___)
70"C, 10 h 43%
OTBDMS (7)
TBDMSO'
144
5 Organometallic Cycloaddition Reactions of Acetylenes
The amount of functionality attached directly to the product ring may also theoretically be increased through the use of heteroatom-substituted alkynes. The reaction is less tolerant in this respect: ynamines, ynediamines, and bis(dipheny1phosphino)acetylene are all known to react with carbenes to ultimately yield indenes [21], alkynyl thioethers do not lead to cycloaddtion products [22], and haloacetylenes can give complex product mixtures [23]. Alkynyl ethers are viable substrates, however, and have been used in the synthesis of the diterpene natural product 12-0-methyl royleanone [24] and in two separate routes to the biologically active furochromone khellin [25]. The key step in one of the routes to the latter involved reaction of a chromium furanalkoxycarbene with an alkoxyalkyne to produce a highly substituted benzofuran derivative [Eq. (S)]. It should be noted that the in-situ inclusion of acetic anhydride in the Dotz reaction can have extremely beneficial effects beyond its obvious role as an acylation agent [25, 261. In the reaction in Eq. (8), cycloaddition will not occur in the absence of acetic anhydride, while failure to also include triethylamine lengthens the reaction time to 72 h with only 36% yield. Even under conditions in which product acylation does not occur [see Eq. (7)], acetic anhydride increases reaction rate and yield, and prevents silanol elimination side-reactions [20b].
Q.
,OMe
ee ,~, AcO
I
PTBDMS
AcpO, NEt3, THF,,
?TBDMS
Stannyl acetylenes present a fascinating regiochemical anomaly in their reactions with vinyl carbenes, as they are capable of reversing the normal outcome of the benzannulation reaction in which the steric demands of the alkyne substituents dictate product regiochemistry 1231. Although the origin of the effect is yet to be clarified, the result is clear: the stannyl group shows a marked propensity to be incorporated adjacent to the methoxy substituent in the phenol product (stannyl-substituted alkyne C-atom next to the original carbene C-atom). The high degree of regioselectivity exhibited, coupled with the high reactivity of aryl-tin bonds, can be exploited to overcome the lack of regiocontrol for internal alkynes, one of the major limitations of the benzannulation reaction. For example, the phenol in Eq. (9) can be produced with total regiocontrol by reaction of the chromium complex with l-(tributylstannyl)1-pentyne, followed by an iodine quench and methylation. By contrast, reaction of the chromium carbene in Eq. (9) with 2-hexyne results in a 44% yield of a 2 : 1 mixture of the expected phenol regioisomers.
Replacing the oxygen atom of the Fischer carbene with other heteroatoms provides a valuable variation. Arylaminocarbenes show a marked chemoselectivity for producing the indane nucleus, which is attributed to the stronger donor character of the amino group relative
5.2 Cycloadditions of Acetylenes with Fischer Carbenes
145
to an alkoxy group, reducing the propensity for CO insertion products [see Section 5.2.21. However, N-acylated aminocarbenes, which have reduced electron-donating abilities, have been shown to provide good yields of naphthalene derivatives [Eq. (lo)] [27]. Sulfur functionality can also be incorporated, as in the production of a 1,4-dihydrothionaphthoquinone visnagan precursor by reaction of a chromium phenyl(alky1thio)carbene with an alkoxyalkyne [Eq. (ll)]. Subsequent steps, including replacement of the alkylthio group with hydrogen through reduction with Raney nickel, yielded the natural product [28]. It should be noted, however, that reactions of unstabilized carbene complexes, i. e., those lacking a heteroatomcontaining substituent, give only poor yields of naphthols lacking the 4-alkoxy group [29].
Me
CHzPh
(1) EtCECEt, toluene, 55-6OoC, 3 h (2) ACpO, DMAP, NEt3 (3) FeCI3-1.5DMF
O+Tot-i3u (CO)&r+
0
(10)
PhCHz'
N\COzt-Bu
56%
AcO
The intramolecular Dotz reaction, in which the alkyne is tethered to the carbene, improves synthetic efficiency and regiocontrol, although internal alkynes work better than terminal alkynes [30]. Such cycloadditions provided the key to a synthesis of the antibiotic deoxyfrenolicin [Eq. (12)] [31] and played a role in work aimed toward the synthesis of the benzofuran angelicin, in which a 3-furylcarbene complex annulates toward C2 of the furan, avoiding forming an isobenzofuran structure [Eq. (13)l [32].
(1) EtzO, 37°C
64 h (12)
(2) DDQ, aq CH3CN
51%
Me0
Cr(C0)5 OH (1) THF, 85°C. 8 h
(2) FeCI3, DMF, THF
&o&c,,,.
(3) CF&OzH, HOAc, C C l c
(13)
Reaction efficiency can also be enhanced by the use of tandem reactions, one-pot combinations of the Dotz reaction with other ring-forming reactions, including Diels-Alder, and nucleophilic aromatic addition reactions [33]. For example, alkynyl groups attached to Fischer
146
5 Organometallic Cycloaddition Reactions of Acetylenes
[
.
,
,
,
i
c OMe
n.r.=l-u
THF, 50"C, 1 d
1-
58% ~r(C0)~
1
OTIPS
M~
TIPS-OTF,
2.6-lutidine. 50-65OC, cr(co)5 CHpClp , 15-24 h (2) LDA, THF, O"C, 1 h * (3) 12, 0 to 25°C
0
[[m]
W
OSiMe3 . r
(14) OMe
OTIPS
(15)
65%
OMeSPh
r(CO)g
Me
+
OMe Ph OMe
147
5.2 Cycloadditions of Acetylenes with Fkcher Carbenes
&(co),
THF, 46'C, 1 d L
ph$Me
(16) Ph
w +pMe
\
n
W(C015 Me
TBDMSO
A
(1) CHsCN, 25"C, 16h, 1 atm CO
(17)
AAJ
(2) llO"C, 23 h
TBDMSO
62%
5.2.2 Indenes Indenes, like cyclobutenones and furans, are common side-products in the reaction of chromium arylalkoxycarbene complexes with alkynes, especially internal alkynes [9]. The indene structure comes about by a process that is very similar to naphthol formation: annulation to the aryl ring still occurs, but without carbon monoxide insertion, and, instead, bond formation takes place directly between an alkyne carbon and the aryl carbon ortho to the metal carbene substituent [Eq. (18)] [4]. Scheme 5-1 shows two pathways that have been suggested for this transformation: beginning from the vinylcarbene intermediate 3, naphthol formation can be diverted to intermediate 8, either by direct cyclization (3 + 8) or through the chromacyclohexadiene (3 + 6 + 8). Aromatization and decomplexation yield the indene [7b, d, 431. More detailed mechanistic analyses consider the roles of the stereochemistry of 3, as an (E)- or (a-vinylcarbene, as well as the coordination of external ligands, in the production of indenes, naphthols, furans, cyclobutenones, and other common side-products [8 a, 9, 13, 441.
+
%Me Cr(C0)5
PheCPh
8OoC,30min heptane, *
30% after decomplexation
& \ h
(18) u
I ,
Me
Certain alkyne and carbene substitution patterns divert the reaction toward predominant indene formation. For example, ynamines and ynediamines both react with chromium arylalkoxycarbenes to give isolable vinyl carbenes, which proceed on to indenes thermally [21a, b]. Chromium aryluminocarbenes also lead to good to excellent yields of indanones after hydrolysis of the enamine [Eq. (19)] [45]. In this latter case, the increased electron-donating
148
5 Organometallic Cycloaddition Reactions of Acetylenes
ability of the amino group apparently strengthens metal-carbonyl interaction, inhibiting carbon monoxide insertion. The reaction is totally regioselective, in the same direction as in naphthol formation. Ortho substitution on the aryl ring of the carbene, in general, also leads to increased production of indenes [9, 131. Substituents capable of internal coordination to the metal give the most dramatic effect, with good selectivity for the indene product seen in reactions with internal alkynes. Modest yields of the steroid skeleton have been obtained in this way [Eq. (20)] [46].
+ n-BuCECH
fl
(2) 95"C, (1) DMF, 125 15 h"C, 5 h
Cr(CO)5
MeOp
95%
PhCeCPh
-
HBU (19)
0
*
heptane, 80%, 1 h 32%
'Me
&+ Me02C "Me
Manipulation of reaction conditions can dramatically and systematically alter the product distribution. For example, use of dimethylformamide (DMF) solvent leads to an 83% yield of the indene from the reactants in Eq. (18) [45]. Similar effects have been particularly well documented for ortho-substituted arylcarbenes, for which the empirical observation is that indene formation is greatest in relatively high-dilution, high-temperature reactions, in polar, coordinating solvents [Eq. (2111 [91.
Q
Et
(1) E t m E t ,
R
O
M
e
Me0 Cr(C0)5
solvent, 45°C
Me0
(2)Ce(NH4)6(N02)
Solvent heptane THF THF
IComDlex] 0.5 M 0.5M 0.005M
yoYield
yoYield
81 61 5
18 (+ 5% indanone) 66 (+ 9% indanone)
<6
The metal employed can also alter chemoselectivity. Early examples in the literature suggested that molybdenum and tungsten arylcarbene complexes can show a reversed chemoselectivity relative to chromium, favoring indene production [47]. This effect has been systematically examined, indicating that indene selectivity decreases as Mo > W > Cr with both internal and terminal alkynes. The reason for this metal-dependent chemoselectivity is not clear; although the increased relative metal - CO bond strength (W - CO > Mo - CO > Cr - CO) has been suggested to explain decreased CO insertion products for molybdenum and tungsten, this does not correlate perfectly with the observed trend [8a]. Group VI vinylcarbene complexes have also been systematically studied, resulting in the first example of a
5.2 Cycloadditions of Acetylenes with Fischer Carbenes
149
carbocyclic five-membered ring annulation (cyclopentadiene) product [48].Not surprisingly, vinyl complexes show a greater chemoselectivity for six-membered ring formation than their aryl counterparts. Five-membered ring production decreases as for the aryl complexes, Mo > W > Cr, and is virtually nonexistent for terminal alkynes. Under certain conditions, however, cyclopentadiene (cyclopentenone after treatment with toluenesulfonic acid) production can be significant [Eq.(22)] [gal. (1) THF, 50"C, 8 h
+
EtCECEt,
%Me
Et
(2) air, HOTS, HzO/THF
O(CO)5
(22)
64%
5.2.3 Cyclobutenones Cyclobutenones are fairly common side-products in the reaction of chromium arylalkoxycarbenes with internal alkynes. As indicated in Scheme 5-1,the branch point in the formation of cyclobutenone versus naphthol products is believed to be vinylketene intermediate 4, which may undergo electrocyclic ring closure to 9, followed by reductive elimination to the product [7a].Cyclobutenone formation occurs only in the presence of internal or external ligands that can coordinate to unsaturated chromium species sufficiently well to prevent complexation to an internal n-system and thus divert the system toward 9. Depending on the alkyne and aryl substitution-patterns and the reaction conditions, cyclobutenone formation can be made to predominate. Thus, solvents of good coordinating ability such as acetonitrile, o-OMe aryl substitution (which allows internal coordination to chromium), and bulky alkynyl substituents all favor cyclobutenone formation [Eq. (23)] [13].In fact, the effect of solvent alone can be even more dramatic: for the reaction partners in Eq. (21), a 0.5 M concentration of the carbene complex in acetonitrile gives instead a 78Yo yield of cyclobutenone and only a combined 17% yield of quinone and indene products [9].
R1
n2
B3
Solvent
%Yield
% Yield
H H OMe H H
Et Et Et f-Bu Ph
Et Et
THF CH3CN CH3CN CH3CN CH3CN
88 30
0 23 43 27 51
Et Me Ph
<2 0 0
Not surprisingly, cyclobutenones have also been isolated from reactions of chromium alkylalkoxycarbenes, where no competing n-system is present [49], and from 40'-disubstituted arylalkoxylcarbenes in which bond formation to the arene is blocked [50]. By contrast,
150
5 Organornetallic Cycloaddition Reactions of Acetylenes
cyclobutenones have virtually never been observed with terminal alkynes; nor are they formed from chromium vinylalkoxycarbenes, in which coordination by the nonaromatic n-system cannot be prevented by other ligands [13].
5.2.4 Cyclopentenones Two distinct approaches are known to the synthesis of cyclopentenonesthrough alkyne/Fischer carbene cycloadditions. In one, chromium cyclopropylmethoxycarbenecomplexes react with alkynes in a formal I4 2 + 1 - 21 cycloaddition, which results in loss of ethylene and formation of a mixture of isomeric cyclopentenones in good overall yield. The reaction is tolerant of alkyne functionality and, similarly to the Dotz process, the larger alkyne substituent ends up a to the carbonyl carbon in the product, totally regioselectively for terminal alkynes, less so for unsymmetrical internal ones [Eq. (24)] [51]. As indicated, reactive cyclopentadienones are the initial product formed in these transformations, with subsequent reduction in the presence of Cr(O)/H,O yielding the observed cyclopentenones [52]. Carrying out this reaction in an intramolecular fashion by using oxygen-tethered acetylenic carbenes, followed by treatment with sodium methoxide/methanol, allows for control of both product regiochemistry and stereochemistry [Eq. (25)l [531.
+
48%
-:\'
1%aq. dioxane
'"4
14%
-A
Ph
H
MeO-/MeOH,
H
S
(25)
Me
74%
bH
In a complementary technique, cyclopentenones may also be produced from the reaction of alkynes with chromium ulkylmethoxycarbenes IEq. (26)] [54]. Compared with the previously discussed methodology, this route generates the same predominant product regiochemistry, and although yields are not as good ( < 30% for terminal alkynes, < 60% for internal alkynes), this route is capable of placing a second substituent on the other a-carbon in the product cyclopentenones [compare Eq. (24)l. The transformation is mechanistically unrelated to the former one, involving instead activation of an a-hydrogen on the carbene alkyl substituent. Interestingly, the C - H-activated cyclopentenone product is consitently favored over non-C - H-activated side-products whether the a-hydrogen is primary, secondary, benzylic, or p-silyl; tertiary H-atoms fail, apparently for steric reasons.
Q T
O
M
e
+
f-BuCFCH 6O-10OoC, hexanes 12-24 h *
Cr(C0)5 33%
'-"*
(26)
OMe
5.2 Cycloadditions of Acetylenes with Fischer Carbenes
151
5.2.5 Cycloheptadienones Unlike their chromium counterpart, molybdenum and tungsten cyclopropylcarbenecomplexes do not yield cyclopentenones on reaction with alkynes, but instead produce cycloheptadienones [ 5 5 ] . Molybdenum carbene complexes, which are known to be more reactive in similar cycloaddition chemistry [56],react efficiently at 65 “C in tetrahydrofuran (THF), while the tungsten complexes require 140“C in p-xylene, with longer reaction times. Under these conditions, molybdenum and tungsten, in similar yields, consistently produce the kinetic and thermodynamic alkene regioisomers, respectively, as the major product isolated [Eq. (27)J [57]. Although internal alkynes react efficiently and regioselectively, the reaction lacks scope, as it will not tolerate terminal alkynes and fails for most substituents on the carbene cyclopropyl group. A phosphine additive is used to boost yield and prevent a significant, if not synthetically useful, production of furanone side-product.
PhC=CMet see text
’6 phG +
Me
OMe
M Mo W
Mti””
-
(27)
-
45% 3%
w
OMe 10% 61%
5.2.6 Cyclopropanes The formation of substituted cyclopropane rings by the reaction of alkenes with Fischer carbenes has been known for some time [58]. More recently, cyclopropyl groups have been produced as parts of bicyclic and tricyclic ring systems by, formally, the reaction of Fischer carbenes with one equivalent of alkene and one equivalent of alkyne. Indeed, this reaction type proceeds through alkene trapping of a metal carbene generated in situ. The various methodologies that have been developed may be divided into three classes : the intermolecular reaction of a,o-enynes with Fischer carbenes, the partially intramolecular reaction of Fischer carbene-tethered alkynes with alkenes, and the fully intramolecular reactions of Fischer carbene-tethered enynes. No fully intramolecular version of this reaction has been reported. In the two separate, initial reports on the reactivity of Fischer carbenes with enynes, one study found cyclobutanone and furan products [59], while the other found products due to olefin metathesis [ a ] . These products have turned out to be the exceptions rather than the rule, as enynes have since been found to react with Fischer carbenes to produce bicyclic cyclopropanes quite generally. The proposed mechanistic pathway is included as part of Eq. (28), in which vinylcarbene 10, produced by insertion of the alkyne into the metal carbene, may then cyclize with the pendant olefin to metallacyclobutane 11, leading to product. The first reported version of this reaction suffered from extreme sensitivity to olefin substitution [Eq.(28); compare R=H, Me] often producing side-products due to metathesis (through 11 to yield dienes) and CO insertion (into 10 to yield cyclobutanones and furans) [61]. Since then, several important modifications have been developed which improve yield, provide greater tolerance for alkene substitution, and increase chemoselectivity for the bicyclic cyclopropane
152
5 Organometallic Cycloaddition Reactions of Acetylenes
product [62]. As previously mentioned, CO insertion can be suppressed by replacing the methoxycarbene substituent with a pyrrolidino substituent 1631, or using an anionic l-oxidoalkylidene-chromium complex [64]. Both complex types give excellent yields of bicyclic cyclopropanes for a variety of olefin substitution patterns. In addition, substituting for chromium in the neutral alkylalkoxycarbene complexes with molybdenum or tungsten also gives much more efficient reactions [65]. The molybdenum and tungsten complexes are even capable of an efficient cyclopropanation of a trisubstituted olefin [66]. Incorporation of an electron-withdrawing group, either as an alkene substituent [Eq. (28); compare R =Me, COzMe] 1611, or as part of the tether [67], is also beneficial: Fischer carbene complexes readily cyclopropanate electron-deficient alkenes [68].
E C R E = C02Me
Me'foMe
*
Cr(C0)5 benzene 80"C, 2 h, N2
11 10
R = H, 69% R = Me, 2% R = C02Me, 64%
(28)
In the partially intramolecular version of this reaction, an electron-deficient alkene reacts efficiently with an alkyne tethered to a molybdenum carbene complex, leading to non-fused bicyclic cyclopropane ring systems [lla, 691. When the alkyne is tethered to the carbene carbon through an all-carbon chain the product is a vinylcyclopropane, while tethering through the carbene oxygen substituent yields cyclopropyldihydrofurans or pyrans, depending on tether length [Eq. (29)]. The fully intramolecular version of this reaction (i. e., with a tethered enyne) may also be accomplished through an all-carbon tether [61], or through the carbene oxygen [70]. This results, in either case, in a tricyclic ring system, with an additional ring now fused to the cyclopropane ring, as indicated in Eq. (30) for an all-carbon tether.
Me
0
+
6 0 2 M e
x(c0)5
I
eoZM
benzene 60"C,2h 59%
(29)
Me
transxis, 2:l
benzene
An interesting expansion of this methodology uses reactions of Fischer carbenes with dienynes to produce bicyclic and tricyclic cycloheptadienes, through cis-divinylcyclopropane intermediates. In the intermolecular version, a molybdenum alkylalkoxycarbene reacts with a substituted dienyne to produce the cis-divinylcyclopropane intermediate which, under the reaction conditions, undergoes a [3,3]-sigmatropic rearrangement to produce a mixture of bicyclo[5.3.0]decanes [Eq. (31)] [71]. The electron-withdrawing nature of the carbethoxy
5.2 Cycloadditions of Acetylenes with Fischer Carbenes
153
substituent on the diene appears to play a dual role: as an electron-deficient alkene to trap the carbene generated in situ, and as the “acceptor” part of a “donor-acceptor” relationship favoring the [3,3]-sigmatropic rearrangement. A fully intramolecular process (i. e., with a tethered dienyne) produces a tricyclic cycloheptadiene ring system as a single diastereomer 1701.
5.2.7 Heterocyclic Ring Systems An increasing variety of heterocyclic products are becoming available through the reaction of Fischer carbenes with alkynes, including butenolides [64, 651, pyrazolopyridine quinones [33 b], lactams [72], pyridines [73], phosphahydroquinones and oxaphospholes [74], among others. As most of the synthetic routes have been developed relatively recently, the area suffers generally from a lack of the systematic studies necessary to make these ring systems broadly available through rational syntheses. Even so, furan synthesis has received a fair amount of attention, in part because furans were among the first side-products isolated from the Ddtz reaction [4]. In fact, their occurrence is quite common, albeit in very low yields [8a, 9, 31 b, 44, 59, 751. Although poor selectivity generally prevents this reaction from being a useful approach to furan synthesis [76], understanding its mechanistic course is relevant to increasing its efficiency. Labeling studies have clarified the origin of the ring atoms in this unusual rearrangement [Eq. (32)] [44], although its mechanistic details have yet to be worked out. The metallacyclobutene intermediate often invoked in the Ddtz mechanism (2, Scheme 5-1) has been proposed as the branch point, the course being governed by the stereochemistry of the electrocycling ring opening [44]. As previously related, however, the existence of such an intermediate is under debate [lo, 771. B
9 ,OMe
F2 OGC-c~r(c0)~ A
~
R’CECH 3 4
-
1
Iron and cobalt carbene complexes are capable of much more selective furan synthesis. Reactions of cobalt methoxycarbenes with internal alkynes give good to excellent yields of 2-methoxyfurans and have been applied to a synthesis of a natural product, bovolide [75b]. Both terminal and internal alkynes are viable substrates in the preparation, of 2-aminofurans from iron (dimethy1amino)carbenes [78], although the nature of the rearrangement of the heteroatom-containing substituent is also as yet unclear [Eq. (33)]. The inclusion of elevated carbon monoxide pressure can divert this reaction to the production of pyrones [75c, 791. Of the nitrogen heterocycles that have been observed as the product of Fischer carbene/alkyne reactions, pyrroles are the most readily obtained, with at least two efficient,
154
5 OrganometaNic Cycloaddition Reactions of Acetylenes Me,
/
,Me
47%
MeCZCMe
60°C. 1 h, 40 atrn CO 47%
complementary syntheses available. In one approach, iminocarbene complexes react with internal or terminal alkynes in hexane, without carbon monoxide insertion, to regiospecifically produce the pyrrole shown in excellent yield. The observed regiochemistry, in which the alkyne carbon with the larger substituent is joined to the carbene carbon, is just the opposite of that observed for carbocyclic annulations of carbene complexes, suggestive of an alternative mechanistic route. The reaction may proceed through initial insertion of the alkyne into the carbene to form the 5-aza-l-chroma-1,3,5-hexatriene intermediate shown, which can subsequently undergo a 6 Ir-electrocyclization, followed by reductive elimination of the metal and a 1,s-hydride shift to give the observed pyrrole [Eq. (34)][73, 801.In a complementary technique, alkynylcarbene complexes undergo a Michael addition with imines to give high yields of isolable 5-aza-l-chroma-1,3,5-hexatrienes which react thermally to produce 2H-pyrroles. Disubstitution of the original imine C-atom obviates the 1,5-hydrogenshift to the pyrrole product [Eq.(35)l [80b].
5.3 The Pauson-Khand Reaction: Cycloadditions of Olefins, Acetylenes, and CO In the presence of octacarbonyldicobalt, three-component cycloaddition of acetylenes occurs with alkenes and carbon monoxide, affording cyclopentenones. Although thermal activation of the organometallic was the original method of effecting this transformation, a variety of reaction “additives” have since been explored, resulting in procedures that permit the
5.3 The Pauson-Khand Reaction: Cycloadditions of Olefins, Acetylenes, and CO
155
Pauson-Khand process to occur at room temperature or below. Similarly, although most examples use stoichiometric cobalt, important progress has been made toward rendering this reaction catalytic in the metal. In addition, a large body of information exists to enable the chemist to apply the Pauson-Khand to synthetic problems with good confidence as to the outcome, both in terms of overall yield as well as stereo- and regioselectivity [81]. Several other transition metal processes give cyclopentenones from similar substrates. Although none of these has yet been shown to possess the scope of the Pauson-Khand, many have advantages that make them valuable alternatives. The interested reader is directed to the literature citations for further information [82-881.
5.3.1 Background and Mechanism Treatment of the yellow- brown octacarbonyldicobalt, CO,(CO)~, with an acetylene in ethereal or hydrocarbon solvent at room temperature under an inert atmosphere generates over several hours' time the red-violet hexacarbonyldicobalt-alkyne complex, C O ~ ( C Oal)~kyne. These species may be isolated and purified under inert atmosphere conditions at room temperature and readily characterized by the usual spectroscopic techniques [89]. Reaction of this complex with an olefin occurs at temperatures ranging from about 60 to 120°C over a period of hours to days; Eq. (36)depicts one of the best examples [l, 901.
+
Me3SiC=CHCo2(CO)6
isooctane, 80-9OoC, 18 h 93%
The olefin is typically in competition against reactions involving additional molecules of acetylene, which lead to a variety of side-products including benzenes. Strained cyclic olefins are the best Pauson-Khand substrates; simple acyclic examples fare less well, as do sterically hindered or less strained cyclic alkenes. Intramolecular versions of the reaction are less susceptible to such difficulties; indeed, the intramolecular Pauson-Khand reaction of derivatives of 1,6-heptenyne to give bicyclo[3.3.O]oct-l-en-3-ones(e. g., Eq. (37) [91]) has been the single most popular application of the reaction in natural product synthesis. Electronic effects also play a role: electron-deficient alkenes usually give acyclic dienes instead of cyclopentenones as products; thus the enones formed in a Pauson-Khand process are generally unreactive toward further cycloaddition [92]. Finally, the Pauson-Khand reaction possesses remarkable functional group tolerance in either the olefin or acetylenic substrate. Compatible groups include alcohols, ethers, thioethers, tertiary amines, amides, esters, ketones, aromatic and heteroaromatic rings, and less reactive unsaturated functions.
MopC O ~ ( C O )heptane, ~, 90°C, 20 h
-
SiMe3
0 (37)
68%
The mechanistic picture developed for the Pauson-Khand reaction is the result of analyses by many groups of, mostly, product selectivities in widely varied systems. Direct evidence is unavailable, because, aside from the C O ~ ( C O alkyne )~. complexes themselves, few observa-
5 Organometallic Cycloaddition Reactions of Acetylenes
156
tions of intermediates on the direct pathway from starting materials to products have been reported [931, Nevertheless, the inferred mechanism has been shown to have excellent predictive value and is generally accepted as a useful working model. Scheme 5-2 shows its essential steps [91, 941.
-
CO,(CO),
-2 c o
+ ?O(CO),
13
yco), --tif+-co y 3 3 3
Co(CO), 12
>----<-
CO(C0)Z
+co Alkene
insertion 14
(‘g (CokicP
Reductive elimination
Decomplexation
- [CO~(CO)~I
Scheme 5-2 Proposed general mechanism for the Pauson-Khand reaction.
Subsequent to formation of Co2(CO),*alkyne complex 12, loss of CO and complexation of the alkene is assumed to occur, giving 13. The first C - C bond-forming step, insertion of the alkene into a formal Co - C bond in 13, then takes place, leading to metallacycle 14. Then follows migratory insertion of CO, and reductive elimination of one metal, followed by decomplexation of the other. The regio- and stereochemical preferences shown in Eq. (36) comprise part of the evidence supporting this Scheme: bicyclic alkenes almost always cyclize on the ex0 face, and terminal acetylenes always react to give products in which the (larger) alkyne substituent ends up at C2 of the cyclopentenone product. These observations may be rationalized as consequences of insertion of the less hindered face of the alkene toward the alkyne carbon with the less bulky substituent (13-14). The steric course of the intramolecular cycloadditions may be rationalized similarly: a strong preference for allylic and propargylic substituents in the original enyne to end up with ex0 rather than endo orientation in the product enone [Eq. (37)] is consistent again with a mode of alkene insertion (13 --t 14) that minimizes steric interference. With few exceptions, this Scheme provides a basis for reliable prediction of favored products. Note that successful Pauson-Khand reaction requires both alkene complexation (12 13) and insertion (13 --t 14) processes to occur. The first requires loss of a CO ligand, achieved either thermally or chemically. Insertion, on the other hand, is promoted by ligands and, as a result, much experimentation has focused on the addition of various ligands. We will address these issues in context. The last step releases a dicobalt fragment, which presumably is not free, coordinatively unsaturated Co2(CO), but contains ligands (e g., solvent molecules) occupying vacant sites on each metal. This fragment is, in principle, capable of complexation by another molecule of acetylene, regenerating 12, implying that the reaction is potentially catalytic in the metal. Achieving a catalytic Pauson-Khand under practical conditions has been an elusive goal, but protocols are now available for both catalytic inter- and intramolecular cycloadditions for certain limited sets of substrates.
-
157
5.3 The Pauson-Khand Reaction: Cycloadditions of Olefins, Acetylenes, and CO
5.3.2 Intermolecular Pauson-Khand Reaction Virtually any acetylene may be successfully employed in an intermolecular Pauson-Khand reaction, although special solvents or promoters are sometimes necessary (e. g., cycloadditions of conjugated alkynones must be carried out in CH,CN [95]). Early work showcased the synthetic utility of thermal cycloadditions of double bonds in four- and five-membered rings. Reaction of the cyclobutene moiety in bicyclo[3.2.0]heptd-ene proceeds with complete regio- and stereoselectivity and constitutes a direct entry into the tricycl0[5.3.0.0~~~]decenone ring system. Eq. (38) depicts a case in which the substrate functionality sets up the product for Grob-type fragmentations to hydrazulene ring systems capable of serving as synthetic precursors to either guaianolide or pseudoguaianolide natural products [96]. A similar example has been employed in an efficient synthesis of the natural product spatol [97]. A third, based on cycloaddition of a bridged 8-oxabicyclo[3.2.1]oct-6-ene, was employed in two syntheses of furanether B [Eq. (39)] [98].
Norbornadiene may undergo cycloaddition either once or twice; the latter process is not usually a major complication; 30-60% yields with simple acetylenes are typical and are further improved by adding a phosphine oxide [3, 991. Under thermal conditions yields of cyclopentenones are limited by side-reactions of both diene and acetylene with the cobalt reagent. Cycloaddition with 4-pentyn-1-01 is such a case [Eq. (a)]. Much better yields, at the expense of some stereoselectivity, are obtained using an amine oxide to facilitate CO loss and alkene coordination [loo]. Prior attachment of the alkynol to a Merrifield polymer is another method to suppress interfering processes and improve yields [Eq.(41)] (101).
-&
+ HO(CH2)3C~CH'Co2(CO)6
(CHd3OH
/
benzene, 80°C, 6 h, 24% CH2C12,02,Me3N0, -78 to 20% 30 m, 93%
only product 8
'
ho (40)
(CH2)3OH 1
1) benzene, 80°C. 6 h 2) KOH, II-Bu~NCI,H20, THF, 80°C. 59%48 h +
RO(CH2)3~CH*cO2(CO)6
R = 2%-crosslinked polystyren
&CH&OH
(41)
158
5 Organornetallic Cycloaddition Reactions of Acetylenes
An early use of the norbornadiene reaction in synthesis employed conjugated addition followed by retro-Diels- Alder to deliver 4,5-disubstituted cyclopentenones [Eq. (42)] [102]. More recently, this approach was adopted with considerable methodological improvement (including use of dimethylsulfoxide (DMSO) as a promoter/solvent) to a preparation of substituted cyclopentadienyl ligands for use in transition metal chemistry [Eq. (43)] [103]. 1) isooctane, 80"C, 48 h 2) Me2CuLi, Et20, ,O"C 3) 6OO0C,vac. distill 38%
1) benzene, DMSO, 50°C 2) PhMgBr, THF 3) NaH, Mel, DMF 4) K, nBuLi, THF
67%
*
T+
isomers (43)
mBu
There are many examples of successful cycloaddition of cyclopentenes, cyclopentadienes, dihydrofurans, and related mono- and polycyclic five-membered ring substrates. Several give better results in the presence of excess acetylene and catalytic Co2(CO), than they do under stoichiometric conditions [Eq. (44)] 11041.
0
benzene, cat C% GO) H C S H . GO, 65'& 48% 70% (cf. 49% with stoichiometric cobalt)
Other cycloalkenes give mixed results. Cyclopropenes, unless heavily substituted, tend to react with each other rather than with acetylenes [105]. Cyclohexenes (and larger rings) lack useful reactivity in general, although an interesting exception is the homoallylic amine derivative shown in Eq. (45)].Evidence from this and related systems suggests that prior complexation of the heteroatom to a cobalt atom in complex 12 (Scheme 5.2) facilitates complexation and, ultimately, cycloaddition of the double bond [106].
28%
W-
Simple acyclic olefins are rather poor Pauson-Khand substrates under thermal conditions. Ethylene reacts moderately well with terminal but less well with internal acetylenes [98, 1071. Usable reaction rates require forcing conditions which, fortunately, can be optimized for catalytic use of the metal [Eq. (46)] [108]. Substituted olefins give very variable results, but reveal interesting regiochemical aspects of the process. Cycloaddition of vinylcyclohexanewith phenylacetylene proceeds to a mixture of 4- and 5-cyclohexyl-2-phenyl-2-cyclopentenones in 45 Vo overall yield. Regioselectivity is total with respect to the acetylene, as expected. However, insertion of the alkene proceeds with little regiochemical preference. Evidently, in going from
5.3 The Pauson-Khand Reaction: Cycloadditions of Olefins, Acetylenes, and CO
159
13 to 14 (Scheme 5.2), the steric demands of bond formation to the terminal acetylenic carbon and to the Co(CO), moiety are quite similar. In contrast, reaction of the same alkene with I-phenylpropyne, an internal acetylene, while proceeding in lower yield, is completely regioselective [Eq. (47)]. In this case, C - C bond formation between two substituted carbons is clearly disfavored in the 13 + 14 process [109]. toluene, cat C O ~ ( C O ) ~ . 40 bar C2H4, 100 bar CO,
n-C5Hj1CeCH
&n-Cd,i
150°C, 16 h 47.49%
(46)
Ally1 alcohol is a substrate for which the usual thermal reaction conditions fail [110]. Cycloaddition with phenylacetylene may be induced by adding either amine oxides or DMSO, each of which favors a different regioisomeric product [Eq. (48)]. The small number of such examples limits generalization of these effects.
&.. 0
HOCH2CH-CH2
+
HOch?&Ph
+
PhC=CHCo2(CO)s
(48)
HOCH2
Me3N0,CH2CI2, pet ether, O°C, 65% DMSO, CH&, 40"C, 30%
2 1
1 2.6
Similarly to the situation shown for cyclic systems above [Eq. (45)], the presence of substituents capable of acting as ligands to cobalt on an acyclic olefin may be highly beneficial. The reaction shown in Eq. (49) is merely one of a large number of examples that explore the effects of the position, nature, and number of heteroatoms on the alkene reaction partner on the yield and regiochemistry of both the intermolecular and intramolecular Pauson-Khand reactions [93, 105, 1111. M ~ Z N ( C H ~ ) ~ C M ~ = ~C H ~~
+
P~C%HCO~(CO)~
77%
" ~
o M e 2Me N(CHAph
+
M&
(49) (CH2)2NMe2
25
1
Development of nonracemic intermolecular Pauson-Khand reactions has followed two directions. The first efforts addressed the generation of dissymmetric complexes of the form CO~(CO)~L* .alkyne, in which L* is an enantiomerically pure ligand. Using (I?)-
160
5 Organometallic Cycloaddition Reactions of Acetylenes
(+)-2,3-O-isopropylideneglycerol-l-diphenylphosphine (“Glyphos”) as the ligand and phenylacetylene as the alkyne, cycloaddition with norbornene proceeds to give the enone product optically pure [Eq. (SO)] [112]. The yield, however, is low for this combination of substrates (up to 86% has been reported in the absence of phosphine [113]), and the ligand is clearly partly responsible: the Pauson-Khand reaction shows little tolerance for phosphines in general. Also detrimental to the outcome of this process is the facile thermal interconversion of the two diastereomers of the initial Co2(CO), Glyphos salkyne complex and the consequent need to employ low reaction temperatures. However, the concept is a very promising one in that high diastereoselectivity requires only restriction of complexation of the alkene in a single orientation at one of the two diastereotopic Co-atoms. The importance of the built-in em-face selectivity associated with reaction of norbornene is illustrated by comparing the cycloaddition of 2,5-dihydrofuran: stereoselectivity is much lower [Eq. (51)] [114].The latter example illustrates another useful variant: the employment of dry-state absorption conditions (DSAC) for cycloaddition. A solution of acetylene complex and olefin is mixed with an absorbent such as silica gel, evaporated to dryness, and gently heated to effect reaction. Results often compare quite favorably to solution-phase thermal conditions [105, 1151. PhGCHCO*(C0)5’(R)-(+)-GlyphOS
(as a single, pure diastereomer) toluene, ultrasound, 45’C, 6 h 31% (product ee = 100%)
PheCHCoZ(CO)S.(R)-(+)-GlyphOS
35-41Yo (product ee = 59%)
A c h i d auxiliary-based method employing heteroatom binding shows considerable promise in nonracemic intermolecular Pauson-Khand chemistry. Scheme 5-3 depicts an acetylene synthetic equivalent bearing the 10-methylthioisoborneol moiety (15), which is capable of significant asymmetric induction upon Pauson-Khand cycloaddition. Loss of carbon monoxide from 15 may be effected thermally, or better yet, by reaction with N-methylmorpholine N-
6 eq NMO, CHZCI,, 20°C
4
A
- co
82% (product de = 92%)
-(c0)2 H ‘
1) Srnl , THF, MedH 2) MeAICI,,
16
1) Me,CuLi, Et,O, Me,SiCI 2) TBAF
H-
86%
, -20%
r\
U
‘SMe
Scheme 5-3 Enantioselective synthesis of 4-methyl-2-cyclopentenone.
65%
5.3 The Pauson-Khand Reaction: Cycloadditions of Olefns, Acetylenes, and CO
161
oxide (NMO), forming chelated complex 16, which has been characterized spectroscopically. Cycloaddition of 16 with reactive alkenes takes place with high yields and diastereoselectivities at - 20 “C. Subsequent conversion of the product by conjugate addition followed by retroDiels- Alder reaction to a simple, enantiomerically enriched, 4-substituted cyclopentenone is illustrated [116].
5.3.3 Intramolecular Pauson-Khand Reaction Tethering the “ene” and “yne” reaction partners for Pauson-Khand cycloaddition was first conceived as a way to improve simultaneously the effective reactivity of simple alkene moieties and to provide access to a wider variety of bi- and polycyclic systems. Indeed, well over 100 examples of such cycloadditions have been published, mostly involving reactions of derivatives of 1,6-heptenyne to give bicyclo[3.3.0]oct-l-en-3-ones[cf. Eq. (37)] together with a smaller number of 1,7-octenynes, which afford bicyclo[4.3.0]non-l(9)-en-8-ones181 c, 1171. The use of dry-state conditions [118] or additives such as amine oxides [119] has virtually revolutionized this field: the cycloaddition shown in Eq. (52) is completed in 15 minutes(!) when promoted by NMO. isooctane, 95”C,4d, I
(52)
35-40%
Most applications of this chemistry have utilized more highly substituted and functionalized systems, for which intramolecular cycloaddition rates and yields are significantly better than in the original prototypical examples. Only substitution at the olefinic C-atoms is generally detrimental to the success of the reaction. For example, while 46-heptenyne cyclizes to give only 31 % of bicyclo[3.3.0]oct-l-en-3-oneafter four days at 95 “C, substrates with C4 disubstitution typically proceed to product in much less time with yields at least twice as high (“gem-dialkyl effect”). Using either dry-state conditions or amine oxide promoters, 55-95Vo yields are routinely obtained after hours or in some cases even minutes. As mentioned earlier, substantial data are available regarding the stereochemical preferences of intramolecular Pauson-Khand reactions. The results in Eq. (53) 11201, together with those already shown in Eq. (37), illustrate the most general principles. A preference for bulky propargylic and allylic substituents in the enyne substrate to end up with e m orientation in the product is evident, with the degree of stereoselectivity dependent upon the size of the substituent on the acetylene terminus: the larger this group is, the greater the preference for exo product. An examination of the possible metallacyclic intermediates in each of these systems rationalizes this result: alkene insertion in such a manner as to place an allylic or propargylic group endo forces a pseudodiaxial steric interaction between that group and the substituent on the acetylenic carbon (Fig. 5-3) [91]. Whether alkene insertion is face-selective to give a ckrather than a trans-ring-fused metallacyclic intermediate, as shown, is not known. The cisfused alternative is, however, more consistent with the stereochemical results observed in these and other, similar systems.
162
5 Organometallic Cycloaddition Reactions of Acetylenes
CQ(CO)~,heptane 1 lo%, 20 h
R = Me, 65% R = Me& 79%
1 1
3.3 26
A
B
R - '
VS.
t-BuMepSiO
t-BuMe,SiO
C
major product of equation 53
(co)~ D
Figure 5-3 Stereocontrolling interactions in the intramolecular Pauson-Khand reaction.
This bicyclo[3.3.0]octenone preparation has been employed in numerous natural product syntheses. Scheme 5-4 shows an early adaption of one of the cycloadditions shown in Eq. (53) toward the synthesis of the linearly fused triquinane system in coriolin. The substitutional and stereochemical characteristics built in by the Pauson-Khand process lend themselves to a very efficient approach [120]. Angularly fused triquinanes have also been prepared by closing the third ring onto a bicyclic Pauson-Khand product [121]. 1) H , PdIC, &OH 2) NaH, DME, &Br
KOt-Bu,
3) 0 2 , PdCIz,
&o
t-BUOH
CuCI,DMF
74%
48%
t-BuMe,SiO
Me
t-BuMezSiO
H
Scheme 5-4 Pauson-Khand-based approach to coriolin.
A totally stereoselective synthesis of an optically pure carbacycline analog utilized the cycloaddition of the D-(+)-ribonolactone-derived substrate shown in Eq. (54). Initial attempts to achieve cyclization on an analogous substrate with a five-membered ring lactone in place of the seven-membered ring ketal failed, presumably due to the strain that would accompany generation of a trans ring fusion [122]. A number of reactions benefit from the addition of a phosphine oxide. Unfortunately, the effectiveness of this additive is neither general nor predictable, nor is its specific mode of action understood [99].
5.3 The Pauson-Khand Reaction: Cycloadditions of Olefins, Acetylenes, and CO
?;$
85%, n Bu3P0, 3 d heptane,-
+g;
51?o'
H
163
0
T SiMe3 c02(co)6
(54)
H SiMe3
The scope of the intramolecular Pauson-Khand has been rapidly expanded as both more complex and heteroatom-containing substrates have been employed. Cycloadditions involving a cycloalkene reaction partner afford the direct construction of tricyclic systems in a single step. Triquinacene derivatives are efficiently obtained from 3-(3-butynyl)cyclopentenes [Eq. ( 5 5 ) ] . An unusual characteristic of this system is the epimerization that occurs at the propargylic position subsequent to cobalt complexation but prior to cyclization. The steric demands on the reaction are evidently so large that one stereoisomer is unable to cyclize and, instead, isomerizes through a cobalt-stabilized propargylic cation [123]. t-BuMe2SiQ
t-BuMe2Si0,
isooctane, 160"C, 3 d
SiMe3 co*(co)6
t-BuMezSiO
0
(55)
76%
The 1-(4-pentynyl)cyclopentenesystem suffers even greater steric resistance to cycloaddition in that the alkene is trisubstituted. Reaction occurs only when the alkyne terminus is unsubstituted [118c, 1241. The process gives angularly fused triquinanes in moderate yields and has been used in the syntheses of pentalenene (Scheme 5-5) and pentalenic acid [125]. The stereochemical outcome of the reaction with respect to the allylic substituent in the substrate is again rationalized mechanistically. Cycloaddition to give the minor isomer experiences steric interference between the allylic methyl group and the propargylic methylene. In the synthesis of pentalenic acid, the latter is substituted, giving a mixture of two diastereomeric enynes. The two possible cyclization modes for one of the enynes experience offsetting steric interaction and therefore give nearly a 1 : 1 mixture of stereoisomeric products. The other substrate affords a single isomer; the approach to the metallacyclic precursor of its stereoisomer is evidently much higher in energy [Eq. (56)].
I
1
C02(CO)6
1) Me3PCH2, Me3PCH3+I-, DMSO, 60%, 5 h 0 2) pTsOH, CH2C12
Li, NH3, EtPO,MeOH M ~ ~ & 89%
8
Me H
45%
Scheme 5-5 Stereoselective synthesis of pentalenene.
M Me eM&
H
Me
164
5 Organornetallic Cycloaddition Reactions of Acetylenes
3
2
5
An approach to the natural product crinipellin B takes advantage of the cycloaddition shown in Eq. (57) [124d, el. In this case structural constraints prohibit the P-hydride elimination that normally interferes with cyclizations of olefins conjugated to electron-withdrawing groups, diverting them towards dienes. This factor also operates in the tandem process illustrated in Eq. (58), which is remarkable despite the low yield [126].
Because of their ease of preparation, ally1 propargyl ethers and amines have been extensively studied. Although their thermal Pauson-Khand reactions often fare poorly [127], a great many proceed well using both dry-state absorption [113] and amine oxide [101, 110, 119, 1281 methodologies [e. g., Eq. (5911.
w
L T
isooctane, 60"C, 1 d, 14% SiO , 0 2 , W C , 3.5 h, 58% c02(co)6 NM6,CH2CIp,25"C, 85%
The combination of Nicholas's cobalt-stabilized propargyl cation chemistry, described elsewhere in this volume (Chapter 4), and Pauson-Khand chemistry has permitted the efficient synthesis of numerous highly complex heteropolycyclics [118b]. The total synthesis of (+)-epoxydictymene in Scheme 5-6 epitomizes many of these and other features of cobaltacetylene chemistry. Reaction of the dienyne acetal with C O ~ ( C Ogives ) ~ the corresponding organometallic complex, in which the former triple bond is bent sufficiently [89] to permit Lewis-acid-catalyzed cyclization to the complexed cyclooctyne shown via the cobalt-stabilized propargyl cation. Subsequent cycloaddition of the complexed ether using an amine oxide at low temperature gives the best stereoselectivity (11 :l), while the best yield is obtained in refluxing acetonitrile (shown). Although the ultimately successful route to the target required unraveling and rebuilding of one five-membered ring to achieve the proper ring fusion stereochemistry, the overall synthesis was the first of a natural product with a rrans-fused bicyclo[3.3.0]octane ring system, and it required the use of no protecting groups [129].
5.3 The Pauson-Khand Reaction: Cycloadditions of Olefins, Acetylenes, and CO
165
1 ) Ph3P, imid, l2
2) t-BuLi, Et,O 3) K, 18-crown-6
50%
Scheme 5-6 Total synthesis of (+)-epoxydictymene combining Nicholas and Pauson-Khand chemistry.
Chiral auxiliaries render the intramolecular Pauson-Khand reaction nonracemic. Attachment of optically pure trans-2-phenylcyclohexanolto 1,6-heptenynehas been carried out both at the alkyne terminus and at the alkene terminus (with both E and Z stereochemistry). Diastereoselectivities are not as high as with the bidentate auxiliary employed in intermolecular examples (Scheme 3), but they are good enough to indicate potential synthetic utility. Linkage of the auxiliary to the substrate in the form of an acetylenic ether affords moderate levels of induction, but best results are obtained using the trans (E) vinyl ether derivative shown in Eq. (60): a diastereomeric excess (de) of 72% accompanies a chemical yield of 55%. The cis isomer gives both lower de and chemical yield. Stereoselectivity is rationalized on the basis of an extended conformation about the vinyl ether oxygen, with the phenyl ring of the auxiliary blocking one face of the olefin [130].
Conformational preferences control regioselectivity in the intermolecular Pauson-Khand reactions of alkenes [e.g., Eq.(47)]. That a conformational bias built into a substrate may affect the stereochemical outcome of an intramolecular reaction is suggested by Eq. (61), in which a symmetrical dienyne gives a 3 :1 ratio favoring the C6 endo-methyl product. Other examples indicate that the usual preference for products with exo substitution, besides being dependent on the size of the group at the alkyne terminus, depends on the relative stereochemistry of the substituted carbons between the double and triple bonds [131]. The direct steric influence of a substituent at C4 of the substrate is probably small [132] and the absence of substitution at the alkyne terminus reduces the steric interaction favoring the insertion mode to the exo product. As a result, energy differences between the substrate conformations
166
5 Organometallic Cycloaddition Reactions of Acetylenes
1
3
that lead to the possible modes of insertion may significantly affect the relative free energies of activation of the competing processes and thus the exo/endo product ratio. Finally, a most promising discovery is the finding that true catalysis of the intramolecular Pauson-Khand reaction is possible to achieve. Referring to Scheme 5-2, perhaps the single most critical issue is the generation of an intermediate which readily undergoes complexation by alkene followed by CO insertion, and which can subsequently by preserved in a sufficiently stable form to trap another acetylene moiety before thermodynamically favored modes of decomposition interfere. An apparently fairly general protocol has been developed which involves reaction of an enyne with 3 mol Vo of Co2(CO), in dimethoxyethane solution, followed by addition of 10 mol Vo to triphenylphosphite, P(OPh)3. Pressurization to 3 atm of CO followed by heating to 120°C for 24 h affords Pauson-Khand products in 51-94070 yields [e. g., Eq. (62)l [133]. These conditions evidently successfully balance several factors, including promotion of initial loss of CO (favored by heat, disfavored by CO pressure and presence of the phosphite), insertions of alkene and CO (favored by CO and phosphite), and )~ for catalytic cycling (phosphite). preservation of C O ~ ( C Ofragments
E$>c",
Co CO)8 3 mol %) PhO)3P (10mol %), D d , CO 13 atrn), d o c , 1 d
82%
Et02C&=o
* EtO2C
(62)
Acknowledgement We thank the National Institutes of Health for support of our work in organometallic chemistry.
Abbreviations DDQ DMF DMSO DSAC Glyphos LDA MOM NMO TBAF TBDMS
2,3-dichloro-5,6-dicyano-l,4-benzoquinone dimethylformamide dimethyl sulfoxide dry-state absorption conditions (I?)-( +)-2,3-O-isopropylideneglycerol-l-diphenylphosphine lithium diisopropylamide methoxymethyl N-methylmorpholine-N-oxide tetra-n-butylammonium fluoride t-butyldimethylsilyl
References
THF TIPS-OTF TMS
167
tetrahydrofuran triisopropylsilyl triflate trimethylsilyl
References [l] (a) N. E. Schore, Chem. Rev. 1988, 88, 1081 - 1119; (b) B. M. Trost, 1. Fleming (Eds.) Comprehensive Organic Synthesis, Pergamon, Oxford, 1991, Vol. 5. [2] K. H. DBtz, Angew. Chem., Int. Ed. Engl. 1975, 14, 644-645. 131 (a) I. U. Khand, G. R. Knox, P. L. Pauson, W. E. Watts, J. Chem. Soc, Chem. Commun. 1971,36 and J. Chem. Soc, Perkin Duns. I, 1973, 975-977; (b) I. U. Khand, G. R. Knox, P. L. Pauson, W. E. Watts, M. I. Foreman, J. Chem. Soc, Perkin nuns. 1. 1973, 977-981. [4] K. H. Dotz, J. Organomet. Chem. 1977, 140, 177-186. [5] (a) K. H. Dotz, H. Fischer, P. Hofmann, F. R. Dreissel, U. Schubert, K. Weiss, Transition Metal Carbene C o m p l m , Verlag Chemie, Deerfield Beach, FL, 1984; (b) K. H. Doh, Angew. Chem., Int. Ed. Engl. 1984, 23, 587-608; (c) C. P. Casey, React. Intermed. 1985, 3, 109-150; (d) K. H. Ddtz in Organornetallics in Organic Synthesis: Aspects of a Modern Interdisc@linaryField (Eds. : A. de Meijere, H. tom Dieck), Springer, Berlin, 1988, pp. 85-104, and in Advances in Metal Carbene Chemistry (Ed.: U. Schubert), Kluwer, Boston, 1989, pp. 199-210; (e) W. D. Wulff in Advances in Metal-Organic Chemistry (Ed.: L. S. Liebeskind), JAI Press, Greenwich, CT, 1989, Vol. 1, pp. 209-393, and in [Ib], pp. 1065-1113; ( f ) P. J. Harrington, Transition Metals in Total Synthesis, Wiley, New York, 1990, pp. 346-399. [6] (a) T. A. Brandvold, W. D. Wulff, A. L. Rheingold, J. Am. Chem. SOC. 1991,113, 5459-5461; (b) T. A. Bandvold, W. D. Wulff, A. L. Rheingold, ibid. 1990, 112, 1645-1647. [7] (a) K. H. Dtitz, B. FUgen-Koster, Chem. Ber. 1980, 113, 1449-1457; (b) C. P. Casey in Reactive Intermediates (Eds.: M. Jones, Jr., R. A. Moss), Wiley, New York, 1981, Vol. 2, pp. 135-174; (c) H. Fischer, J. Mulheimer, R. Markl, K. H. Dt)tz, Chem. Ber. 1982,115, 1355-1362; (d) K. H. Dotz, Pure Appl. Chem. 1983, 55, 1689-1706. [8] (a) W. D. Wulff, B. M. Bax, T. A. Brandvold, K. S. Chan, A. M. Gilbert, R. P. Hsung, Organometallics1994,13, 102-126; (b) B. A. Anderson, W. D. Wulff, J. Am. Chem. SOC. 1990, 112. 8615-8617. [9] M. E. Bos, W. D. Wulff, R. A. Miller, S. Chamberlin, T. A. Brandvold, J. Am. Chem. SOC. 1991, 113, 9293-9319. [lo] (a) P. Hofmann, M. Hammerle, Angew. Chem., Int. Ed. Engl. 1989,28, 908-910; (b) P. Hofmann, M. Hwmerle, G. Unfried, New J. Chem. 1991, IS, 769-789. [ill (a) D. F. Harvey, M. F. Brown, J. Am. Chem. SOC. 1990, 112, 7806-7807; (b) P. F. Krokowski, T. R. Hoye, D. B. Rydberg, ibid. 1988, 110, 2676-2678; (c) W. D. Wulff, R. W. Kaesler, B. A. Peterson, P. Tang, ibid. 1985, 107, 1060-1062; (d) S. 0. Tumer, J. W. Herndon, L. A. McMullen, ibid. 1992, 114, 8394-8404. I121 (a) W. D. Wulff, P. Tang, J. S. McCallum, J. Am. Chem. SOC. 1981, 103, 7677-7678; (b) W. D. Wulff, K. Chan, P. Tang, J. Org. Chem. 1984, 49, 2293-2295. [I31 K. S. Chan, G. A.Peterson, T. A. Brandvold, K. L. Faron, C. A.Challener, C. Hyldahl, W. D. Wulff, J. Organomet. Chem. 1987, 334, 9-56. 1141 J. P. A. Harrity, W. J. Kerr, Btrahedron, 1993, 49, 5565-5576. I151 (a) K. H. Dotz, R. Dietz, Chem. Ber. 1978, 111, 2517-2526; (b) W. E. Bauta, W. D. Wulff, S. F. Pavkovic, E. J. Zaluzec, J. Org, Chem. 1989, 54, 3249-3252. 1161 W. D. Wulff, P.-C. Tang, J. Am. Chem. SOC. 1984, 106, 434-436.
168
5 Organometallic Cycloaddifion Reactions of Acetylenes
I171 (a) K. H.DiStz, M. Popall, G. Miiller, J. Organomet. Chem. 1987,334, 57-75;(b) K. H.Dotz, M. Popall, ibid. 1985, 291, Cl-C4; (c) K. H. Datz, M. Popall, G. Miiller, Tetrahedron 1985, 41, 5797-5802. I181 K.A. Parker, C.A. Coburn, J. Org. Chem. 1991, 56, 1666-1668. [19] (a) M. F. Semmelhack, N. Jeong, Tetrahedron Lett. 1990,31, 605-608;(b) M. F. Semmelhack, N. Jeong, G. R. Lee, ibid. 1990,31, 609-610. [20](a) D. L. Boger, I. C. Jacobson, TetrahedronLeft. 1989,30,2037-2040;(b) D. L. Boger, I. C. Jacobson, J. Org. Chem. 1990, 55, 1919-1928;(c) D. L. Boger, 1. C. Jacobson, ibid. 1991,56,2115-2122. [21](a) K. H.Dotz, D. Neugebauer, Angew. Chem. Inf. Ed. Engl. 1978, 17, 851-852; (b) K. H.Dotz, I. Pruskil, Chem. Ber. 1978,111,2059-2063;(c) K. H.Dotz, I. Pruskil, U. Schubert, K. Ackermann, ibid. 1983, 116, 2337-2343; (d) R. C. Cambie, P. S. Rutledge, M. Tercel, P. D. Woodgate, J. Organomet. Chem. 1986,315, 171-185. [22]T. J. Katz, G. X.-Q. Yang, B. H. Rickman, T. Iwashita, 1 Am. Chem. SOC.1993, 115, 2038-2039. [23]S. Chamberlin, M. L. Waters, W. D. Wulff, J. Am. Chem. SOC. 1994, 116, 3113-3114. [24] J. King. P. Quayle, J.F. Maline, Tetrahedron Lett 1990,31, 5221-5224. [25](a) A. Yamashita, L Am. Chem. SOC. 1985,107,5823-5824;(b) A. Yamashita, A. Toy, T. A. Scahill, J. Org. Chem. 1989,54, 3625-3634. [26](a) A. Yamashita, T. A. Scahill, A. Toy, Tetrahedron Lett 1985,26, 2969-2972; (b) A. Yamashita, J. M. Timko, W. Matt, ibid. 1988, 29, 2513-2516; (c) A. Yamashita, A. Toy, ibid. 1986, 27, 3471-3474. [27]D. B. Grotjahn, E E. K. Droll, T.SchSLfer, K. Harms, K. H. Dbtz, Organometuffics1992,11, 298-310. (281 A. Yamashita, A. Toy, N. B. GhazaI, C. R. Muchmore, J. Org. Chem. 1989,54, 4481-4483. (291 K. H. DiStz, R. Dietz, Chem. Ber 1977, 110, 1555-1563. [30]M. F. Semmelhack, J. J. Bozell, Tetrahedron Lett. 1982,23, 2931-2934. [31](a) M.F.Semmelhack, L. Keller, T. Sato, E. J. Spiess, W. Wulff, J. Org. Chem. 1985,50,5566-5574; (b) M. F. Semmelhack, J. J. Bozell, L. Keller, T. Sato, E. J. Spiess, W. Wulff, A. Zask, Tetrahedron, 1985, 41, 5803-5812. I321 G.A. Peterson, E-A. Kunng, J. S. McCallum, W. D. Wulff, Tetrahedron Lett. 1987,28, 1381-1384. 1331 (a) W. D. Wulff, P.-C. Tang, K . 3 . Chan, J. S. McCallum, D. C. Yang, S . R. Gilbertson, Tetrahedron 1985,4i,5813-5832;(b) K.S. Chan, W. D. Wulff, J. Am. Chem. SOC.1986,108,5229-5236;(c) J.D. King, P. Quayle, Btrahedron Lett. 1991,32, 7759-7762. 1341 (a) W. D.Wulff, D. C. Yang, 1 Am. Chem. SOC. 1984, 106, 7565-7567;(b) K.H.Datz, W. Kuhn, J. Organomet. Chem. 1985,286,C23-C26;(c) B. A. Anderson, W. D. Wulff, T. S. Powers, S. n i b bitt, A. L. Rheingold, L Am. Chem. SOC. 1992, 114, 10784-10798. 1351 For a novel adaption of the latter, see: K. H. Dtitz, M. Popall, G. Miiller, K. Ackermann, Angew. Chem., Int. Ed. Engl. 1986,25, 911-913. [36]Aldol-type syntheses have been Limited in scope until recently: (a) C. P. Casey, R. A. Boggs, R. L. Anderson, J. Am. Chem. SOC.1972,94,8947-8949;(b) C.P. Casey, W. R. Brunsvold, J. Organornet. Chem. 1974, 77, 345-352; (c) C.P. Casey, W. R. Brunsvold, ibid. 1975, 102, 175-183;(d) W. D. Wulff, S. R. Gilbertson, J. Am. Chem. SOC. 1985, 107, 503-505. [37]Reviews: (a) M.E Semmelhack in Comprehensive Organic Synthesis (Eds.: B. M. nost, I. Fleming), Pergamon, New York, 1991, Vol. 4, pp. 517-549; (b) E. P. Kiindig, Pure Appl. Chem. 1985,57, 1855- 1864. (381 S. Chamberlin, W. D. Wulff. J. Am. Chem. SOC.1992, 114, 10667-10669. [39](a) S. Chamberlin, W. D. Wulff, B. Bax, Tetrahedron 1993,49, 5531-5547; (b) S. Chamberlin, W. D. WuLff, J. Org.Chem. 1994, 59, 3047-3054. [40]For a metal carbyne-mediated phenol synthesis, see: T. M. Sivavec, T. J. Katz, TetrahedronLeft. 1985, 26, 2159-2162. [41] W.D. Wulff, Y.-C. Xu, Btrahedron Lett. 1988, 29, 415-418. [42]J. Bao, V. Dragisich, S. Wenglowsky, W. D. Wulff, J. Am. Chem. SOC. 1991, 113, 9873-9878.
References
169
(431 For an alternative mechanism, see: A. Yamashita, Tetrahedron Lett. 1986, 27, 5915-5918. (441 J. S. McCallum, F.-A. Kunng, S. R. Gilbertson, W. D. Wulff, Organometallics, 1988, 7, 2346-2360. [45] See [431. [46] For an interesting case of o,o'-disubstitution in which aromatization to the indene product occurs by alkyl migration, see [21 d]. [47] (a) H. C. Foley, L. M. Strubinger, T. S. Targos, G. L. Geoffroy, J. Am. Chem. Soc. 1983, 105, 3064-3073; (b) N. H. Huy, P. Lefloch, J. Organomet. Chem. 1988, 344, 303-311; (c) K. H . Dbtz, H. Larbig, J Organomet. Chem. 1991,405, C38-C42; (d) K. H. Dtitz, H. Larbig, Bull. Soc. Chim. FK 1992, 129, 579-584. [48] Formation of a five-membered heterocyclic ring from a nitrogen-substituted alkenyl complex was known: M. Duetsch, R. Lackmann, F. Stein, A. de Meijere, Synfett 1991, 324-326. [49] K. H. Dotz, R. Dietz, 1 Organomet. Chem. 1978, 157, C55-C57. [50] K. H. DStz, W. Sturm, J Organomet. Chem. 1986, 310, C22-C24. [51] (a) J. W. Herndon, S. U. Timer, W. F. K. Schnatter, . I Am. Chem. SOC.1988, 110, 3334-3335; (b) S. U. 'hmer, J. W. Herndon, L. A. McMullen, ibid. 1992, 114, 8394-8404. [52] These cyclopentatienones may be isolated through the use of bulky alkynes: J. W. Herndon, S. U. Turner, Tetrahedron Lett. 1989, 34 295-296. [53] J. W. Herndon, J. J. Matasi, J. Org. Chem. 1990, 55, 786-788. [54] C. A. Challener, W. D. Wulff, B. A. Anderson, S. Chamberlin, K. L. Faron, 0. K. Kim, C. K. Murray, Y.-C. Xu, D. C. Yang, S. D. Darling, . l Am. Chem. SOC. 1993, 115, 1359-1376. [55] (a) J. W. Herndon, G. Chatterjee, P. P. Patel, J. J. Matasi, S. U. lkmer, J. J. Harp, M. D. Reid, 1 Am. Chem. SOC. 1991, 113, 7808-7809; (b) J. W. Herndon, M. Zora, Synlett 1993, 363-364; (c) J. W. Herndon, M. Zora, P. P. Patel, G. Chatterjee, J. J. Matasi, S. U. lkmer, Tetrahedron 1993, 49, 5507-5530. [56] D. F. Harvey, M. F. Brown, Tetrahedron Lett. 1990, 31, 2529-2532. [57] For an intramolecular example in which molybdenum and tungsten give similar yields of the same nonconjugated bicycloheptadienone, see [55 c]. [58] (a) E. 0. Fischer, K. D&z, Chem. Ber. 1970, 103, 1273-1278; (b) K. H. Dbtz, E. 0. Fischer, ibid. 1912, 105, 1356-1367. [59] W. D. Wulff, R. W. Kaesler, Organometallics 1985, 4, 1461-1463. [60]T.J. Katz, T. M. Sivavic, J. Am. Chem SOC.1985, 107, 737-738. 1611 P. 6 Korkowski, T. R. Hoye, D. B. Rydberg, J Am. Chem. SOC. 1988, 110, 2676-2678. [62J For example by thermolysis of dry-state-adsorbed enynes on silica gel: T.J. Katz, G. X.-Q. Yang, Tetrahedron Lett 1991, 32, 5895-5898. [63] T. R. Hoye, G. M. Rehberg, Organometallics 1989, 8, 2070-2071. [64] T. R. Hoye, G. M. Rehberg, J Am. Chem. SOC. 1990, 112, 2841-2842. 1651 Manganese Fischer carbenes are also effective: T. R. Hoye, G. M. Rehberg, Organometallics 1990, 9, 3014-3015. [66] T. R. Hoye, J. A, Suriano, Organometallics 1992, 11, 2044-2050. [67] (a) D.F. Harvey, K. P.Lund, D. A. Neil, Tetrahedron Lett. 1991, 32, 6311-6314; (b) D. F. Harvey, K. P. Lund, D. A. Neil, L Am. Chem. SOC.1992, 114, 8424-8434. [681 With electron-rich alkenes, either cyclopropanes or metathesis products form by a different mechanism: C. P. Casey, A. J. Shusterman, Organometallics 1985, 4, 736-744. 1691 D. F. Harvey, M. F. Brown, Tetrahedron Left. 1991, 32, 5223-5226. 1701 D. F. Harvey, M. F. Brown, J 0%.Chem. 1992, 57. 5559-5561. [711 D. F. Harvey, K. P. Lund, J. Am. Chem. SOC. 1991, 113, 5066-5068. 1721 (a) K. H. Detz, R. Weber, Chem. Ber. 1991, 124, 1635-1638; (b) E. Chelain, R. Goumont, L. Hamon, A. Parlier, M. Rudler, H. Rudler, J.-C. Daran, J. Vaissermann, J. Am. Chem. SOC.1992, 114, 8088-8098; (c) E. Chelain, A. Parlier, M. Audouin, H. Rudler, J . 4 . Daran, J. Vaissermann, ibid. 1993, 115, 10568-10580.
170
5 Organometallic Cycloaddition Reactions of Acetylenes
[73]V. Dragisich, C. K. Murray, B. P. Warner, W. D. Wulff, D. C. Yang, . I Am. Chem. Soc. 1990,112, 1251-1253. [74]From phosphaalkynes: K. H. Datz, A. Tiriliomis, K. Harms, Tetrahedron 1993, 49, 5577-5597. [75](a) K. H. Dbtz, R. Dietz, D. Neugebauer, Chem. Ber. 1979,112, 1486-1490;(b) W. D.Wulff, S. R. Gilbertson, J. P. Springer, J. Am. Chem. Soc. 1986,108, 520-522;(c) M. F. Semmelhack, J. Park, Organornetallics, 1986,5, 2550-2552. [76]Molybdenum and tungsten give better selectivity for furans: see [8a] and [47d]. [77]For evidence in favor of a metallacyclobutene intermediate, see: J. M. O’Connor, H.-L. Ji, A. L. Rheingold, J. Am. Chem. Soc. 1993, 115, 9846-9847. [78]Furans are also available from iron alkoxycarbenes: A.4. Rehmann, W. F. K. Schnatter, N. Manolache, J. Am. Chem. SOC. 1993, 115, 9848-9849. [79]For iron-complexed pyrones from iron alkoxycarbenes, see: M. F. Semmelhack, R. Tamura, W. Schnatter, J. Springer, J. Am. Chem. Soc. 1984,106, 5363-5364. [80]For mechanistic discussions see [73]and: (a) R.Aumann, H. Heinen, C. Kriiger, P. Betz, Chem. Ber: 1990, 123, 599-604;(b) F, Funke, M. Duetsch, F. Stein, M. Noltemeyer, A. de Meijere, ibid. 1994, 127, 911-920. 1811 Reviews: (a) P. L. Pauson, I. U. Khand, Ann. N. Z Acad. Sci. 1977, 295, 2-14; (b) P. L. Pauson, Tetrahedron 1985,41,5855-5860 and in Organometallicsin Organic Synthesis. Aspects of a Modern Interdisciplinary Field (Eds.: A. de Meijere, H. tom Dieck), Springer, Berlin, 1988, pp. 233-246; (c) N. E. Schore, Org. React. 1991, 40, 1-90 and in [lb], pp. 1037-1064. (821 Zr-based: (a) E. Negishi, S. J. Holmes, J. M. Tour, J. A. Miller, J. Am. Chem. Soc. 1985, 107, 2568-2569;(b) E. Negishi, Acc. Chem. Res. 1987, 20, 65-71 and in [lb], pp. 1163-1184. 1831 Fe-based: (a) R. Aumann, H. J. Weidenhaupt, Chem. Ber. 1987, 120, 23-27; (b) A. J. Pearson, R. A. Dubbert, J. Chem. Soc., Chem. Commun. 1991, 202-203 and Organometallics 1994, 13, 1656-1661. (841 Ni-based: K. Tamao, K. Kobayashi, Y. Ito, 1 Am. Chem. Soc. 1988,110, 1286-1288 and Synlett 1992, 539-546. [85] Ti-based: (a) R. Grossman, S. L. Buchwald, J. Org. Chem. 1992, 57, 5803-5805;(b) S. C. Berk, R. B. Grossman, S. L. Buchwald, 1 Am. Chem. SOC. 1993, 115, 4912-4913 (catalytic). [86]Pd-based: N. C. Ihle, C. H. Heathcock, 1 Org. Chem. 1993,58, 560-563 (catalytic). [87]Mo-based: N. Jeong, S. J. Lee, B. Y. Lee, Y. K. Chung, Tetrahedron Lett. 1993,34, 4027-4030. [88] W-based: T. R. Hoye, J. A. Suriano, J. Am. Chem. Soc. 1993, 115, 1154-1156. [89]R. S. Dickson, P. J. Fraser, Adv. Organomet. Chem. 1974,12, 323-377. [90]A. de Meijere, A. Kaufmann, R. Lackmann, H.-C. Militzer, 0. Reiser, S. Schomenauer, A. Weier, in Organornetalks in Organic Synthesis 2 (Eds.: H. Werner, G. Erker), Springer, Berlin, 1989, pp. 255-276. [91]P. Magnus, L. M. PrincipG Tetrahedron Lett. 1985, 26, 4851-4854. [92]I. U. Khand, P. L. Pauson, 1 Chem. Soc, Chem. Commun. 1974, 319. [93]M. E. Krafft, I. L. Scott. R. H. Romero, S. Feibelmann, C. E. Van Pelt, J. Am. Chem. SOC.1993, 115, 7199-1207. [94]B. E. La Belle, M. J. Knudsen, M. M. Olmstead, H. Hope, M. D. Yanuck, N. E. Schore, J. Org. Chem. 1985, 50, 5215-5222. [95]T. R. Hoye, J. A. Suriano, 1 Org. Chem. 1993, 58, 1659-1660. [96]V. Sampath, E. C. Lund, M. J. Knudsen, M. M. Olmstead, N. E. Schore, J Org. Chem. 1987,52, 3595-3603. 1971 W. 0. Dauben, B. A. Kowalczyk, Tetrahedron Lett. 1990,31, 635-638. [98]M. E. Price, N. E. Schore, Tetrahedron Lett. 1989, 30, 5865-5868 and J. 0%.Chem. 1989, 54, 5662-5667. [99]D.C. Billington, I. M. Helps, P. L. Pauson, W. Thomson, D. Willison, J. Oganomet. Chem. 1988, 354, 233-242.
References
171
[lo01 N. Jeong, Y. K. Chung, B. Y. Lee, S. H. Lee, S.-E. Yoo, Synlett 1991, 204-206. [loll N. E. Schore, S. D. Najdi, J. Am. Chem. SOC. 1990, 112, 441-442. [lo21 N. E. Schore, Synth. Commun. 1979, 9, 41-47. [lo31 B. Y. Lee, H. Moon, Y. K. Chung, N. Jeong, X Am. Chem. SOC. 1994, 116, 2163-2164. [lo41 D. C. Billington, W. J. Kerr, P. L. Pauson, C. F. Farnocchi, J. Organomet. Chem. 1988, 356, 213-219. [lo51 S. L. Kireev, V. A. Smit, B. I. Ugrak, 0. M. Nefedov, Izv. Akad. Nauk SSSR, Ser. Khim. 1991, 2565-2571. [lo61 M. E. Krafft, X Am. Chem. SOC. 1988, 110, 968-970. [lo71 (a) I. U. Khand, P. L. Pauson, X Chem. Res. Miniprint, 1977, 168-187; (b) D. C. Billington, P. Bladon, I. M. Helps, P. L. Pauson, W. Thomson, D. Willison, ibid. 1988, 2601-2622. [lo81 V. Rautenstrauch, P. Mkgard, J. Conesa, W. Kiister, Angew. Chem., Int. Ed. Engl. 1990, 29, 1413-1416. [lo91 M. E. Krafft, Tetrahedron Lett. 1988, 29, 999-1002. [110] Y. K. Chung, B. Y. Lee, N. Jeong, M. Hudecek, P. L. Pauson, Organometallics1993, 12, 220-223. [lll] (a) M. E. Krafft, C. A. Juliano, I. L. Scott, C. Wright, M. D. McEachin, X Am. Chem. SOC.1991, 113, 1693-1703; (b) M. E. Krafft, C. A. Juliano, J. Org.Chem. 1992, 57, 5106-5115. [112] P. Bladon, P. L. Pauson, H. Brunner, R. Eder, X Organomet. Chem. 1988,355, 449-454. [113] W. A. Smit, S. L. Kireev, 0. M. Nefedov, V. A. Tarasov, Tetrahedron Lett. 1989, 30, 4021-4024. [I141 H. Brunner, A. Niedernhuber, Tetrahedron: Asymmetry 1990, I, 711 -714. [115] (a) V. A. Smit, S. L. Kireev, 0. M. Nefedov, Izv. Akad. Nauk SSSR, Ser. Khim. 1987, 2637; (b) V. A. Smit, V. A. Tarasov, E. D. Daeva, I. I. Ibragimov, ibid. 1987, 2870-2871; (c) S. L. Kireev, V. A. Smit, 0.M. Nefedov, ibid. 1989, 2061-2068. [116] X. Verdaguer, A. Moyano, M. Perids, A. Riera, V. Bernardes, A. E. Greene, A. Alvarez-Larena, J. F. Piniella, J. Am. Chem. SOC. 1994, 116, 2153-2154. [117] N. E. Schore, M. C. Croudace, 1 Org. Chem. 1981, 46, 5436-5438. [I181 (a) W. A. Smit, A. S. Gybin, S. 0. Simonyan, A. S. Shashkov, V. A. Tarasov, I. I. Ibrigamov, Izv. Akad. Nauk SSSR, Ser. Khim 1985, 2650-2651; (b) W. A. Smit, A. S. Gybin, A. S. Shashkov, Y. T. Strychkov, L.G. Kyz’mina, G. S. Mikaelian, R. Caple, E. D. Swanson, Tetrahedron Lett. 1986, 27, 1241-1244; (c) V. A. Smit, A. S. Gybin, S. 0. Veretenov, A. S. Shashkov, Izv. Akad. NaukSSSR, Set Khim. 1987, 232-233; (d) W. A. Smit, S. 0. Simonian, V. A. Tarasov, A. S. Shashkov, S. S. Mamyan, A. S. Gybin, I. I. Ibrigamov, ibid. 1988, 2796-2802. [119] S. Shambayati, W. E. Crowe, S. L. Schreiber, Tetrahedron Lett. 1990, 31, 5289-5292. [I201 C. Exon, P. Magnus, 1 Am. Chem. SOC. 1983, 105, 2477-2478. [121] (a) D. H. Hua, X Am. Chem. SOC.1986,108, 3835-3837; (b) D. H. Hua, M.J. Coulter, 1. Badejo, Btrahedron Lett 1987, 28, 5465-5468. [122] P. Magnus, D. P. Becker, J. Am. Chem. SOC. 1987, 109, 7495-7498. [I231 C. Almansa, E. Carceller, E. Garcia, F. Serratosa, Synth. Commun. 1988, 18, 1079-1089. See also (a) E. Carceller, V. Centellas, A. Moyano, M. A. Pericas, F. Serratosa, Tetrahedron Lett. 1985, 26, 2475-2476; (b) E. Carceller, M. L. Garcia, A. Moyano, M. A. Pericas, F. Serratosa, Tetrahedron 1986, 42, 1831-1839; (c) C. Almansa, E. Carceller, E. Garcia, A. Torrents, F. Serratosa, Synth. Commun. 1988, 18, 381-390; (d) C. Almansa, A. Moyano, F. Serratosa, Tetrahedron 1988, 44, 2657-2662. 11241 (a) M. J. Knudsen, N. E. Schore, X Org. Chem. 1984, 49, 5025-5026; (b) N. E. Schore, M. J. Knudsen, J. Org. Chem. 1987, 52, 569-580; (c) A. L. Veretenov, W. A. Smit, L. G. Vorontsova, M. G. Kurella, R. Caple, A. S. Gybin, Tetrahedron Lett. 1991, 32, 2109-2112; (d) A. S. Gybin, W. A. Smit, R. Caple, A. L. Veretenov, A. S. Shashkov, L. G. Vorontsova, M. G. Kurella. V. S. Chertkov, A. A. Carapetyan, A. Y. Kosnikov, M. S. Alexanyan, S. V. Lindeman, V. N. Panov. A. V. Maleev, Y. T. Struchkov, S. M. Sharpe, X Am. Chem. SOC. 1992, 114, 5555-5566.
172
5 Organometallic Cycloaddition Reactions of Acetylenes
[125] (a) N. E. Schore, E. G. Rowley, J. Am. Chem. SOC. 1988, 110, 5224-5225; (b) E. G. Rowley, N. E. Schore, J. Organomet. Chem. 1991,413, C5-C9; (c) E. G. Rowley, N. E. Schore, J. Organomet. Chem. 1992, 57, 6853-6861. [126] A. van der Waals, R. Keese, J. Chem. Soc, Chem. Commun. 1992, 570. See also: V. A. Smit, S. M. Bukhanyuk, A. S. Shashkov, Y. T. Struchkov, A. I. Yanovskii, Izv. Akad. Nauk SSSR, Ser. Khim. 1988, 2876-2877. 11271 (a) D. C. Billington, D. Willison, Tetrahedron Lett. 1984, 25, 4041-4044; (b) S. W. Brown, P. L. Pauson, L. Chem. Soc, Perkin Trans. I 1990, 1205-1209. [128] (a) N. Jeong, S. Yoo, S. J. Lee, S. H. Lee, Y. K. Chung, Tetrahedron Lett. 1991, 32, 2137-2140; (b) M. E. Krafft, R. H. Romero, I. L. Scott, J. Org. Chem. 1992, 57, 5277-5278; (c) N. Jeong, B. Y. Lee, S. M. Lee, Y. K. Chung, S.-G. Lee, Tetrahedron Letf. 1993, 34, 4023-4026. (1291 T. F. Jamison, S. Shambayati, W. E. Crowe, S. L. Schreiber, J. Am. Chem. SOC. 1994, 116,
5505 - 5506. 11301 (a) J. Castro, H. Sirrensen, A. Riera, C. Morin, A. Moyano, M. A. F’erichs. A. E. Greene, J. Am. Chem. SOC. 1990, 112, 9388-9389; (b) M. Poch, E. Valenti, A. Moyano, M. A. Pericas, J. Castro, A. DeNicola, A. E. Greene, Tetrahedron Lett. 1990, 31, 7505-7508. 11311 (a) W. R. Roush, J. C. Park, Tetrahedron Lett. 1991, 32, 6285-6288; (b) J. A. Casalnuovo, R. W. Scott, E. A. Harwood, N. E. Schore, ibid. 1994, 35, 1153-1156. (1321 P. Magnus, C. Exon, P. Albaugh-Robertson, Tefrahedron1985, 41, 5861 -5869. [133) N. Jeong, S. H. Hwang, Y. Lee, Y. K. Chung, J. Am. Chem. SOC. 1994, 116, 3159-3160.
6 Phosphaalkynes - Starting Point for the Synthesis of Phosphorus- Carbon Cage Compounds’ Manfred Regitz, A . Hoffrnann, U BergstraJer
6.1 Introduction Cage compounds have always been a subject of interest and fascination for chemists [2, 31. The main reasons for this are probably the symmetry characteristics of esthetic molecules such as cubane (1) [4], prismane (2) [ 5 ] , or pentaprismane (3) [a] (Scheme 6-1). In addition, the unusual reactivities of these polycyclic carbon systems have helped to promote the interest and actuality of such compounds over several decades. Finally, these unusual systems have often caused chemists to reconsider the conventional concepts of bonding in regard to atomic separations and bond angles and to develop new theories.
@ 1
3
2
Lq\p
P
b-P’
cp’” 4
\
cp*
5
Scheme 6-1 AU-carbon and all-phosphorus cage compounds. Cp*, pentamethylcyclopentadienyl.
If the CH or CR increments in the carbon cages are substituted by valency isoelectronic phosphorus(II1) moieties, we enter the field of polycyclic phosphanes - an area that has experienced a tremendous development in the past few years. It may now be safely assumed that, second to carbon, phosphorus is the element with the most pronounced ability to form element-element bonds [7]. From the enormous range of fascinating phosphorus cage structures, we need only mention the polycyclic systems 4 [S] and 5 [9] as typical examples. The common link between the two areas of research mentioned above (carbon cages and polycyclic phosphanes) is provided by the phosphorus-carbon cage compounds - the topic of the present chapter. In accord with the philosophy of this book, we will concentrate mainly on the results from our own laboratories and outline their position in the wider perspectives Organophosphorus Compounds, Part 98. For Part 97, see [l].
174
6 Phosphaalkynes
of modern acetylene chemistry. In all cases, starting points for the construction of phosphorus-carbon cage systems are the phosphaalkynes. The synthesis o f these species by 0elimination of hexamethyldisiloxane and their major reaction types will also be discussed.
6.2 Synthesis of Phosphaalkynes [lo] All of the phosphaalkynes 9a-g mentioned in this chapter carry more or less sterically demanding substituents and are thus kinetically stabilized. They are formally prepared by condensation reactions of tris(trimethylsily1)phosphane (7) with the acyl chlorides 6a-g and subsequent cleavage of chlorotrimethylsilane and hexamethyldisiloxane (Scheme 6-2). This reaction was first used by Becker and co-workers for the synthesis of tert-butylphosphaacetylene (9a) [ll]. Later, the process was generalized and optimized with regard to the decisive elimination step 10 -+ 9 in our laboratory [12-141.
7
6
8
'SiMq
L1.31SiMe3-
Me3SiO\ 110-150
R-EP
OC,
NaOH
C=P- S i Me3
- (Me3SilzO
/
R
9
6.8-10
I
a
10
b
C
d
e
g
f
a
Me R
tBu
CMe2Et
iPr
CH2-tBu
I-Ad
M e D
Scheme 6-2 Synthesis of phosphaalkynes 9 by p-elimination from phosphaalkenes. Ad, adamantyl.
From the mechanistic point of view, initial cleavage of chlorotrimethylsilane occurs to furnish the acylphosphanes 8a-g, which can only be detected by "P-NMR spectroscopy in some individual cases [13, 141. In general, a rapid [If]-trimethylsilyl shift follows to furnish a cidtmns-isomeric mixture of the phosphaalkenes 10, which profit from the large bonding energy between oxygen and silicon. The final 0-elimination occurs at 110-150°C in the absence of a solvent but in the presence of solid sodium hydroxide [12]. The two products, i. e., 9 and hexamethyldisiloxane, are then removed from the reaction vessel under vacuum and subsequently separated; in the cases of 9b and 9 c this is not completely possible but the presence of hexamethyldisiloxane does not interfere with the further synthetic applications of these two phosphaalkynes. The adamantyl derivative 9 e is a crystalline solid while the remaining phosphaalkynes are distillable liquids. They all give rise to high-field 31P-NMRresonan-
6.3 Reactivity of Phosphaalkynes [lo]
175
ces in the range 6 = -51.4 to -68.9; the I3C-NMR signals of the sp-hybridized carbon atoms appear in the region 6 = 173.7 to 185.6 with 'J(C,P) couplings of 37.2 to 45.5 Hz. Phosphaalkynes without sterically demanding substituents can also be manipulated either in solution or by freezing out at low temperatures. They are often generated by thermal $elimination of hydrogen halide under vacuum pyrolysis conditions [lo].
6.3 Reactivity of Phosphaalkynes [lo] With regard to their reactivity, the phosphaalkynes 9 reveal more parallels with the alkynes (as is already indicated by the name) than with the nitriles [15] (Scheme 6-3).
PEC-R 9
]xHomo -Diels-Alder
ep=( 9
1
4
R
I
Fp
D
cp
,--
Cyclodimerization,
Scheme 6-3 General reactivity of phosphaalkynes.
Thus, halocarbenes undergo smooth addition to the P/C triple bonds of compounds 9 to furnish 1-phosphirenes (for intramolecular ring closures phosphavinylcarbenes to give the same products, see [16]), which rearrange to the thermodynamically more stable 2-isomers 11 by [1,3]-halogen shifts [17].
176
6 Phosphaalkynes
The cycloaddition behavior of phosphaalkynes toward 1,3-dipoles is particularly pronounced. Thus, reactions with diazo compounds give rise to the 3H-1,2,4-diazaphospholes12 or their 1H-isomers, respectively [16, 181. Azides [18a, 191, nitrilium betaines “a, 201 mesoionic species [21], and sextet dipoles such as selenoxocarbenes [22] react analogously to form heteroatom-substituted phospholes. Phosphaalkynes also play a prominent role as dienophiles in Diels- Alder reactions; hence the h3-phosphinines 13 [23] are formed from cyclic 1,3-dienes such as a-pyrone or cyclopentadienones by way of “extrusion” of C 0 2 or CO, respectively; reactions with anthracene provide an access to the phosphabarrelene series [24]. This type of reaction is also of significance for the construction of phosphorus-carbon cage compounds. The same is true for homoDiels- Alder reactions which, with 2-phosphabicyclo[2.2.2]octa-2,5-diene as reaction partner, lead to the diphosphatetracyclodecenes 14 [25]. Last but not least, ene reactions with phosphaalkynes as enophiles [25] are also valuable for the construction of polycyclic phosphorus-carbon compounds. The reactions of 9 with 2,3-dimethyl-2-butene (-15) [26, 271 emphasize this behavior. Finally, the cyclooligomerization of phosphaalkynes (e. g., -16) [lo, 281 in the coordination sphere of a metal should be mentioned. Zirconium complexes, which will be discussed in detail below, dominate the chemistry of phosphaalkyne cyclotetramers.
6.4 The History of Phosphorus-Carbon Cage Compounds from Phosphaalkynes Phosphorus-carbon cage compounds first saw the light of day in 1989. Three research groups independently prepared the first representatives of phosphacubane or cubane-like classes of compounds (Fig. 6-1) [29]. The tetraphosphacubane skeleton (C4P4cage) was constructed by Regitz and co-workers by the thermal cyclooligomerization of a phosphaalkyne [30]. This nonselective approach was later replaced by a specific “two-step” cyclotetramerization which then made detailed investigations of this fascinating class of compounds possible (see Section 6.5.5.2).
CLPL-cage
C5P6-cage
C5Ps-cage
O n each carbon there is a t-butyl
CgPg-cage
C~Pb-cage
Figure 6-1 History: selection of the first phosphorus-carbon cage compounds.
6.5 Synthesis of Phosphorus- Carbon Cage Compounds
177
Cube-like compounds such as the homotetraphosphapentaprismane (C5P6)cage or, for example, the C5P5cage system were prepared for the first time and their structures confirmed by crystallography in the laboratories of Zenneck [31]. These compounds are formed in low yield when the mixture of sandwich complexes bearing phosphole and phosphete ligands obtained from (q 4-l-methylnaphthalene)(q6-tolyl)iron and tert-butylphosphaacetylene (9a) is subjected to vacuum pyrolysis at 100 "C. Equally worthy of mention at this point are the cage compounds with a C5P5skeleton [32] (an isomer of the species already mentioned) as well as a C,P, cage [33] prepared by Nixon and co-workers, although these - in contradiction to the title of the present review - were not directly obtained from a phosphaalkyne. Starting materials in both cases were the phosphacyclopentadienides.
6.5 Synthesis of Phosphorus-Carbon Cage Compounds 6.5.1 Construction by Cycloaddition Reactions Inter- and intramolecular cycloaddition sequences, in some cases in combination with ene reactions, play a significant part in the construction of phosphorus-carbon cage compounds. Phosphaalkynes are transformed sequentially into phosphaalkenes and finally to cages made up of phosphorus/carbon single bonds in these processes.
6.5.1.1 Diphosphatetracyclodecenes The Diels-Alder reaction of cyclohexa-1,3-diene (17) with the phosphaalkyne 9a only proceeds under drastic conditions in a pressure Schlenk tube but then does provide the 2-phosphabicyclooctadiene 18 [23] in quantitative yield (Scheme 6-4). This compound represents the first member of the previously unknown class of polycyclic phosphaalkenes. From a thermal point of view, 18 is unexpectedly stable; cycloreversion only occurs under flash vacuum pyrolysis conditions (520°C/10-6 mbar) but then it does not give rise to the starting materials 9a and 17; instead it proceeds by cleavage of ethylene to furnish the 2-fert-b~tyl-1~-
IB"
20
Scheme 6-4 Diphosphatetracyclodecenes by Diels-Alder and homo-Diels- Alder reactions with tertbutylphosphaacetylene (9a).
The bicyclic system 18 constitutes a hetero-IP-diene and thus fulfills the prerequisites for a homo-Diels- Alder reaction. The reaction with tert-butylphosphaacetylene (9a) occurs under harsh conditions comparable to those required for the formation of 18 to furnish the
178
6 Phosphaalkynes
diphosphatetracyclodecene19 loco- and orientation-specifically [25]. The cage contains both conventionally and low-coordinated phosphorus atoms, as can be deduced convincingly from the 31P-NMRspectrum (8 = -17 and +315, respectively). The 'J(P,P) coupling constant of 12.6 Hz rules out the formation of the isomer 20. An analogous reaction sequence can also be realized with cyclopentadiene and 9a: the initial [4 + 21-cycloaddition in this case, however, takes place at room temperature (251.
6.5.1.2 Phosphaprismanes and Phosphabenzvalenes Phosphabenzenes (h3-phosphinines) i have been known for a long time and played a major role in the development of the chemistry of low-coordinated phosphorus [34]. In contrast, all previous attempts to prepare their valency isomers, which also include the cage-like phosphaprismanes and phosphabenzvalenes, failed. Success was first possible after the development of syntheses for cyclobutadienes on the one hand, and for phosphaalkynes on the other. n-pentane or CHCls
n-pentane
+ R-CEP
[2+21
tBu 21
tBu
tBu
22
23
+ tBu
24
i P r CH,-fBu
I-Ad
Scheme 6-5 Phosphaprismanes23 from 2-Dewar-phosphininesby intramolecular [2 + 2]-cycloadditions.
When the cyclobutadienecarboxylate 21 [35] stabilized by the presence of bulky substituents, is allowed to react with the phosphaalkynes 9a-d, the 2-phospha-Dewar-benzenesare obtained in quantitative yields [36] (Scheme 6-5). For steric reasons the [4 21-cycloaddition takes place at C1 and C4 and is highly selective with regard to the dienophile orientation. In the cases of the phosphaalkynes 9a-c, products 22a-c dominate over the isomers 24a-c by a ratio of at least 93:7. Within the detection limits of 31P-NMR spectroscopy, the DielsAlder reaction 21 + 9d furnishes 22d exclusively. The structure of 22d was confirmed by X-ray crystallography [36]. The phosphaalkene moieties of the phospha-Dewar-benzenes 2221-d, which are utilized in the subsequent reactions, can be recognized, among others, by their lowfield 3'P-NMR signals (6 = 312-315). In analogy to substituted Dewar benzenes lacking heteroatoms [37a-c], irradiation of the phosphabicyclic compounds 22 a-d induces an intramolecular [2 + 21-cycloaddition to furnish the phosphaprismanes 23a-d [38]. The transformation of X302-phosphorus to h303-phosphorus associated with the conversion 22 --t 23 is reflected in a dramatic high-field shift of the respective 31P-NMR signals by about 450 ppm [38]. The duration of the photolysis
+
6.5 Synthesis of Phosphorus-Carbon Cage Compounh
179
decisively influences the product palette: with increasing irradiation times phosphaprismane yield decreases in favor of the isomeric phosphabenzene and phosphabenzvalene [38]. The first azaphosphaprismanes and -benzvalenes were also prepared by the above-mentioned route. The combination tri-tert-butylazete (25)/fert-butylphosphaacetylene(9a) is initially responsible for the formation of the azaphospha-Dewar-benzene 26; (Scheme 6-6) again the reacting centers are determined by the steric demands of the two reaction partners 1391.
tBwqp t Bu
tBu
hu1>280nrn)
dBU+tBu-EP
n 2ooc - [pLe+n2t 1a n e
tBu
c62;:0:0c
tBu 25
26
9a
t Bu
-
tBu
tBu
tBu
tBu 27
hu l>280nm) C6H6, 2ooc I
tBu
t Bu
t Bu
tBu
28
29
30
Scheme 6-6 Tri-tert-butylazete - starting material for the synthesis of azaphosphaprismanes 27 and -benzvalenes 29.
The subsequent photochemical formation of the tetracyclic species 27 is not as smooth as the isomerization 22 + 23. With shorter irradiation times, only incomplete conversion is achieved, whereas with longer irradiation periods the formation of the azaphosphaprismane is overlapped by its conversion into azaphosphabenzvalene (27 -+ 29). Even so, it is possible by optimization of the reaction time to enrich the content of 27 (together with 26, ratio = 1 : 1) and to characterize it unequivocally by NMR spectroscopy. An irrefutable argument for the formation of the tetracyclic system is again the enormous high-field shift experienced by the "P-NMR signal on changing from 26 to 27 (6 = +202 and -111, respectively) [a]. If the benzene solution of 27 is left at room temperature, complete isomerization to the 1,4-azaphosphinine 30 occurs. Under photochemical conditions, the latter species undergoes a crossed [2 + 21-cycloaddition to give the extremely stable azaphosphabenzvalene 29. The sequence 26 27 -B 30 + 29 is presumably also involved in the above-mentioned long-duration photolysis of the azaphospha-Dewar-benzene [ a ] . The tricyclic compound 29 is converted quantitatively to the 1J-azaphosphinine 28 by way of cleavage of one P/C and one C/C bond in the phosphabicyclobutane part only at 140°C. The product 28 is also accessible from 26 by thermal or proton-catalyzed cleavage of the zero bridge [40]. The polycyclic compounds 27 and 29 (in addition to 26) are the only currently known valence isomers of the azaphosphinines although several representatives of the heteroaromatic system itself have been reported [41]. +
180
6 Phosphaalkynes
6.5.1.3 Diphosphatrieyclooctenes When the reaction of 1,3-butadiene (31) with tert-butylphosphaacetylene (9a) is considered, one would first anticipate the formation of the phosphacyclohexene32 by a Diels-Alder reaction [23, 241. However, the tricyclic product 34 is obtained in optimal yield instead of the monocyclic system when the reactant ratio 31 :9a = 1 :2 (Scheme 6-7) [26, 271.
<
+ tBu-CSP
9 0 ° C , no solvent [L+21
tBu-CnP19a)t
~
ene-react ion
ti3u4i 31
9a
33
32
[L+21, intrarn.
34
Scheme 6-7 Diphosphatricyclooctenes - Diels- Alder and ene reactivity of phosphaalkynes and phosphaalkenes.
,
The 31P-NMR spectrum alone is sufficient to show that the reaction product contains a diphosphirane unit, i. e., the high-field signals of the two phosphorus atoms (6 = - 190.0 and -194.5) and the 'J(P,P) coupling constant of 158.7 Hz [26]. The crystal structure analysis (Fig. 6-2) provided final confirmation of the appearance of the product [27]. On the basis of this knowledge, a reaction mechanism comprising an initial Diels-Alder reaction (+32), a Selected bond lengths [aj PI -P7 P1 - c 2 P7 C6 c3 c4 CS C6
-
-
2.18312) 1.886(4)
1.847(6) 1.317(6) 1.S 11(7)
PI P7 c2 c4 CS
- cs - c2 - c3 - c5 - CB
1.869(4) 1888(4) 1.478(6) 1492(6) 1.553(6)
Selected bond angles ["I
P7-PIC2 C8-P I-cz Pl-P7-C6 Pl-CZ-P7 P7-CZ-C3 c3-c4-c5 C4-CS-CB P7C6-CS
Figure 6-2 Crystal structure of the diphosphatricyclooctene 34.
54.7( 1 ) 100.9(2) 93.8(2) 70.712) 114.713) 118.3(4) 111.2(3) 111 3 3 )
P7-Pl-C8 P 1-P7-CZ CZ-P7-C2 Pl-CZ-C3 CZ-C3-C4 C4-CS-C6 c6-CSC8 PI-CS-cs
94.8(1) 54.6(I) 98.612) 116.3(3) 123.3(3) 108.5(4)
107.1(4) 108.2(3)
6.5 Synthesis of Phosphorus-Carbon Cage Compounds
181
highly selective ene reaction at the phosphaallyl moiety of 32 (+33), and finally an intramolecular [4+ 2]-cycloaddition to furnish the polycyclic system (+34), can be formulated to explain the process [26, 271. Ene reactions in which the phosphaalkyne takes over the role of the enophile are now well known [42]. In harmony with this concept, 1,4-cyclohexadienes also react with one equivalent of a phosphaalkyne to furnish tricyclic products which now contain a phosphirane ring in place of the diphosphirane unit. When unsymmetrically substituted I$-butadienes are allowed to react with a phosphaalkyne, the first step is the formation of isomeric phosphacyclohexadienes and is reflected in the final products formed (exchanged substituents in the C , chain of the original 1J-butadiene) [26, 271. The formation of the diphosphatricyclooctene system is very flexible with regard to the 1,3-diene component; this is probably also valid for the phosphaalkyne dienophiles although no substituent variations have as yet been carried out. 1,2-Bis(methylene)cycIoalkanes 35a-h, which may also contain heteroatoms such as silicon or germanium in the ring, as well as the disiloxy-substituted 1,3-diene 35i react with 9a under the usual, drastic conditions to give diphosphatricyclooctenesin which (with the exception of 36i) the positions 1 and 7 are linked by a chain (36a-h) [27] (Scheme 6-8). Cycloalkenes bearing vinyl groups at the double bond (37, n = 3-6) react with 9a to give the polycyclic products 38 (n = 3-6) [27]. In this context, finally, the analogous reaction of the germene 39 with the phosphaalkyne 9a should be mentioned. The reaction involves participation of one double C6H6, 7O-12O0C Schlenk o r e s s u r e tube
36
35
9a 35, 36
tBu.
1
b
a
c
d
9
h
tBu i
C6H6, 1 0 0 ° C S c h l e n k p r e s s u r e tube
2 tBu-EP +
t Bu
9a
37
38 37,38: n = 3 - 6
2 tBu-CzP +
9a
39
tBu 40
Seheme 6-8 Diphosphatricyclooctenes - scope of application of the reaction sequence Diels-Alder reaction, ene reaction, intramolecular [4 + 2]-cycloaddition. Mes, mesitylene.
182
6 Phosphaalkynes
bond of the fluorenylidene system, i. e., as a 1,3-heterodiene; otherwise the formation of the diphosphatricyclooctene 40 (confirmed by crystal structure analysis) proceeds normally via ene and intramolecular 14 + 21-cycloaddition steps [43]. The use of azadienes enables incorporation of nitrogen into the tricyclic skeleton [44].
6.5.1.4 Diphosphateiracycloundecadienones and Oxadiphosphapentaeyelonadecapentaenones (the Tropone Reaction of Phosphaalkynes)
A rich cycloaddition chemistry with phosphaalkynes 9 may be expected from cross-conjugated carbocyclic compounds of the heptafulvene type (in the present case, tropone (41)). Thermal [n: + 713- and [n,8 + n:]-cycloadditions are allowed by symmetry. Thus, the thermal reaction of tropone (41) with an excess of the phosphaalkynes 9a, 9e (without a solvent) furnishes the diphosphatetracyclic products 43a, b [45]. The initiating Diels- Alder reaction to 42 (there are 31P-NMRspectroscopic indications for its intermediate formation) is accordingly followed by a [2 + 2 + 2]-cycloaddition (=homo-Diels-Alder reaction) with a second equivalent of the phosphaalkyne to yield 43 (Scheme 6-9). The observed peri- and regioselectivities of the two reaction steps can be confirmed by frontier orbital calculations [45].
R - C I P I9a, e) 9 5 T , no solvent
42
41
2 R-CrP (9a, b , e) 120°C, toluene
*
41
43a: R = t B u ; b: R = I-Ad
a. cyclopropyl - a l l y [ -rearrangement b . [1,51-Hc. r i n a inversion
R 4 5 a : R :t B u ; b: R : C M e z E t ; c : R :? - A d
Scheme 6-9 'Ropone as reaction partner for phosphaalkynes in the construction of phosphorus-carbon cage compounds 43 and 45.
On the other hand, when tropone (41) is allowed to react with the phosphaalkynes 9a, 9b, or 9e at a somewhat raised temperature in the molar ratio 1:1 in toluene, the reaction sequence ends with the formation of the diphosphapentacyclic compounds 45a-c [45]. In other words, this means that the initially formed 43 must have undergone a stereoselective, [8+ 21-cycloaddition with a second equivalent of tropone (41). The hexacyclic species 44 thus formed experience a cyclopropyl-ally1 rearrangement with subsequent [1,5 1-H shift and ring inversion, finally to furnish the isolated pentacyclic compounds 45 [451. The semiempirically
6.5 Synthesis of Phosphorus-Carbon Cage Compounds
183
calculated structural parameters (PM3) for 45a agree satisfactorily with the values from the crystal structure analysis [45]. The pentacyclic compound 45a can be smoothly quaternized at the phosphorus atom bearing three C-substituents by treatment with alkylating agents. Oxidizing agents such as tertbutyl hydroperoxide, elemental sulfur, or elemental selenium, on the other hand, preferentially attack at the 0-substituted phosphorus atom. When an excess of the oxidizing agent is employed, both phosphorus atoms are oxidized [45].
6.5.1.5 Diphosphirenes as Intermediates for Phosphorus-Carbon Cage Compounds
The chemistry of the diphosphirenes is an almost completely unexplored field; to date only one stable member of this compound class is known [46].However, tungsten pentacarbonyl complexes of the type 47 at least show a tendency for ring opening (Scheme 6-10). The phosphinidine or carbene intermediates thus formed serve as starting materials for the construction of further novel phosphorus-carbon cage compounds.
Scheme 6-10 Addition of W(CO)s-phosphinidine complexes to phosphaalkynes for the synthesis of phosphorus-carbon polycyclic systems. cHex, cyclohexyl.
-
initiating reaction
When the aminophosphinidine complex 46 (R’= R2 = cyclohexyl) is generated by thermal phosphirane fragmentation [47, 481 in the presence of tert-butylphosphaacetylene (9a), the diphosphirene complex 47 (R’= R2 = cyclohexyl) is initially formed and has been characterized unambiguously in solution by NMR spectroscopy [47]. The formation of the “edgeopened” tetraphosphaprismane 50 commences from 47. It is assumed that reaction of the
184
6 Phosphaalkynes
phosphinidine formed by ring opening of 47 with the still-intact diphosphirene initiates formation of the polycyclic product [47]. If, in contrast, the tungsten pentacarbonyl-phosphinidine complex with o-tolylamino substitution (46, R' = H, R2 = o-tolyl) is the starting material, also generated by phosphirane fragmentation, reaction with four equivalents of tert-butylphosphaacetylene(9a) gives rise to the pentaphosphatetracyclononene-W(CO)5 complex 48, the structure of which was confirmed by crystallography [49]. 1-Adamantylphosphaacetylene (9e) reacts analogously. A fairly obvious interpretation would involve the cleavage of the diphosphirene 47 (R' = H, R2= o-tolyl) to the carbene 49 and [3 + 2]-cycloaddition of the latter with a second equivalent of 9 a to furnish the triphosphole 51. The remaining two equivalents of phosphaalkyne are incorporated by Diels-Alder (-+52) and homo-Diels-Alder reactions (-*48), respectively [49]. 6.5.1.6 Thermal Cyclotetramerization
In spite of their sterically demanding substituents, the phosphaalkynes exhibit an enormous potential for cycloaddition reactions. Even though no intermediates can yet be detected in their thermal cyclooligomerization reactions, the obtained product palette allows the assumption of head-to-head and head-to-tail dimerizations, i. e. [2 + 2l-initiating reactions. When tert-butylphosphaacetylene(9a) is heated in the absence of a solvent at 130"C, a complex mixture of products is formed. The tetraphosphacubane 53a (Scheme 6-11) can be isolated in modest yield (8010) by bulb-to-bulb distillation from this mixture [30, 50b]. The thermal reaction at 180"C is similarly nonselective and the tetraphosphatetracyclooctene 54a can be isolated from the crude mixture in comparable yield [50b]. Last but not least, the tetraphosphacuneane 55 was obtained by thermal cyclotetramerization at the same temperatBu
53 a
L tBu--CP 9a
1
p-' 1 8 0 0 ~ .no s o l v e n t * )
~
uBt ' / P
'+t
Bu
P
w
fBu
tBu 54 a tBu
feu' *) I n Schlenk p r e s s u r e tubes
Scheme 6-11
55
The thermal cyclotetramerization of phosphaalkynes.
6.5 Synthesis of Phosphorus-Carbon Cage Compounds
185
ture but in solution [51]. It is not yet clear at which stage in the process a ref?-butyl group is cleaved as isobutylene. It will be shown further below that both cage compounds 53a and 54a can be prepared specifically. The most interesting by far of these polycyclic compounds is the tetraphosphacubane 53a. The crystal structure analysis (Fig. 6-3) shows constant P/C atomic separations of 1.881 A in the cage. The P - C - P angles in the skeleton are slightly widened to 94.4" while the C -P - C angles are accordingly reduced to 85.6" [30].At first, the high-field 13C-NMR signals of the cage carbon atoms (6 = -29.1) and the similarly surprising low-field 3'P-NMR signals of the cage phosphorus atoms (6 = +257.4) appeared to be unexplainable [30]. A strong participation of the lone electron pairs at phosphorus in the P - C o-bonds of the cage can be deduced from the photoelectron (PE) spectrum of 53a [52]. This leads to a negative partial charge at the carbon atoms and a positive partial charge at the phosphorus atoms; the corresponding net charges were determined from MO calculations [52]. This partial charge situation explains the unusual chemical shifts of the cage atoms as well as the reduced nucleophilicity/basicity of the phosphorus atoms, which will be discussed further below. Selected bond lenghb [A] PI - C I PI - c 4 P2 - C4 P3 - CZ P3 - c 4 P4-Cl
p c 3 2
c33
1
1.880(3) 1.877(3) 1 876(4) 1881(3) 1.886(3) 1.885(3)
PI-CZ PZ cz PZ c 3 P3 c 3 P4 C2 P4 c 3
1.886(3) 1.883(3) 1.891(3) I .875(4) 1.876(3) 1.875(4)
CZ-PI-C4 CI-PZ-C3 C3-PZ-C4 CZ-P3-C4 c 1-P4-CZ CZ-P4-C3 PI-CI-P4 PI -CZ-P3 P3-CZ-P4 PZ-C3-P4 Pl-C4-P2 PZ-C4-P3
85.6(1)
-
~
Selected bond angles to]
c I-PI-C4 P2
A
c4
1
u C 1 4
c1-P1-cz c l-Pz-c4 CZ-P3-C3 C3-P3-C4 c1-P4-C3 PI-c I-PZ PZ-c 1-P4 PI-CZ-P4 PZ-C3-P3 P3-C3-P4 PI-C4-P3
85.7(1) 85.4(1) 85.q I) 85.qZ) 85.1(2)
85.9(1) 94.2(2) 94.1(1) 94.3(2) 94.W
94.9(2) 944(2)
85.5(2)
85.7(2) 85.5(1)
85.6(1) SS.Z(Z)
94.3(2) 942(2) 94.7(1) 94.2(2) 94.3(2) 94.9(2)
Figure 6-3 Crystal structure of the tetraphosphacubane53a.
6.5.2 Construction by Extrusion of Cp,Zr from Phosphaalkyne Dimer Complexes A high-yielding synthesis was required before systematic investigations of the reactivity of the phosphacubane system could be realized. This was achieved by splitting the cyclotetramerization of the phosphaalkyne into two cyclodimerization steps: the first step is the synthesis of the zirconocene complex 59 and the second is the removal of its Cp2Zr fragment with subsequent renewed dimerization to furnish the tetraphosphapentacyclic system 53.
186
6 Phosphaalkynes
6.5.2.1 Cp,Zr-Phosphaalkyne Dimer Complexes When zirconocene dichloride (56)is treated with magnesium as a dehalogenating reagent in the presence of the phosphaalkyne 9a, the tricyclic zirconium complex 59a is formed [53]. It was later found that the metal dehalogenating reagent can be replaced advantageously by nbutyllithium (Scheme 6-12). The red Cp,Zr-phosphaalkyne dimer complexes 59a-e are readily accessible (40-70% yield) by the latter variant [50]. In the first reaction step, the two chlorine atoms are substituted by n-butyl groups (-67). Cleavage of n-butane and I-butene follows (during the warming-up phase) with simultaneous incorporation of the two equivalents of the phosphaalkyne 9 to furnish the zirconadiphosphacyclopentadiene 58. The process ends with a crossed [2 + 21-cycloaddition (-49). The crystal structure of 59a has been determined [53]. The 31P-NMRspectra of all complexes, as is usual for diphosphiranes, exhibit signals at high field (6 = -245.2 to -249.1). The skeletal carbon atoms are also magnetically equivalent and appear at low field (6 = 131.0 to 133.1) with *J(C,P) coupling constants of 62.7 to 70.5 Hz, thus indicating their direct proximity to zirconium [SO].
56
57
lintram.
58,59
1
R a
b
/!Bu
CMezEt
I-Ad
70
65
C
L
R yield[%]
I
70
d
e
50
LO
Mea Meo
R 59
Scheme 6-12 Cp,Zr-phosphaalkyne dimer complexes - starting materials for the preparation of tetraphosphacubanes. Cp, cyclopentadienyl; thf, tetrahydrofuran.
6.5.2.2 Tetraphosphacubanes and Isomeric Cage Compounds It is known that the Cp,Zr fragment of 59 can be cleaved as CpzZrC12by element halides (e. g., PhBCl,, PCl,, GeCI,) with the new heteroatom being incorporated in the heterocyclic system [S4]. On the other hand, halogenating agents (e.g., HgCI,, 19 and hydrogen halides also cause elimination of CpZZrX, and bond cleavage in the residual molecule while being included in the final product 1541. Thus, for the required objective, it was necessary to find a reagent which eliminates the Cp,Zr fragment while not being built into the final product itself. After numerous experiments, hexachloroethane was found to fulfill these requirements perfectly.
6.5 Synthesis of Phosphorus-Carbon Cage Compounds
187
When the zirconocene-phosphaalkyne dimer complexes 59a-e are treated with hexachloroethane, zirconocene dichloride, tetrachloroethene, and the tetraphosphacubanes 53 a-e are indeed formed concomitantly (Scheme 6-13). The yields amount to 50-80% with the exception of 53c (30%), which is lower due to the simultaneous formation of isomers (see below) [Sob]. The pentacyclic products 53b-e exhibit the same unusual NMR spectra as 53a (see Section 6.5.1.6).
59
53
t
12+21 intram.
R
60 53.60.61
R
yield[%]
1
61
I
I1
a
b
tBu
CMeZEt
I-Ad
70
80
30
C
d
e
Meo Meo 60
50
Scheme 6-13 Cp,Zr-cleavage reaction from 59 with hexachloroethane: the specific synthesis of tetraphosphacubanes.
No intermediates can be detected by 3'P-NMR monitoring of the reaction 59 --* 53. A plausible explanation for the formation of the tetraphosphacubanes 53, however, involves formation of the 1,3-diphosphetes 60, hetero-Diels-Alder reaction of the latter to the tetraphosphatricyclic species 61, and final intramolecular [2 21-cycloaddition. The primary formation of diphosphatetrahedranes seems unlikely since these (at least as far as the parent system is concerned) would be thermodynamically more favorable than the 1,3-diphosphetes [ 5 5 ] . The following arguments support the intermediacy of the diphosphetes. When an Fe(CO), unit is attached to a phosphorus atom of 59a before the Cp,Zr fragment is removed by treatment with hexachloroethane, the tricarbonyliron complex of 60a is obtained. Furthermore, crossing experiments with various CpzZr-phosphaalkyne dimer complexes indicate that 60 is the decisive intermediate [50b]. As already mentioned, the modest yield of tetraphosphacubane from the reaction 59c 53c is due to the formation of isomeric polycyclic compounds. The surprising result of the Cp,Zr cleavage reaction of 59c with hexachloroethane in toluene (Scheme 6-14) is, in addition to 53c (30% yield), the formation in equal yield of the tetraphosphatetracyclooctene 54b [50b]. When the more polar solvent dichloromethane is
+
-+
188
6 Phosphaalkynes
used under otherwise identical conditions, the tetraphosphacuneane 62 is obtained in 30% yield together with 53c. The polycyclic skeleton of 62 is isomeric to that of 55 (X-ray structure) 1561. A convincing rationale for this reaction branching cannot yet be given. Cl3C-CCI3 toluene
- C p 2Zr C I 2 1;Ad
I-Ad
?-Ad 54b
1-Ad -CI2C=CCI2 - Cp2ZrC12
I-Ad
62 Scheme 6-14 Adamantyl-substituted tetraphosphabis(homo)prismanes and -cuneanes.
6.5.2.3 P-Functionalization of the Tetraphosphacubane System
As already mentioned (see Section 6.5.1.6), the phosphorus atoms in the tetraphosphacubane system have lost basicity/nucleophilicity on account of the participation of the phosphorus lone electron pairs in the P - C o-bonds. Particularly strongly electrophilic reagents are thus tBu
E t 2 0 , 11O0CIpressure tubel,[-2
Scheme 6-15
Nz]
53 a
I
EtOzC-N+
N-CO2Et Et20, 25°C. [ L + I I
P-functionalization of the tetraphosphacubane system.
1 / 2 SB/NEt3 C6H6. 25OC
189
6.5 Synthesis of Phosphorus-Carbon Cage Compounds
necessary for protonation and alkylation reactions. Hence, monomethylation is only successful with “magic methyl” [57]. For mono- and diprotonation (53a + 63) reactions, fluorosulfonic acid in liquid SO, and “magic acid” (fluorosulfonic acid . SbF, in the same solvent) have proved useful (Scheme 6-15) [57]. With diiron nonacarbonyl, only one Fe(CO), fragment can be introduced (53a -+ 64). The steric situation resulting from the presence of this one unit prevents further complexation by the same reagent [58]. Mono- and bis(phosphazine) formation is possible with diazo compounds (e.g. 53a 65) although the resultant steric situation plays a decisive role [15, 581. Single and double Staudinger reactions have been realized, as has also the linkage of two cubes by means of a difunctional azide (53a 66) [15, 581. The analogous reaction of double phosphazine formation with 1,4-bis(diazomethyl)benzenehas also been achieved [15]. The transformation of h303-phosphorus to h ’ 0 ’-phosphorus in the tetraphosphacubane system can be effected by [4 + 11-cycloadditionof o-quinones or diethyl azodicarboxylate (53a 67) [15, 581. Finally, the stepwise oxidation of the phosphorus atoms in 53a by means of bis(trimethylsily1) peroxide or by elemental sulfur under triethylamine catalysis should be mentioned; these reactions ultimately lead to the tetroxide or tetrasulfide, respectively (53a -P 68) [15, 58, 591. -+
-+
+
6.5.3 Cyclooligomerization with the Aid of Lewis Acids In addition to the above-mentioned Cp2Zr-phosphaalkyne dimer complexes 59, aluminum compounds of the type 71 are also of major significance for the construction of phosphorus-carbon cage compounds, since the metal unit can be removed easily.
6.5.3.1 Spirocyclotrimerization When the phosphaalkynes 9a, 9b, and 9e are allowed to react with aluminum halides in dichloromethane in a molar ratio of 3 : 1, the spirocyclic products 71 are obtained (yields: 85-95% (Scheme 6-16) [60, 611. These compounds represent Lewis acid adducts of a 1,3-diphosphete containing both h30’-and h504-phosphorus atoms; addition compounds of this type are known in cyclobutadiene chemistry [62]. An X-ray crystal structure analysis of 71a has substantiated this new class of compounds [61]. The P - C atomic separations (1.70 and 1.77 A, respectively) in the cationic portion of the molecule justify the formulation of the adducts as the resonance hybrids (71A ++ 71B or 71C). As far as the reaction mechanism is concerned, we assume that the sequence commences with the addition of the Lewis acid to the carbon atom of the phosphaalkyne (9 -+ 69), in accord with the polarization of the P = C bond [lob]. The addition of fluorosulfonic acid to 9a in liquid SO, also begins with C-protonation [63]. The resultant, activated phosphaalkenes undergo [2 + 2]-cycloaddition with a second equivalent of 9 to furnish the adducts 70. A subsequent [2 + 11-cycloadditionwith a third equivalent of 9 is then responsible for the formation of the spirocyclic phosphorus derivatives (471).
190
6 Phosphaalkynes ALX,
CHzClz
PGC-R
12 + 21
70
69
9a, b, e
R
71B
71 A
95
85
R
71C
R
95
Scheme 6-16 Spirocyclotrimerizationof phosphaalkynes with Lewis acids (71).
6.5.3.2 Phosphaalkyne Tetramers from the Spirocyclotrimer 71a
When the original Lewis acid in the spirocyclic species 71a is removed by treatment with dimethyl sulfoxide as a Lewis base, the h3a2,X5a4-diphosphete 72 is liberated; however, it cannot be isolated as such [61]. Even at - 45 "C in dichloromethane, rearrangement to the 1,3,5-triphospha-Der-benzene 73, which again cannot be isolated, occurs after cleavage of the P-P bond in the diphosphirene ring. When phosphaalkyne 9a is added to the reaction mixture as a trapping reagent for the Dewar-benzene 73 before generation of 72, however, the tetraphosphatetracyclic system 75 is indeed formed (37%) by a homo-Diels-Alder reaction (Scheme 6-17) [61]. When the same experiment is carried out with an excess of aluminum chloride at -78°C in dichloromethane, the constitutional isomer 76 is obtained. Under these conditions, the rearrangement of 72 proceeds with P - C bond cleavage to the 1,2,5-triphospha-Dewar-benzene 74 which is finally trapped in a [2 + 2 + 2]-cycloaddition with 9a (-+76; 38% yield) [61]. The function of aluminum chloride or its DMSO-adduct in the last-mentioned reaction has not yet been clarified. A crystal structure analysis of the mesitylnitrile oxide adduct of 76 has been performed and thus the cage structure of the system is confirmed unequivocally [61]. Highly interesting isomerization reactions occur within the phosphaalkyne cyclotetramer system 75/76/54a and 77 (Scheme 6-18), the likes of which have not been previously seen in the chemistry of phosphorus-carbon cage compounds. It is worthy of note that the tetraphosphacubanes 53 do not take part in such processes. First of all it must be mentioned that the two compounds 75 and 76 resulting from the spirocyclotrimerization of phosphaalkyne 9a with subsequent cyclotetramer formation establish a 1 : 1 photochemical equilibrium in benzene solution. When they are heated in-
6.5 Synthesis of Phosphorus- Carbon Cage Compounds
191
71a
72 AtC13, C H ~ C I I - 78OC
CH~CIZ -L5OC
L
73
J
~ B U - C ~( P gal (Homo- Diets-Alder1
tBu-CsP(9al (Homo-Diels-Alder)
P = \(
tBu
75
'
\tBu
76
Scheme 6-17 napping reactions of triphospha-Dewar-benzenes with tert-butylphosphaacetylene: novel tetraphosphatetracyclic systems. DMSO, dimethyl sulfoxide.
dividually at 150°C in the absence of a solvent they both yield the tetraphosphatetracyclic compound 54a already known from the thermal cyclotetramerization of 9 a [15c, 601.When the latter compound is irradiated at 25 "C it sets up an equilibrium with the tetraphosphasemibullvalene 77; when the photolysis is interrupted, the starting material.54a is recovered quantitatively [51]. Long-term irradiation of the photochemically established 1 :1 equilibrium mixture 75/76 also leads to the tetraphosphasemibullvalene 77 [60]. The thermolysis of 75 and 76 to 54a is thermodynamically acceptable; the relative energies of the two cyclotetramers mentioned before (as calculated for the parent compounds with H in place of t-Bu) are higher than that of 54 with H in place of t-Bu [55]. Since it has not yet been possible to detect any intermediates, any comments on the reaction mechanism must remain purely speculative. However, the thermal rearrangement 76 -+ 54 a (loo%), for example, can be interpreted rather simply (and convincingly): if the cleavage of two P-C single bonds (which originate from the P = C double bonds in the precursor 74) is assumed, this would lead to a diphosphirenyldiphosphacyclopentadienein the sense of a retroDiels- Alder reaction. A double [1,5]-sigmatropic shift of the three-membered ring to phosphorus with subsequent [4 + 2]-cycloaddition would then produce the isomer. Under the assumption of a synchronous character, all steps are thermally allowed [m].
192
6 Phosphaalkynes long-time photolysis I
I
0
I
fBu
I
hu lZ280nm)
fBu*tBu p _ " C t B u
i
I
I
(ratio 1:l)
p 76
150°C. n o
4
'"
15OoC, no solvent (100%)
I
,
I I
, I
I
I
I
1
77
I
( 7 5 % . equilibrium)
Scheme 6-18 Isomerization reactions in the system of tert-butylphosphaacetylenecyclotetramers.
6.5.3.3 Hexaphosphapentaprismane from the Spirocyclotrhner 71a
When the cyclic 1,3-diphosphete 72 is liberated from 71a by treatment with dimethyl sulfoxide as Lewis base (Scheme 6-19), unforeseen reaction products are formed in dependence on the reaction conditions. At 35 "C in dichloromethane, the 1-chloro-1H-phosphirene 78 is formed (49'70); this compound had previously been prepared by a specific synthesis in our laboratory [17c, d]. When toluene is used as solvent in place of the chlorinated hydrocarbon, the same product is obtained; this proves that the chlorine in 78 must originate from the aluminum chloride [60]. When the reaction mixture for the preparation of 76 (with excess AlCl,; see Section 6.5.3.2) is allowed to warm up to room temperature from -78"C, a product with the composition C,P,(t-Bu),, i. e., a dimer of 72 which must have lost di-tert-butylacetylene during its formation, is formed together with the already-mentioned cyclotetramer. The structure of this species is still unknown, but it does contain two phosphaalkene units, as shown and I3C-NMR spectroscopy [60]. After the concentrated solution has unambiguously by 31Pbeen exposed to diffuse daylight for about two days the hexaphosphapentaprismane 79 can be isolated in the form of orange-red crystals (40%). The all-carbon analog of 79 has been known for more than ten years [6]. The structure of 79 has been elucidated unequivocally by
6.5 Synthesis of Phosphorus-Carbon Cage Compounds
193
72 DMSO, CHzClz 35oc
CbD6 ( d i f f u s e light1
tBu
tBd
'tBu
79
78
Scheme 6-19 Hexaphosphapentaprismane 79.
X-ray crystallography (Fig. 6-4) [60]. The average P - C bond length of the Cz-symmetrical molecule 79 is 1.884 A and is thus comparable with that of the tetraphosphacubane 53a (1.881 A). The 31P-NMR signal of the atoms of the P,-bridge joining the two triphosphacyclobutane units at 6 = -2.9 ('J(P,P) = 79.6 Hz). The signals for the remaining phosphorus atoms occur at 6 = 180.5 ('J(P,P) = 79.6 Hz) and 6 = 243.7. The low-field positions of the four li3a3-phosphorusnuclei in the 19-diphosphete units can be explained in the same way as for tetraphosphacubane (see above). Counterverification by means of the "C-NMR signals of the carbon atoms of the four-membered rings (high-field signals) has not yet been possible since the substance decomposed during measurement [60].
+
P
F
I
,-
Y
c4
Selected bond lengths P1 -P2 P1- C3 PZ - P6 P3 - c 4 P4 - c3 P4 - c 4 PS - c 4 P6-CI
Selected bond angles P1-PZ-c2 CZ-P2P6 c3-P 1 x 1 P6-P3-C4 C4-P3-CZ C4-P4-C1 C4-PS-CZ c3-PS-cz P3-P6-C1 P6-C 1-P4 P4-c 1-PI PZ-cz-PS PI-C3-P4 P4-C3-P5 P4-C4-P5
Figure 6-4 Crystal structure of the hexaphosphapentaprisrnane79.
[A]
2.21 l(7) 1.864(17) 2.273(7) 1.865(17) 1.912(18) 1.870(18) 1.899(18) 1.880(17)
P1 -c1 PZ c 2 P3 CZ P3 P6 P4 CI PS C2 P5 c 3
-
1.917( 17) 1.883(17) 1.875(18) 2.206(7) 1.879(17) 1.885(17) 1.876(18)
["I
101.4(6) 87.7(6) 88.1(8) 101.2(6) 87.6(8) 104.0(7) 86.3(8) 103.6(8) 101.6(6) 115.6(9) 91.3(8) 115.7(10) 91.9(8) 91.4(8\ 9z.qsj
PI-PZP6 c3-PlPZ PZ-P 1-c1 P6-P3-C2 C4-PK3 C3-PKl C4-PSC3 P3-P6-PZ PZ-P6-C1 PCCI-PI PZ-CZ-P3 P3-CZ-PS P l-C3-Ps P4-C4-P3 P3-C4-P5
80.60) 101.1(6) 89.9(6) 89.9(6) 87.9(8) 87.8(7) 88.1(8)
80.6(3) 89.0(6) 99.6(8) 100.9(8) 92.6(8) 117 l(9) 116 9(9) 92.5(9)
194
6 Phosphaalkynes
6.5.3.4 Phosphorus-Carbon-AIuminum Cage Compounds
In addition to the above-mentioned aluminum halides (Section 6.5.3.1), triorganoaluminum reagents are also able to initiate cyclooligomerization processes of the phosphaalkyne 9a. In these reactions, aluminum is incorporated directly into the cage system. The element combination phosphorus, carbon, aluminum in a polycyclic system was previously unknown. The product palette is highly influenced by the choice of the solvent and the substituents at the metal center. When the reaction mixture comprising triethylaluminum and phosphaalkyne 9a (ratio 2:3.5) in n-hexane is allowed to warm from -50°C to room temperature, phosphaalkyne cyclotrimerization with incorporation of two trialkylaluminum units and formation of the bis(homo)prismane 80 (91 To) occurs (Scheme 6-20) [64]. The structure of the cage compound was elucidated unambiguously by crystallography; all 31P-NMRsignals appear in the highfield region (6 = - 1.2 to - 147.3) [64]. tBu 3 or
1,
9a
1
-
AIEIs, EtZO - 78 +2!j°C
81A
80
818
Scheme 6-20 First phosphorus-carbon-aluminum cage compounds 80, 81.
The reaction of phosphaalkyne 9a with trialkylaluminum (ratio 4: 1) in diethyl ether instead of in n-hexane follows a completely different course: phosphaalkyne cyclotetramerization with incorporation of one equivalent of trialkylaluminum provides the polycyclic compound 81 (74%) [MI. X-ray crystallography revealed that the molecule possesses C, symmetry; the bond lengths in the phosphonium-phosphorane structural element (81A) are indicative of a delocalization of the positive charge according to 81B. The k302-phosphorus atom is easily recognized by its low-field position in the "P-NMR spectrum (6 = 208.0, 'J(P,P) = 320.6 Hz, 2J(P,P)= 10.8 Hz) [64]. When sterically more demanding trialkylaluminum reagents (R = i-butyl or 2-phenylpropyl) are allowed to react with 9a in diethyl ether or n-hexane, the cyclooligomerization process ends at the 1 :3 compounds 82 (Scheme 6-21) 1651. The compounds have the structure of a triphosphametallahomobenzvalene,as has been confirmed for the case of R = CH, - CHPh - CH3 by X-ray crystallography. A plausible mechanistic rationale for the product formation involves a two-fold 12-addition of the triorganoaluminum to two equivalents of phosphaalkyne 9a (-84). This is followed by a crossed [2 + 21-cycloaddition
195
6.6 Outlook
U
Eu
82
83
AlRo
11,Zladdition
R I R L
84
a5
CH2-CHPh-CH3:
n-
hexane I-78+25OCl
Scheme 6-21 Metallatriphosphahomobenzvalenes82 by cyclotrimerizationof tert-butylphosphaacetylene with trialkylaluminum reagents.
between the two phosphaalkene moieties to give 85. Subsequent cycloaddition of the third equivalent of 9a with migration of an organic group from phosphorus to the Lewis acidic metal center finally leads to the product 82 [65]. The ability of phosphorus-carbon-aluminum cage compounds to function as ligands in transition metal complexes will be briefly illustrated by the example of 82 (R = i-Bu) on reaction with diiron nonacarbonyl. Coordination of the Fe(CO), fragment occurs exclusively at the phosphorus atom of the double bond which serves as a two-electron donor (+83,100%) (Scheme 6-21) 1651.
6.6 Outlook On the basis of the cyclooligomer chemistry of phosphaalkynes mentioned in this chapter alone, it is clear that future efforts should be directed toward answering the following questions. (a) Will it be possible to realize substitution reactions on cage compounds with the help of novel, easily accessible phosphaalkynes bearing exchangeable C-substituents (e. g., silyl)? Functionalization reactions at phosphorus are not so difficult. (b) Will it be possible to use further organometallic reagents for the cyclooligomerization of phosphaalkynes and to eliminate the auxiliary groups in a later step (as is the case with Cp,Zr complexes)? (c) Will it be possible to construct further phosphorus-carbon-element polycyclic systems such as those obtained with trialkylaluminum reagents? In light of the current developments in the field of phosphaalkyne chemistry, the authors of the present chapter are hopeful and optimistic that this trend will continue in the future.
196
6 Phosphaalkynes
6.7 Experimental Procedures All reactions described here must be carried out with strict exclusion of moisture and atmospheric oxygen under an argon atmosphere. The solvents used were previously dried by standard procedures and then distilled and stored under argon. All glass apparatus used was previously repeatedly baked out under high vacuum and purged with argon. For further details, see the appropriate text passages in the cited references.
6.7.1 2,2-Dimethyl-l-(trimethylsiloxy)propylidene(trimethylsilyl)phosphane (1Oa) A solution of tris(trimethylsily1)phosphane [66](7;50.0 g, 0.2 mol) in n-pentane (200 mL) was placed in a 500 mL three-necked flask (fitted with a reflux condenser and a 250 mL dropping funnel). A solution of pivaloyl chloride (6a;26.5 g, 0.22 mol) in n-pentane (100 mL) was added dropwise with stirring. The reaction mixture was stirred at room temperature for four days, the solvent was then removed under high vacuum, and the residue distilled at lo-' mbar to afford 10a (47.2g, 90070) as a yellow oil, b.p. 45-48"C/10-' mbar.
6.7.2 (2,2-Dimethylpropylidyne)phosphane (9s) The synthesis of 9a was performed in the apparatus illustrated schematically in Fig. 6-5. Coarsely ground sodium hydroxide (0.5g) was placed in a 100 mL round-bottomed flask and 5 x 10' mbar
I
liquid nitrogen
U
liquid nitrogen
silicone oil bath 160-180 'C
Figure 6-5 Preparation of tert-butylphosphaacetylene(9a).
6.7 Experimental Procedures
197
heated at 160-180°C on a silicone oil bath. The entire apparatus was evacuated to high vacuum (5 x lo-' mbar); cold trap 1 was cooled with acetone/solid CO, and cold trap 2 with liquid nitrogen. Compound 10a (39.9 g, 0.15 mol) was then added dropwise in portions to the sodium hydroxide. The volatile reaction products condensed in cold trap 1 and were frozen out in a 250 mL flask cooled in liquid nitrogen. After complete addition of 10a, the cooling bath was removed and the 250 mL flask allowed to warm to room temperature. After about 3 to 4 h under high vacuum, the (2,2-dimethylpropylidyne)phosphane 9 a formed had distilled off and was frozen out in cold trap 2. The entire apparatus was then purged with argon and the liquid nitrogen removed from cold trap 2. Product 9a (13.5 g, 93%) was obtained as a colorless liquid, b.p. 61 "C/1013 mbar.
6.7.3 Bis(~5-cyclopentadienyl)(2,4-di-tert-butyl-l,3-diphosphabicyclo[l.l.0]butan-2,4-diyl)zirconium (59a) To a suspension of zirconocene dichloride (56; 2.2 g, 7.5 mmol) in tetrahydrofuran (40 mL) cooled to - 78 "C was added dropwise a 1.6 M solution of n-butyllithium in n-hexane (9.2 mL, 15 mmol). The resultant solution was stirred at - 78 "C for 2 h and then (2,2-dimethylpropylidyne)phosphane (9a; 1.5 g, 15 mmol) was added. The reaction mixture was allowed to warm to room temperature and stirring was continued for 12 h at this temperature. The solvent was evaporated under high vacuum, the residue was taken up in diethyl ether (40 mL), and the mixture filtered through Celite. The filtrate was concentrated and cooled to -78 "C to furnish 59a (2.2 g, 70%) as red crystals, m.p. >3OO"C (dec.).
6.7.4 2,4,6,8-Tetra-tert-butyl-l,3,5,7-tetraphosphapentacyclo[4.2.0.0~5.d~*.~~J]octane (53a) To a solution of 59s (3.0 g, 7.1 mmol) in toluene (20 mL) was added, with stirring, hexachloroethane (1.7 g, 7.1 mmol). The reaction mixture was stirred at room temperature for five days, during which time its color changed from red to yellow. The solvent was then evaporated under high vacuum, the residue was taken up in n-pentane, the insoluble zirconocene dichloride was removed by filtration through Celite, and the residue washed with n-pentane. The filtrate and washings were concentrated and then cooled to -78°C to furnish 53a (2.2 g, 78%) as pale yellow crystals, m.p. 241 "C.
6.7.5 2,4,6-~i-tert-butyl-1,5-diphospha-3-phosphoniasp~ro[3.4]hexa-l,4-diene-6-trichloroaluminate(71a) To a suspension of aluminum chloride (0.67 g, 5.0 mmol) in dichloromethane ( 5 mL) was added dropwise at 0°C a solution of 9a (1.5 g, 15.0 mmol) in dichloromethane (5 mL). After 1 h at O"C, the mixture was allowed to warm to room temperature and stirring continued for 2 h. The solvent was removed under high vacuum, the residue was taken up in a small amount of toluene, and cooled to furnish 71a (2.06 g, 95%) as yellow crystals, m.p. 134°C (dec.).
198
6 Phosphaalkynes
6.7.6 2,5,6,8-Tetra-tert-butyl-l,3,4,7-tetraphosphatetracyclo[3.3.0.02~4.03~6]~~t-7-ene (76) To a suspension of 71a (1.16 g, 2.7 mmol) and aluminum chloride (0.59 g, 4.4 mmol) in dichloromethane (10 mL) at -78°C were added with stirring in succession 9a (0.26 g, 2.6 mmol) and then a solution of dimethyl sulfoxide (0.62 mL, 8.7 mmol) in dichloromethane (4 mL). The reaction mixture was allowed to warm to room temperature and stirring was continued for 18 h. After evaporation of the solvent under high vacuum the residue was extracted with five 10 mL portions of n-pentane. Concentration of the pentane extract furnished 76 (0.39 g, 38%). Purification by column chromatography (silica gel 0.063-0.2 mm, n-pentane) to a yellow oil (0.17 g, 16%) was possible but accompanied by high loss in yield.
6.7.7 1,4,6-'Lfi-tert-buty1-2,5,7,7,8,8-hexaethyl-5,8-dialuminato3-phospha-2,7-diphosphoniatetracyclo[3.3.O.O2~4.O3~6]octane (80) To a stirred solution of 9a (0.35 g, 3.5 mmol) in n-hexane (3 mL) at - 5 O O C was added dropwise a solution of triethylaluminum (0.23 g, 2.0 mmol) in n-hexane (3 mL). The reaction mixture was allowed to warm to room temperature within 12 h, the solvent was evaporated under high vacuum, and the residue crystallized from n-hexane at - 78 "C to furnish 80 (0.48 g, 91%) as colorless crystals, m.p. 175OC.
6.7.8 2,5,7,9-Tetra-tert-butyl-3~,4-triethyl-4-aluminato-3,6,8-triphosphal-phosphoniatetracyclo[4.2.1.0'i5.04~9]nona-2,7-diene (81) To a stirred solution of 9a (0.40 g, 4.0 mmol) in diethyl ether (3 mL) at - 78 "C was added dropwise a solution of triethylaluminum (0.23 g, 2.0 mmol) in diethyl ether (3 mL). The reaction mixture was allowed to warm to room temperature within 12 h, the solvent was evaporated under high vacuum, and the residue crystallized from a little diethyl ether at - 78 "C to furnish 81 (0.38 g, 74%) as red-brown crystals, m.p. 168°C.
Acknowledgement We thank the Volkswagenstiftung, the Deutsche Forschungsgemeinschaft (Graduiertenkolleg "Phosphor als Bindeglied verschiedener chemischer Disziplinen"), the Fonds der Chemischen Industrie, and the Landesregierung von Rheinland-Pfalz (Graduierten Stipendien) for generous financial support of our work.
References
199
Abbreviations Ad c-Hex CP CP* DMSO Mes PE PM3 thf
1-adamantyl cyclohexyl cyclopentadieny1 pentamethylcyclopentadienyl dimethyl sulfoxide mesityl photoelectron Parametric Method 3 tetrahydrofuran
References [I] H. Heydt, U. BergstrBBer, R. WRler, E. Fuchs, N. Kamel, T. Mackewitz, G. Michels, W. RCisch, M. Regitz, P. Mazerolles, C. Laurent, A. Faucher, Bull. SOC. Chim. Fr. 1994, in press. [2]F. Vogtle, Reizvolle Molekiile der Organischen Chemie, Teubner, Stuttgart, 1989, p. 28 et seq. [3]E. Osawa, 0. Yonemitsu, Carbocyclic Cage Compounds - Chemistry and Applications, VCH, Weinheim, 1992. [4] P. E. Eaton, T. W. Cole, J. Am. Chem. SOC. 1964,86,3157-3158. [5] T. J. Katz, N. Acton, J. Am. Chem. SOC. 1973, 95, 2738-2739. [6]P. E. Eaton, Y. S. Or, S. J. Branca, J. Am. Chem. SOC. 1981, 103, 2134-2136. [7]M. Baudler, K. Glinka, Chem. Rev. 1994, 94, 1273-1297. [8]M. Baudler, J. Hahn, V. Amdt, B. Koll, K. Kazmierczak, E. DSLrr, Z. Anorg. Allg. Chem. 1986,538, 7-20. [9]P. Jutzi, R. Kroos, A. Milller, H. Bijgge, M. Penk, Chem. Ber. 1991, 124, 75-81. [lo] Reviews: (a) M. Regitz, P. Binger, Angew. Chem. 1988,100. 1541-1565;Angew. Chem., Znt. Ed. Engl. 1988,27,1484-1507;(b) M. Regitz, Chem. Rev. 1990,90,191 -213;(c) M. Regitz in Multiple Bonak and Low Coordination in Phosphorus Chemistry (Eds.: M. Regitz, 0.J. Scherer), Thieme, Stuttgart, 1990, p. 58; (d) W.-G. Veeck, M. Regitz in Functional Group 7hnsformation (Ed.: C . Moody), Pergamon, Oxford, 1994,in press; (e) A. C. Gaumont, J. M. Denis, Chem. Rev. 1994,94,1413-1439. [ll] G. Becker, G. Gresser, W. Uhl, Z. Nuturforsch. Anorg. Chem. Org. Chem. 1981, 368, 16-19. [I21 W. Rtisch, U. Hees, M. Regitz, Chem. Ber. 1987, 120, 1645-1652. [13] (a) G. Becker, Z. Anorg. Allg. Chem. 1977,430,66-76;(b) G. Becker, Z. Anorg. Allg. Chem. 1976, 423, 242-254. [I41 T. Allspach, M. Regitz, G. Becker, W. Becker, Synthesis 1986, 31-36. [I51 Reports on various aspects of phosphaalkyne chemistry: (a) M. Regitz in Heteroatom Chemistry (Ed.: E. Block), VCH, New York, 1990, p. 295; (b) M. Regitz, Bull. SOC. Chim. Belg. 1992, 101, 359-379;(c) M. Regitz in Organic Synthesis via Orgunornetallics (Eds. : D. Enders, H.J.Gais, W. Keim), Vieweg, Braunschweig, 1993, p. 93; (d) M. Regitz, 1 Heterocycl. Chem. 1994,31,663-677. [16] 0.Wagner, G.Maas, M. Regitz, Angew. Chem. 1987,99,1328-1330;Angew Chem., Znt. Ed. Engl. 1987,26, 1257-1259. [17] Reviews: (a) H. Memmesheimer, M. Regitz in Reviews on Heteroatom Chemistry (Ed.: S. Oae), MYU, Tokyo, 1994,p. 61;(b) H. Memmesheimer, M. Regitz in Advances in Curbene Chemistry (Ed.: U. Brinker), Vol. 1, JAI Press, Greenwich, CT 1994, p. 185; (c) 0. Wagner, M. Ehle, M. Regitz, Angew. Chem. 1989,101,227-229;Angew. Chem,, Znt. Ed. Engl. 1989,28,225-226;(d) 0. Wagner,
200
6 Phosphaalkynes
M. Ehle, M. Birkel, J. Hoffmann, M. Regitz, Chem. Ber. 1991, 224, 1207-1213; (e) H. Memmesheimer, J. R. Al-Dulayyimi, M. S. Baird, T. Wettling, M. Regitz, Synlett 1991, 433-435. [18] (a) W. Rtisch, M. Regitz, Angew. Chem. 1984,96, 898-899; Angew. Chem., Int. Ed. Engl. 1984,23, 900-901; (b) W. Rosch, U. Hees, M. Regitz, Chem. Ber. 1987, 120, 1645-1652. [19] W. Rosch, T. Facklam, M. Regitz, Tetrahedron 1987, 43, 3247-3256. [20] W. Rosch, M. Regitz, Synthesis 1987, 689-693. [21] W. Rosch, H. Richter, M. Regitz, Chem. Ber. 1987, 120, 1809-1813. [22] B. Burkhart, S. Krill, Y. Okano, W. Ando, M. Regitz, Synlett 1991, 356-358. [23] W. Rosch, M. Regitz, Z. Naturforsch. Anorg. Chem. Org. Chem. 1986, 4IB, 931-933. (241 U. Annen, M. Regitz, Tetrahedron Lett. 1987, 28, 5141-5144. [25] U. Annen, M. Regitz, Etrahedron Left 1988, 29, 1681-1684. (261 E. P. 0. Fuchs, W. Rosch, M. Regitz, Angew. Chem. 1987, 99, 1058-1059; Angew. Chem,, Inf. Ed. Engl. 1987, 26, 1011-1012. [27] H. Heydt, U. BergstraBer, R. Fafiler, E. Fuchs, N. Kamel, T. Mackewitz, G. Michels, W. Riisch, M. Regitz, P. Mazerolles, C . Laurent, A. Faucher, Bull. Soc. Chim. Fr. 1994, in press. [28] (a) P. Binger, R. Milczarek, R. Mynott, M. Regitz, W. Rosch, Angew. Chem. 1986, 98, 645-646; Angew. Chem., Int. Ed. Engl. 1986,25,644-645; (b) P. Binger, R. Milczarek, R. Mynott, C. Kriiger, Y. H. Tsay, E. Raabe, M . Regitz, Chem. Ber. 1988, 121, 637-645; (c) review: P. Binger in Multiple Bonds and Low Coordination in Phosphorus Chemistry (Eds. : M. Regitz, 0. J. Scherer), Thieme, Stuttgart, 1990, p. 90. [29] Report: J. Emsley, New Scientist November 25th 1989, p. 33. [30] T. Wettling, J. Schneider, 0. Wagner, C. G. Kreiter, M. Regitz, Angew. Chem. 1989, IOI, 1035-1037; Angav. Chem., Znt. Ed. Engl. 1989, 28, 1013-1014. [31] (a) D. Hu, H. Schiiufele, H. Pritzkow, U. Zenneck, Angew. Chem. 1989, 101, 929-931; Angew. Chem., Int. Ed. Engl. 1989,28,900-902; (b) see also U. Zenneck, Angew. Chem. 1990,102, 171-182; Angew. Chem, Znt. Ed. Engl. 1990, 29, 126-137. [32] R. Bartsch, P. B. Hitchcock, J. F. Nixon, J. Organomet. Chem. 1989, 375, C31-C34. (331 R. Bartsch, P. B. Hitchcock, J. F. Nixon, J. Chem. Soc, Chem. Commun. 1989, 1046-1048. [34] G. Mtirkl in Multiple Bonds and Low Coordination in Phosphorus Chemistry (Eds. : M. Regitz, 0. J. Scherer), Thieme, Stuttgart, 1990, p. 220. [35] (a) S. Masamune, N. Nakamura, M.Suda, H. Ona, J. Am. Chem. SOC.1973,95, 8481-8483; (b) optimized procedure: H. Wingert, M. Regitz, Chem. Ber. 1986, 119, 244-256. [36] J. Fink, W. Rdsch, U.-J. Vogelbacher, M. Regitz, Angew. Chem. 1986, 98, 265-266; Angew. Chem., Znt. Ed. Engl. 1986, 25, 280-282. [37] (a) H. Wingert, M. Regitz, Chem. Ber. 1986, 219, 244-256; (b) H. Wingert, M. Regitz, 2. Naturforsch. Anorg. Chem. Org. Chem. 1986, 41& 1306-1310; (c) H. Wingert, G. Maas, M. Regitz, Tetrahedron 1986, 42, 5341-5353. [38] (a) K. Blatter, W. Rosch, U.-J. Vogelbacher, J. Fink, M. Regitz, Angew. Chem. 1987, 99, 67-68; Angew. Chem, Znt. Ed. Engl. 1987, 26, 85-86; (b) K. Blatter, Diploma Thesis, University of Kaiserslautern, 1986. [39] U.-J. Vogelbacher, M. Ledermann, T. Schach, G. Michels, U. Hees, M. Regitz, Angew. Chem. 1988, 100, 304-306; Angew. Chem., Znf. Ed. Engl. 1988, 27, 272-274. I401 S. Haber, Diploma Thesis, University of Kaiserslautern, 1988. [41] (a) G. Msirkl, D. Matthes, Angew. Chem. 1972,84, 1069-1070; Angew. Chem., Znt. Ed. Engl. 1972, I I , 1019-1020; (b) G. Mi-irkl, G. Dorfmeister, Tetrahedron Lett. 1987, 28, 1093-1096. 1421 (a) A. Marinetti, L. Ricard, F. Mathey, M. Slany, M. Regitz, Tetrahedron 1993,49, 10279-10290; (b) T. Mackewitz, Diploma Thesis, University of Kaiserslautern, 1993. [43] M. Lazraq, J. Escudik, C. Couret, U. Bergstmer, M. Regitz, J. Chem. Soc, Chem. Cornmun. 1993, 569-570.
[44]R. FaBler, Thesis, University of Kaiserslautern, 1992.
References
201
[45] M. Julino, Thesis, University of Kaiserslautern, 1994. [46] E. Niecke, R. Streubel, M. Nieger, D. Stalke, Angew. Chem. 1989,101, 1708-1710; Angew. Chem., Znt. Ed. Engl. 1989,28, 1673-1674. [47] F, Mercier, L. Ricard, F. Mathey, M. Regitz, J. Chem. Soc., Chem. Commun. 1991, 1305-1307. (481 F. Mercier, B. Deschamps, F. Mathey, J. Am. Chem. Soc. 1989, 111, 9098-9100. [49] M. Slany, Thesis, University of Kaiserslautern, 1994. (501 (a) T. Wettling, B. GeiBler, R. Schneider, S. Barth, P. Binger, M. Regitz, Angew. Chem. 1992, 104, 761-762; Angew. Chem., Znt. Ed. Engl. 1992,31, 758-759; (b) B. Geinler, T. Wettling, S. Barth, P. Binger, M. Regitz, Synthesis 1994, 1337-1343. [51] T. Wettling, Thesis, University of Kaiserslautern, 1990. [52] R. Gleiter, K.-H. Pfeifer, M. Baudler, G. Scholz, T. Wettling, M. Regitz, Chem. Ber. 1990, 123, 757-760. [53] P. Binger, B. Biedenbach, C. Krilger, M. Regitz, Angew. Chem. 1987,99,798-799; Angew. Chem., Int. Ed. Engl. 1987,26, 764-765. [54] P. Binger, T. Wettling, R. Schneider, F. Zurmiihlen, U. BergstraRer, J. Hoffmann, G. Maas, M. Regitz, Angew. Chem. 1991, 103, 208-211; Angew. Chem., Znt. Ed. Engl. 1991,30, 207-210.
1551 B. GeiBler, S. Barth, U. Bergstraner, M. Slany, J. Durkin, P. B. Hitchcock, M. Hofmann, P. Binger, J. F. Nixon, P. Von Ragu6 Schleyer, M. Regitz, Angew. Chem. 1994, 106, 485-488; Angem Chem., Znt. Ed. Engl. 1994,33, 484-487. [56] B. Geinler, Thesis, University of Kaiserslautern, 1993. [57] K. K. Laali, M. Regitz, M. Birkel, P. J. Stang, C. M. Critell, J. 0%.Chem. 1993,58, 4105-4109. [58] M. Birkel, J. Schulz, U. BerstraBer, M. Regitz, Angew. Chem. 1992, 104, 870-873; Angew. Chem, Znt. Ed. Engl. 1992,31, 879-882. [59] X.-B. Ma, M. Birkel, M. Regitz, Heteroatom Chem. 1994, 4, in press. [60]B. Breit, Thesis, University of Kaiserslautern, 1993. [61] B. Breit, U. BergstrLDer, G. Maas, M. Regitz, Angew. Chem. 1992, 104, 1043-1046; Angew. Chem., Znt. Ed. Engl. 1992,31, 1055-1058. [62] (a) P. B. J. Driessen, H. Hogeveen, J. Am. Chem. Soc. 1978, 100, 1193-1200; (b) H. Hogeveen, R. F. Kingma, D. M. Kok, J. Org.Chem. 1982,47, 989-997; (c) P. B. J. Driessen, H. Hogween, . l Organomet. Chem. 1978, 156, 265-218. [63] K. Laali, B. Geinler, M. Regitz, unpublished results, Kent State University, 1992. [64] B. Breit, A. Hoffmann, U. BergstrhBer, L. Ricard, F. Mathey, M. Regitz, Angew. Chem. 1994, 106, 1541-1543; Angew. Chem, Znt. Ed. Engl. 1994,33, 1491-1493. [65] A. Hoffmann, Projected Thesis, University of Kaiserslautern, 1995. [66] F. Uhlig, S. Gremler, M. Dargatz, M. Scheer, E. Herrmann, Z. Anorg. Allg. Chem. 1991, 606, 105-108.
This Page Intentionally Left Blank
7 The Enediyne Antibiotics K. C. Nicolaou, Adrian L. Smith
7.1 Introduction Since the beginning of history mankind has striven to emulate the wonders of Nature, developed over millions of years of evolution. According to Greek mythology Icarus, after observing the flight of birds, became the first person to fly by fashioning wings out of feathers and wax and attaching them to his arms. Although Icarus’s early attempts were brought to a premature end when he flew too close to the sun and the wax in his wings melted, the idea of emulating the birds remained alive and was instrumental in the pioneering days of powered flight at the turn of the twentieth century when human science and technology had progressed sufficiently to enable a scientific analysis of the problem. The following century has seen man progress from emulating Nature, through jet-powered flight, to being in a position to reach for the stars. If Nature had not first solved the problem of flight, when would mankind have hat the imagination to develop flying machines and would the human race be where it is today? Who knows? There is one major scientific story however - that of the enediyne antibiotics (Fig. 7-1) - which almost certainly would not have developed without the guiding hand of Nature [l-51. This is a rapidly developing field which may have major implications for revolutionary biomedical applications, most notably in the treatment of certain types of cancer. As will be shown in the following chronicle, which reviews the field up to the end of 1993, the groundwork for this story was laid in the early 1970s. It was then that Robert Bergman first studied the thermal cycloaromatization of a (Z)-1,5-diyn-3-ene system (hereafter called an “enediyne”), demonstrating the intermediacy of a highly reactive 1,4-benzenoid diradical species in the process [6-91. For a quarter of a century little was made of this seminal work until the mid-1980s when the first members of a new series of potent naturally occuring anticancer antibiotics, the enediyne antibiotics, were discovered. As their name implies, these molecules contain systems related to the enediyne studied by Bergman, and their phenomenal biological profiles are, at least in part, attributable to the same cycloaromatization reaction. The diradical intermediate is a powerful weapon used by these antibiotics to literally rip hydrogen atoms from the sugar phosphate backbone of a target cell’s DNA, leading to the destruction of the cell’s genetic material and ultimately its death. Once the structures of the first natural enediyne antibiotics had been elucidated, the scientific community was quick to respond. The mechanisms by which these molecules exert their effects continue to be probed ever deeper, with surprising new details still emerging. The hitherto unprecedented and structurally complex molecular architecture of the natural products has required the development of new synthetic methodology in order to synthesize functional models which would shed new light upon the natural enediyne antibiotics. Researchers in the field have pushed the art of organic synthesis to its present limits with total syntheses of calicheamicin yi (2), the most prominent member of the natural enediyne antibiotics, and a derivative of dynemicin A (4) having been achieved. Furthermore, the factors affecting the
204
7 The Enediyne Antibiotics
cycloaromatization and DNA-damaging properties of simple model systems have been examined, leading to the development of a second generation of designed enediyne antibiotics with, perhaps, improved potential over the natural products for the treatment of cancer.
Esperamicin A1 (3)
Neocarzinostatin
chrornophore(1)
"'"y;X 0
0
(2-1027 chromophore (6) Figure 7-1 Some naturally occurring enediyne antibiotics.
Kedarcidin chromophore (5)
7.2 The Aromaticity Era
205
7.2 The Aromaticity Era 7.2.1 The Cycloaromatization of Conjugated Polyenyne Systems The story of the enediyne antibiotics really starts back in the 1960s and 1970s in the laboratories of three groups led by Sondheimer, Masamune, and Bergman. Probably the first observation of relevance came in a report by Mayer and Sondheimer in 1966 in which the thermal base-induced elimination of the bis-mesylate 7 (Scheme 7-1) was found to be accompanied by a cycloaromatization to give initially 8 and finally 9 [lo]. The second observation of relevance came shortly afterwards, in 1971, when Masamune reported the spontaneous cycloaromatization of the system 11, formed by base-induced elimination of the bis-mesylate 10 (Scheme 7-2) [ll]. Although neither report detailed the mechanisms of these observations, both certainly involved diradical intermediates, and it was Robert Bergman who first studied the mechanism of the cycloaromatization, thus giving his name to the reaction. Bergman's findings, reported in 1972, are summarized in Scheme 7-3 and clearly demonstrate that the
Scheme 7-1 The original observation by Mayer and Sondheimer (1966).
c
10
=
\
(a),
c
L
J
11
12
(b)
Scheme 7-2 The observations of Masamune et al. (1971).
13
14
15
16
Scheme 7-3 The system designed by Bergman to study the enediyne cycloaromatization reaction (1972).
206
7 The Enediyne Antibiotics
cycloaromatization of the enediyne 13 occurs via a 1,4-benzenoid diradical [6-91. Subsequently, Wong and Sondheimer demonstrated the facile cycloaromatization of the strained cyclic enediyne 19 (Scheme 7-4) [12]. When the reaction was carried out in the presence of d,-THF, deuterium was incorporated into the product 21, thus clearly demonstrating the presence of a highly reactive, 1,4-benzenoid diradical intermediate (20) capable of stripping hydrogen atoms from the solvent molecules.
+
[m] 20
D
21
Scheme 7-4 Observation by Wong and Sondheimer (1980).
7.2.2 Application to the Synthesis of Aromatic Systems Given the efficiency with which polyenyne systems can be synthesized and induced to undergo cycloaromatization in the presence of suitable reagents to quench the intermediate diradicals, there have been surprisingly few reports detailing the use of such reactions for the specific purpose of synthesizing aromatic systems. Grissom and Calkins utilized the Bergman cycloaromatization of 22 (Scheme 7-5) as a means of generating a radical for carrying out a subsequent radical ring annulation reaction to give 24 [13, 141. Another approach by Magriotis and Kim utilized a tandem Ireland-Claisen/Bergman rearrangement strategy to effect a
<*1,rl-cyclohexadiene
PhCl I210 “C I24 h
\\
22
23
24
(a) n = 1; R = H (72%) (b) n = 1; R =CHpOTBS (58%) (c) n = 2; R = H (53%) Scheme 7-5 Tandem Bergman/radical cyclization strategy for ring annulation (Grissom and Calkins).
7.3 The Discovery of the Enediyne Antibiotics
207
stereocontrolled tetrahydronaphthalene synthesis (Scheme 7-6) [15]. A particularly novel approach to systems related to the antitumor agent taxol has been proposed by Fallis (Scheme 7-7), in which the cycloaromatization of the enediyne 29 (termed a taxamycin by Fallis) is anticipated to give the aromatic taxane 30 [16]. The feasibility of this approach has yet to be demonstrated, however.
Go -a. TBSO Me
TBSO
pMe
LiHMDS
\ \
(a) R (b) R
0 25 = H (50%) = SPh (45%)
\
'Pr3SiOTf THF I -78 oc
-
26
&C02Si'Pr3
\
OSi'Pr3
1,4-cyclohexadiene PhH I 140 "C I 5 h
6 \ \
Co2SiiPr3
27
28
Scheme 7-6 Tandem ClaisedBergman rearrangement strategy for stereocontrolled tetrahydronaphthalene synthesis (Magriotis and Kim).
RO 29
30
Scheme 7-7 A proposed Bergman cycloaromatization approach to taxol-related systems (Fallis et d.).
7.3 The Discovery of the Enediyne Antibiotics In 1987 the announcement of the structures of two new classes of antitumor antibiotics, the calicheamicins [17-191 and esperamicins [20,211, aroused a great deal of interest in chemical and biological circles. The newly discovered compounds combined unprecedented and intriguing molecular structures with striking biological activities. At the same time a remarkable mechanism of action was proposed for the molecules to account for their phenomenal biological profiles [17-211, recalling the earlier work of Bergman. For the molecules contained an enediyne unit embedded within their complex architectural frameworks, and it was proposed that these molecules delivered the enediyne portion within the minor groove of a target
208
7 The Enediyne Antibiotics
cell’s DNA and then initiated a series of reactions leading to cycloaromatization of the enediyne and lP-benzenoid diradical formation. The highly reactive 1,4-benzenoid diradical would then be perfectly positioned to strip hydrogen atoms from the sugar phosphate backbone of adjacent strands of DNA, causing scission of the DNA double helix. Synthetic chemists throughout the world immediately became interested in these structures. Furthermore, the related neocarzinostatin chromophore (l),whose structure had been elucidated two years previously but which had failed to fully capture the chemist’s imagination, was now viewed with renewed interest because of its structural, biological, and mechanistic similarities with the newly discovered enediynes [22, 231. The ensuing years have seen the emergence of further classes of enediyne antibiotics - the dynemicins in 1989 [24, 251, kedarcidin in 1992 [26-281, and C-1027 [29-341 and maduropeptin [35, 361 in 1993. Our present understanding of these different classes of enediyne antibiotics are detailed in the following sections. No doubt the coming years will reveal further details about these and other, as yet undiscovered, systems.
7.3.1 Neocarzinostatin Neocarzinostatin (NCS), first reported by Ishida et al. in 1965 [22], is a 1 : 1 noncovalently associated mixture of a protein component (NCS apoprotein) and a chromophoric molecule (NCS chromophore). The mixture was separated somewhat later into its component parts [37-391 and eventually characterized structurally. The chromophoric component was shown in 1985 to have the novel bicyclic polyeneyne skeleton 1 by Edo et al. [23], the apoprotein has been characterized as a 113 amino acid polypeptide based upon the gene base sequence [40] and apoprotein crystal structure [41], the three-dimensional solution structure of intact neocarzinostatin has been determined by Hirama and co-workers using two-dimensional NMR techniques (Fig. 7-2) [42], and the crystal structure of holo-NCS has been reported by
Figure 7-2 Space-fillingmodel of the best refined structure of the NCS complex. Apo-NCS residues with significant backbone and/or side-chain chemical shift changes ( 50.1 ppm) upon chromophore binding are yellow, residues with intermolecular NOE with NCS-chr are red, and the remaining residues are white. NCS-chr is shown in blue. Only heavy atoms are shown for NCS-chr. [Reproduced with permission from: T. Tanaka, M. Hirama, K.4. Fujita, S. Imajo, M. Ishiguro, J Chem. SOC, Chem. Commun. 1993, 1205.1
7.3 The Discovery of the Enediyne Antibiotics
209
the groups of Myers and Rees [43]. Until recently it appeared that the mechanism-based biological activity of NCS resides primarily in the chromophore, whilst the apoprotein serves to both stabilize the chemically sensitive chromophore and act as a transporter. Recent studies, however, suggest that the apoprotein may also contribute actively to the cytotoxicity through selective proteolytic activity [44]. The NCS apoprotein binds tightly and specifically to NCS chromophore (KD= 1 x lo-'' M) [45] and delivers the active chromophore to its target, DNA, by controlled release [46]. The biological activities of NCS include potent antitumor and antibacterial actions and are exerted by DNA cleavage. The DNA-damaging activity of the free NCS chromophore results primarily in single-strand DNA cuts and proceeds via an oxygen-dependent reaction [47]. Thiols [48] and UV radiation [49] greatly enhance the DNAcleaving properties of NCS chromophore. The mechanism by which NCS chromophore exerts its DNA-damaging properties was first suggested by Myers in 1987 [50]. According to this proposal (Scheme 7-8, path A) the cascade of reactions leading to DNA damage is initiated by stereospecific nucleophilic attack at C (12). This triggering event is accompanied by rearrangement of the ring skeleton with epoxide opening and formation of cumulene 31 as shown in Scheme 7-8. This highly strained and reactive intermediate then undergoes rapid cycloaromatization to form diradical 32, which proceeds to attack DNA by hydrogen atom abstraction, resulting in 33. Corroboration of this scenario has been provided by using HSCH2C02Me [50-531 and NaBH, [53] as nucleophiles in invitro experiments. The cumulene intermediate 31 has been observed by NMR at low temperature [52], and the methyl thioglycolate adduct has been isolated and fully characterized, including its absolute configuration [51]. Evidence that the basic methylamino side chain on the sugar residue assists in the thiol addition at C(12) through base catalysis was provided by Myers, who derivatized the amine as the corresponding nitrosamine; the NCS chromophore derivative thus produced was inert to thiols below 0°C [54]. Valuable additional information has been provided by the three-dimensional solution structure of intact NCS [42] - the aminomethyl group of the sugar appendage is forced into close proximity to C(12) (4.3 A) due to a salt bridge with Asp-33, suggesting that nucleophilic attack at C(12) may be assisted by this basic nitrogen; together with additional steric hindrance at C(12) from the side chains of Ser-98, Asp-33, and Phe-52 and the positioning of the epoxide in a hydrophobic pocket away from an acid catalyst, this suggests how the apoprotein serves to stabilize the chromophore. Furthermore, the bicyclic core lies on a disulfide bridge which may stabilize the strained uninteraction. saturated system of the chromophore through a HOMOdienediyne-LUMOdisulfide This disulfide bridge is conserved in all chromoprotein antibiotics and may be a common stabilizing feature. A second, distinct, cycloaromatization pathway has been recently observed when NCS chromophore is incubated with 2-mercaptoethanol in the presence of the apoprotein in which the zwitterionic intermediate 35 (Scheme 7-8, path B) is indicated [55, 561, although this mechanism probably does not operate for the free chromophore and Chin and Goldberg reported that this pathway is not responsible for DNA cleavage [57]. It is generally accepted that the NCS chromophore intercalates into DNA via its naphthoate side chain, which positions the rest of the molecule within the minor groove [58-601. Another role for the a-hydroxynaphthoate has been suggested in the activation of NCS chromophore (Scheme 7-9) [61]. Here, the hydroxyester participates in epoxide opening and cumulene formation. Again, the three-dimensional solution structure of intact NCS provides invaluable information, showing that this naphthoate is kept well away from the bicyclic core until the chromophore is released from the apoprotein [42].
LE
I
7.3 The Discovery of the Enediyne Antibiotics
211
Much work has gone into identifying the details of DNA damage by the NCS chromophore diradical 32. It has been demonstrated that at least 80% of the DNA cleavage leads to the 5’-aldehyde of A and T residues selectively [621, The chemistry leading to these breaks involves hydrogen atom abstraction from C(5‘) of deoxyribose and reaction with molecular oxygen as outlined in Scheme 7-10. Less than 20% of the strand breaks result from hydrogen atom
7
7
HO-P=O
0
HO-P=O
[5
HO-P=O
1.02
H
O
O
V
ArH
&l
17
Reduction
HO-P=O
7
HO-Y=O OH
+
? HO-P=O
H29 ?
HO-P=O
&I
&I
Scheme 7-10 DNA strand cleavage initiated by C (5’) hydrogen atom abstraction.
abstraction at C(4’) [63-671 and C(1’) [a] (Schemes 7-11 and 7-12). Chin and Goldberg have shown that the radical at C(2) of 32 is particularly susceptible to both internal and external quenching (up to 70% under physiological conditions) [68], accounting for the observation that NCS chromophore effects primarily single-stranded DNA cuts by the C (6) radical at C (5’) of deoxyribose, whilst those double-stranded lesions which are observed involve additional hydrogen abstraction by the C(2) radical from C(1’) or C(4‘) of the deoxyribose on the complementary strand. Further insight into the interaction of NCS chromophore with DNA has recently been provided with the observation that a thiol-independent cleavage mode is possible with single-stranded DNA bulges (regions where double-stranded structures are generated intramolecularly) [69].These findings imply that the DNA in these cases is an active participant in its own destruction, since DNAs containing point mutations which disrupt the bulge are not cleavage substrates. Biosynthetic studies [70]with NCS chromophore showed that the C1, dienediyne ring skeleton incorporates six intact acetate units and two terminal acetate units which suffer C - C
212
7 The Enediyne Antibiotics
bond cleavage, and suggested that the C14chain is derived from degradation of oleate via the oleate-crepenylate pathway for polyacetylenes. Similar proposals have been made for the C,, ring skeleton of the calicheamicin/esperamicin families [70].
7.3.2 The Calicheamicins The calicheamicins (Table 7-1) are a family of enediyne antibiotics isolated from Micromonosporn echinospora ssp. calichensis and were discovered by Lee et al. (Lederle Laboratories) through a program aimed at identifying microbial fermentation products active in the biochemical induction assay (an assay exquisitely sensitive to certain DNA-damaging antitumor agents) [19]. The structures were first elucidated in 1986 and reported the following year [17-191; one stereochemical center was revised in 1989 [71, 721. Calicheamicin 7 ; (2) is the most prominent member of this class of compounds. The iodine-containing calicheamicins were discovered when sodium iodide was added to the fermentation broth, resulting not only in the new compounds containing iodine rather than bromine, but also in much improved yields [71, 721. The calicheamicins are active in the biochemical induction assay at concentrations below 1 pg/mL, extremely active against Gram-positive bacteria, and
P
HO-F=O
? HO-P=O
&
I
ArH
2. [H*]
? HO-P=O &I
OH HO-P=O
+
I
? HO-P=O
I/ Reduction
HO-Y=OO
n
&I
--OH
HO-kO
U
I P
HO-&O
c3 Scheme 7-11 DNA strand cleavage initiated by C(4’) hydrogen atom abstraction.
7.3 The Discovery of the Enediyne Antibiotics
oy1~ -
HO-Y=O
HO-P=O
HO-Y=O 0
[A?
1.02
2.[Ha]
ArH
HO-P=O
+1
l3
0
,
Reduction
7
7
7
HO-P=O
HO-P=O
HO-P=O
0
0 HO-i=O
I
HO-bO
9
9 Z H e0 - g47
6
6
6
+
HO-P=O
H? HO-P=O
r3
ki
BH
0 1 HO-P=O
A
Scheme 7-12 DNA strand cleavage initiated by C(1') hydrogen atom abstraction. Table 7-1 The calicheamicin family
Calicheamicin
X
Rl
RZ
Calicheamicin bIBr Calicheamicin yl Br Calicheamicin a,I Calicheamicin a3I Calicheamicin p1I Calicheamicin y1I (2) Calicheamicin 6, I
Br Br I I I I I
Rha (a) Rha H Rha Rha Rha Rha
Ami Ami Ami H Ami Ami Ami
R3
0)
CHMe, Et Et CHMe, Et Me
213
214
7 The Enediyne Antibiotics
highly active against Gram-negative bacteria [17, 18, 71, 721. Most importantly, they exhibit extraordinary potency against murine tumors such as P338 and L1210 leukemias and solid neoplasms such as colon 26 and B-16 melanoma with optimum doses of 0.15-5 pg/kg [73]. These compounds are thought to exert their biological activites by damaging DNA. Indeed, the calicheamicins are highly potent DNA-cleaving agents, giving rise primarily to sequencespecific double-strand cuts [74, 751. There is also some suggestion that proteins might be a second target for calicheamicin y: (2), although the evidence is not convincing [76]. Calicheamicin y: (2) is a remarkable piece of engineering by Nature, which has perfectly constructed the molecule to endow it with its extraordinary chemical and biological properties. The molecule contains two distinct structural regions. The larger of the two consists of an extended sugar residue comprising four monosaccharide units and one hexasubstituted benzene ring which are joined together through a highly unusual series of glycosidic, thioester, and hydroxylamine linkages. The second structural region, the aglycone (termed calicheamicinone), contains a compact, highly functionalized bicyclic core housing a strained enediyne unit within a bridging ten-membered ring. The structure of the drug, including its absolute configuration, has been confirmed by total synthesis by Nicolaou et al. [77]. The aryloligosaccharide serves to deliver the drug to its target, binding tightly in the minor groove of double-helical DNA (see Fig. 7-3) and a displaying high specificity for sequences
Figure 7-3 Computer-generated models of free calicheamicin y: (2) (top) and DNA-bound 2 (bottom). [Reproduced with permission from: K. C. Nicolaou, W.-M. Dai, Angew. Chem, Int. Ed. Engl. 1991,30, 1387.1
7.3 The Discovery of the Enediyne Antibiotics
215
such as 5’TCCT-3‘ and 5’-TTTT-3’ through significant hydrophobic interactions and other forces [74, 75, 78-84]. This binding is thought to be facilitated by substantial preorganization of the oligosaccharide into a rigid, extended conformation [81], and the DNA appears to distort upon binding in order to widen the minor groove and accommodate the drug (induced fit) [79, 82, 831. The specificity of binding appears to be associated with the hydrophobicity of the entire drug; however, it would be wrong to assume that the sequence recognition involves contacts made only by the aglycone or the aryloligosaccharide [85-871. Molecular modeling calculations by Schreiber and co-workers suggest that a significant portion of the sequence selectivity for 5’-TCCT-3’ arises from a favorable interaction between the large and polarizable iodo substituent of the hexasubstituted aromatic ring and the exocyclic amino substituents of the two guanines in the 3‘-AGGA-5’ tract [88]. Nicolaou et al. provided experimental support for this idea of the importance of the iodine in the DNA-binding affinity of the oligosaccharide by carrying out DNA footprinting experiments with synthetic oligosaccharide analogs [87]. As mentioned above, the aglycone is a rigid, highly functionalized bicyclic core. The enediyne moiety is locked within a rigid ten-membered bridging ring awaiting activation to undergo the Bergman reaction [7]. Also forming part of the aglycone is an allylic trisulfide, which serves as a trigger. Once the molecule is in the vicinity of DNA (it has yet to be proved that binding occurs first), a series of chemical events unfolds which eventually leads to DNA cleavage. A nucleophile (e. g., glutathione) attacks the central sulfur atom of the trisulfide group, causing the formation of a thiol which adds intramolecularly to the adjacent a , P-unsaturated ketone embedded within the framework of the aglycone (Scheme 7-13) [18, 891. This
H
H 0 i0I - C
-
4 ‘
O
“Sugar
i
“Sugar
WSMe
HS:
38
\-Nu Calicheamicin y,’ (2)
Hot..
I
0
b
J
,/ SE ‘’Sugar
diradical
40
41 strand cleavage
Scheme 7-13 Mechanism by which calicheamicin y: (2) cleaves DNA.
39
216
7 The Enediyne Antibiotics
reaction, converting a trigonal bridgehead position to a tetragonal center, causes a significant change in structural geometry which imposes a great deal of strain on the ten-membered enediyne ring. This strain is completely relieved by the enediyne undergoing the Bergman reaction [7], generating the highly reactive lP-benzenoid diradical. The calicheamicin diradical abstracts hydrogen atoms from duplex DNA at the C(5‘) position of the cytidine in 5’-TCCT-3’ and the C(4’) position of the nucleotide three base pairs removed on the 3’ side of the complementary strand [90-921, leading to cleavage of both strands of DNA [93]. Carefully designed kinetics experiments by Townsend and co-workers [94-961 have revealed considerable information about the sulfur chemistry involved in the thiol-induced conversion of 2 into 38, indicating complex chemistry involving the formation of intermediate mixed disulfides and trisulfides. The rate-determining step in the process appears to be thiolate formation from 38, with subsequent rapid 1P-addition to give the dihydrothiophene 39, which has been observed at low temperature by NMR and is estimated to cycloaromatize at 37 “C with a half-life of 4.5 f 1.5 s and a free energy of activation from 39 + 40 of 19.3 & 0.2 kcal/mol [94]. It has been suggested that the ethylamino side chain of the E-ring of the calicheamicin y: oligosaccharide plays a role as a general-base catalyst in the rate-limiting thiolate formation step [75, 891, although recent evidence suggests that this is not the case and that the amine simply enhances the affinity of the drug for DNA through an energetically favorable ionic interaction 1961.
7.3.3 The Esperamicins The esperamicins (Table 7-2) are a subclass of naturally occurring enediynes with extremely high activities as broad-spectrum antibiotics and antitumor agents. They were isolated by Konishi et al. (Bristol-Myers) from the fermentation broth of Actinomuduru verrucososporu and first reported in 1985 [97]. The producing organism was collected at Pto Esperanza in Argentina, from which the series derives its name. Their structural elucidation was reported two years later [20, 211 simultaneously with the calicheamicins, to which they are closely related, sharing an almost identical bicyclic enediyne core (esperamicin contains an additional hydroxyl group), and having sugar appendages with similarly unusual structural motifs. The absolute configuration of esperamicin A, (3) and its relatives has been determined [98- 1001, and is the same as that of the calicheamicins. The esperamicins, together with the calicheamicins, are among the most powerful antitumor agents known, exhibiting powerful activity against a number of murine tumor models such as P388, B16, M5076 at injected doses in the 0.1 pg/kg range [loll. The esperamicins are thought to exert their biological action through cleavage of DNA in an almost identical manner to the calicheamicins (see Fig. 7-4). Esperamicin A, (3), however, exhibits less sequence selectivity than calicheamicin y i (2) and shows a preference for T > C > A > G [102-1041. As in the case of calicheamicin y: (Z), the DNA-cleaving ability of the esperamicins is significantly enhanced by thiols, resulting in a mixture of single- and double-stranded cuts. It has been demonstrated that C(5’) and C(4‘) hydrogen abstractions from DNA are the major chemical events initiated by the esperamicins. Esperamicin Al (3) itself effects almost exclusively single-stranded DNA cuts, and it appears that this is due to the fucosyl-anthranilate side chain inhibiting the C(4’) hydrogen atom abstraction [104]. Light-induced DNA cleavage by esperamicin Al (3) has also been demonstrated [49].
7.3 The Discovery of the Enediyne Antibiotics
217
Table 7-2 The esperamicin family
~
Esperamicin
n
Rl
R2
R3
Esperamicin A, (3) Esperamicin A,, Esperamicin A,, Esperamicin P Esperamicin A2 Esperamicin AZb Esperamicin A2,
3 3 3
H H H H
Ar (a) Ar Ar
Ar
H H H
CHMe, Et Me CHMez CHMe, Et Me
4
3 3 3
Ar Ar
Ar
7.3.4 The Dynemicins Dynemicin A (4), the first member of the dynemicin class of enediyne antibiotics to be discovered, is a violet-colored solid isolated from the fermentation broth of Micromonospora chersina. The structure was first reported by Konishi et al. in 1989 [24, 251 and immediately attracted attention due to its novelty, combining an anthraquinone (reminiscent of the anthracycline antiobiotics [lOS]) with a ten-membered bridging enediyne ring. The X-ray structure of dynemicin A (4) [25] shows that the anthraquinone portion of the molecule is puckered rather than flat. Dynemicin A (4) exhibits high potency against a variety of cancer cell lines and significantly prolongs the life span of mice inoculated with P388 leukemia and B16 melanoma cells [25]. Furthermore, dynemicin A (4) and its derivatives exhibit promising invivo antibacterial activity with low toxicity [24]. Subsequently, a second member of this family, deoxydynemicin A (42, Table 7-3), a bioactive compound with a similar profile to dynemicin A (4), was isolated from Micrornonosporaglobosa MG331-hF6 [106]. The absolute
218
7 The Enediyne Antibiotics
Figure 7-4 Computer-generated models of free esperamicin A, (3) (top) and o 3 on approaching DNA (bottom). [Reproduced with permission from: K. C. Nicolaou, W.-M. Dai, Angew. Chem., Int. Ed. Engl. 1991, 30, 1387.1
Table 7-3 The dynemicin family
R
0
OH
Compound
Name
R
4 42
Dynemicin A Deoxydynemicin A
OH H
configuration of these systems, although inferred from computer modeling studies [107], has not yet been determined. Biosynthetic studies indicate that dynemicin A (4) is biosynthesized from two heptaketide chains, which form the bicyclic enediyne core and the anthraquinone moiety, respectively
7.3 The Discovery of the Enediyne Antibiotics
219
[108]. Both are formed from seven head-to-tail coupled acetate units, whilst the carboxyl group is derived from one C-atom on an acetate unit and the 0-Me group from methionine. Dynemicin A (4) cleaves duplex DNA, causing both single- and double-stranded cuts [103, 109-1141. The potency as a DNA-cleaving agent is significantly enhanced by thiols [lo91 and by visible light irradiation [110], and it preferentially attacks the 3’ side of purine bases such as 5‘-AG, 5’-GC, 5’-GT and 5-AT with clear selection for G and, to a lesser extent, A [109, 1101. Intercalators and minor-groove binders interfere with DNA cleavage, suggesting both intercalation and minor groove binding for this agent [109]. It is suggested [lo91 that intercalation of the anthraquinone portion of dynemicin A (4) into the target DNA via the minor groove is the first step in a series of events leading to DNA damage by dynemicin A (see Fig. 7-5). This intercalation is accompanied by a local distortion of the DNA double helix in order to accommodate the drug [115, 1161, with the molecule recognizing conformationally flexible regions of DNA and acting as a “molecular wedge” [117]. The anthraquinone then undergoes bioreduction (Scheme 7-14) to give the anthraquinol 43. The electron-rich anthraquinol is then able to open the epoxide moiety by electron push as shown, perhaps being assisted by transfer of the acidic phenolic proton to the neighboring basic nitrogen atom, to generate a quinone methide intermediate (44). This is then either trapped by a nucleophile such as H,O (path A) or protonated (path B), resulting in an overall cis opening of the epoxide to give 45 or 49, respectively. Opening the epoxide introduces a great deal of strain into the system which is rapidly relieved by the molecule undergoing the cycloaromatization reaction to generate a 1P-benzenoid diradical species (46/50) which strips hydrogen atoms from the DNA, resulting in its cleavage. Both 48 and 51 have been identified as reaction products indicating that, in path A, a reoxidation step is involved (e. g., 47 -+ 48).
Figure 7-5 Computer-generated models of free dynemicin A (4) (top) and DNA-bound 4 (bottom). [Reproduced with permission from: K. C. Nicolaou, W.-M. Dai, Angew. Chem., Znt. Ed. Engl. 1991,30, 1387.1
220
7 The Enediyne Antibiotics
It is also possible that epoxide opening is initiated by electron push from the nitrogen atom rather than the phenol, and this will be discussed in a later section. In either case, epoxide opening is the trigger for cycloaromatization, diradical formation and DNA damage 1118-1201.
OH OH OH
I
l
fl*r
OH 1 OH OH
l
OH OH OH
I
Nuckophilic altack
44
Proton iranstsr
Cycloaromalizeiion
QMe COOH
OH OH OH
46
DNA dindlul
OH OH OH
47
DNA double Qrend cleavage
Scheme 7-14 Proposed mechanism of cleavage of DNA by dynemicin A (4).
OH 0
OH
50
7.3 The Discovery of the Enediyne Antibiotics
221
7.3.5 The Chromoprotein Enediyne Antibiotics The extensive body of knowledge relating to neocarzinostatin and its mechanism of action is presented in Section 7.3.1. It is now becoming clear, however, that there is a substantial family of closely related enediyne antibiotics, a number of which had been isolated several years ago but remained structurally uncharacterized. They share the common properties of being noncovalenty associated complexes between unstable chromophores and stabilizing proteins; they possess DNA-cleaving properties (both single- and double-stranded) associated with the chromophore, and are potent antitumor agents. The distinctiveness of the various complexes is demonstrated, however, by a general specificity of binding of a particular chromophore to its own apoprotein. The revelation of the structure of the kedarcidin chromophore (5) in 1992 [26-28) thus heralded the arrival of this new class of chromoprotein enediyne antibiotics, and was followed shortly afterwards by C-1027 [29-341 and maduropeptin [35, 361, the structures of whose chromophores were revealed in 1993. It seems likely that actinoxanthin (121, 1221 and macromomycidauromomycin [123- 1281 are set to join them.
7.3.5.1 Kedarcidin
Kedarcidin, isolated by Zein et al. (Bristol-Myers), was first reported in 1991 as the fermentation product of a novel actinomycete strain obtained from soils collected in India [129]. It exhibits potent in-vivo antitumor activity similar to that of the other enediyne antibiotics, and pronounced activity against Gram-positive bacteria. Kedarcidin was separated relatively easily by reversed-phase HPLC into the apoprotein and chromophore components, and is found to be a varying complex depending upon fermentation conditions [28).
Kedarcidin chromophore (5) The kedarcidin apoprotein exists as three main variants [28]; the major variant consists of 114 amino acid residues and a further two minor variants lack one or both of the first two amino acids (an alanine and a serine) of the major variant. Similarly, three kedarcidin chromophores have been identified having molecular weights of 1029, 1015 and 1001. Unlike the 1 :1 apoprotein/chromophore ratio observed with neocarzinostatin, this ratio varies from 1 : 1 to 18 : 1 for kedarcidin. Using pilot-scale fermentations (lo00 L) with fish emulsions in
222
7 The Enediyne Antibiotics
the production media, a complex containing only the chromophore of molecular weight 1029 was produced, and this was used for structural characterization of the chromophore. The kedarcidin chromophore (form I) (5) bears a striking resemblance to the neocarzinostatin chromophore (1). It is a highly unstable molecule and organic solutions of it rapidly darken upon concentration. For the first time observed in an enediyne antibiotic, the enediyne unit is contained within a highly strained nine-membered ring which is “locked” by an allylic epoxide forming part of a fused bicyclic system. As with the other enediyne antibiotics, there is an assortment of sugar [130] and aromatic appendages; there is also a peptidic linkage associated with a macrocyclic structure. The DNA-damaging properties of kedarcidin reside principally in the chromophore, resulting in highly sequence-specific single-stranded cuts [131]. The principal DNA recognition sequence is S’-TCCTN-3’, similar to calicheamicin y { (2), raising intriguing questions as to why two such structurally dissimilar molecules should recognize the same sequence. The cleavage chemistry requires reducing agents and oxygen, similarly to the other enediyne antibiotics, and is enhanced by the presence of thiols. In contrast to calicheamicin and esperamicin, however, DNA cleavage by the kedarcidin chromophore is inhibited by the addition of divalent ions such as Ca” and Mg2+ which chelate with the 2-hydroxynaphthoate moiety, and NMR experiments implicate this moiety in being involved in binding to DNA [131]. The mechanism of activation of the kedarcidin chromophore is thought to be similar to that of the neocarzinostatin chromophore, with nucleophilic addition at C(12) initiating epoxide opening (Scheme 7-15) [28]. The change in structural geometry then facilitates the
Y
Nucleophilic Attack Me c
5
I
52
+
O2 DNA single strand cleavage
Scheme 7-15 Proposed mechanism by which kedarcidin chromophore cleaves DNA.
7.3 The Discovery of the Enediyne Antibiotics
223
cycloaromatization reaction, leading to 1,Cbenzenoid diradical formation, hydrogen abstraction from DNA, DNA strand cleavage and cell death. Despite the observation that the DNA-cleaving properties of kedarcidin reside primarily in the chromophore, cytotoxicity assays using human colon cancer cell lines HCT 116 showed that the chromophore and apoprotein exhibit similar IC,, values of M, suggesting that the apoprotein contributes actively to the cytotoxicity of kedarcidin. This led to the finding [44] that the kedarcidin apoprotein, a highly acidic polypeptide, exhibits selective proteolytic activity against peptides which are most opposite in net charge, such as histones (the proteins around which chromosomal DNA is coiled to form chromatin). Preliminary experiments indicate that the apoproteins of other chromoprotein antitumor antibiotics such as neocarzinostatin [44] and macromomycin [124] also exhibit proteolytic activity, suggesting that this dual DNA-cleaving/proteolytic mechanism for attacking chromatin, the packaged genetic material of a target cell, is a common feature of all these chromoprotein enediyne antibiotics.
7.3.5.2 C-1027
The antibiotic C-1027, first reported in 1988 [31], was isolated from a culture filtrate of Streptomyces globisporus C-1027 and shown to consist of an extremely labile nonprotein chromophore tightly bound noncovalently to a 110 amino acid residue apoprotein with a 1 : 1 stoichiometry [29-34, 132, 1331. This new antibiotic displays extremely potent antineoplastic activity against a panel of transplantable tumors such as leukemia L 1210, P 388 and ascites hepatoma H 22, and its cytotoxic effect is much stronger than even that of neocarzinostatin [34]. The primary cause of the cytotoxicity appears to be DNA cleavage brought about by the chromophore. The nature of the DNA cleavage is somewhat different from that by the previously described enediyne antibiotics since the antibiotic (and similarly the isolated chromophore) efficiently cleaves DNA in a double-stranded manner even in the absence of thiol compounds or reducing agents. Furthermore, the sites of cleavage in the two DNA strands are two base pairs apart (rather than three base pairs apart as observed for other double-strand cleavers previously described), and are specific for sequences such as 5’-TAr-3‘/ 3’-ATA-S and 5’-AGA-3’/3’-sT-5’ in the two strands [132]. The DNA cleavage chemistry is primarily C(4’) hydrogen atom abstraction from deoxyribose [132]. The chromophore is
C-1027 chromophore (6)
224
7 The Enediyne Antibiotics
readily extracted from the apoprotein by organic solvents such as methanol, and was shown through careful NMR studies to have the structure 6,in which the enediyne unit is contained within a strained nine-membered ring forming part of a bicyclic system resembling the core structures of NCS and kedarcidin chromophores [29, 301. Unlike all the previously characterized enediyne antibiotics, however, the system contains no triggering mechanism and is already primed to undergo the cycloaromatization reaction and produce the DNA-damaging 1,4-benzenoid diradical species. The chromophore is clearly demonstrated to be stabilized by its association with the apoprotein since the DNA-cleaving properties of intact C-1027 are retained for extended periods, whilst those of the isolated chromophore rapidly decay under similar conditions (tl,z = 10 h at ambient temperatures) [132, 1331. A study of the interaction between the chromophore and apoprotein of C-1027 reveals a deep hydrophobic pocket in the apoprotein which is thought to bind the benzoxazine side chain of the chromophore and may be partly responsible for the tight and specific binding of the chromophore to the apoprotein [133]. As with neocarzinostatin, an interaction between a disulfide bridge (Cys-36 and Cys-45) and the chromophore acetylene bonds may contribute to the stabilization of the chromphore. Structurally, C-1027 has some features in common with some other chromoprotein antibiotics currently awaiting full structural characterization. Auromomycin, the holoprotein of macromomycin, is an antitumor chromoprotein antibiotic which has been isolated from Streptomyces macromyceticus [123-1281. It displays potent cytotoxicity against a range of tumor cell lines, and DNA cleavage is implicated in its mechanism of action. Degradation studies on the auromomycin chromophore indicate that it contains a benzoxazine side chain identical to that found in C-1027 [128]. Similarly, actinoxanthin [121, 1221 is an antitumor chromoprotein antibiotic isolated from Actinomyces globisporus No. 1131 in 1957 [134, 1351. The actinoxanthin apoprotein has been shown to have a high degree of sequence homology (95 %) with the C-1027 apoprotein [133].
7.3.5.3 Maduropeptin Maduropeptin is a complex of new chromoprotein antitumor antibiotics isolated from Actinomadura madurae (ATCC 39144) in 1990 [35, 361. It exhibits potent inhibitory activity against Gram-positive bacteria and tumor cells and strong in-vivo antitumor activity in P388 leukemia and B16 melanoma implanted mice. The structure of the chromophore has been recently elucidated [36], confirming it as a member of the family of enediyne antibiotics, and will be published shortly.
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics 7.4.1 Neocarzinostatin Chromophore Model Systems 7.4.1.1 Theoretical Considerations The underlying reaction behind the biological activity of neocarzinostatin chromophore is the cycloaromatization of the (a-cumulene-ene-yne 55 to generate the DNA-damaging diradical 56 (Scheme 7-16) [50]. In order to obtain some insight into the nature of this reaction, Myers
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
225
56
55
Scheme 7-16 Cumulene-ene-yne cycloaromatization observed in NCS chromophore.
studied the cycloaromatization of the related (2)-allene-ene-yne 58 (Scheme 7-17) [136, 1371, (MTAD) produced by dinitrogen extrusion from 57 using 4-methyl-l,2,4-triazoline-3,5-dione under anaerobic conditions. The allene-ene-yne 58 is a volatile liquid which required storage and subsequent reactions to be carried out under oxygen-free conditions in order to prevent rapid decomposition. Thermolysis of 58 in benzene- 1Q-cyclohexadiene produced 60 and 61 as a mixture of products, clearly implicating the diradical 59 as an intermediate Whilst the neocarzinostatin-type diradical 56 is constrained geometrically to be a o,a-diradical, the diradical59 is a o,n-diradical. The highly reactive benzenoid o-radical of 59 strips a hydrogen atom from the 1,Ccyclohexadieneto form a radical pair. The less reactive aliphatic n-radical from 59 can then either recombine with the n-radical derived from the 1,Ccyclohexadieneto give 61, or abstract a second hydrogen atom to give toluene (60). The existence of ionic
1p
H
61
60
Scheme 7-17 The Myers cychation.
character in the cycloaromatization intermediate 59 (62, Scheme 7-18: Cf. Scheme 7-8, path B) was demonstrated by thermolysis experiments carried out in CH30D and CD30H in which deuterated methoxymethylbenzenes 63 and 64 respectively were identified as the major products. Myers measured the kinetic parameters relating to these reactions, showing that, for the reaction 58 -+ 60 + 61, A H z = 21.8 f 0.5 kcal/mol, AS* = -11.6 f 1.5 eu, E, = 22.5 kcal/mol and log A = 10.7. The enthalpy of activation for the cyclization of 58 is about 10 kcal/mol lower than that measured in Bergman’s original enediyne system (13 + 16, Scheme 7-3). The rather large negative entropy of activation is due, at least in part, to the loss of rotational freedom about the ene-allene a-bond of 58, which is not the case with cyclic systems such as 55. The kinetics of cycloaromatization in methanol (Scheme 7-18) were virtually identical, suggesting that the radical and ionic pathways share a common rate-limiting step.
226
7 The Enediyne Antibiotics
Q 59
CH2
H
D
I
Scheme 7-18 Ionic characteristics associated with the Myers cyclization.
Myers has also studied the cycloaromatization of the related aromatic hydrocarbon 1,6-didehydro[lO]annulene (65, Scheme 7-19),demonstrating that the reaction has a half-life of -25 min at -51 "C and making it one of the most rapid diradical-forming reactions known [138, 1391.
Scheme 7-19 Cycloaromatization of 1,6-didehydro[lO]annulene(Myers et al.).
7.4.1.2 Synthetic Studies
Much synthetic work has been carried out on systems relating to the neocarzinostatin chromophore, some of it aimed toward the synthesis of the chromophore and other parts inspired by the mechanism of action of the chromophore resulting in novel systems and triggering devices. Some of the strategies successfully utilized toward these goals are described in this section. The various approaches used in constructing the bicyclic core of the neocarzinostatin chromophore can be broadly categorized according to the means by which ring closure is effected to generate the unsaturated nine-membered ring of the chromophore. One of the most successful approaches, utilizing an acetylide anion-aldehyde ring closure, has been described by Myers in the synthesis of the epoxy dienediyne core of the chromophore (78, Scheme 7-20) [54, 140, 1411. Key features in this synthesis include Pd(O)/Cu(I)-catalyzed construction of the enediyne moiety (67 -+ 68), Sharpless asymmetric epoxidation (68 691, diastereomeric resolution by crystallization (72, obtained from 70), intramolecular acetylide addition to an aldehyde to close the nine-membered ring (75 76), a suprafacial transformation (76 -+ 77), and a 1,4 conjugate elimination (77 78). The epoxy dienediyne 78 is considerably less stable than the neocarzinostatin chromophore itself, decomposing in seconds upon concentration in the absence of radical inhibitors. Unlike the chrornophore, however, an acidic solution of 78 -+
-+
+
227
Z4 Theoretical and Synthetic Studies on the Enediyne Antibiotics 1. DIBAL (82%)
2.Na(OMe),EH H20. THF (60%)
PivO
Br
TMS 3. Asymmetric
(Ph2P),PdC12, Cul
67
(88%)
70
'
TMS
6
TMS
Epoxidation (-)-LET 4. 'EuCOCI, base (8396,93% ee)
E1.N
8
72
1. mCP0A
2. 'Pr2NEt I A (84%)
PivO,
PivO,
Scheme 7-20 Synthesis of the epoxy dienediyne core of NCS chromophore (Myers et al.).
is completely inert to methyl thioglycolate; however, co-addition of triethylamine with the methyl thioglycolate results in a rapid reaction (t,,, 15 min at 23 "C) to give indene 79 as the major product. This result adds support to the hypothesis that the NCS chromophore provides intramolecular base catalysis for its own activation through its amino sugar [54]. Another notable approach to the NCS chromophore nine-membered ring based upon an acetylide-aldehyde ring closure, although ultimately unsuccessful, comes from the labora-
-
228
7 The Enediyne Antibiotics
tories of Terashima and is shown in Scheme 7-21 [142-1471. The precursor 82 for their studies came from the Pd(O)/Cu(I)-catalyzed coupling of the enol triflate 80 and the sugar derived acetylene 81. Whilst it was possible to ring-close the epoxide 83 under basic conditions to give a ten-membered ring analog (84), attempted ring closure of the aldehyde 85 under similar conditions resulted in extensive decomposition and none of the desired nine-membered ring 86 was observed. This probably reflects the considerable instability associated with ninemembered systems such as 86, and it is thus not surprising that many of the “neocarzinostatin chromophore”-related model systems reported have opted for a synthetically more convenient ten-membered ring.
1. TBAF (61%) 2. NalO,
-
OH
LiHMDS
BF,.OEt,
THF, -70 “C
84
83
85
86
Scheme 7-21 Terashima’s approach to the NCS chromophore.
Nicolaou et al. utilized an intramolecular acetylide-aldehyde ring closure in the design of the system 87, named golfomycin A by the authors, as a novel DNA-cleaving agent inspired by the neocarzinostatin chromophore [148]. The synthesis of golfomycin A (87) and some interesting chemistry associated with it are shown in Schemes 7-23 and 7-24. The rationale behind golfomycin A is shown in Scheme 7-22, allowing for two competing mechanisms of DNA cleavage. Conjugate addition of a nucleophile to the acetylenic ketone could generate the allene-ene-yne type intermediate 88, which would undergo the Myers cycloaromatization to give the DNA-damaging diradical89. If the adding nucleophile comes from the DNA itself
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
229
(Nu-DNA), then this would lead to alkylation of the DNA (90) and provide an alternative pathway for DNA cleavage. Golfomycin A exhibited moderate single-stranded DNA-cleaving properties and growth inhibition of MB49 murine bladder carcinoma cells (IC50 = 3.4 pM). It is most probable that these properties result from a DNA alkylation pathway.
fi 1
NuH or Nu-DNA
-
NuH
-
OH
~
87
OH
88 Cyclization
NU-DNA
&> - m - m 00
OH Nu
QNU-DNA
90
OH
*
OH
89
Scheme 7-22 Mechanism-based design of golfomycin A (87) (Nicolaou et al.).
Q \\
OR
1. KHMDS, THF, -78' C
(51%)
2.MnO2 (82%)
9l:R=TBS
+& -
OR
92: R = TBS 87: R = H
DBU (75%)
0
93
NaBHd
(97%)
-
Me 95
Scheme 7-23 Synthesis and reactions of 87 (Nicolaou et al.).
96
230
7 The Enediyne Antibiotics
HSCH2C02Me DBU
-
OTBS
92
J
99 (20%)
97
98 (55%)
Scheme 7-24 Reaction of the golfomycin A precursor 92 with methyl thioglycolate (Nicolaou et ale).
Although a Pd(O)-catalyzed intramolecular coupling has not proven successful in closing a nine-membered ring in studies relating to the neocarzinostatin chromophore, it has been successfully used by Hirama in closing a ten-membered ring leading to the formation of novel cyclic dienediyne systems with their own means of activation leading to diradical formation. Scheme 7-25 shows how the application of a Pd(0)-catalyzed intramolecular coupling of the vinyl bromide/alkynylstannane 100 led to the synthesis of 102 and 103 [149]. The reaction of ketone 103 with methyl thioglycolate in AcOH/EtOH under aerobic conditions proved to be quite interesting, leading to the formation of 106-108, presumably via 104 and 105. The acetate 102 reacted similarly under the same conditions to give the compounds l l l a and 111 b via the postulated intermediates 109 and 110 (Scheme 7-26). Under anaerobic conditions the compounds 114a-d were isolated; 114a-c are presumably formed via the pathway 102b + 112 113. The regioisomeric ketone 115 (Scheme 7-27) was also synthesized by Hirama and co-workers [150] and its reaction with methyl thioglycolate explored leading to the products 117a-c under aerobic conditions and 120a-c under anaerobic conditions. Hirama has also synthesized the cumulene-ene-yne122 shown in Scheme 7-28 [151]. This compound exhibited the expected reactivity toward a Myers-type cycloaromatization, leading to 124.The product 126 was also isolated, resulting from a formal [2+2] intramolecular cycloaddition (122b 125). The half-life of 122 at 80°C was found to be 1.1 h 1151, 1521. PalIadium(0)-catalyzed couplings have also been used extensively in constructing the unsaturated framework of other neocarzinostatin-related systems. A particularly elegant example by Nuss et al. is shown in Scheme 7-29, in which the diiodoenyne 127 was coupled in a one-pot process with an alkynylstannane to stereospecifically give the dienediyne 128 [153]. This reaction effected no less than three consecutive Pd(0)-mediated C-C bond-forming processes. A similar strategy was reported independently by Torii et al. [154]. In a related study, Brilckner and Suffert generated a similar dienediyne through a Pd(0)-catalyzed coupling of a bis(eno1triflate) with two equivalents of alkyne 1155- 1571. Petasis and Teets reported the study shown in Scheme 7-30[158]. The silylallene ester 130 was metalated with LDA to give an alkynyllithium enolate intermediate which was reacted with the propargyl aldehyde 129, and the resulting dkoxysilane eliminated under acidic condi-+
-+
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
231
1. Pd(PhjP)d, 50 "C (72%) 2. AcOH. H20 (86%)
* 101 (COCl)p, DMSO EhN (72%)
102
I
103 HSCH2C02Me AcOH, EtOH 02.20 "C. 2 h
104
/C02Me
@$+* 0
/COpMe
R
0
106: R = H (28%) 107: R = SCH&O$4e (7%)
0
0
OH
108 (10%)
Scheme 7-25 Neocarzinostatin chromophore related studies by Hirama et al.
tions in a Peterson olefination to give the enediyne 131. Although this strategy was clearly aimed at closing the nine-membered ring through an intramolecular Pd(0)-mediated coupling between the vinyl bromide and the terminal acetylene, this is not reported. An interesting and successful alternative approach to the transition-metal-catalyzed construction of the nine-membered ring of the neocarzinostatin chromophore has been reported by Wender et al. (Scheme 7-31)[159] in which the ally1 bromide 132 was induced to condense in an intramolecular fashion with the propargyl aldehyde in a chromium(I1)-nickel(I1)-mediated reaction [la,1611. Related approaches in the construction of ten-membered ring analogs have involved BF, OEt,-induced allylsilane-propargyl aldehyde (Suffert) [162] and TiC1,-induced silyl enol ether-propargyl acetal (Krebs and co-workers) [I631 ring closures.
-
232
7 The Enediyne Antibiotics
102
Nu- = RS., HOO.
1
o=o
NU-= EtO-. Me02CGH2s
HSCH2CO.#e AcOH, EtOH
02.25
I
HSCH2C02Me AcOH, EtOH A 112 h
"C.22 h
(&
I
00.OAc
/*
I
OAc
109
112
113 1 114a: R' = R2 = H (3%) EtO 114b R' = SCH2C02Me R2 = H (0.5%) 114c: R' = H R2 = SCH2C02Me(3%)
e@
Meo2 R2
OAc
0
l l l a : X = 0 15%) 1 l?b: X = H, 'SCHzCOzMe (5%)
OAc
SVC02Me
11 4d (6%)
Scheme 7-26 Neocarzinostatin chromophore related model studies by Hirama et al. Cycloaromatization of 102 under aerobic and anaerobic conditions.
An elegant alternative approach to the construction of the nine-membered ring of the neocarzinostatin chromophore has been reported by Magnus and Pitterna (Scheme 7-32) employing a boron-mediated aldol condensation for ring closure [164]. Recognizing the instability associated with the nine-membered enediyne ring 136, these researchers cleverly masked the enediyne by forming a cobalt complex from one of the acetylenes, thus enabling the isolation of the nine-membered ring 135. In-situ generation of the parent nine-membered cyclic enediyne 136 under oxidative conditions resulted in cycloaromatization to give the tricyclic compound 137. In a related study by Mikami et al. 11651, the propargyl aldehyde 138 (Scheme 7-33) readily underwent a thermal intramolecular ene reaction to give initially the ten-membered ring 139. Under the conditions of the reaction, however, this spontaneously
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
115
1
NU= RS., HOO.
233
1
o=o
HSCH~COZME EtOH. AcOH 02.25 "C
OfJ 00.
117a: X = o (5%) 117b: X = H, OH (5%) 117c: X = H,SCH2C02Me (4%)
I
119
120a: R1 = H, R2 = SCH2COzMe (20Y0) 120b: R' = R2 = H (14%) 120c: R1 = SCH2C02Me, R2 = H (4%)
Scheme 7-27 Neocarzinostatin chromophore related model studies by Hirarna et al. Cycloaromatization of ketone 115.
eliminated H20to give the dienediyne 140, which itself spontaneously underwent cycloaromatization to give 143, presumably via 141 and 142. Deuterium labeling experiments also supported the formation of 145, indicating partial hydrogen shift (142 -,144). A similar example in a monocyclic sulfur-containing system was previously reported by Toshima, Tatsuta and co-workers [166]. A number of researchers have sought access to the nine-membered ring of the neocarzinostatin core structure by first constructing a more readily accessible larger ring and then effecting a ring contraction. One approach developed by Wender was to carry out a Wittig rearrangement of an allylic ether contained within a 12-membered ring (Scheme 7-34) [ISS]. A similar approach was adopted by Takahashi [167]. Wender also demonstrated the feasibility of contracting a ten-membered ring through photolysis of the bis-propargylic sulfone 148 (Scheme 7-35) [168]. Elimination of H,O from 149 then gave the parent hydrocarbon 150 representing the carbocyclic framework of the neocarzinostatin chromophore, which w a s found to be rather labile (tl,2= 48 h at ambient temperature) and rapidly polymerized upon removal of solvent. An alternative use was found for bis-propargylic sulfones by Nicolaou et al., who hypothesized that such systems may be DNA-cleaving agents through their ability to form
234
7 The Enediyne Antibiotics
C S R
HSCH2C02Me EtzN, CH$N 25 “C, 2 h
-
L
(46%)
Scheme 7-28 Synthesis and cycloaromatization of enyne[3]cumulene 122 (Hirama et al.).
fi’
Pd(Ph3P)d (Cat.)
J
-
R
_____)
HO
127
“Bu3Sn-=-R
(32%)
- R HO 128
Scheme 7-29 Palladium-mediatedconstructionof a dienediyne related to NCS chromophore(Nuss et al.).
wf TMSf02Me
TBSo 129
LDA, -78°C THF
Ir
then
130
H+ (62%)
TBSO 131
Scheme 7-30 Construction of an enediyne related to NCS chromophore (Petasis and Teets).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
HO
HO
132
235
133
Scheme 7-31 Chromium-mediated construction of the NCS chromophore core structure (Wender et a],).
Scheme 7-32 Intramolecular aldol-based construction of a model system related to NCS chromophore by Magnus and Pitterna.
allenic sulfones under basic conditions (Scheme 7-36) 11691. This would then allow for the possibility of either diradical formation/DNA cleavage (path A), or DNA alkylation/DNA cleavage (path B). Indeed, such systems were found to be modest DNA-cleaving agents, resulting in single-stranded cleavage probably arising through the DNA alkylation pathway (path B). Similar ideas have also been investigated by Shibuya and co-workers [170], Toshima et al. [171], and Lown and co-workers 11721. In a related study, Nicolaou et al. designed the ene-yne-allene phosphine oxides 159 (Scheme 7-37) [173], prepared by a [3,2]-sigmatropic rearrangement (158 159), as novel DNA cleavage agents with the potential for cleaving DNA through both diradical (path A) and DNA alkylation (path B) mechanisms. Mechanistic studies confirmed the dual mode of action of these compounds. Another system utilizing an intramolecular SN2’-typeaddition to a propargyl alcohol derivative to generate the ene-yneallene system was developed by Myers and co-workers (Scheme 7-38) (137, 1741. In this example, the compound 164 was triggered under basic conditions to form the (Z)-1,2,4-heptatrien6-yne derivative 165, which underwent the Myers cycloaromatization reaction below ambient temperature. +
236
7 The Enediyne Antibiotics
HO
138
1
J-
144
145
139
141
14*
143
Scheme 7-33 “Ene reaction” approach to NCS chromophore (Mikami et al.),
”eUU
(429%)
OH
133
146
Scheme 7-34 Wittig rearrangement strategy for construction of the NCS chromophore core structure (Wender et at.).
1. NazS(69%)
-
o s2Q ;-
OMS
HO
147
48
-1
hv, PhCOPh (9-15%)
MsCI. DMAP
150
149
Scheme 7-35 Synthesis of the parent carbocycle 150 of the NCS chromophore by photolytic ring contraction of the bis-propargylic sulfone 148 (Wender et al.).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
1
I 152
151 Path A
1
237
155 1. H20
2.pH >7 Cleavage
7153 '
154
156
Scheme 7-36 Mechanistic rationale for the design of DNA-cleaving molecules of the propargylic and allenic sulfone type (Nicolaou et al.).
DNA-NU
0
162
Ph
163
161
Scheme 7-37 Design of DNA-cleaving molecules with dual mode of action. Path A follows a radical mechanism; path B follows an alkylation mechanism (Nicolaou et d.).
Although most work relating to the neocarzinostatin chromophore has concentrated upon the dienediyne core of the molecule, a number of studies concerning the naphthoate moiety have also appeared [175-1791.
238
7 The Enediyne Antibiotics
Et3N, DMSO ~
\ \ \ 164
167
~~~
7
1.4-cyclohexadiene 25 "C
166
Scheme 7-38 Model studies related to NCS chromophore (Myers and co-workers).
7.4.2 Calicheamicin/Esperamicin Theoretical and Synthetic Studies 7.4.2.1 Synthetic and Theoretical Studies on the Bergman Cycloaromatization of Cyclic Enediynes The disclosure in 1987 of the structures of the calicheamicins and esperamicins, together with their unique and intriguing mode of action, stimulated considerable activity in the laboratories of many synthetic and theoretical chemists. Section 7.3.2 described the triggering of the bicyclic enediyne core of calicheamicin, in which a change in geometry of the molecule resulted in the termini of the enediyne unit being forced together and thus imposing further strain into the system. This was accompanied by rapid cycloaromatization. It therefore occurred to Nicolaou et al. that, in the absence of other factors affecting strain, it might be possible to predict the reactivity of a cyclic enediyne system toward cycloaromatization from the distance c . . - d between the ends of the 1,5-diyn-3-ene system, and they undertook the study of a series of monocyclic enediynes of varying ring size [180, 1811. The parent series of ten- through sixteen-membered ring enediynes 169 b-h were conveniently prepared [180, 181) via the Ramberg-BPcklund reaction of the corresponding achlorosulfones 168 b-h (Scheme 7-39). The ten-membered ring enediyne 169 b readily underwent the Bergman cycloaromatization reaction at room temperature with a half-life of 18 h (Table 7-4), while the larger ring enediynes 169c-h were found to be stable. By contrast, the
Scheme 7-39 Preparation and cycloaromatization of monocyclic enediynes (Nicolaou et al.).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
239
Table 7-4 Calculated c * - d distances and stabilities of cyclic enediynes
Compound
n
Ring size
c . . . d distance
9 10 11 12 13 14 15 16 10 10 10 10
2.84 3.25 3.66 3.90 4.14 4.15 4.33 4.20 3.20 3.29 3.34 3.42
(A)
Stability
-
169a 169b 169c 169 d 169e 169f 169g 169h 171 175 176 177
Unknown tl12= 18 h at 25°C Stable at 25 "C Stable at 25 "C Stable at 25 "C Stable at 25 "C Stable at 25 "C Stable at 25 "C t1,2= 11.8 h at 37°C t,,z = 4 h at 50°C t,,2 = 2 h at 50°C Stable at 25°C
nine-membered ring 169a could not be prepared although products formally arising from a Bergman reaction were identified. Comparison of the distances c . . . d between the termini of the enediyne moiety of these systems and the ease with which they underwent the Bergman reaction (Table 7-4)showed a clear trend in which a decreased c - - -d distance reflected, in addition to a closer intimacy between the acetylenic groups, an increasing ring torsion and hence an increased tendency to undergo the Bergman reaction in order to relieve the strain. For these simple systems a critical upper limit for the c . . d distance of around 3.2-3.3 i% appeared to be required for the Bergman reaction to occur at a measurable rate at ambient temperatures. Thus, whilst this empirical c . - . d distance rule is not strictly applicable to complex systems such as those found in calicheamicin or dynemicin where geometrical constraints prevent cycloaromatization prior to triggering, it does provide a convenient means of assessing the likely stability of many systems toward cycloaromatization. More sophisticated calculation and kinetics experiments by Snyder [182-1851 and Magnus et al. [185, 1861 have subsequently demonstrated that the crucial factor in determining the ease with which a particular system undergoes the Bergman reaction is the relative strain energies of the ground and transition states for the reaction. These findings should always be borne in mind when applying the Nicolaou c...d distance rule. Since the simple ten-membered ring enediyne 169b underwent the Bergman reaction at physiological temperatures, Nicolaou proceeded to mimic the DNA-cleaving action of the calicheamicins and esperamicins by using simple systems such as these. The diol 171 was designed in order to endow the molecule with some degree of water solubility and also to provide for the option of attachment to delivery systems (Scheme 7-40)[181, 1871. It was correctly
171
172
DNA
Cleavage
173
Scheme 7-40 Enediyne 171 as designed DNA-cleaving agent (Nicolaou et al.).
240
7 The Enediyne Antibiotics
predicted from the calculated c . . . d distance of 3.20 A that this molecule would be sufficiently stable for isolation and handling at ambient temperatures but would undergo the Bergman reaction at physiological temperature at a sufficient rate to cause DNA cleavage. Thus enediyne 171 caused significant cleavage of phage (OX174 double-stranded supercoiled DNA in the absence of any additives at concentrations as low as 10 pM at 37"C, with the extent of cleavage being dependent upon concentration, incubation time, and temperature. As a control, it was demonstrated that the corresponding Bergman cyclized product 173 caused no DNA cleavage. The cleavage data are therefore consistent with a Bergman cyclization of 171 leading to a diradical species 172 which proceeds to abstract hydrogen atoms from DNA in a mechanistic mode similar to the one proposed for the calicheamicins and esperamicins. In a subsequent collaboration between Nicolaou and Bergman [MI, the thermally reactive diols 175 and 176 (Scheme 7-41) were prepared via a pinacol coupling of the dialdehyde 174 and similarly demonstrated to effect DNA cleavage. By contrast, the conformationally locked and thermally stable derivative 177 failed to cleave DNA. Under basic conditions, however, compound 177 became active via the release of diol 176, thus exhibiting both DNA cleaving and cytotoxic properties.
< a 1 h. Smlp, 42% (tr;d THF, cis 25 = "C20:
\\
CHO
Ti&. Zn-Cu, DME 25'C, 1 2 h, 45% (trandcis = 1:2.6)
174
-
OH
\
175
+
177
176
Scheme 7-41 Model enediyne studies of Nicolaou, Bergman and co-workers.
Since the naturally occurring enediyne antibiotics are triggered to exert their biological actions by bioreductive processes, Nicolaou et al. designed the system shown in Scheme 7-42 in order to control the Bergman cyclization by a hydroquinone quinone redox process (1891. It was postulated that a hydroquinone such as 179, prepared via an intramolecular Nozaki-type coupling of the iodoalkyne 178 1190. 1911, should be rather more stable toward cycloaromatization than the corresponding quinone 180 due to its lower activation energy for the process. This was borne out by the measurement of the half-lives of these compounds: 179 (t1,2 = 74 h at IlO"C), 180 (t,,, = 2.6 h at 55 "C), 181 (tl,2= 32 min at 55 "C) "91. Furthermore, interaction of compounds 179-181 with phage (OX174 DNA at pH 7.4 and 37°C revealed that 179 had no DNA-cleaving activity whilst 180 and 181 showed significant DNAdamaging properties. This concept of activation of enediyne systems through redox processes
-
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
241
was taken a step further by Myers and Dragovich [192], who designed the system shown in Scheme 7-43. Enzyme-mediated reduction of the anthraquinone 182 led to the elimination of succinic acid followed by tautomerization and oxidation to reveal the enediyne system in 185. This then slowly underwent the Bergman reaction at 37 “C 2 days). Most recently, Semmelhack et al. demonstrated that the steric effects of substituents on calichemicin models can also be used to moderate the reactivity of these systems by raising the energy of the conformation most suited to cycloaromatization relative to the lowest-energy conformation [193]. J
-
‘BuCOO
‘BuCOO
CCl2 - NiCll
‘BuCOO
178
0
I
0
181
179
180
-
Scheme 7-42 Control of cycloaromatization through a hydroquinone quinone redox process (Nico-
laou et al.).
red.
OH
Scheme 7-43 Myers’ approach to redox activation of enediyne systems.
7.4.2.2 Synthetic Approaches to the Calicheamicin Aglycone
Following the first reports of the structure of the calicheamicins and esperamicins, many synthetic chemists turned their attention to the synthesis of the bicyclic enediyne core of these molecules (the aglycone). The unprecedented structures of the aglycones, differing only in the presence of an extra hydroxyl group in esperamicin, presented many challenges since new
242
7 The Enediyne Antibiotics
synthetic methodology would have to be developed in order to tackle its various structural features. Most pressing was the need to find ways of constructing the bicyclic enediyne framework, and then the question of introducing the other unusual features such as the allylic trisulfide trigger and the carbamate would have to be addressed. Kende and Smith were first off the mark, reporting a synthetic approach to the basic bicyclic enediyne framework of the calicheamicin aglycone in 1988 (Scheme 7-44) [194]. In this pioneering work, a number of key strategies were demonstrated for the first time. Firstly, Pd (0)-catalyzed coupling of the acetylene 187 with the (a-chloroenyne 188 stereospecifically introduced the (Z)-1,5-diyn-3-ene unit into the molecule leading to 189. Treatment with base then deprotonated the enediyne and effected ring closure onto the aldehyde to generate the bicyclic enediyne structure as a 3 : 1 mixture of epimers at the newly generated hydroxyl center (major epimer shown, corresponding to that of calicheamicin). Hydrolysis of a ketal finally revealed the bridgehead enone. This work was also instrumental in correcting the stereochemistry of the corresponding hydroxy center in the natural product from that which was originally reported.
186
OH 1.188, Pd(PhjP)4 Cut. "BuNH~(59%) 2. (COC1)2. DMSO. E1,N 3."&lqNF (60%)
0
1. LiHMDS (42%)
w
*
.
C"
L
190 3 :1 mixture of epimers
2. H30+ (90%)
189
Scheme 7-44 The Kende synthesis of a model of the calicheamicin bicyclic core.
An alternative approach to the bicyclic enediyne core of esperamicin utilizing an intramolecular Diels- Alder reaction was reported by Schreiber and co-workers (Scheme 7-45) [195- 1971. This approach was initially hampered by an unfavorable regiochemical outcome in the Diels-Alder reaction (195-+ 196), but was successfully corrected by a clever maneuver involving a Tsuchihashi pinacol rearrangement [198, 1991 with concomitant diastereoselective acyloin shift to secure the desired bicyclic ring skeleton (197 198 199). A third approach to the enediyne core of calicheamicin was reported by Magnus et al. (Scheme 7-46) [I86, 200-2061. Cleverly exploiting some interesting organocobalt chemistry, they prepared the acetylenic cobalt complex 201 [201] and subjected it to a Nicholas-type reac-+
-+
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
kcno 3. "Bu4NF
193
2.
OPMB
191
1
OPMB
OTBS
4. TBSOTf. Et3N (64%)
243
I
Cul, Pd(MeCN)& Et,N (49%)
OT>Br 194
OPMB
01".
-
.
OPMB
-' 4
6TBS
196
120°C (40%)
195
~TBS
1. K2CO3 (83%)
2. MsCl (73%) 3.D W (91%)
199
~TBS
Scheme 7-45 An intramolecular Diels-Alder approach to the esperamicin enediyne core (Schreiber and
co-workers). tion [207] in which Lewis acid treatment promoted the formation of the propargylic cation and intramolecular reaction with the enol ether to generate the bridging ten-membered enediyne ring masked as its cobalt complex 202. The enediyne was subsequently unmasked under oxidative conditions and the bridgehead double bond introduced via selenoxide elimination chemistry (202 203 204). Magnus et al. then went on to demonstrate the introduction of the allylic trisulfide trigger for the first time [204]. The keto group was submitted to a regioselective Horner-Emmons Wittig olefination (204 + 205), and the thioester 206 prepared via thioacetate displacement of the corresponding allylic mesylate. The trisulfide was finally introduced onto the free thiol utilizing an N(alky1dithio)phthalimide(207) [2081. Magnus has also demonstrated the feasibility of closing the bridging ten-membered enediyne ring via an aldol reaction involving a propargylic aldehyde masked as its organocobalt complex (209 + 210, Scheme 7-47) [203], as well as studying the cycloaromatization of a number of related systems [185, 186, 201, 202, 2061. Tomioka et al. [209], Kadow et al. (210, 2111 and Maier et al. (2121 have subsequently reported very similar studies involving organocobalt chemistry. A number of other strategies have been reported for constructing the enediyne core of the calicheamicin aglycone. These include the oxidative formation of the enediyne unit from a +
244
7 The Enediyne Antibiotics
Ti&. DABCO -40°C (50%)
T
B
S
-' L 4
O
12
T
(70%)
B
S
0
203
202
1. TBSOTf. KHMDS 2. PhSeCl 3. H202
(Et0)2P(O)CH2CN NaH. DME (90%)
204
O c0:co (C013 (
k
W 3
+TB CN 1. DIBAL, then H30'.
fhenDIBAL 2. MsCI. EbN 3. NaSAc (77%)
206
xsss X = Me (-90%.ca. 1.3:1 mixture with disulfide) X = CHzPh (-92%)
207
Seheme 7-46 An approach to the calicheamicin aglycone by Magnus et al.
1,s-diyne precursor 12131, base-induced elimination in a 3-OMs-1,5-hexadiyne system to generate the enediyne unit [214, 2151, and an intramolecular Pd(0)-catalyzed coupling of a terminal acetylene with a (a-chloroenyne to generate an 11-membered enediyne bridging ring [216]. The first total synthesis of the calicheamicin aglycone (222, calicheamicinone) was finally achieved in 1990 by Danishefsky and co-workers 1217-2231. Their synthesis, shown in Scheme 7-48,was a maiw achievement in this field. Key steps in the sequence, which begins
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
245
21 1
212
Scheme 7-47 Intramolecular aldol-based construction of the calicheamicin core (Magnus et al.),
with 213, are: (a) the formation of the quinone epoxide 215 from 214; (b) the regioselective addition of dilithioenediyne to ketoaldehyde 215; (c) an intramolecular acetylide addition to the aldehyde group of 216 to close the ten-membered enediyne ring and give 217 in which the generated hydroxyl center is primarily of the correct relative stereochemistry for the natural product; (d) utilization of the a, P-unsaturated ketone to allow introduction of the nitrogen as an azide through a 1,4-addition/elimination reaction (218 219); (e) an intramolecular Horner-Emmons condensation (218 219) which introduced the allylic functionality with the correct regiochemistry; and (f) stabilization of an intermediate primary enamine as a vinylogous carbamate (219 220). Danishefsky and co-workers have also resolved several intermediates along the route in their racemic synthesis [224, 2251. Simulation of the calicheamicin/esperamicin cascade has been amply demonstrated by Danishefsky and co-workers with a variety of compounds (Scheme 7-49) [219-2211. Furthermore, comparisons of calicheamicin yi (2) and calicheamicinone (222) in DNA-cleavage experiments pointed to the importance of the carbohydrate fragment of the natural product in the molecular recognition of its target sequence [SS]. The latest chapter in the story of the calicheamicin aglycone came in 1991 with the first enantioselective synthesis by Nicolaou and co-workers [226, 2271. Key features of this conceptually different synthetic approach (Scheme 7-50) include: (a) the use of an asymmetric allylboration reaction to introduce asymmetry into the molecule (223 + 224 225); (b) the incorporation of the N-atom of the urethane through an intramolecular 1S-dipolar cycloaddition reaction of an alkenyl nitrile oxide (226 227); (c) the stereospecific introduction of the enediyne moiety through alkylation of ketone 228; (d) an unusual one-pot double oxidation of a hydroxy isoxazoline to a keto isoxazole (229 230); (e) the stereospecific introduction of the alkylidene side chain through W h i g olefination of 230; (f) the unveiling of the key enamine-aldehyde functionalities through reductive ring opening of isoxazole 231; (g) the bridging of the enediyne ring through an intramolecular acetylide-aldehyde addition (232 + 233); and (h) an unusual lactonization to correct an errant stereocenter (233 + 234). The synthetic calicheamicinone thus obtained had an optical rotation [a]g- 472" (c 0.21, -+
-+
-+
+
+
+
246
7 The Enediyne Antibiotics
1. NES, MeCN
2. CI,CHOMe. TiCI4
CozMe
Me0
3.BC$ (65% from 2l3) 4. DlBAL
Meo
OH
Br
213
2. Dess-Main periodinane (40% from 214)
Me0
Br 215
214
0
I
217
n*
216
1. NaN,. MeOH 2. (EtO),P(0)CH2COCI, pyr. 3. Base (41% from 218)
HopI
m0
0n 0 O% u
2. 1. Triphosgene. H2S, piperidine. pyr.MeOH
*
3.MeOH. 3. (76%) pyr. (76%)
n 0
219
M
0
e
o z + -x
c
o
k
&"n 0
I
220
1. DlBAL 2. MeCOSH, Ph3P 'PQCN=NCO;PI (19%)
0
HoIt 0-0
NHChMe 2. 1. DlBAL N-(Methyldithiio)phlhallmide 3.CSA, THF, H& ~
MHe S S S O
(*I-=
I
F
(42%)
AcS 221
Scheme 7-48 Total synthesis of the calicheamicin aglycone (Danishefsky and co-workers).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
247
NaBH,. MeOH 25 "C
6H (50%)
2[
Hob&b
or base, 25 "C
H6 I R1
m:R' = SSSMe, R2 = NHC0,Me R' R' R'
= SSSMe, R2 = H
H
O
I
i
k
SiH
T
R IH, AC (16-71%)
SSSBn, R2 H SAC, R2 = H
I I
Scheme 7-49 Cycloarornatization of calicheamicinone (222) and related analogs (Danishefsky et al.).
CH2C12)and was subsequently proved to be of the same absolute stereochemistry as that in calicheamicin yi (2) by conversion of an advanced intermediate through to the natural product itself 1771 (see Section 7.4.2.4). Perhaps the key feature of this synthesis is that it readily yielded the multigram quantities of enantiomerically pure advanced intermediates necessary for the completion of the total synthesis of calicheamicin y: (2).
248
7 The Enediyne Antibiotics
-
1. TBSCI. imidazde
EM
0 MEMO
oMEM
2. PhCOCI. pyr. 3."Bu4NF
HO
THF. -78 "C (87%)
223
4. (COCI)2. DMSO. Et3N CH2CIz.-78 "C 5.N H P H (98% from 225)
2225
226
I NaOCl CH2C12IH@ (65%, 4 :1 mixture)
n
- -
t
n
n
U+TMS
1. NaOMe. MeOH
THF, -78 "C then A@
ACO&O
2. Jones' reagent (95%)
(6756)
@
OMEM
TMS
I
BZO'"
OMEM
OMEM
228
227
1. ZnBr,. CH2C12 2. (C0CI)p. DMSO Et3N,CH2C12.-78 "C (54% from 229)
n
n 1. Ph3P=CHC02Me
2. NaOMe
1. Mo(CO)~.MeCN 2. NaOMe 3. Phthaloyl chloride ____)
TMS
230
3.TESOTf, 2,blutidine 4.188. Pd(PPh& CUI "BuNHZ. PhH (59% from 230)
4. S i i i i gel, CH,CI, 5. Ac20
-
TMS
232
231
KHMDS toluene. -90"C (44%)
n
1. MeNHNH2 2. Triphosgene pyr.. MeOH 3. DIBAL. CH2C12 4. NaBH4.MeOH 5. Pivaloyl chloride
-
6. TESOTf 7. AcSH, PPh,, DEAD 8. DiBAL 9. N-(Methyldithio)phthalimide 10. TsOH. THF. H-0 (1 7% irom i34j
NPhth
(90%)
0
234
OMe
Scheme 7-50 Enantioselective total synthesis of the calicheamicin aglycone (Nicolaou et al.).
233
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
249
7.4.2.3 Synthetic Approaches to the CalicheamicidEsperarnicin Carbohydrate Fragments The complexity of the carbohydrate fragments of calicheamicin y: (2) and esperamicin A, (3) with their multitude of highly unusual structural components, together with their importance for binding the antibiotics within the minor groove of DNA, made them compelling synthetic targets. In particular, the DNA sequence selectivity observed for calicheamicin y f (2) focused synthetic attention on the oligosaccharide of this antibiotic. Nicolaou and co-workers, in achieving the first total synthesis of the calicheamicin yf oligosaccharide [228-2321, identified the following novel and challenging structural features shown in the model compound 235 (Scheme 7-51): (a) the unusual alkoxyamine bond (B) forming a link between the carbohydrate units A and B through bonds a and y ; (b) the Bconfiguration of the glycoside bond, y, which, taken in combination with the 2-deoxy nature of saccharide B, presents a unique challenge to synthetic construction; (c) the sulfur bridge, linking carbohydrate unit B with the fully substituted aromatic ring via bonds 6 and E ; and (d) the a-configuration of the N- and S-bearing stereogenic centers of carbohydrate units A and B, respectively.
V NH2
~TBS
Scheme 7-51 Retrosynthetic analysis of the model 235 for the ABC rings of the calicheamicin y:
oligosaccharide (Nicolaou and co-workers).
In tackling the model system 235 (Scheme 7-51), Nicolaou et al. provided rather novel solutions to the synthetic challenges described above [230]. The retrosynthetic disconnections for the synthesis of 235 led to thiocarbonyldiimidazole as the sulfur source, N-hydroxyphthalimide as the origin of the alkoxyamino group, and precursors to rings A, B and C as potential starting materials. Scheme 7-52 outlines the strategy utilized in this synthesis which, in addition to solving the above problems, avoided a potentially difficult deoxygenation step to generate the desired methylene group of the B ring. Thus, intermediate 236 was designed with an ester group at position 2 in order to ensure the correct j3-configuration from the glycosidation reaction through participation (236+237) as well as to serve as a device to deliver the sulfur atom stereoselectively at position 4 via a sigmatropic rearrangement (237 + 238- 239). Compound 239 was then expected to serve as a precursor to 240. The final
250
7 The Enediyne Antibiotics
240
239
Scheme 7-52 Synthetic strategy for the construction of the central ring region 240 of the calicheamicin y: oligosaccharide (Nicolaou et al.).
outcome of the synthesis of the model 235 for the ABC rings of the calicheamicin y f oligosaccharide is presented in Scheme 7-53. In addition to the above-mentioned strategies, other highlights included efficient stereoselective reductions of both the C = O and C = N bonds (250 +235). The development of the synthesis of the model 235 for the ABC rings of the calicheamicin yf oligosaccharide by Nicolaou and co-workers provided the synthetic methodology necessary for the construction of the natural oligosaccharide itself, and its total synthesis as the methyl glycoside 251 followed swiftly from the same group (228, 229). Scheme 7-54 indicates the strategic bond disconnections which allowed the tracing of the requisite intermediates to the readily available starting materials: L-rhamnose (ring D); 3,4,5-trimethoxytoluene (ring C ) ; D-ghlCOSe (ring B); N-hydroxyphthalimide (0- NH group); D-galactose (ring A); and L-serine (ring E). The thioester linkage (CO - S ) was reserved as the key bond for the final coupling reaction. Schemes 7-55-7-57 summarize the total synthesis of 251 by Nicolaou et al. utilizing the intermediates 252, 253, 260, 261, and 265 as key building blocks. Despite the encouraging results with model system 235, the final reduction of the C=N bond in 274 proved problematic and the conditions originally reported in the preliminary communication 12281 proceeded with rather low stereoselectivity to afford a mixture of products (1 :2 in favor of the wrong isomer). This was subsequently optimized 12291 as shown in Scheme 7-57, however, to give an 86% yield of an epimeric mixture of hydroxylamines in an a/B ratio of ca. 6 : 1 in which the desired a-component predominated. Having successfully completed the synthesis of the calicheamicin yi oligosaccharide, Nicolaou et al. turned their attention to the carbohydrate fragments of esperamicin A, (3) [233]. The arylsaccharide portion was readily obtained as shown in Scheme 7-58 in the form 279 suitable for direct coupling to the esperamicin aglycone. The trisaccharide unit (282), being essentially identical to the ABE rings of the calicheamicin yi oligosaccharide, was obtained through a very slight modification of the previous synthesis as outlined in Scheme 7-59.
251
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
242 R = fflCIGH4CO
241:R = fflCIC&CO
OH
243: Ar = fflCIc&
1. NH2NH2 4
2. CSA.
MeQ.,,OBn 245
246 R
OTBS = fflCIGH.&O
(71%)
0
‘“oBn
OTBS 244: Ar = mCIGH4
OH
1. Slylation 2. DlBAL 3. Thiocarbonyldiimidazole(77%)
PhMe. 110°C
OTBS
(100%)
24% 1. DIBAL
247
Me 3.TBAF, THF W H . H20
,,OBn 1. K-selectride, DME (65%. -71 ratio)
-*80Bn,Z. “BuqNF, THF
3.EHpNH,, PPTS (81%)
Me
Me
Me
250
Scheme 7-53 Synthesis of the calicheamicin y: oligosaccharide model system 235 (Nicolaou and cO-
workers).
252
7 The Enediyne Antibiotics
L-Serine phthalimlde
HO
D-Glucose
n
3,4,5-trimethoxyj.S.J-llIIIIIIIIUxvtoluene
L-Rhamnose
Calicheamicinyll (2): R = 251: R =Me = Me
Scheme 7-54 Strategic bond disconnections and retrosynthetic analysis of the calicheamicin y: oligosac-
charide derivative 251 (Nicolaou and co-workers).
Me,
0
AcO#LAC 6Me I
OMe
252
253 AgC104 I SnCI2 (80%)
a
254: R = Ac, X = C02Me 255: R = H, X = COzMe 256: R = SiEt3, X = C02Me 257: R = SiEt3, X = CH20H 258: R = SiEb, X = CQH 4 259: R = SiEt3, X = COCI
4 F
Me,
6Me
Scheme 7-55 Synthesisof the CD ring system 259 of the calicheamiciny: oligosaccharide(Nicolaou et al.). (a) K2C0,. MeOH (100%); (b) Et,SiOTf, 2,dlutidine (92Oro); (c) DIBAL (90%); (d) RuC1, * H,O, NalO,, CC14/MeCN/H20 (75 Yo); (e) (COCI), (95 %).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics Me
0
OMe
QMe YMOC
"'OH
QMe FMOC AgCiO,. SnClz
M e q r u * E
~
(7%)
260
253
RO OR
261
NaH.(g3W2
R,R = CO L,262: 263: R = H 1
(imid)+O
(87%)
QMe TMOC
Et
(98%)
D,BAL
L,270: X = lmidazole
(88~~)
271: X = H NaSMe (95%)
QMe TMOC
OSiEt3 HS'
272
OTBS
Scheme 7-56 Synthesis of the EAB ring system 272 of the calicheamicin y j oligosaccharide (Nicolaou et al.).
254
7 The Enediyne Antibiotics
'Et OSIEt3 6Me
259 Et3N. DMAP
(44%)
OMe 1 . "BU4NF. AcOH 2. K-Selectride (75% from 273) 3.HF-Py (7%) 4. Et2NH (85%)
NaCNBH,. BFpOEt2 CH2CI2. -60 -+ -40"C (86%. 6:l mixture of isomers)
OMe
Total synthesis of the calicheamicin y: oligosaccharide as its methyl glycoside 251 (Nicolaou et Scheme 7-57
6H 251
6Me
al.).
1 . NH2NH2
NPhth
Me0
278 279
277
Scheme 7-58 Synthesis of
esperamicin A, sugar fragment 279 (Nicolaou et al.).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
255
9""
GMe TMOC
M
e
~
O
OSIEt3 ~ ~
~
w
I
o
N
~
M
e
HS' OTBS
280 1. Mel, iPr2NEt, CH2CI2 (90%) 2. "Bu~NF,AcOH, THF 3. K-Selectride (67%)
281 1. HF*Py (85%) 2. NaCNBH3, BF3.OEt2 (90%) 3. EtzNH, THF (81%)
6H
282
Scheme 7-59 Synthesis of the esperamicin Al trisaccharide 282 (Nicolaou et id.).
Danishefsky and co-workers were next on the scene with syntheses first of the esperamicin A, trisaccharide [234, 2351 and then the calicheamicin y: oligosaccharide. Highlights of the Danishefsky synthesis of the esperamicin A, trisaccharide, shown in Scheme 7-60, include (a) the utilization of glycals in glycosidation reactions, and @) the use of a urethane anion to construct the crucial C - N bond (286 + 287 +288a, b), as demonstrated by the Kahne group [236]. An interesting observation in the Danishefsky work was the rearrangement of the central pyranose ring to a pyrrolidine ring (288 b + 290) upon liberation of both the anomeric and the NH -0 groups of the central sugar unit. The following synthesis of the calicheamicin yi oligosaccharide by Danishefsky and co-workers shown in Scheme 7-61 then built upon the
256
-'r uNpht
7 The Enediyne Antibiotics
OMe
1. Dimelhyldioxirane
203 OPMB
(X
PMB-OH
285
1. ItC10,'(symcollidine)2
(49%) 2. PhaSnH, AIBN, A (84%) 3. TfZO. Py.
0-
OPMB 286: R = Me,PMB
288b:R = PMB ~TBS
287a
1. NH,NH, 2. NaCNBH3, Me2C0. 'PrOH MgSO, (85%)
12888)
3.DDQ(99%) 4. "Bu~NF(92%)
MeS'
I
289
6H
HO
1. NH2NH2 2. NaCNBH3. Me2C0, 'PrOH
M9.504 3. DDQ 4. "BU4NF
290
6H
Scheme 7-60 Synthesis of the esperamicin A, trisaccharide 288 (Danishefsky and co-workers). (PMB, p-methoxybenzyl; TEOC,2-(trimethylsilyl)ethoxycarbonyl).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
257
OM43
NaH, DMF
(80%)
I)Me
OMe
~TBS 293
dMe Scheme 7-61 Synthesis of the calicheamicin yi oligosaccharide derivative 293 (Danishefsky and COworkers).
chemistry developed in his esperamicin trisaccharide synthesis using the same basic strategy [237]. The Kahne group was the first to publish a synthesis of the 4-ethylamino sugar (ring E) of calicheamicin y! (2) and to assign its absolute configuration [238]. The same group has reported a new method for the construction of N - 0 linkages for oligosaccharides based upon urethane anion chemistry and applied it to the construction of the calicheamicin/esperamicin core trisaccharide 300 as shown in Scheme 7-62 [236]. Subunits of the calicheamicin and esperamicin oligosaccharides have also been synthesized in the laboratories of Scharf [239-2421, Mash [243] and Beau [244].
258
7 The Enediyne Antibiotics
Mevr~ (a$> 12:1,70%)
1. TsOH, H@ MeOH (90%)
DMAP (75%)
OCOPh
E)Me FOCF3
4
298
297
E)Me FOCF3
Ar
Ar
6COPh
6COPh 299
Scheme 7-62 The Kahne synthesis of the calicheamicin/esperaicin core trisaccharide fragment 300.
7.4.2.4 Total Synthesis of Calicheamicin
4
The previous two sections demonstrated that by 1990- 1991 both the Nicolaou and the Danishefsky groups were within sight of achieving total syntheses of the natural enediyne antibiotic calicheamicin 7: (2) [245]. In January 1990 the Danishefsky group reported the first racemic synthesis of the calicheamicin aglycone (222, Scheme 7-48) [218]. The first total synthesis of the carbohydrate fragment of calicheamicin 7 ; was reported by the Nicolaou group in July 1990 (Scheme 7-57) [228]. In October 1991 the Danishefsky group reported their synthesis of the calicheamicin y! oligosaccharide (Scheme 7-61) and its coupling with an immature precursor to the calicheamicin aglycone using Schmidt trichloroacetimidate methodology [237] (Scheme 7-63). Simultaneously Nicolaou, at the time lacking the calicheamicin aglycone, reported the feasibility of coupling his oligosaccharide with a model compound based upon his work relating to dynemicin A (see Section 7.4.3) and using the same trichloroacetimidate methodology (Scheme 7-64) [246]. Shortly afterwards, in December 1991, the Nicolaou group successfully completed their enantioselective synthesis of the calicheamicin aglycone (222, Scheme 7-50) [226] and, being in possession of multigram quantities of enantiomerically pure advanced precursors to both the aglycone and oligosaccharide, were in a position to push on to the end. Success finally came to the Nicoaou group in September 1992 when they completed the first total synthesis of calicheamicin 7: (2) as outlined in Scheme 7-65 (77, 227, 229, 2471. Utiliz-
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
259
TB ~
I
UID3
OMe TBSO
Scheme 7-63 The coupling of precursors to the calicheamicin y! oligosaccharide and aglycone (Danishefsky et al.).
1. hv (8%) 2. NaH. C13CCN
X = Pnitrobenzyl E303: 304:X = C(=N)CCl3
I
Scheme 7-64 Synthesis of the calicheamicin/dynemicin hybrid 306 (Nicolaou et al.).
305
I
260
7 The Enediyne Antibiotics
AcS
310
HO
'I
1 . TESOTl. 'PrflEl. CH&C 2 . AcOH. EIOAc. H20 H20 (75%) 3. DIBAL
1 4 . K ( ~ 6 m i o ) p h l h a l i m i d e(57%)
Scheme 7-65 The total synthesis of calicheamicin -f! (2) (Nicolaou et al.).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
261
ing the Schmidt trichloroacetimidate methodology [248, 2491, the suitably protected advanced precursors 304 and 307 to the oligosaccharide and aglycone were coupled to give 308 containing the complete framework of calicheamicin y f (2) [247]. Subsequent introduction of the thioacetate precursors for the trisulfide trigger (308 + 309), reduction of the oxime to the alkoxyamine 310 and introduction of the trisulfide gave the fully protected calicheamicin 311. Finally, removal of the carefully chosen protecting groups in the correct order completed the synthesis to give calicheamicin yi (2), which was identical in all respects to the naturally occurring compound, and hence confirmed its absolute structure [247].
7.4.3 Dynemicin Synthetic Studies When the structure of dynemicin A (4) was revealed in 1989 [24, 251, the combination of its intriguing novelty, potent antitumor and antibiotic properties, and low toxicity by comparison with other enediyne antibiotics such as calicheamicin yi (2) were sufficient to ensure the attention of a scientific community already fascinated by its forebears. Nicolaou et al. were first to identify and study the features of the natural product essential for its activation through the design and synthesis of model systems based on 312 (Scheme 7-66) [250, 2511. These were traced retrosynthetically to tetrahydrophenanthridine (313) and were synthesized as summarized in Scheme 7-67. Highlights of this synthetic approach include: (a) regiospecific functionalization of the tetrahydrophenanthridine ring (313 + 314 and 314 + 315); (b) palladium-catalyzed construction of the enediyne (316 +317); (c) an intramolecular acetylide- ketone condensation to close the ten-membered ring (317 --* 318); and (d) radical-induced deoxygenation (319 + 320).
dOO 0
\
OH
0
OMe
OH
Dynemlcln A (4) Functionalbe
312
Functlonalize
Scheme 7-66 Synthetic strategy for dynemicin A model systems (Nicolaou et d.).
262
7 The Enediyne Antibiotics
1. mCPBA (80%) 2. AC20. A (77%)
Phococl 4. 'BuMe+iOTf (92%)
313
1
314
I
318
317
1.2;B"A
(85%)
2. "EU~NF(95%)
316
(l~nid)~C=S. DMAP CH2CI2 (95%)
319
320
Scheme 7 4 7 Synthesis of dynemicin A model systems (Nicolaou et d.).
The chemistry of these dynemicin A models proved to be quite fascinating [250, 2511. The dynemicin model 318 was readily triggered to undergo the reaction cascade shown in Scheme 7-68by treatment with acid. This caused acid-catalyzed epoxide opening to give 322a [calculated c . . d distance = 3.19 and was simultaneously accompanied by spontaneous cycloaromatization and pinacol rearrangement leading to 324. It is interesting to speculate about the timing of the bond migration leading to ring contraction and whether the sequence included the ring-contracted enediyne 325. Blocking the final rearrangement step (R = H or Ac) resulted in the isolation of the structures 323b-e (Scheme 7-68). However, if the enediyne moiety of 318 was first protected as the cobalt complex 326 (Scheme 7-69)and thus prevented from undergoing cycloaromatization, acid-induced epoxide opening was still accompanied by the pinacol rearrangement to give the masked enediyne contained within the bridging nine-membered ring of 327 [251]. The structure of 327 was confirmed by X-ray crystallographic analysis of its acetate derivative, indicating that not only had ring contraction occurred, but also that one of the acetylenic dicobalt complexes had been regiospecifically lost. Oxidative decomplexation of 327 was accompanied by spontaneous cycloaromatization leading to 324, presumably via 328.
A],
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
Acid, [ H I source
263
*
Epoxide opening
--..
318: R = OH 320: R = H 321: R = OAC
--- 322b: 322a: X = R = OH X = OH; R = H L
Bond ‘..migration
‘\ r
Bond migration ,**
1 8 k
1
PhO
L
.a-
322~: X = OH; R = OAC 322d: X = CI; R OAC 322e: X = CI; R = H
Bergman Cycloarornatiration
325
PhO
324
[32%: X = R = OH] 323b: X = OH;R = H 323~: X = OH;R = OAC 3234: X = CI; R = OAC 3 2 3X ~ = CI; R = H
Scheme 7-68 Chemistry of dynemicin A model systems. Triggering the Bergman cyclization/pinacol rearrangement cascade (Nicolaou et al.).
Nicolaou et al. then proceeded to study a second generation of “designed enediynes” 329 (Scheme 7-70) based upon the above dynemicin model systems. These new compounds were equipped with acid-, base-, and photosensitive triggering devices which could be activated under mild and potentially physiological conditions to undergo the reaction cascades depicted in Scheme 7-70.Noteworthy is the fact that these two cascade scenarios closely mimic the proposed activation cascade of dynemicin A (4) itself as described in Section 7.3.4, involving iminoquinone methide species 331a (related to dynemicin-derived 44 a) and quinone methide species 331 b (related to dynemicin-derived 44). The enediyne 336 was designed in order to give easy access to the parent compound 330a as shown in Scheme 7-71 [119, 2521. The dynemicin models such as 320 and 336 in which the nitrogen is protected as a carbamate are robust molecules showing no tendency to undergo epoxide opening and cycloaromatization. However, the enediyne 330a proved to be too labile for isolation. The reactivity of 330a and its ability to cause double-stranded DNA breaks supports the notion of electron push from the free nitrogen participating in the formation of an
264
7 The Enediyne Anfibiotics
318
324
Fe(NO& of Me3N+-O‘ 4
CH,C12,25 “C (&%J L
327
Scheme 7-69 Synthesis and chemistry of cobalt complexes of dynemicin A model systems. Triggering of the Bergman cycloaromatization by decomplexation (Nicolaou et aL)-
ortho-iminoquinone methide species (331a, Scheme 7-70). This suggests that a similar contribution from the nitrogen in the dynemicin A cascade (compound M a , Scheme 7-70)may, at least in part, account for the triggering of the natural product following bioreduction. The anthraquinone portion of dynemicin A can thus be envisaged as acting as a (‘lock’’ for the epoxide by withdrawing electron density from the N-atom and hence preventing epoxide opening in a similar manner to the way the carbamate acts as a “iock” for these dynemicin model systems. The structure of 330a, which was also generated in solution by LiAIH, reduction of 320 (Scheme 7-72), was deduced from a number of trapping experiments as summarized in Scheme 7-71 11191 and Scheme 7-72 [251]. By contrast to 330a, the methoxy compound 338 (Scheme 7-71) proved to be a stable, crystalline compound whose structure was confirmed by X-ray crystallographic analysis and which underwent the expected dynemicin-iype cascade upon treatment with acid (Scheme 7-71}. Here, the methoxy substituent is presumably having an electronically destabilizing effect upon tbe formation of the adjacent carbonium ion involved in the activation cascade. The instability of the enediyne-containing dynemicin model compounds in which the epoxide had been opened prompted Nicolaou et al. to seek a means by which they could tame the reactivity of the cis-diols sufficiently to observe them [253,254]. By comparing the resonance energies of benzene (36 kcal/mol), naphthalene (61 kcal/mol) and anthracene(84 kcal/mol) it was reasoned that there would be less of a driving force for cycloaromatization of the diols 346 and 347 compared with 345 (Scheme 7-73). Indeed, treatment of 343 and 344 with silica
265
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
-
Hydroxylation
Hydrolysis
___l__t
= on)
(R' = H)
R2
HO
329
I"'
1-
OMe
&)
.H
44
OH
OH OH
II
i l l
& 0
0
HO
I 332a
iiU
I
332b
333b
333a
Scheme 7-70 Simulation of the dynemicin cascade by model compounds (Nicolaou et al.).
gel in wet benzene led smoothly to the cis-diols 346 and 347. The benzene diyne 346 was stable enough to be detected by TLC and 'H NMR spectroscopy, undergoing cycloaromatization with a half-life of -2.5 h at 20°C,whilst the naphthalene diyne 347 displayed enhanced stability with a corresponding half-life of 44 h at 37 "C. The second activation cascade indicated in Scheme 7-70 for the Nicolaou dynemicin model systems was demonstrated through the triggering of the phenol derivative 352, cleanly produced by photolytic deprotection of the o-nitrobenzyl ether 351 (Scheme 7-74) [119, 1201. "keatrnent of 352 with nucleophiles (e.g., EtOH, EtSH, n-PrNH,) provided the aromatization
-
266
7 The Enediyne Antibiotics
320: R = H 321: R = OMe
336:R=H 337: R = QMe Cs$03 or DBU
PhOH orPhSH, C+C03
*
0
.25 "c (for 33Oa)
or TsOH. 60 "C
0
, (for 338)
$)
"'N
0
R
339a:X=OPh;R=H 339b: X = SPh; R = H 339c: X = OH;R = QMe
33Oa: R = H 338: R = OMe (97%)
34Oa: X = OPh; R = H
341a: X = OPh; R = H (25%) 341b: X = SPh; R = H (33%) 341c: X = OH; R = OMe (20%)
340b: X = SPh; R = H 34W. X = OH; R = OMe
Scheme 7-71 Chemistry of dynemicin A model systems. Triggering the dynemicin cascade with base or acid (Nicolaou et d.).
products 355. An important observation in these reactions was the isolation of the quinone epoxide 356 from the reaction with ethanol under aerobic conditions. The isolation of 356 provided direct evidence for the intermediacy of quinone methide 353,being formed by trapping 353 with molecular oxygen [255]. The 2-(pheny1sulfonyl)ethylcarbamate nitrogen-protecting group contained in enediyne 336 (Scheme 7-71) proved to be an important discovery. The facile manner in which it could be removed under mildly basic conditions, even undergoing slow release at physiological pH,
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
UAIH,.THF
0 "C. 30 min
267
'& 00:::3%
Scheme 7-72 Chemistry of dynemicin model systems. Triggering the dynemicin cascade by reduction (Nicolaou et al.).
HN
330a: n = 0 343: n = 1 344:n=2
345: n = 0 346: n = 1 347: n = 2
348: n = 0 349: n = 1 350: n = 2
Scheme 7-73 Dynemicin A model systems with tempered reactivity (Nicolaou et al.).
resulted in the possibility of it acting as a prodrug for the unstable and cytotoxic free amine 330a (ICs0 -1.6 x lo-'' M against the MOLT-4 leukemia cell line). Thus, whilst the stable M), the phenyl carbamate displayed weak cytotoxicity against MOLT-4 leukemia (3.1 x corresponding 2-(phenylsulfony1)ethylcarbamate 336 displayed enhanced cytotoxicity with an IC,, of -2.5 x lo-" M comparable with that of dynemicin A itself. Further structural modifications led to the finding that compound 357 (Scheme 7-75) had an ICs0 of 2.0 x M against the same cell line [256], making it one of the most potent in-vitro antitumor agents which has been reported to date. Scheme 7-75 shows a number of other
268
7 The Enediyne Antibiotics
1
353
Scheme 7-74 Photoinduced simulation of the dynemicin A cascade. Evidence for a quinone methide intermediate (Nicolaou et al.). NBn, o-nitrobenzyl.
1
R 336:R=H 3$1:R = OCH2CH20H
360
OMe 358
361
359
362:R, = Me; Rp = H 363: R1 = H; R2 = Me 364: R, = R2 = Me
Scheme 7-75 Designed enediynes with novel triggering devices and tempered reactivity (Nicolaou et d.).
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
269
designed enediynes with triggering devices and tempered reactivity synthesized by Nicolaou et al. (358 [5], 359 [253], 360 [253], 361 [252], and 362-364 [257], some of which demonstrated significant DNA-cleaving properties. The o-nitrobenzyl carbamate 358 is particularly interesting since it contains a photosensitive triggering device and cleaves DNA upon irradiation. Table 7-5 shows the ICso values determined for enediyne 357 with 21 different cell lines [256]. There are significant differences in cytotoxicities, ranging from M for the highly M for the highly sensitive leukemia cell line. Parresistant melanoma cell lines to ticularly important is the high cytotoxicity against the multiple-drug-resistant TCAF-DAX cell line (ICso = 1.7 x M). Another striking feature is its relatively low cytotoxicity against a number of nontransformed cell lines; preliminary in-vivo studies with animals infected with leukemia and solid tumors also show encouraging results [258]. In order to confirm that the remarkable cytotoxicity of enediyne 357 is indeed due to DNA damage, MOLT-4 leukemia cells were treated with ethidium bromide, which intercalated into the DNA rendering it M led to fluorescent. Exposure of these cells to enediyne 357 at a concentration of rapid DNA strand breakage as determined by fluorimetry, resulting in 95 % destruction after Table 7-5 Cytotoxicities of designed enediyne 357 against a panel of 21 tumor cell lines (top group) and four normal cell lines (bottom group)
Cell type
Cell line
IC,, (MI
Melanoma Melanoma Melanoma Colon carcinoma Ovarian carcinoma Ovarian carcinoma Astrocytoma Glioblastoma Breast carcinoma Lung carcinoma Lung carcinoma Lung carcinoma Lung carcinoma Pancreatic carcinoma T-cell leukemia T-cell leukemia (a) Myeloma Mouse leukemia Mouse leukemia Promyeocytic leukemia T-cell leukemia
SK-Mel-28 M-14 M-21 HT-29 Ovcar-3 Ovcw-4 U-87 UG U-521 MG MCF-7 H-322 H-358 H-522 UCLA P-3 Capan-1 TCAF TCAF-DAX RPMI-8226 P-388 GI210 HG60 Molt-4
3.1 x 1.6 x 1.6 x 1.6 x 7.8 x 7.8 x 7.8 x 3.9 x 7.8 x 3.9 x 2.0 x 9.8 x 9.8 x 3.1 x 1.1 x 1.7 x 7.7 x 4.6 x 1.3 x 3.6 x 2.0 x
HNBM
5.0 x 1 0 - ~ 6.3 x 5.0 x 3.1 x
Bone marrow Human mammary epithelial cells Normal human dermal fibroblast Chinese hamster ovary (a) Multiple drug resistant cell line.
HMEC NHDF CHO
10-7
10-~ lo-' 10-7 10-7 10-~ lo-' lo-' 10-9
10-9 10-~ 10-9 10-9 10-~ lo-" 10-l~
270
7 The Enediyne Antibiotics
4 h at 37 "C. Cell death showed an approximately 2 h delay relative to DNA strand breakage, implicating DNA damage as the direct cause of cell death in these experiments. It was also shown that enediyne 357 severely impairs the ability of MOLT-4 leukemia cells to synthesize DNA (inhibition of [3H]thymidine uptake), RNA (inhibition of [3H]uracil uptake), and protein (inhibition of 13H]leucineuptake). Treatment of MOLT-4 cells with enediyne 357 under appropriate conditions followed by observation of cell morphology and cell death revealed the phenomenon of programmed cell death (apoptosis) I2591 as the prevailing cause of cell destruction [260]. Furthermore, competition experiments using enediynes with relatively low toxicities resulted in the identification of certain inhibitors of apoptosis. Specifically, the methoxy enediyne 361 (Scheme 7-75), which displayed diminished tendency to undergo the Bergman reaction, inhibited the cytotoxic action of compound 357. Thus, when MOLT-4 cells were preincubated with enediyne 361 at M for 1 h prior to treatment with the cytotoxic compound 357, a dramatic reduction by a factor of lo5 was observed in the cytotoxicity of 357. Similar reductions by factors of lo2 - lo4 were observed in the cytotoxicities of the naturally occurring enediynes dynemicin A (4) and calicheamicin y: (2). Particularly intriguing was the observation that 361 inhibits apoptotic morphology of cell death by powerful inducers of apoptosis such as actinomycin D and cycloheximide, although cell viability was not affected in these cases. Further insight into the remarkable cell-type selectivity observed with these designed enediynes was obtained by comparing the cytotoxicities (Table 7-6) of enantiomerically pure compounds (+)-336 and (-)-336 [261] and the methyl-substituted compounds 362-364 (Scheme 7-75). These experiments demonstrated dramatic differences in potencies depending upon the enantiomeric form of the enediyne 336 and the degree and stereochemistry of methyl substitution in compounds 362-364, raising intriguing questions: Is there an intracellular receptor for these enediynes other than DNA? Could this putative receptor serve as a capturing and delivery system for these enediynes to specific sequences of DNA? Is there a prevailing biological mechansim in certain cell types which facilitates the I)-elimination? These questions raise even more interesting issues concerning the regulation of cell death. Exploitation of such observations and elucidation of the mechanism of action of these agents may lead to new approaches to drug design. Table 7-6 Cytotoxicities (IC5,JM) of designed enediynes containing p-sulfone triggers (Nicolaou et a!.) Cell line
Cell type
SK-Mel-28 Melanoma Capan-1 Pancreatic carcinoma MCF-7 Breast carcinoma HL60 Prornyeocytic leukemia Molt-4 T-cell leukemia
(+)-336
6.3 x
10-6
(-)-336
362
6.3 x 10-6
< 1.0 x 10-4
3.9 x 10-7 7.8 x
1.6 x 10-6
>9.8 x 1.0 x 1 0 - l ~
7.8 x lo-' 1.0 x
1.6 x
6.3 x <1.0
x
364
363
10-6
5.0 x 10-5
3.1 x
10-6
2.5 x lo-'
1.6 x 7.8 x lo-' 1.0 x I O - ~ 1.0 x
< 1.0 x 10-4 5.0 x 10-5 2.5 x lo-'
2.5 x 1.0 x
Wender et al. have also synthesized dynemicin A model systems (Scheme 7-76). In this approach, closure of the ten-membered enediyne ring was achieved by fluoride-induced desilylative condensation of a silyl-protected alkyne with an aldehyde (365 366) [262]. Following the example of Nicolaou et al., these researchers have also explored simulation of the -+
271
7.4 Theoretical and Synthetic Studies on the Enediyne Antibiotics
dynemicin cascade (366 + 367) [262] and photochemically triggered systems through the use of an o-nitrobenzyl carbamate [263]. Isobe and co-worker have also synthesized similar systems [264-2661. Magnus and Fortt synthesized and studied the core tetrahydroquinoline enediyne structure 372 utilizing a cobalt-acetylene complex as a precursor, as summarized in Scheme 7-77 [2671.
365
366
367
Scheme 7-76 Synthesis and activation of dynemicin A model 366 (Wender et al.).
'OTHP
374
L
373
372
Scheme 7-77 Synthesis and chemistry of dynemicin A model 372 (Magnus et d.).
Schreiber et al. have been able to apply their enediyne intramolecular Diels-Alder approach to the synthesis of dynemicin model systems [268-2701, culminating in the total synthesis of di- and tri-0-methyl dynemicin A methyl esters 388 and 389 (Scheme 7-78) [271], derivatives of the natural product itself. Highlights of this synthetic approach include: (a) intramolecular lactonization and concomitant Diels- Alder cyclization (380 -P 381); (b) allylic hydroxylation followed by an allylic diazene rearrangement in order to regiospecifically isomerize a double bond (381 + 382); (c) a-hydroxylation of the lactone 381 and subsequent conversion to the pketoester 383; (d) annelation of the anthraquinone unit (383 384 + 385 386); (e) mild base-induced p-elimination of the N-protecting group of 386 to give the free amine 387; and ( f ) a final oxidation to complete the anthraquinone (387 * 388). +
+
272
7 The Enediyne Antibiotics
,
c0::
Me
"BurSn
Pd(PPhJ), (85%)
OMe 377
OMe
375
I
1, BrCH=CHC02Me
Pd(PPh& Cul
1. DBU (92%) 2. CAN (97%) 7 3. Me2AICI 4. Mesitylenesulfonoh drazide (92%) 5. &MDS. M ~ P H OMe
6Me
382:R = (CH2)30Bz (720A)
.
.
381: R = (CH2)sOBz 'PyBroP
1 . NaBH4 2. NaOMe 3. (Cl~CO),Co 4. Dess-MarlinDeriodinane
= Bromo-trir-pynolldino-phosphonium herafluoro-phosphate
#'
Me0
'. Me0 OM I
.AgOTf
Br
. )
383:R = (CH2)&02Me MeAICI2. Et,SiH
3.D W (51%)
co2
4. mCPBA (73%)
4
5. DBU
\
WO
, 0
OMe 385: R = (CH2)2COzMe
387:R=H
OMe
~0
0
OR
Cs2CQG389:R=Me (50%. 3 steps)
Scheme 7-78 Total synthesis of di- and tri-0-methyl dynemicin A methyl esters 388 and 389 (Schreiber et al.).
7.5 Medical Applications of the Enediyne Antibiotics
273
7.4.4 The Chromoprotein Enediyne Antibiotics The similarities of the core enediynes of the kedarcidin and C-1027 chromophores with the neocarzinostatin chromophore structure means that many of the synthetic studies relating to NCS chromophore also, by implication, relate to the other chromoprotein enediyne antibiotics, although no examples specifically relating to these new antibiotics have appeared to date. However, a synthesis of kedarosamine, a sugar component of the kedarcidin chromophore, has appeared recently [272].
7.5 Medical Applications of the Enediyne Antibiotics The studies described in this chapter have shown how chemists have tackled the mechanistic and synthetic challenges resulting from the discovery of the enediyne antibiotics. This has led to the development of “designed enediynes” such as the dynemicin A models of Nicolaou et al. described in Section 7.4.3 which show great promise both in vitro and in vivo as antitumor agents. A number of other “designed enediynes” have also been prepared, such as 306 (Scheme 7-64)[246] and those shown in Scheme 7-79 with the aim of producing new antineoplastic agents. The hybrids 306 [246], 390 [l], and 391 [l] were prepared by Nicolaou et al. in order to attach delivery systems (i.e, minor-groove binders and intercalators) to the dynemicin models and thus improve the DNA affinityhelectivity. The synthetic calicheamicin 0: (392) (Nicolaou et al.) [273] containing a thioacetate triggering device is considerably more potent than any previously known enediyne (natural or synthetic). The systems 393, 394, and 395 have also been reported by Toshima et al. [274], Hirama and co-workers [275], and Boger and Zhou [276], respectively. There has been considerable interest among the natural products themselves as anticancer agents in the pharmaceutical industry, with a number of clinical trials having been carried out. Esperamicin A, (3) is currently undergoing Phase I clinical trials (Bristol-Myers). Calicheamicin 7 : (2), by contrast, is too toxic for use as a drug, having an MED greater than the LD,, . Calicheamicin-antibody immunoconjugates offer great promise, however, with the most advanced product currently in Phase I trials against ovarian cancer (Celltech/American Cyanamid) and others in preclinical trials against acute myeloid leukemia and colorectal cancer. Perhaps the most studied systems relate to neocarzinostatin, however, since this has been available the longest. This has been shown to possess antitumor activity in patients with liver cancer, bladder cancer, stomach cancer, and leukemia, as well as in various animal tumors [277]. SMANCS is a product of conjugation of neocarzinostatin with poly(styrene-co-maleic acid) which has shown good antitumor activity in animal models following oral administration (2781, and immunoconjugates of neocarzinostatin such as A7-NCS [279] have undergone clinical evaluation, showing increased survival times when administered to post-operative cancer patients (both with and without metastases) when compared with other chemotherapies.
274
7 The Enediyne Antibiotics
pho$JNp& $$h”/Me2
N
390 (Nicolaou)
H
OMe
HO 392 (Nicolaou)
& O $
0
391 (Nicolaou)
g P
o 395 (Boger) 394 (Hirama)
Scheme 7-79 Some designed enediynes.
7.6 Concluding Remarks With their unprecedented molecular structures, fascinating mode of action, and phenomenal biological activities, the enediyne anticancer antibiotics have excited the creative impulses and imagination of many researchers and thus established a vibrant field of investigations. Research in this rapidly evolving field spans areas from computational chemistry, chemical synthesis, molecular recognition, DNA chemistry, and medicine. In the short period between this review and the first comprehensive review in the field [l] much new information has appeared, including the first successful syntheses of two of the
7.6 Concluding Remarks
215
natural products themselves. Indeed, virtually every week, important new information appears in the latest issues in our libraries. It is likely that the family of natural enediyne antibiotics will be much expanded a few years from now, and already there are synthetic analogs which rival the natural products themselves, paving the way for important advances in the fight against cancer. And yet it is likely that these exciting developments would have had to wait many years had it not been for the guiding hand of Nature. This should serve as a constant reminder to those with little appreciation for natural products chemistry that this discipline continues to provide valuable leads which become the guiding forces in our search for new scientific ventures and the “wonder drugs” of tomorrow.
Abbreviations AIBN CAN CSA DABCO DBU DDQ DEAD DET DIAD DIBAL DME DMSO HMDS HMPA HOMO
2,2’-azobisisobutyronitrile ceric ammonium nitrate camphorsulfonic acid 1,4-diazabicyclo[2.2.2]octane 1,S-diazabicyclo[5.4.01undec-Sene 2,3-dichloro-5,ddicyano-l,Cbenzoquinone diethyl azodicarboxylate diethyl tartrate diisopropyl azodicarboxylate diisobutylaluminum hydride 1,Zdimethoxyethane dimethyl sulfoxide hexamethyldisilazide hexamethylphosphormide highest occupied molecular orbital median inhibitory concentration G o dissociation constant KD median lethal dose LDSO lithium diisopropylamide LDA LUMO lowest unoccupied molecular orbital mCPBA rn-chloroperbenzoic acid mean effective dose MED methoxyethylmethyl MEM methanesulfonyl Ms MTAD 4-methyl-1,2,4-triazoline-3,5-dione NBS N-bromosuccinimide NCS neocarzinostatin NCS-chr neocarzinostatin-chromophore N-methylmorpholine N-oxide NMO NOE nuclear Overhauser effect PCC pyridinium chlorochromate Phth phthaloyl PMB p-methoxybenzyl
276 PPTS PY PYr S,2' TBAF TBS TES Tf TFA Thexyl THF THP TMS
7 The Enediyne Antibiotics
pyridinium p-toluenesulfonate pyridine pyridine conjugate addition tetra-n-butylammonium fluoride tert-butyldimethylsilyl triethylsilyl trifyl (trifluoromethanesulfonyl) trifluoroacetic acid tetrahydrofuran tetrahydropyranyl trimethylsilyl
References [l] K. C. Nicolaou, W.-M. Dai, Angew. Chem., Int. Ed. Engl. 1991,30, 1387. [Z] M. D. Lee, G. A. Ellestad, D. B. Borders, Acc. Chem. Res. 1991, 24, 235. 131 I. H. Goldberg, Acc. Chem. Res. 1991,24, 191. [4] K. C. Nicolaou, A. L. Smith, Acc. Chem. Res. 1992,25, 497. [5] K. C. Nicolaou, A. L. Smith, E. W. Yue,Roc. Nutl. Acud. Sci. USA 1993,90, 5881. [6] R. R. Jones, R. G. Bergman, J. Am. Chem. SOC. 1972, 94, 660. [7] R. G. Bergman, Acc. Chem. Res. 1973, 6, 25. [8] T. P. Lockhart, P. B. Comita, R. G. Bergman, J. Am. Chem. SOC. 1981, 103, 4082. (91 T. P. Lockhart, R. G. Bergman, J. Am. Chem. SOC.1981, 103, 4091. [lo] J. Mayer, F. Sondheimer, J. Am. Chem. SOC. 1966, 88, 602. [I11 N. Darby, C. U. Kim, J. A. Salatin, K. W. Shelton, S. Takada, S. Masamune, J. Chem. SOC, Chem. Commun. 1971, 1516. 1121 H. N. C. Wong, F. Sondheimer, Etrahedron Lett 1980, 21, 217. (131 J. W. Grissom, T. L. Calkins, Tetmhedron Lett. 1992, 33, 2315. [14] J. W. Grissom, T. L. Calkins, J. Org. Chem. 1993,58, 5422. [15] P. A. Magriotis, K. D. Kim, J. Am. Chem. SOC. 1993, 115, 2972. 1161 Y.-F. Lu, C. W. Harwig, A. G. Fallis, I Org. Chem. 1993, 58, 4202. [17] M. D. Lee, T. S. Dunne, M. M. Siegel, C. C. Chang, G. 0. Morton, D. B. Borders, J. Am. Chem. SOC. 1987,109, 3464. [18] M. D. Lee, T. S. Dunne, C. C. Chang, G. A. Ellestad, M. M. Siegel, G. 0. Morton, W. J. McGahren, D. B. Borders, J. Am. Chem. Soc,1987, 109, 3466. (191 M. D. Lee, T. S. Dunne, C. C. Chang, M. M. Siegel, G. 0. Morton, G. A. Ellestad, W. J. McGahren, D. B. Borders, J. Am. Chem. SOC. 1992, 114, 985. [20] J. Golik, J. Clardy, G. Dubay, G. Groenewold, H. Kawaguchi, M. Konishi, B. Krishnan, H. Ohkuma, Am. Chem. Soc. 1987, 109, 3461. K. Saitoh, T. W. Doyle, .l [21] J. Golik, G. Dubay, G. Groenewold, H. Kawaguchi, M. Konishi, B. Krishnan, H. Ohkuma, K. Saitoh, T. W. Doyle, I Am. Chem. SOC. 1987, 109, 3462. [22] N. Ishida. K. Miyazaki, M. Kumagai, M. Rikimaru, J. Antibiot. 1965, 18, 68. I231 K. Edo, M. Mizugaki, Y.Koide, H. Seto, K. Furihata. N. Otake, N. Ishida, Tetmhedron Lett. 1985, 26, 331.
References
217
[24] M. Konishi, H. Ohkuma, K. Matsumoto, T. Tsuno, H. Kamei, T. Miyaki, T. Oki, H. Kawaguchi, G. D. VanDuyne, J. Clardy, J. Antibiot. 1989, 42, 1449. [25] M. Konishi, H. Ohkuma, T. Tsuno, T. Oki, G. D. VanDuyne, J. Clardy, J. Am. Chem. Soc. 1990, 112, 3715. (261 S. J. Hofstead, J. A. Matson, A. R. Malacko, H. Marquardt, J. Antibiot. 1992,45, 1250. [27] J. E. Leet, D. R. Schroeder, S. J. Hofstead, J. Golik, K. L. Colson, S. Huang, S. E. Klohr, T. W. Doyle, J. A. Matson, J. Am. Chem. SOC. 1992, 114, 7946. [28] J. E. Leet, D. R. Schroeder, D. R. Langley, K. L. Colson, S. Huang, S. E. Klohr, M. S. Lee, J. Golik, S. J. Hofstead, T. W. Doyle, J. A. Matson, J. Am. Chem. SOC. 1993, 115, 8432. [29] Y. Minami, K.4. Yoshida, R. Azuma, M. Saeki, T. Otani, Tetrahedron Lett. 1993, 34, 2633. [30] K.-i. Yoshida, Y. Minami, R. Azuma, M. Saeki, T. Otani, Tetrahedron Lett. 1993, 34, 2637. [31] J. Hu, Y.-C. Xue, M.-Y. Xie, R. Zhang, T. Otani, Y. Minami, Y. Yamada, T. Marunaka, J. Antibiot. 1988, 41, 1575. [32] T. Otani, Y. Minami, T. Marunaka, R. Zhang, M.Y. Xie, J. Antibiot. 1988, 41, 1580. [33] T. Otani, Y. Minami, K. Sakawa, K.4. Yoshida, J. Antibiot. 1991, 44, 564. [34] T. Otani, J. Antibiot. 1993,46, 791. [35] M. Hanada, H. Ohkuma, T. Yonemoto, K. Tomita, M. Ohbayashi, H. Kamei, T. Miyaki, M. Konishi, H. Kawaguchi, S. Forenza, J. Antibiot. 1991, 44, 403. [36] D. R. Schroeder, presented at the 3rd International Conference on the Biotechnology of Microbial Products, Society for Industrial Microbiology, Rohnert Park, CA, April 1993; Paper S4. [37] M. A. Napier, B. Holmquist, D. J. Strydom, I. H. Goldberg, Biochem. Biophys. Res. Commun.1979, 89, 635. [38] W. Koide, F. Ishii, K. Hasuda, Y. Koyama, K. Edo, S. Katamine, F. Kitame, N. Ishida, J. Antibiot. 1980,33, 342. [39] H. Suzuki, K. Miura, K. Kumada, T. Takeuchi, N. Tanake, Biochem. Biophys. Res. Commun. 1980, 94, 225. [40] N.Sakata, S. Minamitani, T. Kanbe, M. Hori, M. Hamada, K. Edo, Biol. Pharm. Bull. 1993, 16,26. [41] A. Teplyakov, G. Obmolova, K. Wilson, K. Kuromizu, Eur. J. Biochem. 1993,213, 737. [42] T. Tanaka, M. Hirama, K.-i. Fujita, S. Imajo, M. Ishiguro, J. Chem. SOC,Chem. Commun. 1993, 1205. [43] K.-H. Kim, B.-M.Kwon, A. G. Myers, D. C. Rees, Science 1993,262, 1042. [44] N. Zein, A. M. Casazza, T. W. Doyle, J. E. Leet, D. R. Schroeder, W. Solomon, S. G. Nadler, Proc. Natl. Acad. Sci. USA 1993, 90, 8009. [45] L. F. Povirk, I. H. Goldberg, Biochemistry 1980,19, 4773. [46] G. Jung, W. Ke)hnlein, Biochem. Biophys. Res. Commun. 1981, 98, 176. [47] L. F, Povirk, I. H. Goldberg, Biochemistry 1984, 23, 6304. 1481 L. S. Kappen, I. H. Goldberg, Nucleic Acids Res. 1978,5, 2959. [49] Y. Uesawa, J. Kuwahara, Y. Sugiura, Biochem. Biophys. Res. Commun.1989, 164, 903. [SO] A. G. Myers, Tetrahedron Lett. 1987, 28, 4493. [511 A. G. Myers, P. J. Proteau, T. M. Handel, J. Am. Chem. Soc. 1988, 110, 7212. 1521 A. G. Myers, P. J. Proteau, J. Am. Chem. SOC. 1989, 111, 1146. [531 0. D. Hensens, I. H. Goldberg, J. Antibiot. 1989, 42, 761. [54] A. G. Myers, P. M. Hamngton, B.-M. Kwon, J. Am. Chem. SOC. 1992, 114, 1086. [521 H. Sugiyama, K. Yamashita, M. Nishi, I. Saito, Tetrahedron Lett. 1992,33, 515. 1561 H. Sugiyama, T. Fujiwara, H. Kawabata, N. Yoda, N. Hirayama, I. Saito, J. Am. Chem. SOC. 1992, 114. 5573. 1571 D.-H. Chin, I. H. Goldberg, J. Am. Chem. SOC. 1993, 115, 9341. [581 L. F. Povirk, N. Dattagupta, B. C. Warf, I. H. Goldberg, Biochemistry 1981, 20, 4007. 1591 S. H. Lee, I. H. Goldberg, Biochemistry 1989, 28, 1019. @[I) I. H. Goldberg, Free Radicals Biol. Med. 1987,3, 41.
278
7 The Enediyne Antibiotics
M. Lamothe, P. L. Fuchs, I Am. Chem. SOC. 1993, 115, 4483. L. S. Kappen, I. H. Goldberg, Biochemistry 1983, 22, 4872. D.-H. Chin, L. S. Kappen, I. H. Goldberg, Proc. Natl. Acad. Sci. USA 1987, 84, 7070. L. S. Kappen, I. H. Goldberg, S. H. Wu, J. Stubbe, L. Worth, Jr., J. W. Kozarich, J. Am. Chem. SOC. 1990, 112, 2791. [65] L. S. Kappen, I. H. Goldberg, B. L. Frank, L. Worth, Jr., D. F. Christner, J. W. Kozarich, J. Stubbe, Biochemistry 1991, 30, 2034. [66] I. Saito, H. Kawabata, T. Fujiwara, H. Sugiyama, T. Matsuura, J. Am. Chem. SOC. 1989, 111, 8302. [67] B. L. Frank, L. Worth, Jr., D. F. Christner, J. W. Kozarich, J. Stubbe, L. S. Kappen, I. H. Goldberg, J. Am. Chem. SOC. 1991, 113, 2271. [68] D.-H. Chin, I. H. Goldberg, Biochemistry 1993, 32, 3611. [69] L. S. Kappen, I. H. Goldberg, Science 1993, 261, 1319. [70] 0. D. Hensens, J.-L. Giner, I. H. Goldberg, J. Am. Chem. SOC.1989, I l l , 3295. [71] M. D. Lee, J. K. Manning, D. R. Williams, N. Kuck, R. T. Testa, D. B. Borders, J. Antibiot. 1989, 42, 1070. [72] W. M. Maiese, M. P. Lechevalier, H. A. Lechevalier, J. Korshalla, N. A. Kuck, A. Fantini, M. J. Wildey, J. Thomas, M. Greenstein, J. Antibiot. 1989, 42, 558. [73] B. Zhao, S. Konno, J. M. Wu, A. L. Oronsky, Cancer Lett. 1990, 50, 141. [74] N. Zein, A. M. Sinha, W. J. McGahren, G. A. Ellestad, Science 1988, 240, 1198. [75] N. Zein, M. Poncin. R. Nilakantan, G. A. Ellestad, Science 1989, 244, 697. [76] N. Zein, W. Solomon, A. M. Casazza, J. F. Kadow, B. S. Krishnan, M. M. Tim, D. M. Vyas, T. W. Doyle, Bioorg. Med. Chem. Lett. 1993, 3, 1351. [77] K. C. Nicolaou, C. W. Hummel, E. N. Pitsinos, M. Nakada, A. L. Smith, K. Shibayama, H. Saimoto, J. Am. Chem. SOC. 1992, 114, 10082. [78] N. Zein, W. J. McGahren, G. 0. Morton, J. Ashcroft, G. A. Ellestad, J. Am. Chem. SOC. 1989, 111, 6888. [79] S. Walker, R. Landovitz, W.-d. Ding, G. A. Ellestad, D. Kahne, Proc. Natl. Acad. Sci USA 1992,89, 4608. [SO] W.-d. Ding, G. A. Ellestad, 1 Am. Chem. SOC. 1991, 113, 6617. [Sl] S. Walker, K. G. Valentine, D. Kahne, J. Am. Chem. SOC. 1990, 112, 6428. [82] S. Walker, J. Murnick, D. Kahne, 1 Am. Chem. SOC.1993, 115, 7954. [83] M. Uesugi, Y. Sugiura, Biochemistry 1993, 32, 4622. [84] S. Walker, D. Kahne, J. Am. Chem. SOC. 1991, 113, 4716. [85] J. Drak, N. Iwasawa, S. Danishefsky, D. M. Crothers, Proc. Natl. Acad. Sci. USA 1991, 88, 7464. [86] J. Aiyar, S. J. Danishefsky, D. M. Crothers, d Am. Chem. SOC. 1992, 114, 7552. [87] K. C. Nicolaou, S.-C. Bay, T. Suzuki, G. F. Joyce, J. Am. Chem. SOC. 1992, 114, 7555. [88] R. C. Hawley, L. L. Kiessling, S. L. Schreiber, Proc. Natl. Acad. Sci. USA 1989, 86, 1105. [89] G. A. Ellestad, P. R. Hammann, N. Zein, G. 0. Morton, M. M. Siegel, M. Patel, D. B. Borders, W. J. McGahren, Btrahedron Lett. 1989,30, 3033. [90] N. Din, W. J. McGahren, G. 0. Morton, J. Ashcroft, G. A. Ellestad, J. Am. Chem. SOC. 1989, I l l , 6888. [91] J. J. De Voss, C. A. Townsend, W.-d. Ding, G. 0. Morton, G. A. Ellestad, N. Zein, A. B. Tabor, S. L. Schreiber, J; Am. Chem. Soc. 1990, 112, 9669. [92] J. J. Hangeland, J. J. De Voss. J. A. Heath, C. A. Townsend, W.-d. Ding, J. S. Ashcroft, G. A. Ellestad, J. Am. Chem. SOC. 1992, 114, 9200. [93] P. C. Dedon, A. A. Salzberg, J. Xu, Biochemistry 1993, 32, 3617, 7064. [94] J. J. De Voss, J. J. Hangeland, C. A. Townsend, J. Am. Chem. SOC. 1990, 112, 4554. [95] K. D. Kramer, C. A. Townsend, Tetrahedron Lett. 1991, 32, 4635. I961 M. Chatterjee, K. D. Cramer, C. A. Townsend, J. Am. Chem. SOC. 1993, 115, 3374. I971 M. Konishi, H. Ohkuma, K. Saitoh, H. Kawaguchi, J. Golik, 0. Dubay, G. Groenewold, B. Krishnan, T. W. Doyle, J. Antibiot. 1985, 38, 1605. [61] [62] [63] [64]
References
279
[98] J. Golik, H. Wong, D. M. Vyas, T. W. Doyle, Rtrahedron Lett. 1989,30, 2497. [99] J. Golik, T. W. Doyle, G. VanDuyne, J. Clardy, Tatrahedron Lett. 1990,31, 6149. [lo01 J. Golik, B. Krishnan, T. W. Doyle, G. VanDuyned, J. Clardy, Tetrahedron Left. 1992, 33, 6049. [loll B. H. Long, J. Golik, S. Forenza, B. Ward, R. Rehfuss, J. C. Dabrowiak, J. J. Catino, S. T. Musial, K. W. Brookshire, T. W. Doyle, Proc. Natl. Acad. Sci. USA 1989,86,2. [lo21 Y. Sugiura, Y. Takahashi, Y. Uesawa, J. Kuwahara, Nucleic Acids Res. Symp. Ser. 1988, 20, 63. [lo31 Y. Sugiura, Y. Uesawa, Y. Takahashi, J. Kuwahara, J. Golik, T. W. Doyle, Proc. Natl. Acad. Sci. USA 1989,86,7672. [lo41 D. F. Christner, B. L. Frank, J. W. Kozarich, J. Stuhbe, J. Golik, T. W. Doyle, I. E. Rosenberg, B. Krishnan, J. Am. Chem. Soc. 1992, 114, 8763. [lo51 H. S. El Khadem (Ed.), Anthraqdine Arzfibiotics, Academic Press, New York, 1982. [lo61 K. Shioma, H. Iinuma, H. Nakagawa, M. Hamada, S. Hattori, H. Nakamura, T. Takeuchi, Y. Itaka, J. Antibiof. 1990, 43, 1000. [lo71 D. R. Langley, T. W. Doyle, D. L. Beveridge, J. Am. SOC. 1991, 113, 4395. [I081 Y. Tokiwa, M. Miyoshi-Saitoh, H. Kobayashi, R. Sunaga, M. Konishi, T. Oki, S. Iwasaki, J. Am. Chem. Soc. 1992, 114, 4107. [lo91 Y. Sugiura, T. Shiraki, M. Konishi, T. Oki, Proc. Natl. Acad. Sci. USA 1990, 87, 3831. [110] T. Shiraki, Y. Sugiura, Biochemistry 1990,29, 9795. [lll] T. Arakawa, T. Kusakabe, J. Kuwahara, M.Otsuka, Y. Sugiura, Biochem. Biophys. Res. Commun. 1993, 190, 362. [112] Y. Sugiura, T. Arakawa, M. Uesugi, T. Shiraki, H. Ohkuma, M. Konishi, Biochemistry 1991, 30, 2989. [I131 A. Ichikawa. T. Kuboya, T. Aoyama, Y. Sugiura, Biochemisfty 1992, 31, 6784. [114] T. Shiraki, Y. Sugiura, Nucleic Acids Res. Symp. Ser. 1989, 21, 53. [115] P. A. Wender, R. C. Kelly, S. Beckham, B. L. Miller, Roc. Natl. Acad. Sci. USA 1991, 88, 8835. [116] M. G. Cardozo, A. J. Hopfinger, Biopolymers 1993,33, 377. [117] T. Kusakabe, K. Maekawa, A. Ichikawa, M. Uesugi, Y. Sugiura, Biochemistry 1993,32, 11669. [118] M. E Semmelhack, J. Gallagher, D. Cohen, Rtrahedron Lett. 1990, 31, 1521. [119] K. C. Nicolaou, W.-M. Dai, S. V. Wendeborn, A. L. Smith, Y. Torisawa, P.Maligres, C.-K. Hwang, Angew. Chem., Inf. Ed. Engl. 1991,30, 1032. [I201 K. C. Nicolaou, W.-M. Dai, J. Am. Chem. SOC. 1992, 114, 8908. [121] A. S. Khokhlov, B. Z. Cherches, P. D. Reshetov, G. M.Smirnova, I. B. Sorokina, T. A. Prokoptzeva, T. A. Koloditskaya, V. V. Smirnov, S. M. Navashin, I. P.Fomina, J. Anfibiot. 1969, 22, 541. [122] A. S. Khokhlov, P. D. Reshetov, L. A. Chupova, B. Z. Cherches, L. S. Zhigis, I. A. Stoyachenko, J. Anfibiof.1976,29. 1026. [123] H. Chimura, M. Ishizuka, M.Hamada, S. Hori, K. Kimura, J. Iwanaga, T. Takeuchi, M. Umezawa, J. Antibiot. 1968,21, 44. [124] A. Zaheer, S. Zaheer, R. Montgomery, J. Biol. Chem. 1985,260, 11787. [125] P. V. Roey, T. A. Beerman, Proc. Nafl. Acad. Sci. USA 1989, 86, 6587. [126] T. Hidaka, Y. Yano, T. Yamashita, K. Watanabe, J. Antibiof. 1979, 32, 340. [127] T. Yamashita, N. Naoi, T. Hikada, K. Watanabe, Y. Kumada, T. Takeuchi, H. Umezawa, L Antibiot. 1979, 32, 330. [128] Y. Kumada, T. Miwa, N. Naoi, K. Watanabe, H. Naganawa, T. Takita, H. Umezawa, H. Nakamura, Y. Itaka, J. Antibiot. 1983,36, 200. [I291 K. S. Lam, G. A. Hesler, D. R. Gustavson, A. R. Crosswell, J. M. Veitch, S. Forenza, K. Tomita, 1 Anfibiof. 1991, 44, 472. [130] J. E. Leet, J. Golik, S. J. Hofstead, J. A. Matson, A. Y. Lee, J. Clardy, Rfrahedron Left. 1992,33, 6107. I1311 N. Zein, K. L. Colson, J. E. Leet, D. R. Schroeder, W. Solomon, T. W. Doyle, A. M.Casazza, Proc. Natl. Acad. Sci. USA 1993,90, 2822.
280
7 The Enediyne Antibiotics
Y. Sugiura, T. Matsumoto, Biochemistry 1993, 32, 5548. T. Matsumoto, Y. Okuno, Y. Sugiura, Biochem. Biophys. Res. Commun. 1993, 195, 659. N. M. Vichrova, T. 1. Kruchkova, E. V. Preobrazhenskaya, A. S. Khokhlov, Antibiotiki 1957,2, 21. A. A. Melnikova, V. A. Semjonova, N. K. Solovjova, L. P. Sneznova, G. N. Ginsburg, Antibiotiki 1958, 3, 18. (1361 A. G. Myers, E. Y. Kuo, N. S. Finney, J. Am. Chem. SOC. 1989, I l l , 8057. [137] A. G. Myers, P. S. Dragovich, E. Y. Kuo, J. Am. Chem. SOC. 1992, 114, 9369. [138] A. G. Myers, N. S. Finney, J. Am. Chem. SOC. 1992, 114, 10986. [139] A. G. Myers, P. S. Dragovich, J. Am. Chem. SOC. 1993, 115, 7021. [140] A. G. Myers, P. M. Harrington, E. Y. Kuo, J. Am. Chem. SOC. 1991, 113, 694. [141] A. G. Myers, M. M. Alauddin, M. A. M. Fuhry, P. S. Dragovich, N. S. Finney, P. M. Harrington, Tetrahedron Lett. 1989, 30, 6997. (1421 K. Nakatani, K. Arai, S. Terashima, Tetrahedron 1993, 49, 1901. [143] K. Nakatani, K. Arai, S. Terashima, J. Chem. SOC, Chem. Commun. 1992, 289. [144] K. Nakatani, K. Arai, N. Hirayama, F. Matsuda, S. Terashima, Tetrahedron 1992, 48, 633. [145] K. Nakatani, K. Arai, K. Yamada, S. Terashima, Tetrahedron 1992, 48, 3045. [I461 K. Nakatani, K. Arai, K. Yamada, S. Terashima, Tetrahedron Lett. 1991, 32, 3405. [147] K. Nakatani. K. Arai, N. Hirayama, F. Matsuda, S. Terashima, Tetrahedron Lett. 1990, 31, 2323. (1481 K. C. Nicolaou, G. Skokotas, S. Furuya, H. Suemune, D. C. Nicolaou, Angew. Chem, Znt. Ed. Engl. 1990, 29, 1064. [149] M. Hirama, K. Fujiwara, K. Shigematsu, Y. Fukazawa, J. Am. Chem. SOC.1989, 111, 4120. [150] K. Fujiwara, A. Kurisaki, M. Hirama, Tetrahedron Lett. 1990, 31, 4329. [151] K. Fujiwara, H. Sakai, M. Hirama, J. Org. Chem. 1991, 56, 1688. [I521 K. F. Fujiwara, M. Hirama, Gendai Kagaku 1990, 14. [153] J. M. Nuss, R. A. Rennels, B. H. k i n e , J. Am. Chem. SOC. 1993, 115, 6991. I1541 S. Torii, H. Okumoto, T. Tadokoro, A. Nishimura, M. A. Rashid, Tetrahedron Lett. 1993,34,2139. [155] R. Brlickner, S. W. Scheuplein, J. Suffert, Tetrahedron Lett. 1991, 32, 1449. (1561 J. Suffert, R. Bruckner, Tetrahedron Lett. 1991, 32, 1453. [157] S. W. Scheuplein, K. Harms, R. Brtickner, J. Suffert, Chem. Ber. 1992, 125, 271. I1581 N. A. Petasis, K. A. Teets, Tetrahedron Lett. 1993, 34, 805. [I591 P. A. Wender, J. A. McKinney, C. Murai, J. Am. Chem. SOC. 1990, 112, 5369. (1601 Y. Okude, S. Hirano, T. Tiyama, H. Nozaki, J. Am. Chem. SOC. 1977, 99, 3179. [161] P. Wender, J. W. Grissom, U. Hoffmann, R. Mah, Tetrahedron Lett. 1990, 31, 6605. [162] J. Suffert, Tetrahedron Lett. 1990, 31, 7437. [163] T. Wehlage, A. Krebs, T. Link, Tetrahedron Lett. 1990, 31, 6625. [164] P. Magnus, T. Pitterna, J. Chem. Soc, Chem. Commun. 1991, 541. [165] K. Mikami, H. Matsueda, T. Nakai, Synlett. 1993, 23. [166] K. Toshima, K. Ohta, T. Ohtake, K. Tatsuta, TefrahedronLett. 1991, 32, 391. (1671 T. Doi, T. Takahashi, J. Org. Chem. 1991, 56, 3465. (1681 P. A. Wender, M. Harmata, D. Jeffrey, C. Mukai, J. Suffert, Tetrahedron Lett. 1988, 29, 909. (1691 K. C. Nicolaou, G. Skokotas, P. Maligres, G. Zuccarello, E. J. Schweiger, K. Toshima, S. Wendeborn, Angew. Chem, Znt. Ed. Engl. 1989, 28, 1272. [170] Y. Sakai, Y. Bando, K. Shishido, M. Shibuya, Tetrahedron Lett. 1992, 33, 957. [171] K. Toshima, K. Ohta, A. Ohtsuka, S. Matsumura, M. Nakata, 1 Chem. Soc., Chem. Commun. 1993, 1406. I1721 G. Xie, A. R. Morgan, J. W. Lown, Biorg. Med. Chem. Lett. 1993, 3, 1565. [173] K. C. Nicolaou, P. Maligres, J. Shin, E. de Leon, D. Rideout, J. Am. Chem. SOC. 1990, 112, 7825. I1741 A. G. Myers, P. S. Dragovich, J. Am. Chem. SOC. 1989, 111, 9130. [1751 M. Shibuya, K. Toyooka, S. Kubota, Tetrahedron Lett. 1984, 25, 1171. 11761 K. Shishido, A. Yamashita, K. Hiroya, K. Fukumoto, T. Kametani, Tetrahedron Lett. 1989,30, 111. [132] [133] [134] [135]
References
281
[I771 K. Shishido, A. Yamashita, K. Hiroya, K. Fukumoto, T. Kametani, Tetrahedron 1989, 45, 5791. [I781 K. Takahashi, T. Suzuki, M. Hirama, Tetrahedron Lett. 1992,33, 4603. [I791 M. Hirama, T.Gomibuchi, K. Fujiwara, Y. Sugiura, M. Uesugi, J. Am. Chem. SOC.1991,113,9851. [I801 K. C. Nicolaou, G. Zuccarello, Y. Ogawa, E. J. Schweiger, T. Kumazawa, J. Am. Chem. SOC. 1988, 110, 4866. [I811 K. C. Nicolaou, G. Zuccarello, C. Riemer, V. A. Estevez, W.-M. Dai, 1 Am. Chem. SOC.1992, 114, 7360. [182]J. P. Snyder, J. Am. Chem. SOC. 1989,111, 7630. [183] J. P. Snyder, J. Am. Chem. SOC. 1990, 112, 5367. [184] J. P. Snyder, G. E. Tipsword, J. Am. Chem. SOC. 1990, 112, 4040. [185]P. Magnus, S. Fortt, T. Pitterna, J. P. Snyder, J. Am. Chem. SOC. 1990,112, 4986. [186] P. Magnus, P. Carter, J. Elliot, R. Lewis, J. Harling, T. Pitterna, W. E. Bauta, S. Fortt, J. Am. Chem. SOC. 1992, 114, 2544. [I871 K. C. Nicolaou, Y. Ogawa, G. Zuccarello, H. Kataoka, J. Am. Chem. SOC.1988, 110, 7247. [188]K. C. Nicolaou, E. J. Sorensen, R. Discordia, C.-K. Hwang, R. E. Minto, K. N. Bharucha, R. G. Bergman, Angew. Chem., Int. Ed. Engl. 1992,31, 1044. [189]K. C. Nicolaou, A. Liu, Z. Zeng, S. McComb. . I Am. Chem. SOC. 1992, 114, 9279. [190] K. Takai, M. Takashiro, T. Kuroda, K. Oshima, K. Utimoto, H. Nozaki, J. Am. Chem. SOC. 1986, 108, 6048. [191] C. Crevisy, J.-M. Beau, Tetrahedron Lett. 1991,32, 3171. [I921 A. G.Myers, P. S. Dragovich, J. Am. Chem. SOC. 1992, 114, 5859. [193]M. F. Semmelhack, J. J. Gallagher, T. Minami, T. Date, J Am. Chem. SOC, 1993, 115, 11618. [194]A. S. Kende, C. A. Smith, Tetrahedron Lett. 1988, 29, 4217. [195] S. L. Schreiber, L. L. Kiessling, 1 Am. Chem. SOC. 1988, 110, 631. [196] S. L. Schreiber, L. L. Kiessling, Tetrahedron Lett. 1989, 30, 3765. [197]F. J. Schoenen, J. A. Porco, Jr., S. L. Schreiber, G. D. VanDuyne, J. Clardy, TetrahedronLett.1989, 30, 3765. [198]K. Suzuki, E. Katayama, G. Tsuchihashi, Tetmhedron Lett. 1983, 24, 4997. [199]K. Suzuki, T. Ohkuma, M. Miyazawa, G. Tsuchihashi, Tetmhedron Lett. 1988,29, 373. [200]P. Magnus, R. Lewis, F. Bennett, J. Am. Chem. SOC. 1992, 114, 2560. [201] P. Magnus, R. T. Lewis, J. C. Huffmann, J. Am. Chem. SOC. 1988,110, 6921. [202]P. Magnus, P. A. Carter, J. Am. Chem. SOC.1988, 110, 1626. [203]P. Magnus, H. Annoura, J. Harling, J. Org. Chem. 1990, 55, 1709. [204]P. Magnus, R. T. Lewis, F. Bennett, J. Chem. SOC, Chem. Commun. 1989, 916. [205]P. Magnus, F. Bennett, Tetrahedron Lett. 1989, 30, 3637. [206]P. Magnus, R. T. Lewis, Tetrahedron Lett. 1989,30, 1905. [207]K. M.Nicholas, Acc. Chem. Res. 1987,20, 207. [208]D. N. Harpp, K. Steliou, T. H. Chen, J. Am. Chem. SOC. 1978, 100, 1222. [209]K. Tomioka, H.Fujita, K. Koga, Tetrahedron Lett. 1989,30, 851. [210]J. F. Kadow, M. G. Saulnier, M. M. lkn, D. R. Langley, D. M. Vyas, Tetrahedron Lett. 1989,30, 3499. [211] J. F. Kadow, M. M. Tun, D. M. Vyas, M. D. Wittmann, T. W. Doyle, Tetrahedron Lett. 1992,33, 1423. [212]M. E. Maier, T. Brandstetter, Liebigs Ann. Chem. 1993, 1009. [213]M. E. Maier, T. Brandstetter, Tetrahedron Lett. 1991,32, 3679. [214]H. Audrain, T. Skrydstrup, G. Ulibarri, D. S. Grierson, Synlett 1993,20. [215]T. Skrydstrup, H.Audrain, G. Ulibarri. D. S. Grierson, Tetrahedron Lett. 1992, 33, 4563. [216] D. Schinzer, J. Kabbara. Synleft 1992, 766. [217]J. N. Haseltine, M. P. Cabal, N. B. Mantlo, N. Iwasawa, D. S. Yamashita, R. S. Coleman, S. J. Danishefsky, G. K. Schulte, J. Am. Chem. SOC. 1991, 113, 3850.
282
7 The Enediyne Antibiotics
12181 M. P. Cabal, R. S. Coleman, S. J. Danishefsky, 1 Am. Chem. SOC. 1990, 112, 3253. I2191 J. N. Haseltine, S. J. Danishefsky, J. Org. Chem. 1990, 55, 2576. (2201 J. N. Haseltine, S. J. Danishefsky, G. Schulte, J. Am. Chem. SOC.1989, 111, 7638. 12211 N. B. Mantlo, S. J. Danishefsky, J. Org. Chem. 1989,54, 2781. [222] S. J. Danishefsky, N. B. Mantlo, D. S. Yamashita, J. Am. Chem. SOC, 1988,110, 6890. [223] S. J. Danishefsky, D. S. Yamashita, N. B. Mantlo, Tetrahedron Lett. 1988, 29, 4681. 12241 D. S. Yamashita, V. P. Rocco, S. J. Danishefsky, Tetrahedron Lett. 1991, 32, 6667. [225] V. P. Rocco, S. J. Danishefsky, G. K. Schulte, Tetrahedron Lett. 1991, 32, 6671. 12261 A. L. Smith, C.-K. Hwang, E. Pitsinos, G. R. Scarlato, K. C.Nicolaou, J. Am. Chem. SOC. 1992, 114, 3134. 12271 A. L. Smith, E. N. Pitsinos, C.-K. Hwang, Y. Mizuno, H. Saimoto, G. R. Scarlato, T. Suzuki, K. C. Nicolaou, J. A. Chem. SOC. 1993, 115, 7612. [228] K. C. Nicolaou, R. D. Groneberg, T. Miyazaki, N. A. Stylianides, T. J. Schulze, W. Stahl, J. Am. Chem. SOC.1990, 112, 8193. [229] R. D. Groneberg, T. Miyazaki, N. A. Stylianides, T. J. Schulze, W. Stahl, E. P. Schreiner, T. Suzuki, Y. Iwabuchi, A. L. Smith, K. C. Nicolaou, J. Am. Chem. SOC. 1993,115, 7593. 12301 K. C. Nicolaou, R. D. Groneberg, J. Am. Chem. SOC.1990, 112, 4085. 12311 K. C. Nicolaou, R. D. Groneberg, N. A. Stylianides, T. Miyazaki, J. Chem. SOC,Chem. Commun. 1990, 1275. [232] K. C. Nicolaou, T. Ebata, N. A. Stylianides, R. D. Groneberg, P. J. Carrol, Angew. Chem., Znt. Ed. Engl. 1988, 271, 1097. [233] K. C. Nicolaou, D. Clark, Angew. Chem., Znt. Ed. Engl. 1992, 31, 855. [234] R. L. Halcomb, M. D. Wittman, S. H. Olson, S. J. Danishefsky, J. Golik, H. Wong, D. Vyas, J. Am. Chem. SOC. 1991, IN, 5080. (2351 M. D. Wittman, R. L. Halcomb, S. J. Danishefsky, J. Org. Chem. 1990, 55, 1979. 12361 D. Yang, S.-H. Kim, D. Kahne, I Am. Chem. SOC. 1991, 113, 4715. 12371 R. L. Halcomb, S. H. Boyer, S. J. Danishefsky, Angew. Chem., Znt. Ed. Engl. 1992, 31, 338. I2381 D. Kahne, D. Yang, Tetrahedron Lett. 1990, 31, 21. [239] K. V. Laak, H.-D. Scharf, Tetrahedron Lett. 1989, 30, 4505. [240] K. V. Laak, H.-D. Scharf, Etrahedron 1989, 45, 5511. [241] H. Rainer, H.-D. Scharf, Liebigs Ann. Chem. 1993, 117. (2421 A. ClaRen, H.-D. Scharf, Liebigs Ann. Chem. 1993, 183. [243] E. A. Mash, S. K. Nimkar, Tetrahedron Lett. 1993,34, 385. [244] E X Dupradeau, S. Allaire, J. Prandi, J.-M. Beau, Tetrahedron Lett. 1993, 34, 4513. [245] K. C. Nicolaou, Angew. Chem., Int. Ed. Eng. 1993,32, 1377. [246] K. C. Nicolaou, E. P. Schreiner, Y. Iwabuchi, T. Suzuki, Angew. Chem., Znt. Ed. Engl. 1992,31,340. 12471 K. C. Nicolaou, C. W. Hummel, M. Nakada, K. Shibayama, E. N. Pitsinos, H. Saimoto, Y. Mizuno, K.-U. Baldenius, A. L. Smith, I Am. Chem. SOC. 1993, 115, 7625. 12481 G. Grandler, R. R. Schmidt, CarbohydK Res. 1985, 135, 203. [249] R. R. Schmidt, Angav. Chem., Znt. Ed. Engl. 1986,25, 212. 12501 K. C. Nicolaou, C.-K. Hwang, A. L. Smith, S. V. Wendeborn. J. Am. Chem. SOC.1990, 112,7416. [251] K. C. Nicolaou, A. L. Smith, S. V. Wendeborn, C.-K. Hwang, J. Am. Chem. SOC. 1991, 113, 3106. I2521 K.C. Nicolaou, P. Maligres, T. Suzuki, S. V. Wendebom, W.-M. Dai, R. K. Chadha, J. Am. Chem. SOC. 1992, 114, 8890. [253] K. C. Nicolaou, Y.-P. Hong, Y. Torisawa, S.-C. Tsay, W.-M. Dai, J. Am. Chem. SOC.1991,113,9878. [254] K. C. Nicolaou, W.-M. Dai. Y. P. Hong, S.-C. Tsay, K. K. Baldridge, J. S. Siegel, J. Am. Chem. SOC.1993, 115, 7944. [255] 0.Gandiano, T. H. Koch, J. Am. Chem. SOC. 1990, 112, 9423. 12561 K. C. Nicolaou, W.-M. Dai, S.-C. Bay, V. A. Estevez, W. Wrasidlo, Science 1992, 256, 1172. [257] K. C. Nicolaou, W.-M. Dai, S.-C. Tsay, W. Wrasidlo, Bioorg. Med. Chem. Lett. 1992, 2, 1155.
References
283
[258] K. C. Nicolaou, W. Wrasidlo, unpublished results. [259] D. L. Vaux, Proc. Natl. Acad. Sci. USA 1993, 90, 786. [260] K. C. Nicolaou, P. Stabila, B. EsmaeliAzad, W. Wrasidlo, A. Hiatt, Proc. Nutl. Acad. Sci. USA 1993, 90, 3142. [261] K. C. Nicolaou, Y . 2 Hong, W.-M. Dai, Z.-J. Zeng, W. Wrasidlo, J Chem. Soc, Chem. Commun. 1992, 1542. [262] P. A. Wender, C. K. Zercher, J. Am. Chem. SOC. 1991, JJ3, 2311. [263] P. A. Wender, C. K. Zercher, S. Beckham, E.-M. Haubold, J. Org. Chem. 1993, 58, 5867. [264] T. Nishikawa, A. Ino, M. Isobe, T. Goto, Chem. Lett. 1991, 1271. [265] T. Nishikawa, M. Isobe, T. Goto, Synlett 1991, 393. [266] T. Nishikawa, M. Isobe, T. Goto, Synlett 1991, 99. [267] P. Magnus, S. M. Fortt, J Chem. Soc., Chem. Commun. 1991, 544. [268] J. A. Porco, Jr., F. J. Schoenen, T. J. Stout, J. Clardy, S. L. Schreiber, J. Am. Chem. SOC. 1990, 112, 7410. [269] H. Chikashita, J. A. Porco, Jr., T. J. Stout, J. Clardy, S. L. Schreiber, J. Org. Chem. 1991,56, 1692. [270] J. L. Wood, J. A. Porco, Jc, J. Taunton, A. Y. Lee, J. Clardy, S. L. Schreiber, J. Am. Chem. SOC. 1992, 114, 5898. [271] J. Taunton, J. L. Wood, S. L. Schreiber, J. Am. Chem. SOC. 1993, JZ5, 10378. [272] M. Hornyak, I. F. Pelyviis, F. J. Sztaricskai, Tetrahedron Let?. 1993, 34, 4087. (2731 K. C. Nicolaou, T. Li, M. Nakada, C. W. Hummel, A. Hiatt, W. Wrasidlo, Angew Chem, Int. Ed. Engl. 1994, 33, 183. [274] K. Toshima, K. Ohta, A. Ohashi, A. Ohtsuka, M. Nakata, K. Tatsuta, J. Chem. Soc., Chem. Commun. 1992, 1306. [275] M. Tokuda, K. Fujiwara, T. Gomibuchi, M. Hirama, M. Hirama, M. Uesugi, Y. Sugiura, Dtmhedron Lett. 1993, 34, 669. [276] D. L. Boger, J. Zhou, J. Org. Chem. 1993, 58, 3018. [277] H. Maeda, Anticancer Res. 1981, I , 175. [278] D. A. Schmitt, K. Kisanuki, S. Kimura, K. Oka, R. B. Pollard, H. Maeda, F. Suzuki, Anticancer Res. 1992, 12, 2219. [279] T. Takahashi, T. Yamaguchi, K. Kitamura, A. Noguchi, M. Honda, E. Otsuji, Jpn. J. Cancer Res. 1993, 84, 976.
This Page Intentionally Left Blank
8 Cyclic Alkynes :Preparation and Properties Rolf Gleiter, Roland Merger
8.1 Introduction In linear acetylenes the two n-MOs of the triple bonds are degenerate. This degeneracy is lifted when the triple bond is incorporated in a ring system. Thus, in cyclic alkynes, three different “sides” of the triple bond, which correspond to three different modes of reactivity, can be distinguished (see Fig. 8-1):
Figure 8-1 Ring closure removes the degeneracy of the x MOs of the triple bond. As a consequence three different modes of reactivity can be imagined: 0 , out-of plane attack; 0 , intermolecular in-plane attack; 0 ,intramolecular in-plane attack. X is a functional group and R a reagent.
(1) The triple bond can be attacked at the part of its x-system which is “out of plane” relative
to the ring system. (2) The reagent can approach the triple bond “in the plane” of the ring. (3) A second functional group can undergo a transannular reaction with the triple bond. It can also alter the reactivity of the acetylenic unit by transannular orbital interactions. If the cyclic alkyne is strained, i. e., if the triple bond is bent, electron density is pushed to the outside of the ring. Thus, if the “in-plane” orbitals of the triple bond are involved, the exocyclic and the transannular attack of the triple bond are electronically different reactions (Fig. 8-2).
Figure 8-2 Bending of a triple bond as in a strained ring increases the electron density outside the ring.
286
8 Cyclic Alkynes: Preparation and Properties
Of course, a bent triple bond will often not behave in the same way as a linear triple bond. Thus, the consequence of ring strain and the different modes of reactivity [l] make cyclic alkynes fascinating synthetic targets [2]. Furthermore, the study of the electronic nature of the strained triple bond and its electronic interaction with other functional groups, particularly other n-systems, along and across the ring, has dramatically improved our understanding of such phenomena as aromaticity, homoconjugation, and deinteraction [3]. In this chapter, we will focus on cyclic mono- and diacetylenes, as other topics, such as cyclic oligo- and polyacetylenes or dehydroannulenes [2a], have either been reviewed in detail [I -31 or are covered by the corresponding contributions to this publication. Extremely strained systems which have been detected only by trapping reactions have been reviewed very recently [3c] and are not included here.
8.2 Synthesis of Cyclic Acetylenes There are two main strategies for the preparation of cyclic alkynes. One is to perform a ringclosure reaction on an acyclic precursor already containing the triple bond. As the four carbon atoms of a C - C = C - C unit prefer a linear arrangement, this strategy is usually limited to the preparation of large- and medium-size ring systems. Heavily strained cyclic systems where the deviation from linearity at the acetylenic carbon atoms is greater than 10" can (with few exceptions) only be prepared by elimination reactions on cyclic compounds.
8.2.1 Cyclic Alkynes from Ring-Closure Reactions As this chapter is concerned with the properties of the carbon-carbon triple bond, we will mainly discuss ring-closure reactions which make use of the special reactivity of acetylenic systems - either at the sp-hybridized carbon atoms or at the propargylic centers. It is obvious, however, that virtually all ring-closure reactions can be employed to close a ring with a triple bond. Earlier reviews have dealt with these possibilities in quite some detail [I-31.
8.2.1.1 Using Acetylenic Reactivity: Nucleophilic Substitution with Metal Acetylides and Related Reactions One of the oldest and still one of the most powerful methods of making cyclic alkynes uses the reaction of metal acetylides with halogen compounds [4].If the halogen (or a similar leaving group) is situated at an sp3-hybridized carbon atom, the reaction can usually be carried out without the help of a catalyst in moderately polar solvents such as THE Recently, it has been found that lithium acetylides give better yields than sodium acetylides in the preparation of cyclic dialkynes (Scheme 8-1) [S]. This method has been used for the synthesis of a series of skipped cyclic ene- and dienediynes [5]. Remarkably high yields are obtained in some cases even for strained compounds, (6)can be obtained in 32 @lo yield by this simple e. g., 4,9-diisopropylidene-1,6-cyclodecadiyne cyclization reaction. The power of this method is further demonstrated by the remarkable synthesis of a tetrasilacyclohexyne (9) by Barton et al. [Eq. (l)] 161.
8.2 Synthesis of Cyclic Acetylenes
1
2
3
281
35%
Grignard derivatives of acetylenes have also been used for ring-closure reactions, as is illustrated by the preparation of 1,6-dioxa-3,8-cyclodecadiyne(12) from bis(bromomagnesium)acetylene (11) and di(chloromethy1)ether (10) Fq. (2)] [7]. 2 c :+ 10
2 BrMgCECMgBr 11
12
In cases where stable carbocations can be generated, e. g., tertiary propargylic cations, the rather low nucleophilicity of silylated terminal alkynes is still sufficient for a coupling reaction. Ring closures have been carried out in this fashion to give the permethylated pericyclynes, which are discussed by L. T. Scott and M. J. Cooney in Chapter 9 of this monograph 181. If the leaving group is localized at an sp2-hybridized carbon center, however, the nucleophilic power of a metal acetylide is often too weak for a direct displacement reaction. In this case, the leaving group must be activated, usually via oxidative addition to a transition metal. So far, the Heck reaction and similar Pd-mediated coupling reactions have been most successful in this field [9]. Pd-complex-catalyzed coupling reactions are especially popular in the calicheamicinheocarcinostatin field for the construction of enediyne systems, to which Chapter 7 is devoted. A remarkable success of this type of chemistry is the short synthesis of the dehydroannulene 15 when “deprotection” (by liberation of acetone) of a terminal alkyne and two Pd - catalyzed coupling steps can be carried out in one pot; this makes 15, an interesting n-system and ligand, readily available [Eq. (3)l [loa]. In some cases, the Pd catalyst
13
a
288
8 Cyclic Alkynes: Preparation and Properties
can be replaced by a copper salt, e.g., in the synthesis of 17, the thiophene analog of 15 [Eq. (4)l [lob]. Although copper acetylides seem to be able to perform a nucleophilic substitution reaction at the sp-carbon atom of a bromo- or iodoacetylene (Cadiot-Chodkievicz reaction), this reaction has only rarely been used for the preparation of cyclic 1,3-diacetylenes. Copper-mediated oxidative coupling reactions (Glaser, Hay and Eglinton coupling) are more popular in this area and have attracted much attention in the construction of carbon-rich cyclic and polycyclic systems (see Chapter 13). One of the earliest carbon-rich systems of this type was the CZoH8 system 19 [ll, 161 [Eq. (511.
d8’H 1 .CUSO,, 2.
16
N H I O H . NHZOH. HCI, EtOH (74%)
Py,reflux (21%)
c
17
18
19
8.2.1.2 Employing Propargylic Cations, Anions, and Radicals The high reactivity of propargylic halides toward nucleophiles allows the preparation f som rather strained cyclic alkynes and especially dialkynes, in surprisingly simple ways. Thus, the ten-membered ring 22 can be prepared in one step from methylamine and 1,4-dibromo2-butyne (20), 1121 (Scheme 8-2). Similarly, Misumi cyclization of 20 (treatment of 20 with 0.5 equiv. of KSeCN followed by NaBH,) affords the diselenacyclodecadiyne 24 in a one-pot procedure [13] (Scheme 8-2).
-
/c”-7
,v/Br H,NMe C H r N +
Br
20
I Scheme 8-2
21
uN-c 22
KSeCN
8.2 Synthesis of Cyclic Acetylenes
289
Particularly useful in cyclic dialkyne chemistry is sulfide as a nucleophile. This is due to a discovery made by Regen and co-workers, who found that the yield of dialkyl sulfides is strongly enhanced when the sulfide reagent is first adsorbed onto A1,0, ; thus, the substitution reaction takes place on the surface of the aluminum oxide. As the preparation of cyclic sulfides from dibromides is a stepwise process, these conditions were expected to increase the yield of cyclic sulfides dramatically as one end of the chain is immobilized after displacement of the first bromine [14]. Nicolaou first used this principle in the preparation of a series of cyclic thiadiynes [Eq. (6)] [15]. The cyclic thiadiynes prepared in this way are very useful as they can be converted to cyclic enediyne systems via the Ramberg-Backlund ring-contraction reaction (see below). Br
Bf 26
25
n=1-8
However, not only heterocyclic but also carbocyclic systems can be made via nucleophilic attack at propargylic centers. This is usually done intramolecularly as shown in Eq. (7), when a malonic ester anion displaces a propargylic chloride [16]. E E
-*
/\ E E 27
E E \/
A
E E
E E \/
A
A
E E
E E
29
28
As in the case of reactions with metal acetylides (see Section 8.2.1.1), the electrophilic center must be activated if the nucleophile is too weak to overcome the activation barrier for ring formation. This, again, can be achieved by metalorganic reactions. A very important reaction in this sense is the Magnus cyclization reaction (which can be considered a kind of intramolecular Nicholas reaction) [Eqs. (8) and (9)] [17]. Complexation of the triple bond with a C O ~ ( C Ofragment )~ stabilizes the positive charge at the propargylic center; another complexation effect is the reduction of ring strain during the cyclization step by "bending" the triple bond toward bond angles typical for sp2-hybridizedcarbon atoms. The triple bond can subsequently be liberated by oxidative decomplexation.
30
31
32
290
8 Cyclic Alkynes: Preparation and Properties
0
33
34
35
A very interesting and powerful new cyclization method employing intramolecular attack of a propargylic-allylic anion at a propargylic aldehyde is the Nozaki cyclization [Eq. (lo)] : even the heavily strained nine-membered ring 37 can be obtained by this procedure [18].
36
37
In a similar way, the very interesting and long-sought 1,6-dehydro[lOjannulene (40) was made (Scheme 8-3) [19].Treatment of the iodoaldehyde 38 with CrCI2 doped with NiCI, affords the strained diol 39 in a remarkably high yield (40%). Dehydration of 39 at low temperature gives a solution of 40, which has been detected by NMR spectroscopy. Compound 40 is only stable up to -90°C; above this temperature it cylizes to give naphthalene via a biradical (see below). CHO
40
Scheme 8-3
Propargylic anions, although not yet frequently used, seem to be very promising candidates for new ring-closure reactions. This is illustrated by the reaction shown in Eq.(11) when 1,9-bis(bromomagnesium)-2,7-nonadiyne (41) is cyclized with Me2SiCI2 to afford the silacyclodecadiyne 42 [20].
41
42
Propargylic radicals are generated on reduction of metal-complex-stabilized Nicholas cations with Zn. This principle, which has been found very recently [21],can be used for ringclosure reactions via intramolecular radical recombination [Eq. (12)l. A special kind of propargylic reactivity has also been discovered only very recently. The propargylic systems shown in Scheme 8-4react in a SN2' fashion exclusively, as the propargylic
8.2 Synthesis of Cyclic Acetylenes
291
43
48
47
49
Scheme 8-4
centers are sterically hindered. Thus, 2,5-dimethyl-2,5-dichlorohexyne(45) does not give the expected dithia-cyclodecadiynederivative 46 when treated with Na2S on A1203.Instead, the radialenes 48 and 49 are produced in this reaction. Consequently, the dimethyldichlorooctadiyne derivative 50 yields 2,5,7,10-tetraisopropylidene-1,6-dithia-3,8-~clodecadiyne(51) in the same fashion [Eq. (13)] [22]. As shown in Eqs. (14) and (15), this reactivity pattern seems to apply for a variety of nucleophiles 1231.
y-y
No2 S .A1203
CI
CI
+,3-t &-
(13)
51
50
%--t
0:; - a;t)o 50
52
+--$ 53
(14)
292
8 Cyclic Alkynes: Preparation and Properties
8.2.2 Cyclic Alkynes from Elimination Reactions Several thorough reviews of this topic have been published [l -31. Here we will focus on new methods, typical examples, and particularly interesting molecules which show considerable ring strain. Molecules whose existence has been postulated only after isolation of trapping products [3c] are not included.
8.2.2.1 1,2-Eliminetion
The combination of probably the oldest synthetic procedure for formation of a triple bond, i. e., the dehydrobromination of a vinyl bromide, with modern crown ether chemistry has resulted in one of the simplest yet very powerful methods for making highly strained cycloalkynes. Thus, 1,5-cyclooctadiyne (56) can be made by treating 1,5-dibromo-1,5-cyclooctadiene (55) with potassium tert-butanolate in nonpolar solvents in the presence of 18-crown-6 [3b, 24). The nonpolar solvent protects the bent triple bond from nucleophilic attack by tertbutanol (Scheme 8-5). 1J-Cyclooctadiyne had previously been made in very low yield by dimerization of butatriene (57), which is not a readily accessible compound [25]. Other important 1,2-elimination reactions generating cyclic alkynes are the oxidative degradation of
-0"'
KO'Bu c 18-crown-6
Br
55
58
-
59
L
61
N2
HOSOTol.
Scheme 8-6
-dimerizotion
56
Scheme 8-5
60
101
8
63
II
i
ii 57
8.2 Synthesis of Cyclic Acetylenes
293
1J-bishydrazones with lead tetraacetate in dichloromethane (and the closely related oxidation
of l-amino-1,2,3-triazoles) and the Eschenmoser fragmentation of the tosylhydrazones of aepoxyketones which are illustrated in Scheme 8-6 [26].
8.2.2.2 Cycloelimination Reactions Photolysis of annelated cyclopropenones [e. g., Eq. (16)] is an elimination method mainly used for matrix isolation of unstable cycloalkynes, e. g., cyclopentyne (67), acenaphthyne (68) and some highly strained cyclohexyne and cycloheptyne derivatives such as 68 [3c, 271.
65
64
66
67
68
Other important cycloelimination procedures correspond to an elimination of H,O from cyclic ketones. Thus, the a-hydrogen atoms of semicarbazones of cyclic ketones are removed by oxidative cyclization with thionyl chloride or selenium dioxide (Scheme 8-7). The 1,2,3-thiadiazoles (71)or 1,2,3-selenadiazoles (72) which result from these reactions can be cleaved in a second step to yield cyclic alkynes (Scheme 8-8) [28]. Several fragmentation conditions are known, among them thermal decomposition and base-induced cleavage. The mechanism of these reactions has been studied in detail [29]. It has been noted that the crucial step is the cleavage of the carbon-sulfur or carbon-selenium bond, as in this step the geometrical strain is introduced into the system. Clearly, due to the weakness of the C-Se
69
72
Scheme 8-7
294
8 Cyclic Alkynes: Preparation and Properties
73
74
75
Scheme 8-8
bond, the selenadiazoles (cleavage of which is called the Lalezari reaction [30]) are synthetically more useful than their sulfur congeners. Unfortunately, large quantities of seleniumcontaining side products and toxic wastes are usually produced by this procedure. The Lalezari reaction and other elimination procedures have been used by Meier et al. to produce an impressive series of cyclooctenynes such as 76-79 [3 b].
76
77
7a
79
8.2.2.3 Ring Contraction
The excellent yields of cyclization reactions with sulfides and mercaptans are well known, particularly in cyclophane and host-guest chemistry. It is often possible to synthesize a rather unstrained thiacycle in high yield and to introduce the ring strain, which would be disadvantageous during the cyclization procedure, in a second step via sulfur-eliminating ring-contraction procedures, such as treatment with triphenylphosphine, SOz extrusion or the Ramberg-Biicklund reaction [31]. The latter method has been used by Nicolaou et al. in their synthesis of a series of cyclic enediynes, which serve as models for calicheamicin-type antitumor antibiotics [Eq.(17)] [15].
81 n=2-8
80
Another interesting ring-contraction method is the photochemical extrusion of dimethylsilylene from cyclic oligosilanes. Ando et al. used this method to cut out two MezSi units from dodecamethylhexasilacyclooctyne to yield octamethyltetrasilacyclohexyne (Scheme 8-9) [32, 331.
\
7'
-.
Si<
-\
-
hv
h i
/
-Me2Si
hv
-Me2Si
r 82 Scheme 8 9
83
9
8.2 Synthesis of Cyclic Acetylenes
295
8.2.3 Ring-Enlargement Reactions The Fritsch-Buttenberg- Wiechell rearrangement provides a method to expand a cyclic ketone to a cyclic acetylene with an additional carbon atom in the ring (341. The ketone is first converted to a vinyl bromide or a vinylidene dibromide, from which the corresponding vinylidene carbene is produced with butyllithium. The vinylidene carbene then rearranges to a cyclic alkyne (Scheme 8-10).
84
Scheme 8-10
86
Another possibility is to use the high reactivity of cyclic dialkyne systems toward electrophiles in order to add the additional carbon atom to the ring. When treated with dichlorocarbene at low temperature, cyclic dialkynes give mainly cyclic diynones [Eq.(18)]. In the same fashion, cyclononyn-3-one (91),the smallest conjugated cycloalkynone yet, has recently been made [Eq. (19)] [35].
A w
(CH2)n
m
1. CHCIJBuLi
THF.-78OC--
(CH3"
0
-3&
(CH3,
8a
+
c
2.HZO
A
87
(W. Qdn
89
90 92
(18)
296
8 Cyclic Alkynes: Preparation and Properties
8.3 Structural and Spectroscopic Properties 8.3.1 Structures of Cyclic Mono- and Dialkynes In Table 8-1 we have listed the lengths of the triple bonds and the angles at the sp centers for cyclic monoalkynes as derived by X-ray and electron diffraction measurements 16, 33, 36-41], It is seen that bending the triple bond even to 150" does not have a significant effect on the bond lengths. The deviation from 1.20 A - the value of the C - C triple bond in unstrained systems - is small. Table 8-1 Comparison of selected bond lengths and bond angles in cyclic monoynes
Transannular distance (A)
Compound
94(c)
0 s
95
0 1
Angle@)(")
Deformatiodb) (")
Ref.
146.8 150.5
33.2 29.5
I371
145.5
34.5
[361
1.194
146.0 148.5
34.0 31.5
1.22
159.6 162.2
20.4 17.8
1.19
154
26
1.232
158.5
21.5
160.2 (107)
19.8
[MI
4.4 4.8
[41I
175.6 175.2
Bond angles at sp centers. @) Cisoid deformation at the triple bonds.
(')
Electron diffraction data.
8.3 Structural and Spectroscopic Properties
297
There is a correlation between the I3C chemical shift of the sp center and the angle at this center (see Fig. 8-3). This has been pointed out by Meier for monocyclic acetylenes [3b]. In the case of cyclic diynes more structural data are available, probably due to the fact that diynes crystallize better. The most relevant structural parameters of cyclic diacetylenes have been collected in Tables 8-2-8-4. The most strained diynes are those with an eight-membered ring, 1,5-cyclooctadiyne (56) [25] and 5,6,11,12-tetradehydrodibenzo[a,e]cyclooctene(98) [42]. The sp centers in both hydrocarbons are strongly bent like those listed in Table 8-1. The transannular distances between the triple bonds amount to 2.597 (56) and 2.617 A (98), respectively (Table 8-2). A further consequence of the imposed strain is an elongation of the other bonds; the bonds between the sp3 atoms in 56 are stretched to 1.57 A, those between the sp2 centers in 98 to 1.43 A. In the tetrasila compound 99 most of the strain is relieved in the long Si-Si bonds [43]. Consequently the bond angles at the sp centers deviate only 11-16' from linearity.
A
Table 8-2 Comparison of selected bond lengths and bond angles in eight-membered cyclic diynes Compound
Transannular distance (A)
Angle@)(")
Deformatiodb)(")
2.597
159.3
20.7
155.7
24.3
168.6 164.1
11.4 15.9
Ref,
>Si >$='?i< Sic
\=/ (a)
99
3.223 3.269
Bond angles at sp centers. @) Cisoid deformation at the triple bonds.
A somewhat smaller deviation from the linear arrangement (between 2" and 13") is found in the ten-membered 1,6-diynes (Table 8-3). With the exception of diyne 109,all of these rings are found in the chair conformation in the solid state. Depending on the heteroatom incorporated in the bridging unit, the transannular distance between the triple bonds varies between 2.91 and 3.5 A. One should also note that the substituents on the nitrogen atoms in 22 and 103 adopt the axial position in the crystalline state. This also holds for the 1,6-diazacyclodeca-3.8-diyne ring, even when it is substituted with bulky isopropyl and tert-butyl groups. As anticipated from the previous discussion, we notice only small deviations (4" and 9") at the sp centers in the case of the 11-and 12-membered rings (Table 8-4). In Fig. 8-3 we show a correlation of the 13C chemical shift of the sp-hybridized carbon atom with the bending angle a at this center for a number of strained monoynes and diynes with sp3 centers next to the triple bond. As can be seen from this Figure, a linear correlation is obtained [57].
298
8 Cyclic Alkynes: Preparation and Properties
Table 8-3 Comparison of selected bond lengths and bond angles in diacetylenes with ten-membered rings Compound
(+> -
Transannular distance
Angle'') (") Deformation(b)(")
Distance(c)c Ref.
171.2 (2)
8.3 (2)
5.141 (2)
[44]
169.6 (1)
10.4 (1)
4.935 (2)
1451
171.4 (3)
8.8 (3)
5.244 (4)
[45]
172.7 (2) 173.9 (2)
7.3 (2) 6.1 (2)
5.422 (3)
[45]
174.2 (2)
5.8 (2)
5.623 (2)
1451
3.085 (2)
173.9 (2)
6.1 (2)
5.958 (2)
[22]
3.161
176.1
3.9
5.77
1461
2.983 (3)
171.1 (2) 111.9 (2)
8.5 (2)
2.976 (3) 3.068 (3)
26
*f!
H1 +-$
Se 5 4
T0s-d
-
(A)
1121
\S 103
5.352 (2) 3.072 (2)
173.7 (2) 173.7 (2) 170.8 (1) 169.8 (1)
=(TP
104 2.996 (2)
>(yh -
6.3 (2)
9.2 10.2
172.8 (2) 171.2 (2)
7.2 (2) 8.8 (2)
3.229 (3)
173.5 (2)
7.5
3.065 (3)
173.9 (3)
7.1
105
5.137 (2)
[12]
5.106
[471
5.324 (2)
[20]
8.3 Structural and Spectroscopic Properties
299
Table 8-3 (continued).
Transannular distance (A) Angle") (") Deformation(b)(")
Compound
87) \ /
-
2.795 (3)
107 2.917 (3)
a
108
2-706 3.261
3.471 (2) 3.449 (2)
) 109
3.449 (2)
A
3.446 (2)
175.1 (1) 173.8 (1)
4.9 6.2
177.9 (2) 170.4 (2)
2.1 (2) 9.6 (2)
166.7 174.8
Distance(c)c Ref.
5.662
[481
13.3 5.2
176.6(d) 176.9@) 177.0(d) 176.9") 178.3'd) 176.5@ 175.9(d) 177.3")
5.906'*)
0
1511
6.142(@
0
Bond angles at sp centers. @) Cisoid deformations at the triple bonds. Distance between positions 1 and 6. ( d ) Distance in the boat conformation. (e) Distance in the chair conformation.
(a)
Table 8-4 Comparison of selected bond lengths in cyclic diacetylenes with 11- and 12-membered rings.
Transannular distances u,b (A) Angle deviation'') (") Twist angle (") Ref.
Compound
7.4 (4) 6.8 (4)
):8
111
(T) 3
0
1521
2.824 3.291
5.2 4.7
25.6
WI
4.06
6.2
24
[441
8 Cyclic Alkynes: Preparation and Properties
300
Table 8-4 (continued).
Transannular distances a,b
Compound
R
$
#$ +
(a)
116
117
(A)
Angle deviatioda) (") Twist angle (") Ref.
3.356 (1)
5.6 (1) 6.8 (1)
0
4.850 (1)
6.2 (1) 6.5
0
3.036 3.108
8.7
48
3.051 3.411
4.1 4.6
58
3.011 3.408
5.3
58
153,551
2.85
6.2
53
1561
Deviations from 180" at the sp centers.
1471
[53, 541
8.3 Structural and Spectroscopic Properties
301
170'-
1 60°-.
1 SO0--
L
I
80
90
100
110
6
Figure 8-3 Correlation of the I3C chemical shift ( 6) of the sp-hybridized carbon atoms with the bending angle a of some cyclic acetylenes. The values of a are taken from experiments (Tables 8-1-8-3).
8.3.2 Photoelectron Spectra of Cyclic Diacetylenes Photoelectron (PE) spectroscopy is the proper tool for investigating the influence of the structural parameters on the electronic structure of the triple bonds such as bending, transannular interactions (through-space), and the influence of the length of the bridging unit (throughbond) [58]. The investigation of the PE spectra of the strained species 93 [59] and 90 [60] shows a splitting of the x-band, which is degenerate in linear acetylenes. The splitting amounts to 0.3 eV for 93 and 0.1 eV in the case of 90. For 97 a considerable interaction between the x-system of the triple bond and that of the cyclopropenone unit has been encountered [41]. A neat example for the importance of both the through-space and through-bond interaction in cyclic diynes is demonstrated in Fig. 8-4. In the PE spectrum of 56 [60]all four x-bands are closer together as compared with 100 [44]although in the latter compound the distance between the triple bonds is by 0.4 A larger than in 56. This apparent contradiction can be resolved by assuming two interaction mechanisms, a through-space and a through-bond interaction. The first type depends on the overlap integral between the two triple bonds and thus affects mainly the out-of-plane x-orbital (x,) while the in-plane x-MOs (xi) are mainly influenced by the through-bond interaction. In Fig. 8-5 this is demonstrated by comparing the PE spectra of medium-sized cyclic diynes (571. Most noticeable is that the splitting of the ni levels reaches its maximum at 1,6-cyclodecadiyne and then diminishes with increasing bridge
302
8 Cyclic Alkynes: Preparation and Properties
I" eV
eV
Figure 8-4 Comparison of the first bands of the PE spectra of 56 and 100. I.E. I
9 -.
10 -eV d
w
...... ......
......
- -\...... - ... /=-
2.56
.....
2.88
2.99
3.36
4.06
Figure 8-5 Comparison of the first PE bands of some cyclic diynes of ring size CBto C12.
length. The splitting observed for the I[, level decreases strongly with the transannular distance and remains virtually constant at distances greater than 3 A between the triple bonds. The ten- and twelve-membered cyclic diynes such as 104 and 112 provided the first models to test the concept of homoconjugation between double and triple bonds [47].In Fig. 8-6 the first PE bands of 4-methylene-l,6-cyclodecadiyne(118), 4,9-dimethylene-l,6-cyclodecadiyne (104) and 4,9-diisopropylidene-1,6-cyclodecadiyne (6) are correlated with those of 1,6-cyclodecadiyne (100). The most interesting feature in this correlation is the significant shift of n: toward higher energy in 104 (0.33 eV) and 118 (0.21 eV). This shift i s a clear-cut evidence for a homoconjugative interaction between the double bond(s) and the triple bonds in 118 and 104. The large effect is due to the fact that the basis orbital energy of the double bond(s) (broken lines in Fig. 8-6) and the triple bonds are close in energy. In the case of 6 this is not the case any more, and thus the homoconjugative effects between the double bonds and ni+ are very weak. This is a clear counterproof of the proposed conjugative interaction in 104 and 118. For the sake of completeness it shoud be mentioned that the PE spectra of a number of
8.4 Organic Reactions of Cyclic Alkynes
8
9
10
11
303
I (ev)
Figure 8-6 Correlation of the first PE bands of 100 with those of 6, 104 and 118 to demonstrate homoconjugation between double and triple bond(s) in 104 and 118.
1,6-dihetero-3,8-dyclodecadiyneshave been investigated [45].These investigations revealed that in the case of the dioxa compound 12 the sequence is x < n. In the other cases, such as 102 and the corresponding 1,6-diselena-3,8-cyclodecadiyne24, the ionizations from the lone pairs at the heteroatoms correspond to the first ionization energies. The split between niand ni+ varies from 0.9 to 1.5 eV.
8.4 Organic Reactions of Cyclic Alkynes A characteristic feature of the reactions of cyclic alkynes is the tendency to release ring strain by changing the hybridization at the alkyne carbons from sp to sp2. This is achieved either by rearrangement reactions or by intra- or intermolecular addition reactions.
8.4.1 Rearrangement of Cyclic Alkynes Since an allenic system is built of only three collinear carbon atoms, whereas an acetylenic unit requires two bond angles of 180",it is expected that the allene isomers of cyclic alkynes will become increasingly more stable than their acetylenic analogs with decreasing ring size. This has been demonstrated for the ring sizes 9-11 by Moore and Ward [61](Table 8-5). In their experiments, the equilibrium compositions of cyclic allenekyclic acetylene mixtures were
304
8 Cyclic Alkynes: Preparation and Properties
Table 8-5 The cycloalkyne/cycloallene equilibrium in t-BuOH/KO-tBu at 100“C
Ring size
Cycloalkyne in the equilibrium (Yo)
11
74
10
35 7
9
determined in t-BuOH/KO-t-Bu at 100 “C. However, the equilibrium ratios strongly depend on the medium used for this equilibration reaction [61]. Later, Dale used the extensive triple-bond migration, which takes place easily in macrocyclic dialkynes (Cl2-Cz0)when they are treated with KO-t-Bu in DMSO, to confirm his predictions for the relative stability of cyclic diyne isomers [4c, 621. These predictions were based on conformational considerations. Thus, 1,7qclotetradecadiyne (119) is almost completely rearranged to the 1,8-isomer 120 because in the latter compound both angle and terminal strain are absent, i. e., an “ideal” conformation is possible (Scheme 8-11). Dale’s predictions, which were made long before the routine use of X-ray crystallography, were summarized in a seminal paper [63] and later confirmed by various X-ray investigations. The triple bond migration in cyclic diynes proceeds via allenic intermediates, which, however, are found only in small quantities in the reaction mixture.
f-=7 (CH2)4 L - /
(CH2)2
119
KO‘BU
DMSO
/--=(CH2)3 L - /
(CH2)3
II I
120
Scheme 8-11
In the case of 1,6-cyclodecadiyne (loo), which already has an “ideal” geometry in Dale’s sense (it can be considered a “stretched” cyclohexane), no rearrangement is observed even after prolonged heating with KO-t-Bu/t-BuOH [Eq.(20)] [64]. The reactivity of the propargylic hydrogen atoms in a 1,6-cyclodecadiyne system can, however, be dramatically enhanced if exocyclic double bonds are attached to the central carbon atom of the two C3-units between the triple bonds. Thus, both 4,9-dimethylene-1,6-cyclodecadiyne(104) and 4,9-diisopropylidene-1,6-cyclodecadiyne(6) readily isomerize to the corresponding cyclic diallenes when treated with KO-t-Bull-BuOH in THF at low temperature [5a] [Eq. (21)l.
305
8.4 Organic Reactions of Cyclic Alkynes
R=H 104 R=CHj 6
R-H
121
R=CH,
122
8.4.2 Transannular Reactions Proximity is a general chemical principle which is responsible for such fundamental phenomena as the chelate effect or enzymic catalysis. In cyclic systems, proximity of functional groups can cause unusual reactions, which otherwise proceed only with difficulty or not at all. If a triple bond in a cyclic alkyne is faced by a carbonyl group on the opposite side of the ring, a facile rearrangement to a bicyclic enone, involving the formation of a transannular C - C bond, is possible. The mechanism of this reaction has recently been studied by Grunwell et al. [65]. Based on labeling experiments and molecular modeling, they proposed the mechanism shown in Scheme 8-12. The carbonyl oxygen is positioned directly above the C-5 of the triple bond in the most stable conformation of 5-cyclodecynone (123). Thus, nucleophilic attack of a carbonyl oxygen(!) at a triple bond is possible as the two “reactants” are forced into a geometry just right for a start into the path along the transition state of this highly unusual reaction, which has never been observed in cases where the triple bond and the carbonyl group are located in two different molecules. Another transannular C - C bond formation reaction which starts with the attack of an electrophile at the triple bond is the formation of bicyclic systems from cyclic diynes shown in Scheme 8-13, where H2S04 induces the “transannular addition” of H 2 0 to a cyclic diyne
m-m-m-m -
7” H’
123 Scheme 8-12
Scheme 8-13
7*
H
H +*
-H’
II
0 124
306
8 Cyclic Alkynes: Preparation and Properties
system [12]. Of course, transannular C-C bond formation can also be induced by nucleophilic attack at the triple bond, which is illustrated as well in Scheme 8-13. One recent interesting example of this type of reactions is the “zipper” reaction of cyclic o-ethynylbenzenes discovered by Youngs et al. (Scheme 8-14) [66]. The products of these reductive cyclizations show a helical ribbon geometry; this chemistry might lead to extended helical aromatic ribbon structures when applied to larger oligobenzocyclynes.
Li THF
0 127
1
r
i
L
0 .aJ 128 Scheme 8-14
-
8.4 Organic Reactions of Cyclic Alkynes
307
One of the most important transannular reactions of cyclic alkynes is the Bergman cyclization [67] of enediynes (Scheme 8-15). Nicolaou et al. have shown that the crucial distance between the termini of the enediyne system, allowing spontaneous Bergman cyclization at room temperature, is in the range of 3.1-3.2 A [9c]. Questions about the energy and structure of the biradicaloid intermediate and the mechanism of the biradical formation have stimulated considerable theoretical and experimental work [68]. These investigations and the importance of the Bergman cyclization reaction for the action of a new class of antitumor antibiotics will not be discussed here, as the enediynes are dealt with in considerable detail in Chapter 7.
129
J
L
131
130
Scheme 8-15
Related cyclization reactions are the formation of zethrene (134) 1691, which was the product of all attempts to prepare the corresponding bisnaphthalenediyne system 132 (Scheme 8-16). Also, 1,6-dehydro[lO]annulene(40) [19], already an aromatic system, spontaneously cyclizes to give naphthalene via 1,s-dehydronaphthalene (135) at low temperature [19] (Scheme 8-17).
-
2 RH -2RH.
134
133
132 Scheme 8-16
r
40
1
A
L
135
136
Scheme 8-17
The diyne cyclizations discussed so far involve the formation of an aromatic system together with the biradical. Surprisingly, though, a new kind of diyne- to -biradical cyclization has been discovered only recently, where the transannular interaction of two triple bonds in a diazacy-
308
8 Cyclic Alkynes: Preparation and Properties
clodecadiyne system (137) results in the formation of a nonaromatic 1,Cdehydrobutadiene system (Scheme 8-18) [70]. The scope and limitations of this reaction are not yet fully explored. The importance of transannular electronic interactions is beautifully shown in a reaction found by S. Misumi, T. Ogawa and T. Kaneda [Eq. (22)], when transannular interaction between a diyne system and an aromatic ring results in completely unusual reactivity [3a].
137
138
139
140
Scheme 8-18
1NC " 142
8.4.3 Addition Reactions of Cyclic Alkynes The triple bonds in cyclic alkynes can, of course, be subjected to all known addition reactions of acetylenes. Here, we will discuss examples which either lead to particularly interesting addition products or demonstrate unusual reactivity of bent triple bonds.
8.4.3.1 Homonuclear Addition Reactions
The addition of dihydrogen to triple bonds, which can be readily achieved either using ionic reagents such as LiAlH, to give trans double bonds or catalytically affording cis-alkenes, has found important applications in the synthesis of cyclic alkenes with interesting n-parameters. Thus, Sondheimer's famous synthesis of [18]annulene (145) employs the hydrogenation of an acetylenic precursor in the last step (Scheme 8-19). Recently, hydrogenation of cyclic dienediynes has been used for the preparation of interesting cyclic tetraenes, such as tetrahomocyclooctatetraene 146 [Eq. (23)l 15 a]. The high reactivity of bent triple bonds toward both nucleophiles and electrophiles is demonstrated by the reaction of cyclooctyne with lithium and iodine, both reactions affording a homonuclear addition product [3b] (Scheme 8-20).
8.4 Organic Reactions of Cyclic Alkynes
3 HCECCHZCHZCZCH
309
c u (Pyridine O A C ) ~0 H ~ ~
143
- 144
145
Scheme 8-19
(13'
H2 Li n d I o r catalyst
112
146
Scheme 8-20
8.4.3.2 Heteronuclear Addition Reactions Heteronuclear addition reactions to alkynes usually produce the corresponding tram-alkene derivatives. However, in the case of strained cyclic alkynes, these would be even more heavily strained than the starting alkyne. Thus, in most reactions, cis-products are observed. The initial addition of the electrophile may, however, still proceed in an anti-manner, as was shown by Krebs et al. [l b] for the reaction of a cycloheptyne derivative with trichloromethylsulfenyl chloride (Scheme 8-21). This reaction produced the first seven-membered trans-cycloalkene derivative to be isolable at room temperature.
8.4.3.3 Cycloaddition Reactions Qpical reactions of strained cycloalkynes are the [2 + 11 cycloaddition reaction with isonitriles and the [3 + 21 cycloaddition reaction with CS2 (Schemes 8-22 and 8-23) [l b]. As the latter
310
8 Cyclic Alkynes: Preparation and Properfies
-6O’C 94
149
0 2 q s - c c 1 3
-SOP. + 2 0°C + CDCI,L
150
CHCI, reflux
Scheme 8-21
-
02Qscc’3
CI
151
reaction leads to electron-rich tetrathiafulvalene systems, this concept is still popular in the design of new materials. Heterocyclic alkynes thus can lead to particularly electron-rich tetrathiafulvalene systems [3b].
X=S.CHz 93.59
Scheme 8-22
154
+
153
Scheme 8-23
Other cycloaddition reactions are the dimerization and oligomerization of strained cycloalkynes, which proceed via cyclobutadiene systems (Scheme 8-24) [l, 3 b]. Usually, however, these reactions require the assistance of transition metal complexes, which are dealt with in the following section.
8.5 Reactions of Cyclic Alkynes with Metal Compounds
311
157
Scheme 8-24
160
Scheme 8-25
8.5 Reactions of Cyclic Alkynes with Metal Compounds This section is short, for two main reasons: (1) there are two other chapters in this book (Chapters 4 and 5) devoted to the metalorganic chemistry of acetylenes, and (2) there are several recent reviews dealing with metal compounds [l b, c, 72-77]. Therefore we 411 deal here only with very recent reactions which are typical for cyclic mono- and diynes. Highly strained cyclic alkynes can be stabilized by complexing them with metal compounds [lb, c 75, 761. Thus, even reactions with elusive species such as cyclohexyne are possible, as exemplified in Scheme 8-26 [78]. The reaction of cyclic diynes with C ~ C O ( C O or ) ~ CpRh(CO)2, or in some cases with Fe(CO),, leads to the tricyclic cyclobutadiene complexes, as demonstrated for the reaction of 1,7-cyclododecadiyne (3) with CPM(CO)~in Eq. (24). This reaction was discovered and explored 20 years ago [77, 791. The availability of cyclic diynes with uneven chains such as 1,Bcyclodecadiyne (loo), 1,8-~yclotetradecadiyne(120) and 1,lOqclooctadecadiyne (168) led to the superphanes 169-171 and to the intramolecular complexes 172-174 [Eq. (25)] [l C, 801. While the one-pot reaction is limited so far only to diynes with uneven chains and to CpCoL2 complexes, a route has been designed which allows the synthetic of superphanes
312
8 Cyclic Alkynes: Preparation and Properties
R e R
161
I--:
162
163
k:
164
165
Scheme 8-26
3
M=Co, Rh 166 167
with different chain lengths [81b] and with different metals as complexing units [81a]. The easy access to 169 has opened a new way to bridged cage systems [82]. It is interesting to note that the reaction of the sulfur-containing diynes shown in Eq.(26) with CpCo(CO), leads not to the anticipated cyclobutadiene complexes but to the thiophenophanes [83]. This reaction works even with catalytic amounts of the CpCo complex.
XaCH, ,C,H,, S. NR
26.1 75.101.102
176-179
8.5 Reactions of Cyclic Alkynes with Metal Compounds
313
A further reaction which finally gives bridged cage compounds is that of some cyclic alkynes with AlCl,. It goes back to the discovery that dialkyl-substituted acetylenes react with AlCl, at low temperature to yield a cyclobutadiene-A1C13 complex [84].This reaction has been used recently to synthesize doubly bridged Dewar benzenes as shown in Scheme 8-27 and Eq. (27) by reacting the complex 180 either with dimethylacetylene dicarboxylate (DMAD) or tert-butylsulfonylacetylene[MI. The resulting Dewar benzenes are remarkably stable and can be used for further reactions. Thus, the ester groups can be reduced and the tert-butylsulfonyl substituents replaced. Some of the resulting Dewar benzenes have been converted into prismanes [85].
20
6
1
90
DMAD
'C02Me 181
Scheme 8-27
'C02Me 11-5.6
182
314
8 Cyclic Alkynes: Preparation and Properties
8.6 Conclusions In the last 20 years many efficient protocols have been developed which allow the synthesis of cyclic mono- and diynes from ring size six to sixteen. The availability of these molecules has led to new insights into the response of triple bonds toward bending and orbital interactions. The availability of energy-rich cyclic systems has paved the way to synthesis of strained rings and cages.
8.7 Experimental Procedures 8.7.1 Preparation of Cyclic Dialkynes of Ring Size CI2-C,, 8.7.1.1 General Procedure [5 b]
To a solution of the a,w-diyne (85 mmol) in anhydrous THF (1.5 L) at -20°C was added 2.5 M BuLi in hexane (68 mL) over a period of 10 min under an Ar atmosphere. A white precipitate was observed, and the solution changed to yellow after all the BuLi had been added. The mixture was allowed to warm to room temperature (rt) and the stirring was continued for another 15 min. Finally the dihalogenide (usually diiodide) (90 mmol) was added. The resulting mixture was refluxed for 3-7 d until the precipitate had disappeared. The reaction was terminated as soon as the gas-chromatographic analysis showed no starting material. After cooling, the solution was poured into a mixture of petroleum ether (bp 30-75°C) (300 mL) and 2 M HCI (400 mL). The organic layer was separated and the aqueous layer was extracted with petroleum ether (2 x 100 mL). The combined organic layers were neutralized with saturated NaHCO, solution, dried (Na2S04) and concentrated in vacuo. The crude products were worked up as follows. 8.7.1.2 1,7-Cyclododecadiyne(3) After chromatography (silica gel, CC14, 30 cm x 60 mm) the CCI, phase was treated with an aqueous solution of Na2S20, (to remove Id. All fractions containing more than 70% 3 were concentrated in vacuo and recrystallized from EtOH (-25°C); yield 4.75 g (35%); mp 36-37 "C (Lit. [4b]) 37-38 "C.
8.7.1.3 1,8-Cyclotetradecadiyne (120) The raw material was filtered through silica gel (10 cm x 60 mm, CC14) and recrystallized from EtOH (100 mL) at - 15 "C. This gave 10- 12 g of white crystals with a characteristic odor. From the mother liquor a further fraction of 1.5-3 g of 120 was obtained by chromatography on silica gel with CCl,; total yield: 13.5 g (85 Vo), mp 98 "C (Lit. [4c, 5 b] mp 97-98 "C). The final fractions of column chromatography contained 0.32 g (2Vo) of the tetramer. 'H-NMR (200 MHz, CDC1,): 6 = 2.4-2.0 (m, H), 2.0-1.4 (m, 4H), 1.4-1.2 (m, 8H). 13C-NMR (50.32 MHz, CDC1,): S = 80.7, 28.0, 26.7, 18.4.
8.7 Experimental Procedures
315
8.7.2 General Procedure for Dewar Benzenes 181 and 182 [85] A slurry of 1.34 g (10mmol) of AlCl, (sublimed) and 10 mL of CH2C12 was cooled to -40 "C and a solution of 10 mmol of the cycloalkadiyne (3, 120,or 90) in 10 mL of CH2Cl, was added slowly. The solution turned red: the AlCl, was dissolving. The solution was allowed to warm to rt, stirred for 30 min, and cooled to -20°C. Solutions of 2.84 g (20 mmol of dimethylacetylene dicarboxylate (DMAD) in 10 mL of CH2C12,and 5 mL of dimethylsulfoxide (DMSO) in 20 mL of CH2C12,were added successively. The mixture was poured into 80 mL of ice-water/80 mL of pentane and separated; the aqueous layer was extracted twice with 40 mL of pentane, dried over Na2S0,, and concentrated on a rotary evaporator. The crude products were purified by column chromatography (silica gel, CH2C12). The two isomeric Dewar benzenes (181 and 182) usually required a second column-chromatography step to be separated from each other. The second chromatography afforded 30% 181 (n = 5), 13% 182 (n = 5) and 17% mixture.
8.7.2.1 Dimethyl Tetracyclo[l2.2.0.0'~7.0S~'4]hexadeca-7,15-diene15,16-dicarboxylate182; (n = 5) Yield 425 mg (13 Olo), viscous oil; 'H-NMR (200 MHz, CDCL,) 6 = 3.77 (s, 6H), 2.4-1.2 (m, 20H); I3C-NMR (50.32MHz, CD2ClZ) 6 = 163.5,150.3, 146.5,60.4,51.6, 31.1, 28.6, 28.4, 28.1,27.2;UV/vis (acetonitrile) X , [nm] (log E) = 196 (4.08),244 sh (3.05);high-resolution MS (HRMS) calcd. for C20H2604 (M') 330.1831,found 330.1842.
8.7.2.2 Dimethyl TetracycIo[7.5.2.0.0z~E]hexadeca-2,15-diene-l5,l6-dicarboxylate (181; n = 5) Yield 1000 mg (30%), viscous oil; 'H-NMR (200MHz, CDCl,) 6 = 3.76 (s, 6H), 2.4-1.2(m, 20H); I3C-NMR (50.32MHz, CDCl,) 6 = 162.8, 151.2, 147.7,61.6, 51.5, 33.2, 29.2, 28.5, 28.4,27.6,26.5; UV/vis (acetonitrile) ,,X [nm] (log E) = 190 (4.41),240 sh (3.84); HRMS calcd. for C2,H2604 (M') 330.1031,found 330.1862.
8.7.3 Cyclonon-2-ynone (91) and Bicyclo[6.1.0]non-1(8)-en-9-one (92) A solution of 1.02 g (9.4 mmol) of cyclooctyne and 3.7 mL (45.4mmol) of trichloromethane in 150 mL of THF was stirred at -78"C, and 6.4 mL (102 mmol) of 1.6 M n-BuLi in hexane was added dropwise within 45 min. The mixture was stirred for 1 h at - 78"C, warmed to -10°C and treated with 50 mL of water. The organic layer was washed several times with brine until the aqueous phase was separated on neutral alumina with cyclohexane/ethyl acetate (10:1 to 0:1) as eluent which gave 427 mg (33.2%) of cyclonon-2-ynone 91 and 230 mg (17.9%) of bicyclo[6.1.0]non-1(8)-en-9-one 92 [86].An analytically pure sample of 91 was prepared by Kugelrohr distillation at 120"C/O.8 mbar. 91: 'H-NMR (300 MHz, C6D6): 6 = 2.13-2.09 (m, 2H), 1.79-1.75(m, 2H), 1.44-1.16(m, 8H); 13C-NMR (75.47MHz, C6D6): 6 = 189.03, 112, 14, 85.69,44.16,27.70, 27.21, 25.78, 22.76, 19.34. IR (neat): v (cm-') = 2930, 2862, 2184, 1670, 1457, 1230, 1187,975.
316
8 Cyclic Alkynes: Preparation and Properties
Acknowledgements We thank the Deutsche Forschungsgemeinschaft, the Fonds der Chemischen Industrie and the BASF AG, Ludwigshafen, for financial support. R. M. thanks the Humboldt foundation for a Feodor-Lynen Fellowship. We are grateful to Petra Schlickenrieder and Sandro Silverio for typing the manuscript and making the drawings, respectively.
Abbreviations CP DMAD PE rt THF
cyclopentadienyl dimethyl acetylenedicarboxylate photoelectron room temperature tetrahydrofuran
References [I] For earlier reviews on the reactivity of cyclic alkynes see: (a) R. W. Hoffmann, Dehydrobenzene and Cycloalkynes, Verlag Chernie, Weinheim, 1967; (b) A. Krebs, J. Wilke, Top. Curc Chem. 1983, 109, 189-233; (c) R. Gleiter, Angew. Chem. 1992, 104, 29-46; Angew. Chem., Znt. Ed. Engl. 1992, 31, 27-44; (d) see also [2a], [3b]. [2] Mainly concerned with synthetic aspects are the following reviews: (a) M. Nakagawa, “Cylic Acetylenes”, in The Chemistry of the Carbon-Carbon Diple Bond (Ed.: S. Patai), Wiley, New York, 1978, pp. 635-712; (b) H. Meier, N. Hanold, T. Molz, H. J. Bissinger, H. Kolshorn, J. Zountsas, Tetrahedron 1986, 42, 1711-1719; (c) see also [lc]. [3] Structural and electronic properties are the main topics of the following reviews: (a) S. Misumi, T. Kaneda, Proximity interactions of Acetylenes in The Chemistry of the Carbon-Carbon Triple Bond (Ed.: S . Patai), Wiley, New York, 1978, pp. 713-737; (b) H. Meier, Advances in Strain in Organic Chemistry Vol. 1, J. A. I . , 1991, pp. 215-272; see also [2a], [l c]; (c) W. Sander, Angew. Chem. 1994, 106, 1522-1524; Angew. Chem., Int. Ed. Engl. 1994, 33, 1455-1456. [4] (a) A. J. Hubert, J. Dale, Chem. Ind. 1961, 249, 1224-1225; (b) J. H. Wotiz, R. F. Adams, C. G. Parsons, J. Am. Chem. SOC. 1961, 83, 373-376; (c) J. Dale, A. J. Hubert, G. S. D. King, J. Chem. SOC. 1963, 73-86; (d) A. J. Hubert, J. Dale, ibid. 1963, 86-93; (e) A. J. Hubert, M. Hubert, Tetrahedron Lett. 1966, 5779-5782; (f) G. Schill, U. Keller, Synthesis 1972,621-622; (9) S. F. Karaev, M. M. Movsumzade Zh. Org.Khim. 1974,14880; J. Org.Chem. USSR 1974,10,886; (h) G. Schill, E. Logemann, H. Fritz, Chem. Ber. 1976,109,497-502; (i) A. Nissen, H. A. Staab, ibid. 1971,104. 1191-1198; (k) N. Darby, C. U. Kim, J. A. Salaiin, K. W. Shelton, S. Takada, S. Masamune, Chem. Commun. 1971, 1516-1517. [5] (a) R. Gleiter, R. Merger, B. Nuber, J. Am. Chem. SOC. 1992, 114, 8921-8927; (b) R. Gleiter, R. Merger, B. Treptow, W. Wittwer, 0. PflBsterer, Synthesis 1993, 558-560. [6] Y. Pang, A. Schneider, T. J. Barton, M. S. Gordon, M. T. Carroll, J. Am. Chem. SOC. 1992, 114, 4920-4921. [7] (a) M. Lespieau, C. R. Acad. Sci. 1929, 188, 502-503; (b) F. Sondheimer, Y. Gaoni, J. Bregman, Tetrahedron Lett. 1960, 26, 25-29. [8] L. T. Scott, G. J. DeCicco, J. L. Hyun, G. Reinhardt, J. Am. Chem. SOC.1985, 107,6546-6555 and literature cited therein.
References
317
(91 (a) K. C. Nicolaou, W.-M. Dai, Angew. Chem. 1991, 103, 1453-1481; Angew. Chem., Znt. Ed. Engl. 1991, 30, 1387-1416 and literature cited therein; (b) R. Gleiter, D. Kratz Angew. Chem. 1993, 105, 884-887; Angew. Chem., Int. Ed. Engl. 1993,32, 842-845 and references therein; (c) K. C. Nicolaou, A. L. Smith, Acc. Chem. Res. 1992, 25, 497-503 and references cited therein. [lo] (a) C. Huynh, G. Linstrumelle, Tetrahedron 1988, 44, 6337-6344; (b) D. Solooki, J. D. Bradshaw, C. A. Tessier, W. J. Youngs, Organometallics 1994, 13, 451-455. [I11 (a) R. Wolovsky, F. Sondheimer, J Am. Chem. Soc. 1962,84,2844-2845; (b) R. Wolovsky, F. Sondheimer, ibid. 1965, 87, 5720-5727. (c) F. Sondheimer, R. Wolovsky, P. J. Garratt, I. C. Calder, ibid. 1966, 88, 2610-2610. [I21 R. Gleiter, J. Ritter, H. Irngartinger, J. Lichtenthtiler, Tetrahedron Lett. 1991, 32, 2883-2886 and references cited therein. 113) R. Gleiter, S. Rittinger, H. Langer, Chem. Ber. 1991, 124, 357-363. [I41 For a detailed discussion see: L. C. Tan, R. M. Pagni, G. W. Kabalka, M. Hillmyer, J. Woosley, Tetrahedron Lett. 1992, 33, 7709-7712 and references cited therein. [I51 K. C. Nicolaou, G. Skokotas, P. Maligres, G. Zuccarello, E. J. Schweiger, K. Toshima, S. Wendeborn, Angew. Chem. 1989, 101, 1255-1257; Angew. Chem, Znt. Ed. Engl. 1989, 28, 1272-1274. [I61 D. Brillon, P. Deslongchamps, Tetruhedron Lett. 1986, 27, 1131-1134 and references cited therein. [17] P. Magnus, H. Annoura, J. Harling, L Org. Chem. 1990,55, 1709-1711 and references cited therein. (181 (a) P. A. Wender, J. A. McKinney, C. Mukai, L Am. Chem. SOC.1990,112,5369-5370; (b) K. 'bkai, T. Kuroda, S. Nakatsukasa, K. Oshima, H. Nozaki, Btrahedron Lett. 1985, 26, 5585-5588; (c) T. D. Aicher, Y.Kishi, Tetrahedron Lett. 1987,28, 3463-3466; (d) P. J. Proteau, M. S. Thesis, California Institute of Technology 1990; (e) C. Crevisy, J.-M. Beau, Tetrahedron Lett. 1991, 32, 3171-3174 and references therein. [I91 A. G. Myers, N. S. Finney, L Am. Chem. SOC. 1992, 114, 10986-10987. [20] R. Gleiter, H. Stahr, publication in preparation. (211 G. G. Melikyan, R. C. Combs, J. Lamirand, M. Khan, K. M. Nicholas, Tetrahedron Lett, 1994,35, 363-366. (221 R. Gleiter, H. Reckel, H. Imgartinger, T. Oeser, Angew. Chem., 1994, 106, 1340-1342; Angew. Chem., Int. Ed. Engl. 1994, 33, 1270-1272. (231 R. Gleiter, H. Rockel, B. Nuber, Tetrahedron Lett. 1995, 36, 1835-1838. (241 H. Detert, B. Rose, W. Mayer, H. Meier, Chem. Ber. 1994, 127, 1529-1532. [25] (a) E. Kloster-Jensen, J. Wirz, Helv. Chim. Acta 1975, 58, 162-177; (b) idem, Angew. Chem 1973, 85. 723; Angew. Chem, Znt. Ed. Engl. 1973, 12, 671. [26] (a) A. Eschenmoser. D. Felix, G. Ohloff, Helv. Chim. Acta 1967, 50, 708-713; (b) M. %abe, D. F. Crowe, R. L. Dehn, Tetrahedron Lett. 1967, 3943-3946. (271 0. L. Chapman, J. Gano, P. R. West, M. Regitz, G. Maas, L Am. Chem. Soc. 1981, 103,7033-7036. (281 C. D. Hurd, R. I. Mori, L Am. Chem. Soc. 1955, 77, 5359-5364. [29] H. Meier, K.-P. Zeller, Angew. Chem. 1977, 89, 876-890; Angew. Chem., Znt. Ed. Engl. 1977, 16, 835-852. [30] (a) I. Lalezari, A. Shafiee, M. Yalpm-, Tetrahedron Lett. 1969, 5105-5106; Angew. Chem. 1970, 82, 484; Angew. Chem., Znt. Ed. Engl. 1970, 9, 464-465; (b) H. Meier, H. Petersen, Synthesis 1978, 596-598; (c) H. Buhl, H. Gugel, H. Kolshorn, H. Meier. Synthesis 1978, 536-537. [31] For a review of the Ramberg-BBklund reaction see: L. A. Paquette, Org. React. 1977, 25, 1-71. [32] W. Ando, N. Nakayama, Y. Kabe, T. Shimizu, Tetrahedron Lett. 1990, 31, 3597-3598. [33] (a) W. Ando, F. Hojo, S. Sekigawa, N. Nakayama, T. Shimizu, Organometal/ics1992,II, 1009-1011; (b) S. Sekigawa T. Shimizu, W. Ando, Tetruhedron 1993, 49, 6359-6366. (341 T. Matsuoka, T. Negi, T. Otsubo, Y. Sakata, S. Misumi, Bull. Chem. SOC.Japan 1972,45,1825-1833; (b) T. Matsuoka, T. Negi, S. Misumi, Synth. Commrtn. 1972, 2, 87-92. [35] R. Gleiter, M. Merger, Synthesis. in press. (361 J. Haase, A. Krebs, Z. Naturforsch. Pi1 A 1972, 27, 624-627.
318
8 Cyclic Alkynes: Preparation and Properties
[37] H. H. Bartsch, H. Colberg, A. Krebs, Z. Kristallogr. 1981, 156, 10-12. [38] R. A. G. de Graaff, S. Gorter, C. Romers, H. N. C. Wong, E Sondheimer, J. Chem. SOC, Perkin %m. 1 1981, 478-480. [39] J. Haase, A. Krebs, Z. Naturforsch., Teil A 1971, 26, 1190-1193. [40] V. 'Qpke, J. Haase, A. Krebs, J. Mol. Struct. 1979, 56, 77-86. 1411 R. Gleiter, M. Merger, H. Irngartinger, J. Org. Chem., in press. [42] R. Destro, T. Pilati, M. Simonetta, Acta Crystallogr., Sect. B 1977, 33, 447-456. [43] H. Sakurai, Y. Nakadaira, A. Hosomi, Y. Eriyama, C. Kabuto, J Am. Chem. SOC. 1983, 105, 3359-3360. [44]R. Gleiter, M. Karcher, R. Jahn, H. Irngartinger, Chem. Ber. 1988, 121, 735-740. [45] R. Gleiter, S. Rittinger, H. Irngartinger, Chem. Ber. 1988, 124, 365-369. (461 R. Gleiter, H. Rdckel, B. Nuber, Tetrahedron Lett. 1994, 35, 8779-8782. [47] R. Gleiter, R. Merger, H. Irngartinger, J Am. Chem. SOC. 1992, 114, 8927-8932. [48] E. Kloster-Jensen, C. Rdmming, Acta Chem. Scand. Ser.B 1986, 40, 604-605. (491 H. Irngartinger, A. E. Jungk, Chem. Ber. 1977, 110, 749-759. (501 M. J. Bennett, R. A. Smith, Acta Crystallogr., Secf.B 1977, 33, 1123-1126. [51] G. A. Eliassen, E. Kloster-Jensen, C. Rdmming, Acra Chem. Scand., SexB 1986, 40, 574-582. [52] K. C. Nicolaou, G. Zuccarello, Y. Ogawa, E. J. Schweiger, T. Kumazawa, J Am. Chem. Soc. 1988, 110, 4866-4868. [53] H. Irngartinger, Chem. Ber. 1977, 110, 744-748. [54] H. Irngartinger, Chem. Ber. 1972, 105, 1184-1202. [55] H. Irngartinger, Chem. Ber. 1973, 106, 751-760. [56] H. Irngartinger, Chem. Ber. 1973, 106, 761-772. (571 R. Gleiter, D. Kratz, W. Schafer, V. Schehlmann, J. Am. Chem. SOC. 1991, 113, 9258-9264. [58] R. Gleiter, W. Schafer in The Chemistry of Triple-BondedFunctional Groups (Ed.: S. Patai), Suppl. C, Vol. 2, Wiley, New York, 1994, 153-189. [59] H. Schmidt, A. Schweig, A. Krebs, Tetrahedron Left. 1974, 1471-1474. [a01 G. Bieri, E. Heilbronner, E. Kloster-Jensen. A. Schmelzer, J. Win, Helv. Chim. Acta 1974, 57, 1265-1283. (611 W. R. Moore, H. R. Ward, J. Am. Chem. SOC.1963, 85, 86-89. [62] J. Dale, 1 Chem. SOC. 1963, 93-111. [63] J. Dale, Angew. Chem. 1966, 78, 1070-1093. (641 R. Gleiter, R. Merger, M. Karcher, unpublished results. 1651 J. R. Grunwell, M. F. Wempe, J. Mitchell, J. R. Grunwell, Tetmhedron Left. 1993, 34, 7163-7166; M. F. Wempe, J. R. Grunwell, J Org.Chem. 1995, 60, 2714-2720. [66] J. D. Bradshaw, D. Solooki, C. A. Tessier, W. J. Youngs, X Am. Chem SOC.1994, 116, 3177-3179. [67] T. P. Lockhart, P. B. Comita, R. G. Bergman. J. Am. Chem. SOC. 1981, 103, 4082-4090. [68] See, for example: P. G. Wenthold, R. R. Squires, J. Am. Chem. SOC.1994,116, 6401-6412 and the literature cited therein. [69] (a) R. H. Mitchell, F. Sondheimer, Tetmhedron 1970,26, 2141-2150; @) H. A. Staab, J. Ipaktschi, A. Nissen, Chem. Ber 1971, 104, 1182-1190. 1701 R. Gleiter, J. Ritter Angew. Chem. 1994, 106, 2550-2552; Angew. Chem. Int. Ed. Engl. 1994, 33, 2470-2472. [71] (a) E Sondheimer, R. Wolovsky, J. Am. Chem. SOC. 1962, 84, 260-269; (b) idem, J. Chem. SOC, Dalton nuns. 1982, 89-94. [72] K. P. C. Vollhardt. Acc. Chem. Res. 1977,10,1-8; Angew. Chern. 1984,96,525-541; Angew. Chem., Znt. Ed. Engl. 1984, 23, 539-555. (731 H. BBnnemann, R. Goddard, J. Grub, R. Mynott, E. Raabe, S . Wendel, Organometullics,1989, 8, 1941- 1958. [74] N. E. Shore, Chem. Rev. 1988, 88, 1081-1119.
References
319
[75] M. A. Bennett, H. P. Schwemlein, Angew. Chem. 1989, 101, 1349-1373; Angew. Chem., Int. Ed. Engl. 1989,28, 1296-1320. (761 S. Buchwald, R. B. Nielsen, Chem. Rev. 1988,88, 1047-1058. [77] A. Efraty, Chem. Rev. 1977, 77, 691-744. [78] S. L. Buchwald, R. T. Lum, J. C. Dewan, X Am. Chem. SOC.1986, 108, 7441-7442. [79] (a) R. B. King, A. Efraty, X Am. Chem. SOC.1972,94,3021-3025; (b) R. B. King, X Ind. Chem. SOC. 1977,55, 169-175; (c) R. B. King, 1. Haiduc, C. W. Eavenson, X Am. Chem. SOC.1973,95, 2508-2516. [80]R. Gleiter, D. Kratz, Acc. Chem. Res. 1993,26, 311-318 and references therein. [81] (a) R. Gleiter, H. Langer, B. Nuber, Angew. Chem. 1994,106, 1350-1352; Angew. Chem., Int. Ed. Engl. 1994, 33, 1272-1274; (b) R. Gleiter, V. Schehlmann, Angew. Chem. 1990, 102, 1450-1452; Angew. Chem, Int. Ed. Engl. 1990,29, 1426-1427. [82] R. Gleiter, G. Pflftsterer, H. Irngartinger, Chem. Ber. 1993, 126, 1011-1013. Angew. Chem., [83] (a) R. Gleiter, M. Karcher, B. Nuber, M. L. Ziegler, Angew.Chem. 1987,99,805-806; Int. Ed. Engl. 1987, 26, 763-764; (b) R. Gleiter, S. Rittinger, H. Langer, Chem. Ber. 1991, 124, 357-363. [84] H.Hogeveen, D. M. Kok in Chemistry of Triple-Bonded Functional Groups, SuppLC, Part 2 (Eds. : S. Patai, Z. Rappoport), Wiley, New York, 1983,pp. 981-1013, and references therein. [85] (a) R. Gleiter, B. lleptow, X Org. Chem. 1993,58, 7740-7750; (b) R. Gleiter, F. Ohlbach, X Chem. Soc, Chern. Commun. 1994,2049-2050. 1861 (a) J. M. White, M. K. Bromley, Tetrahedron Lett. 1993,34,4091-4094; (b) G. Wittig, J. J. Hutchison, Liebigs Ann. Chem. 1970, 741, 79-88.
This Page Intentionally Left Blank
9 Macrocyclic Homoconjugated Polyacetylenes Lawrence T. Scott, Mark J. Cooney
9.1 Introduction Early attempts to synthesize the intriguing cyclonona-l,4,7-triyne molecule, 1 [l], grew from a recognition that the in-plane p-orbitals of the three acetylenes around its perimeter should encroach upon each other’s space while the out-of-plane p-orbitals should consitute an essentially ideal trishomobenzene. The tantalizing prospect that a [2 2 21 cycloaddition requiring very little atomic motion might transform this monocylic triyne into tricyclopropabenzene (2, Fig. 9-1) added further incentive to prepare 1.
+ +
1
2
Figure 9-1 Cyclonona-l,4,7-triyne, 1, and tricyclopropabenzene, 2.
Although cyclonona-1,4,7-triyneremains unknown and a worthy target to this day, as does 2, higher homologs of 1and many related macrocyclic polyacetylenes with a capacity for cyclic
homoconjugation have been brought into being since the early 1980s, largely as a result of the systematic program in our laboratory aimed at understanding cyclic homoconjugation, particularly in neutral molecules [l-251. This chapter presents a broad overview of our work on the synthesis and properties of macrocyclic homoconjugated polyacetylenes. The synthetic methods discovered, invented, developed, and used in this arena, of course, have applicability in the wider field of modem acetylene chemistry, and it is our hope that others will benefit from the findings described in the following sections of this chapter. Our story begins with a family of compounds we refer to as the [Npericyclynes [7, 81 a name intended to capture the structural essence of molecules composed of N alkyne units distributed symmetrically around the perimeter of a cycle with Nvertices (Fig. 9-2). Hydrocarbon 1would be called “[3]pericyclyne” and represents the smallest member of the family. To preclude experimental complications arising from labile hydrogens on the carbon atoms between acetylenes in the parent molecules, we have generally appended methyl groups at the saturated centers. Section 9.3 moves next to “exploded” pericyclynes, in which each side of the polygon consists of a longer 1,3-diyne unit. This is followed by hybrid systems wherein some sides are alkynes and others are diynes, and we conclude with a return to [Nlpericyclynes, introducing heteroatoms at one or more of their vertices.
322
9 Macrocyclic Homoconjugated Polyacetylenes
3
4
mp 201 "C
mp 249 "C
5
6
mp 173 "C
mp 189 "C
Figure 9-2 Permethylated pericyclynes of order [5], [6], [7], and [8].
9.2 Pericyclynes Pericyclynes 3-6 with gem-dimethyl groups at every corner were all synthesized [7, 81 from a single starting material, 7, the readily available adduct of acetone and acetylene. Preliminary conversion of 7 to two complementary building blocks, 8 and 9,set the stage for a coppermediated coupling reaction to give diyne 10. This reaction, which makes a quaternary center with remarkable efficiency, became the workhorse for a convenient homologation protocol that could be repeated all the way up to pentayne 11 (Fig. 9-3). Though even higher oligomers of 9 could undoubtedly be prepared by further repetition of this procedure, a somewhat more streamlined, convergent synthesis proved more practical, even for the preparation of pentayne 11 (Fig. 9-4). Noteworthy here is the utility of acetyl chloride for the mild conversion of a methyl ether to an alkyl chloride (and methyl acetate), presumably via an S,1 process that is facilitated by the adjacent triple bond; Lewis acids accelerate the reaction but are not essential. Treatment of pentayne 11 with acetyl chloride under the same conditions likewise produced propargylic chloride 14, and this could be cyclized to decamethyl[5]pericyclyne, 3, in 35%
9.2 Pericyclynes
-
Me0
b , 0 Y H
Mcf Me
Me Me
TMS
Me" Me 10
9
7
323
/
d,c,d,c,d,c
c y T M S Mlb Me
Me0 Me Me
8
M Z Me 11
Figure 9-3 Homologation sequence for the construction of synthetic precursors to perrnethylated pericyclynes: (a) 2 equiv n-BuLi, then 2.5 equiv TMSCl, then conc HCl, Cu powder, CaCI,, 71 Vo overall; (b) NaH, Me2S0,, SO%, (c) EtMgBr, then CuCl, then 8, 83% for 9 10, 62-83% for each coupling in 10 11; (d) KOH, MeOH, 76-80% for each desilylation in 10 -+ 11. +
+
a, b, a
Me0 Me"' Me
Me"' Me 13
Me0
TMS
M Z Me 10
Me0 Me" Me
Me Me 11
Figure 9-4 A more convergent strategy for the construction of synthetic precursors to permethylpericyclynes: (a) KOH, MeOH, 77-80%; (b) EtMgBr, then CuCl, then 8, 62%; (c) CH,COCl, SnCI,, 82%; (d) EtMgBr, then CuC1; (e) combine, 69% for 12 + 11.
324
9 Macrocyclic Homoconjugated Polyacetylenes
yield by the action of AICI, (Fig. 9-5). Though unprecedented at the time as a method for ring closure, the Friedel- Crafts-like alkylation of a trimethylsilyl acetylene seemed reasonable on mechanistic grounds for our molecule (Fig. 9-9, so we tried it, and it worked [26].
Me
b
Me'
, Me , . ) = = [Me, ,Me 14
presumably
via
3
Me
Me
Me
Me
Figure 9-5 The first synthesis of a pericyclyne and the likely intermediatesinvolvedin the cyclization step: (a) CH,COCI, SnCI,, 82%; (b) AICI,, CS,, 35%.
Convergent syntheses completely analogous to the one illustrated in Fig. 9-4 were then carried out to give the higher homologs of 11 containing six, seven, and eight homoconjugated C = C triple bonds [8]. Cyclizations according to the steps in Fig. 9-5 gave the higher pericyclynes, 4,5, and 6,with gem-dimethyl groups at every corner. Not surprisingly, the yield in the cyclization step dropped steadily as the ring size increased up to 24 carbon atoms: 4 (22%), 5 (6.2%), and 6 (1.5%). Macrocycles 3, 4, 5 and 6 are all colorless, crystalline solids that are stable in the air and light indefinitely at room temperature. None has ever exhibited any shock sensitivity. X-ray crystal structures of 3, [9] 4, [27] and 5 [27] have all been determined, and the structural resemblance between these macrocycles and the corresponding cycloalkanes with the same number of vertices is striking: the [Slpericyclyne, 3, adopts an envelope conformation, the [6]pericyclyne, 4, adopts a chair conformation; and the [7]pericyclyne, 5, adopts a tub conformation. Unfortunately, the [8]pericyclyne, 6, crystallizes as fine hairs that are unsuitable for X-ray diffraction analysis. Of particular interest to us was the question of cyclic homoconjugation in the pericyclynes. The existence of strong orbital interactions among the acetylenic units in such molecules is
9.2 Perkyclynes
325
revealed most dramatically by photoelectron spectroscopy, which shows at least five discernible x-ionization potentials spread over a range of approximately 1 eV for decamethyl[5]pericyclyne, 3 [7,91. If there were no orbital mixing, five equivalent alkynes should give a much simpler photoelectron spectrum with nearly degenerate x-ionization potentials. A large degree of orbital mixing is likewise seen in the photoelectron spectrum of dodecamethyl[6]pericyclyne, 4 [9]. Electron transmission spectroscopy of 3 and 4, which measures gas-phase electron affinities, reveals an even larger splitting (ca. 1.6 eV) of the nonbonding orbitals in these pericyclynes [9]. The lowest unoccupied molecular orbitals (LUMOs) are stabilized by 0.4-0.7 eV relative to the LUMO of acetylene. The heat of hydrogenation of decamethyl[5]pericylyne, 3, was also measured and found to be -340.7 kcal/mol [16]. To determine whether or not this value might reflect any thermodynamic consequence of cyclic homoconjugation, we synthesized a series of acyclic homoconjugated polyacetylenes (15, n = 2, 3, 4, and 5) to use as reference compounds [15, la]. Fig. 9-6 shows how the requisite building blocks were prepared; our standard copper-mediated coupling reaction was then used to join terminal acetylenes 16, 17, and 18 with appropriate propargylic chlorides, e. g., 18 with 20 to give 15, n = 5. From the heats of hydrogenation of these acyclic homoconjugated polyacetylenes we were able to derive an additivity value of - 69.8 kcal/mol for the contribution of an interior alkyne
I
15 n = 2 , 3 , 4 , 5
Me
Me
- AH M“, Me
a,b
” / ,M e Me Me
c
Me
Me&OH
Me”‘ Me 16
M i Me
u”x\G”’” Me, Me
g
Me‘ Me
Me“ Me 19
17
Me
Me, Me
Me, Me Me&Ci
M i Me
Me‘ Me
Me‘ Me 20
18
Figure 9-6 Acyclic homoconjugated polyacetylenes synthesized as reference compounds : (a) EtMgBr, then CuCI, then 8; (b) KOH, MeOH, 46% for 16 + 17, 54% for 17 18; (c) EtMgBr then acetone, 77%; (d) conc. HCI, Cu powder, CaCI,, 71%. +
326
9 Macrocyclic Homoconjugated Polyacetylenes
21
la
24 25
Figure 9-7 The first syntheses of a [4]pericyclyne and a quinone of [IJpericyclyne:(a) EtMgBr, then HCOOEt, then NH,CI, 10% 22, 6% 23; (b) Jones Reagent, 54% for 22 + 24, 32% for 23 -+ 25.
321
9.2 Pericyclynes
to the overall heat of hydrogenation, i. e., the contribution of an alkyne that is homoconjugated at both ends but not in a cycle [16]. Five times this contribution gives the “expected” heat of hydrogenation for decamethyl[5]pericyclyne, 3 ( - 349.0 kcal/mol). Even after correcting for the calculated strain energy in the hydrogenation product of 3 ( 52 kcal/mol), there remained a little more than 6 kcal/mol that we ascribe to “homoaromatic stabilization” of 3. Several rounds of theoretical calculations on the pericyclynes have been published over the years [9, 21, 281, and the most recent ones stand in conflict with our experimental demonstration of a weak but positive homoaromatic stabilization in 3 [29]. Whether the thermodynamic stabilization resulting from cyclic homoconjugation in the pericyclynes really is greater than zero or not, everyone agrees that it must be small compared with the aromatic stabilization of benzene. Thus, although the homoconjugated acetylenic units in the pericyclynes feel a strong electronic influence from the neighboring triple bonds, as revealed by photoelectron and electron transmission spectroscopy, this orbital mixing has, at best, only minimal thermodynamic consequences. Squeezing the C = C triple bonds closer together, as in [4]pericyclyne and [3]pericyclyne, would be expected to enhance the through-space orbital interactions of the in-plane p-orbitals 19, 281. Unfortunately, all attempts to synthesize these ring systems by the routes that yielded the larger, strain-free pericyclynes failed. Eventually, however, a route to the [4]pericyclynering system was found using a completely different cyclization strategy (Fig. 9-7) [22]. Cyclic
TMS
27
8
29
28
Mes. Me
H 26
AH
i
Mfi Me,
TMS
Me
Me
~
f‘[
e
TMS
H
H 21
30
Figure 9-8 TWOways to assemble the symmetrical tetrayne 21: (a) Mg, then acetaldehyde; (b) DMSO, CICOCOCI, then triethylamine, 71% for 8 28; (c) LDA. then (EtO),POCI; (d) LDA, 63% for 28 -t 30; (e) EtMgBr, then CuCl; (0 combine, 56’70,then KOH, MeOH, 69%; (9) EtMgBr, then CuCl, then 8; (h) KOH, MeOH, 37% for 26 21. -+
+
328
9 Macrocyclic Homoconjugated Polyacetylenes
1 31
Me
+
Me
II
II
Me 32
OH 33
Ib
b
II
It
34 0 35
Figure 9-9 Syntheses of [5]pericyclynone and a quinone of [IO]pericyclyne: (a) EtMgBr, then HCOOEt, then NH,CI, 27% 32; 33 converted directly to the dione; (b) Jones Reagent, 18% 34, 2.2% for 31 --t 35.
329
9.2 Pericyclynes
dimers with a 24-membered ring were also formed in this reaction. Oxidation of the cyclic alcohols gave the corresponding “pericyclynones” ; macrocycle 25 can be viewed as a quinone of [8]pericyclyne. The symmetrical acyclic tetrayne 21 used as a precursor to these new pericyclynes was easily assembled by two different methods (Fig. 9-8) [22]. the lengthier of the two started with the familiar building block 8, whereas the shorter route relied on the more difficultly-accessible 3,3-dimethylpenta-l,Cdiyne, 26. An analogous cyclization of 31 gave the [5]pericyclyne ring system by this new route and also yielded small quantities of the first 30-membered ring in this family, 33 [22]. Oxidation of the cyclic alcohols as before gave the [5]pericyclynone, 34, and the quinone of the [lolpericyclyne, 35. An X-ray crystal structure of octamethyl[5]pericyclynone,34, showed the ring to be perfectly planar [22, 271. nKo methods were devised for construction of the symmetrical acyclic pentayne 31 (Fig. 9-10) [22]. One started at the center and built outward, adding two acetylene units at a time, while the other took advantage of the availability of tetrayne 38, an intermediate from Fig. 9-3 used in our first synthesis of pericyclynes. H
H
a,b Me
1
Me Me
Me
$.
36
37
a,b TMS
Me
Me
M 8 Me
38 Me
Me
Me
31
Figure 9-10 ’ h o ways to assemble the symmetrical pentayne 31: (a) TMS - C C - MgBr, CuCI; (b) KOH,MeOH. 32% for 36 37; 36% for 38 31; (c) EtMgBr, then CuCI, then 8, 39% for 37 + 31. +
+
As an alternative route to [5]pericyclynone 34, we also prepared octamethyl[5]pericyclyne, 41 [8], and oxidized it to the corresponding ketone (Fig. 9-11) [22]. This approach followed the earlier route outlined in Fig. 9-3, deviating only by the omission of methyl groups on the final propargylic alkylating agent. As expected, all the pericyclynes with one or more hydrogen atoms at a doubly-propargylic position (22, 23, 32, 33, and 41) exhibited marked sensitivity to air and base, and this prevented detailed examination of their electronic properties by UV or photoelectron spectroscopy. The ketones and quinones (24, 25, 34, and 35), however, proved relatively easy to
330
9 Macrocyclic Homoconjugated Polyacetylenes
handle [22]. Each gave rise to two UV absorption maxima around 239 and 250 nm, as do acyclic diethynyl ketones; however, [4]pericyclynone, 24, gave two “new/extra” bands of even greater intensity at 224 and 228 nm. Though it is premature to draw firm conclusions about the origins of these extra bands, we suspect that enhanced homoconjugation must play some contributing role. Presumably, a [3]pericyclyne would show the greatest effects, but all attempts in our laboratory to access the cyclonoatriyne ring system, 1, have been consistently thwarted [1,8].
a Me0
39
Me0 Me Me
40
41
34
Figure 9-11An alternative synthesis of [Slpericyclynone: (a) EtMgBr, then CuCI, then BrCH,C = CTMS, 60%; (b) CH,COCl. SnCI,, 90%; (c) AlCI,,CS2,26%; (d) (CH,),COOH, CrO,, p-TsOH, 63%.
9.3 “Exploded” Pericyclynes To relieve some of the angle strain at the saturated vertices of the smallest pericyclynes, we decided to explore the “big brother” compounds with 1,3-diyne units along each side in place of the shorter alkyne linkages. In the same sense that “jpericyclynes can be thought of as “exploded” cycloalkanes with acetylene spacers inserted between the corners [9], the new family with 1,3-diyne spacers can be thought of as “exploded” pericyclynes.
331
9.3 ”Exploded ” Pericyclynes
Besides doubling the number of sp-hybridized carbon atoms over which to spread the strain, this structural modification opened the door to completely different synthesis strategies. Furthermore, and perhaps equally importantly, it provided a convenient UV chromophore to use as a reporter for the similarities and differences in electronic properties among members of the family. The most obvious approach to such compounds is the one we call the “shotgun” synthesis [18, 221, in which a symmetrical, difunctional monomer is subjected to coupling conditions in the hope that some fraction of the growing chains will cyclize in competition with further chain elongation (Fig. 9-12) [30]. In fact, this approach actually works, albeit to only a modest degree.
a
____)
H
H 26
A2
+
+
...
43 44 Figure 9-12 “Shotgun” synthesis of “exploded” pericyclynes: (a) CuCl,, CuCI, pyridine, 4.2% of 44 isolated; lesser amounts of 42 and 43 identified only by G U M S .
Exposure of 3J-dimethylpenta-l,rl-diyne, 26, to oxidative coupling conditions gave a product mixture that contained, in addition to copious amounts of high-molecular-weight material, the first three members of the family, 42,43, and 44, as revealed by gas chromatography/mass spectral (GUMS) analysis [18, 221. In all probability, higher cyclic oligomers are also formed, but we know from an independent synthesis of the cyclic hexamer and octamer (see below) that they fail to vaporize sufficiently for GC-analysis. Laborious column chromatography of the crude product from the shotgun synthesis gave only the five-sided macrocycle, 44, and that was isolated in just 4.2% yield. The same cyclic pentamer was prepared much more efficiently by the cross-coupling of a preformed trimer, 45, with a preformed dimer, 47 [22, 311. This was accomplished by separately converting the trimer to the corresponding bis-cuprate, 46, and the dimer to the bisbromoalkyne, 48, and then mixing the two in pyridine under dilute conditions (Fig. 9-13). In this way, the five-sided macrocycle, 44,could be isolated in the greatly improved yield of 54%.
332
9 Macrocyclic Homoconjugated Polyacetylenes
45
t-
46
44
47
40
Figure 9-13 A two-component route to an “exploded” [5]pericyclyne: (a) n-BuLi, THF, then CuCI; (b) n-BuLi, THF, then TsBr, 69%; (c) combine in pyridine, 54%.
a
cu 26
Me
51 R = TMS
TMS 30
45R=H
Br
TMS 50
R e
- Me
Me
52
=
TMs2
47R=H
Figure 9-14 Syntheses of the acyclic dimer and trimer of 3,3-dimethyl-1,4-pentadiyne:(a) n-BuLi, THF, then CuC1; (b) n-BuLi, THF, then TsBr, 76%; (c) combine in pyridine, 53%; (d) KOH, MeOH, 87% for 51 -+ 4 5 ; 88% for 52 47; (e) n-BuLi, THF, then CuCI, then 50, 86%. +
9.3 ”Exploded” Pericyclynes
333
Stepwise preparations of the acyclic dimer and trimer of 3,3-dimethylpenta-I,4-diyne are outlined in Fig. 9-14 [22, 311. These two compounds occupied pivotal positions in many of our synthetic routes to macrocycles containing 1J-diyne units. Oxidative cyclization of acyclic trimer 45 under high dilution conditions gave us the first rational synthesis of the “exploded” [3]pericyclyne, 42, which was isolated in 39% yield (Fig. 9-15) [18]. Varying amounts (up to 8%) of the cyclic hemmer, 54, were also obtained
r
b. c
45
42 t
II II
II II
1
334
9 Macrocyclic Homoconjugated Polyacetylenes
under certain conditions [22]. Acyclic trimer 45 could also be elongated at both ends by familiar technology to make acyclic pentamer 53, the oxidative cyclization of which gave the 25-membered ring macrocycle, 44, in the amazingly good isolated yield of 89% [18]! The contrast between the success of this intramolecular macrocyclization (89%) vs. the two-component route (54%) vs. the five-component route (4.2 To) bears noting. The same strategy also led to successful syntheses of “exploded” [4]-, [6]-, and [8]pericyclynes (43, 54, and 55, Fig. 9-16) [18, 221. Thus, elongation of the acyclic dimer, 47, gave acyclic tetramer 56, which on oxidative cyclization gave both the cyclic tetramer, 43 (5 Vo), and the cyclic octamer, 55 (2.3 To). Repeated attempts to improve the yield of this cyclization went unrewarded, and we do not understand why it should be so inferior to the cyclizations leading to “exploded” [3]- and [5]pericyclynes; the parallel to problems frequently encountered in the closing of cyclobutane rings is difficult to ignore. Fortunately, another cycle of elongation converted the acyclic tetramer, 56, to the acyclic hexamer, 57, which underwent smooth oxidative cyclization to the “exploded” [6]pericyclyne, 54, in 32% yield 118, 221. An X-ray crystal structure of the cyclic trimer, 42 [22, 321 shows a bending at the acetylenic carbon atoms to 169” and a compression of the endocyclic bond angle at the tetrahedral carbon atoms to 103’. UV absorption spectra of the “exploded” pericyclynes were virtually all superimposable, with a long-wavelength absorption maximum at 259 nm, except for that of the “exploded” [3]pericyclyne,42, fur which the lung-wavelength absorption came at 277 nm [18]. Bending a 1,3-diyneto the degree observed in 42 is known to have an insignificant effect on the UV absorption maximum [18], so we attribute this large bathochromic shift largely, if not entirely, to the effects of enhanced cyclic homoconjugation, principally as a result of stronger through-space interactions of the in-plane p-orbitals. In an attempt to enhance the through-bond interactions of the out-of-plane p-orbitals, we embarked on a collaborative venture with Professor A. de Meijere to synthesize “exploded” pericyclynes that were spirocyclopropanated at every corner. Fig. 9-17 shows how the six-sided and the nine-sided members of this family were prepared [24]. Noteworthy here are (1) the direct formation of a bromoalkyne, 60, from a 2,2-dibromovinyl group; (2) the desilylation of a TMS-protected acetylene in the presence of a bromoalkyne; and (3) the absence of the three-sided compound in the cyclization (contrast with Fig. 9-15). An X-ray crystal structure of the 18-membered ring compound, 63 [%I, shows that it adopts a giant chair conformation not unlike that of cyclohexane, albeit slightly less puckered, with essentially linear diynes and internal angles of 115.5’ at the corners. Unlike the permethylated derivatives of “exploded” pericyclynes, crystals of the perspirocyclopropanated derivative, 63, explode when struck [33]. The most dramatic change in the UV absorption spectrum on going from a simple reference compound, 1,4-dicyclopropylbutadiyne, to the homoconjugated macrocycles 63 and 64 appears not in the long-wavelength region but at short wavelength, between 200 and 220 nm, where 1,4-dicyclopropylbutadiyneshows only end absorption; in this region, macrocycles 63 and 64 show striking new bands [24]. The acyclic trimer, 62, in which electronic interactions among the diynes should also be expected, likewise shows one new band at 210 nm; however, the new bands in the spectra of macrocycles 63 and 64 are characterized by further splitting and enormous amplitudes (E = 250000-450000). Clearly, some unusual electronic effects are associated with the cyclic alternation of 1,3-diynes and spirocyclopropanes. No such effect is seen in macrocycle 66 (Fig. 9-18), where two spirocyclopropanes separate the 1,3-diyne units [201.
9.3 "Exploded" Pericyclynes
335
Me 47
54
Figure 9-16 Oxidative cyclization routes to "exploded" [4]-, 161-, and [Ilpericyclynes: (a) n-BuLi, THE then CuCI, then 50; (b) KOH, MeOH, 73% for 47 56; 46% for 56 57; (c) CuCl, Cu(OAc)z, pyridine, 5% 43, 2.3% 55; (d) CuCl, CuCI,, pyridine, 32%. +
+
336
9 Macrocyclic Homoconjugated Polyacetylenes
TMS
TMS
TMS
Br 59
58
Br 60
A
H
Br
61
f
II
II
II
I1
.___)
62
63 (n = 6) 64 (n = 9)
Figure 9-17 The first perspirocyclopropanated “exploded” pericyclynes: (a) n-BuLi, Et20, then DMF then H,O;(b) Zn, Ph,P, CBr,, 70% for 58 -+ 59; (c) t-BuOK, THF, 95%; (d) KF . 2H20,DMF, 74% for 60 --t 61; 84% 62; (e) 2 equiv n-BuLi, Et20, then CuC1, then combine with 2 equiv 60 in pyridine, 54%; (f) CuCl, Cu(OAc),, pyridine, 39% 63, 8% 64.
a
65
66
Figure 9-18 Separation of the 1,3-diynes by two spirocylcopropanes: (a) CuCl, CuCI,, pyridine, 2.1 %.
9.4 Homoconjugated Mixed Polyalkyne/Diyne Macrocycles
337
9.4 Homoconjugated Mixed Polyalkyne/Diyne Macrocycles With precursors in hand and reactions all worked out for the construction of both pericyclynes and “exploded” pericyclynes, the temptation to prepare hybrid macrocycles containing simple alkynes along some of the edges and 1,3-diynesalong the others eventually became irresistible. Fig. 9-19 shows the route to one such crossbreed, 68, a 23-membered ring with nine triple bonds (cf. Fig. 9-13) [22, 311.
46
Me
Me
37
T Me
Me
Me 68
Me 67
Figure 9-19 A macrocyclic polyacetylene with 4N + 2 electrons in both the out-of-plane and the in-plane x-systems: (a) n-BuLi, THF, then TsBr, 79% or NBS, AgNO,, 90%; (b) combine in pyridine, 39%.
This macrocycle was prepared to determine whether any difference in electronic properties could be detected between homoconjugated macrocycles having 4N + 2 electrons in both the out-of-plane and the in-plane n-systems, as 68 has, and macrocycles having 4N electrons in both the out-of-plane and the in-plane n-systems, as the “exploded” [5]pericyclyne 44 has. The answer, at least as far as UV absorption spectroscopy is concerned, is definitely negative; all five bands in the UV spectrum of 68 are virtually superimposable on those in the UV spectrum 44 (f1.5 nm) [22, 311. Hiickel’s rule does not operate in these systems. Elongation of acyclic dimer 47 at both ends, using the method encountered so frequently in Section 9-2, followed by oxidative cyclization gave the 16-membered ring hexayne 70 in 149’0 yield (Fig. 9-20) [22]. An alternative synthesis of the same macrocycle via a bimolecular coupling between the bis-cuprate of 37 and dibromo compound 67, according to the method in Fig. 9-19, worked in only 2.6% yield [22]. Hexayne 70 is one of the highest melting (m.p. 240 OC with decomposition) and least soluble macrocycles we have encountered in this work. It survives sublimation under vacuum at 160°C and is barely soluble enough in cyclohexane to give a UV spectrum. Unlike most of the other macrocyclic polyacetylenes, it shows poor solubility even in benzene. Unfortuantely, the UV
338
9 Macrocyclic Homoconjugated Polyacetylenes
47
69
70
Figure 9-20 A homoconjugated macrocycle with alternating alkyne and diyne units: (a) EtMgBr, then CuC1, then 8; (b) KOH, MeOH, 56% for 47 --t 69; (c) Cu(OAc),, 6 : 1 pyridine/ether, 14%.
absorption spectrum of the 16-membered ring, 70,is disappointingly similar to those of the unstrained "exploded" pericyclynes 1221; apparently, the internal bond angles at the vertices in such molecules must be reduced even more to cause a significant shift in the UV absorption spectrum, as we saw for the 15-membered ring polyacetylene, 42 (and for the 14-membered ring pentayne, 71, described next). The final pair of compounds in this section provides a nice illustration of how structural features in a molecule can be designed to enhance either the through-space or the throughbond components of homoconjugation. Fig. 9-21 shows the oxidative cyclization routes to two compounds of interest, 71 and 73 [25]. Despite the obvious strain in these two 14-membered ring pentaynes, the yields in the macrocyclizations are surprisingly good: 67% in the case of 71 and 45% in the case of 73. Precursor 31 has been seen before in Fig. 9-10, and precursor 72 was built up from dicyclopropylacetyleneas shown in Fig. 9-21. Of special note are the two different methods used for turning cyclopropane carbon atoms into quaternary centers. An X-ray structural analysis of the permethylated macrocycle, 71 [25] shows a geometry in the vicinity of the 1,3-diyne unit that is nearly identical to that in the "exploded" [3]pericyclyne, 42,i. e., compression of the internal C - C - C bond angles to 103- 104" at the saturated carbon atoms flanking the diyne in both cases and comparable outward bowing of the 1,3-diynes (168-169" C - C - C bond angles in both cases). A bathochromic shift is likewise seen in the long-wavelength absorption maximum of 71,although it is only half as large as that seen in the UV absorption spectrum of 42, relative to the long-wavelength absorption maxima of the less strained homoconjugated cyclic diynes. We ascribe the bathochromic shifts in the spectra of both 71 and 42 to an enhancement of homoconjugation resulting from stronger through-space interactions of the in-plane p-orbitals; the lesser effect in 71 presumably reflects the poorer match of orbital energies when a diyne interacts with a simple alkyne. In the perspirocyclopropanated 14-membered ring pentayne, 73,the internal C - C - C bond angles at the corners are no longer so small (109" by X-ray analysis) [25], yet the longwavelength absorption maximum is shifted even further ro the red than it is in the spectrum of 71. Here it is the cyclopropanes between the diynes that cause the shift (through-bond orbital interactions); similar bathochromic shifts, though lower in magnitude, are seen in the UV
9.4 Homoconjugated Mixed Polyalkyne/Diyne Macrocycles
339
absorption spectra even of acyclic 1,3-diynes that are “conjugated” with adjacent cyclopropanes [24, 251. Thus, spirocyclopropanes serve as “conjugating” links between diynes connected to the same carbon atom. Still stronger interactions would be expected, of course, through a sulfur atom or some other heteroatom with appropriate orbitals, and that is the subject of the next section.
Li+
.f
tc 77
76
Figure 9-21 Two 14-membered ring pentaynes: (a) CuCl, Cu(OAc),, pyridine, 67% 31 71; 45% for 72 + 73; (b) t-BuLi, TMEDA, pentane then DMF, 38%; (c) Zn,Ph3P, CBr,, 99%; (d) t-BuOK, THF, go%, (e) combine in Et,O, THF, 63%; ( f ) NaOH, MeOH, 96%. +
340
9 Macrocyclic Homoconjugated Polyacetylenes
9.5 Heterocyclic Cognates of Pericyclynes Many of our syntheses of "jpericyclynes containing heteroatoms in the cycle have followed closely the strategy developed for preparing pericyclynones and quinones (cf. Fig. 9-7 and 9-9). For example, the ends of pentayne 31 could be joined to the same sulfur atom to make a mono-thia[5]pericyclyne (78, Fig. 9-22) [22].
Me
a
___)
"'Me Me
Me 78
31
Figure 9-22 A mono-thia[S]pericyclyne: (a) n-BuLi, then (PhSO,),S, 21 To.
csi] 21
Me2
Me!,. Me
Me
Me
MeW" Me
*
Me
'WMe
si
Me
81
Me2
83 82
Figure 9-23 Heteroatom derivatives of [4]-and [Ilpericyclyne: (a) n-BuLi then (PhS0J2S, 14% 79, 2 % 80; (b) n-BuLi then t-BuPCI,, 1.3% 81, 2.5% 82; (c) n-BuLi, then Me,SiCl,, 6%.
9.5 Heterocyclic Cognates of Pericyclynes
341
With the lower homolog of 31, thia[4lpericyclyne 79 was formed as the major product, but a small amount of cyclic dimer 80 was also isolated (Fig. 9-23) [22]. The same outcome was seen when the ends were joined to a phosphorus atom (Fig. 9-23) [22]; however, with dichlorodimethylsilane as the cyclization agent, only the cyclic dimer was formed (83, Fig. 9-23) [22]. In general, the yields of these [4]pericyclynesand their cyclic dimers were rather low ( < 15%). Attempts to make a [3]pericyclyne by this strategy, starting from the symmetrical acyclic triyne 37 (from Fig. 9-10), failed in every case, which is not surprising, but they did not even give cyclic dimers, except for the disila[6]pericyclyne 84, which could be isolated in only 0.5 Vo yield (Fig. 9-24) [22]. It is not clear to us why 84 and the other [6]pericyclynes having two heteroatoms at opposite corners should be so difficult to form under these conditions.
a
H 37 84
Figure 9-24 A disila[6]pericyclyne: (a) n-BuLi, then MqSiCl,, 0.5 To.
From the dianion of 3,3-dimethylpenta-l,Cdiyne,we obtained [4]pericyclynesand [6]pericyclynes bearing heteroatoms at alternate corners, both with sulfur and with phosphorus (Fig. 9-25) [22] ;however, silicon gave the [6lpericyclyne, 89, and the [8]pericyclyne,90, accompanied by none of the [4]pericyclyne (Fig. 9-25) [22]. We have no rational explanation for the change in behavior observed in going from sulfur and phosphorus to silicon. A closer look at these compounds reveals some special properties of the phosphorus heterocycles. Owing to the high energy barrier for pyramidal inversion at phosphorus atoms, the di-t-diphospha[4]pericyclyne, 87, exists as two, noninterconverting diastereoisomers, and these were separable by careful preparative thin layer chromatography. The tri-t-butyl-tnphospha[6]pericyclyne, 88, likewise exists as two, noninterconverting diastereoisomers, and these were also separated, both from the [4]pericyclynes and from each other. We presume that the previously mentioned tetraphospha[8]pericyclyne, 82, consists of a mixture of diastereoisomers, but no attempt was made to separate them. All of these phospha[N]pericyclynes suffer rapid oxidation to phosphine oxides on handling in the air. Fig. 9-26 outlines stepwise syntheses of both the dithia[4]pericyclyne, 85, first prepared in Fig. 9-25, and the previously elusive dithia[6]pericyclyne, 93, which was not accessible from triyne 37 (cf. Fig. 9-24) [22]. As expected, the yield of the cyclization to dithia[4]pericyclyne 85, but not the total yield, improved significantly (from 6.5 To to 11Yo) when more of the ring atoms started out already joined together (contrast Fig. 9-25 and 9-26). The yield of the dithia[6]pericyclyne,93, however, was still unexpectedly low (2.4%). Once formed, 93 appears to be perfectly normal (m.p. 225OC, dec), but the cyclization step itself seems to be mysteriously unfavorable.
342
9 Macrocyclic Homoconjugated Polyacetylenes
-
Me
b
+
i+ Me Me 86
MrMep> C
Me
+
___)
26
a7 88
>M si(
Me2
-
Me
Me
d
+
MepSi
SiMep
89 90
Figure 9-25 Rricyclynes of order [4], [6], and [8]with heteroatoms at every other corner: (a) n-BuLi; @) (PhSO&,S, 6.5% 85, 9% 86;(c) t-BuPCI,, 3.5% 87, 1.5% 88; (d) Me,SiCl,, 2% 89, 17% 90.
From sulfide 91, we were able to make [4]pericyclynes and [I]pericyclyneswith two different heteroatoms in the rings (Fig. 9-27) [22]. To our surprise, however, the [4]pericyclyne with one phosphorus and one sulfur, 94, showed no tendency to air-oxidize, even over a period of one year; such behavior stands in sharp contrast to that of all the other phospha[Mpericyclynes mentioned above. Silicon, on the other hand, showed the same reluctance we had seen before (cf. Fig. 9-23 and 9-25) to form [4]pericyclynes.
9.5 Heterocyclic Cognates of Pericyclynes
30
91
85
92
93
343
Figure 9-26 Dithia[4]- and dithia[6]pericyclynes: (a) n-BuLi, then (PhS02)2S, 2.3% for 92 -, 93; (b) KOH, MeOH, 76% for 30 91; 21% for 91 92; (c) EtMgBr, then CuCI, then 2 equiv 8; (d) LiNFMS),, then (PhSO&S, 11%. -+
+
95
Figure 9-27 Mixed heteroatom derivatives of [4]- and [8]-pericyclyne: (a) n-BuLi; (b) t-BuPCI,, 5.6%; (c) Me2SiCI2,5%.
344
9 Macrocyclic Homoconjugated Polyacetylenes
One reason for the low yields in these reactions was eventually traced to the attack by butyllithium at the sulfur center in competition with deprotonation of the terminal acetylenes. By using the more hindered mesityllithium, we were able to improve the yield of the 24-membered ring, mixed heterocyclic 95 from 5 % (butyllithium as base) to 24% (mesityllithium as base) [22]. The earlier reactions outlined in Fig. 9-26 that involve deprotonation of sulfur-containing substrates by butyllithium have not been reexamined using mesityllithium as the base. Voronkov [34] and Sakurai [35] reported the first pericyclynes with silicon atoms at every corner, in large rings and small ones, respectively, and others have developed the chemistry further [36-381. We have had good luck with the phosphorus counterparts (but not sulfur, despite considerable effort [23]). Our work in this direction was motivated by the fact that, unlike carbon and silicon, phosphorus and sulfur have preferred bond angles in the 90-100" range, and this should relieve significant angle strain in the smallest pericyclynes. In fact, we found that the tetraphospha[4]pericyclynering system self-assembles with relatively little difficulty (Fig. 9-28) [19]. Even the corresponding [3]pericyclyne can be made, with phosphorus at every corner (Fig. 9-28).
96
98
99
Figore 9-28 N-phospha[N]pericyclynes: (a) EtMgBr; (b) combine, 11 O7o 97, 1-16'70 99.
X-ray crystal structures of both 97 (all-tmns isomer) and 99 (tmns,cis,frunsisomer) have been obtained [39], and they confirm the relief of angle strain at the acetylenic carbon atoms that the phosphorus atoms provide in these small-ring pericyclynes (actual geometries shown in Fig. 9-28: average phosphorus atom endocyclic bond angles = 96" in 97 and 91" in 99; average acetylene carbon bond angles = 174" in 97 and 163" in 99). The [4]pericyclyne, 97, has the potential for existing in four diastereoisomeric forms, and all four are generated in the synthesis, though not in equal abundance [19]. By careful chromatography, we were able to separate and assign unambiguous structures (by 'H-, I3C-, and
9.6 Experimental Procedures
345
31P-NMR)to each diastereoisomer [40]. The [3]pericyclyne, 99, has the potential for existing in two diastereoisomeric forms, but we see only the trans,cis,frans isomer. The UV absorption spectra of 97 and 99 show multiple bands with tails extending out almost to 300 nm [19]. Though it is difficult to say how much electronic interaction among the acetylenic units occurs via orbitals on the phosphorus atoms and how much occurs via through-space overlap of the in-plane p-orbitals, there can be little doubt that these N-phospha[Nlpericyclynes enjoy a substantial degree of cyclic electron delocalization. The real appeal of the phosphapericyclynes to us, however, was their potential for building into the third dimension, i. e., homoconjugated cage structures with phosphorus atoms at the corners of a polyhedron and acetylenes along every edge. Tetrahedron 100, for example, represents an “exploded” analog of elemental phosphorus, each face of which corresponds to the now-known ring system of 99. The cube, 101, should have even less angle strain, and both should exhibit electron delocalization across the entire surface of the polyhedron, much like that seen in fullerenes. As a more modest intial synthetic target, we selected the bicyclic ring system of 102. In this connection, we were delighted to find that deprotonation of the diethynylphosphine 96 with one equivalent of ethylmagnesium bromide gave quite selectively the monoanion, 103, which could be trapped with PCl, to give the three-armed hexayne 104 in 71 Yo yield (Fig. 9-29) [41]. Triple deprotonation of 104 with ethylmagnesium bromide followed by quenching with PCl, then gave us our first fully homoconjugated cage molecule, but it turned out not to have the bicyclic structure of 102. Quenching the same trianion of 104 with POC1, also gave a cage molecule, the structure of which was deduced from spectroscopic evidence and subsequently confirmed by X-ray crystallography [42] to be that of a 2 : 2 adduct, 107, rather than that of the 1 :1 adduct 106 (Fig. 9-29). Spectroscopic data for the cage formed with PC1, points to an analogous “cyclophane” structure in that case as well, but X-ray quality crystals have been elusive. We presume that the transformations in the final reaction proceed according to the original plan up to the last step (104 + 109,but the last ring closure apparently does not compete successfully with intermolecular coupling that leads ultimately to the observed tricyclic cage (Fig. 9-29; geometry of 107 taken from the X-ray coordinates). Nevertheless, the first steps into the third dimension of pericyclyne chemistry have now been taken, and the electronic properties of many marvelous molecules await discovery.
9.6 Experimental Procedures 9.6.1 Conversion of a Methyl Ketone to a Terminal Acetylene (28
+
30, Fig. 9-8)
Under a nitrogen atmosphere, 22.9 g (226 mmol) of diisopropylamine in 500 mL of dry THF was cooled to 0 “C, and 94 mL (226 mmol) of 2.5 M n-butyllithium in hexane was added. The mixture was stirred for 30 min at OOC, followed by cooling to -78°C. Next, 38.2 g (209 mmol) of ketone 28 in 35 mL of THF was added over a 25 min period. A white solid precipitated out of solution after about two-thirds of the ketone solution had been added. The mixture was stirred for 1 h at -78 “C, then 37.9 g (220 mmol) of diethyl chlorophosphate was added. The solution of en01 phosphate was allowed to warm to room temperature while a se-
346
9 Macrocyclic Homoconjugated Polyacetylenes
101
100
102
H P -,
’
F
a ___f
) p\
ErMg-
H
L
b
H 103
96
104
MgBr
-
105
intra
107
Figure 9-29 Fully homoconjugated cages; the tricyclic “cyclophane” 107 has been isolated and characterized by X-ray crystallography: (a) 1 equiv EtMgBr; (b) 0.33 equiv PCl,, 71%; (c) 3 equiv EtMgBr; (d) POCI,, 14%.
9.6 Experimental Procedures
347
cond solution of lithium diisopropylamide (LDA) was prepared, as above, from 196 mL (470 mmol) of 2.5 M n-butyllithium in hexane and a solution of 47.6 g (470 mmol) of diisopropylamine in 390 mL of THE After being stirred for 30 rnin at 0 "C, the LDA solution was cooled to -78 "C. The enol phosphate solution was transferred to a pressure-equalizing dropping funnel with a cannula under nitrogen pressure and then added to the cold LDA solution over a 90 rnin period. The reddish-colored reaction mixture was stirred at - 78 "C for 30 min. It was allowed to warm to room temperature, and the reaction was quenched by the addition of 200 mL of water. The layers were separated, and the aqueous phase was extracted with 3 x 100 mL of pentane. The combined extracts were washed with 200 mL of 10% aqueous hydrochloric acid and with 200 mL of saturated aqueous sodium bicarbonate, dried over magnesium sulfate and concentrated under reduced pressure. Fractional distillation gave 21.7 g (63070) of monosilylated 3,3-dimethylpenta-l,4-diyne(30) as a colorless liquid: bp 53 OW20 torr.
9.6.2 Conversion of a Terminal Acetylene to a Bromoalkyne Using Tosyl Bromide (30 + 50, Fig. 9-14) To 4.24 g (25.8mmol) of monosilylated diyne 30 in 110 mL of dry THF was added dropwise, under a nitrogen atmosphere, 11 mL (28 mmol) of 2.5 M n-butyllithium in hexane at - 78 "C. The mixture was stirred for 30 min, then 7.12 g (30.3 mmol) of p-toluenesulfonyl bromide in 30 mL of THF was added over a 10 min period. The solution was stirred for 30 min at -78 "C and at room temperature for 1 h. Next, 150 mL of water was added. The layers were separated, and the aqueous layer was extracted with 3 x 50 mL of pentane. The combined extracts were washed with 150 mL each of 10% aqueous sulfuric acid, saturated aqueous sodium bicarbonate, and water, dried over magnesium sulfate, and concentrated under reduced pressure. Column chromatography on silica gel with 10: 1 hexane/ethyl acetate gave 4.90 g (76%) of bromoalkyne 50 as a colorless liquid that was 90% pure by GC (contaminants were not characterized). It is not advisable to purify the product by distillation due to the possibility of explosive decomposition - CAUTION [44]!
9.6.3 Preparation of a 1,3-Diyne by Cross-Coupling of a Preformed Copper Acetylide with a Bromoalkyne - 2: 1 Example (49 + 50 + 51, Fig. 9-14) Under a nitrogen atmosphere, 2.11 g (19.7mmol) of 3,3-dimethylpenta-lP-diyne (26) was dissolved in 250 mL of dry T H E The solution was cooled to O"C,and 16.4 mL (41 mmol) of 2.5 M n-butyllithium in hexane was added dropwise. The mixture was stirred for 30 min, and then 4.06 g (41 mmol) of cuprous chloride was added. After stirring for 1 h at room temperature, the solvent was removed (not to dryness; dry copper acetylides are explosive CAUTION!) with reduced pressure and replaced with 650 mL of dry, oxygen-free pyridine. Next, 10.06 g (41.4mmol) of bromoalkyne 50 in 35 mL of THF was added with a syringe pump over 6 h at room temperature. Stirring was continued overnight at room temperature. The dark green solution was poured slowly into 600 mL of ice-cold 25% aqueous hydrochloric acid, and the aqueous mixture was extracted with 2 x 300 mL of pentane. The
348
9 Macrocyclic Homoconjugated Polyacetylenes
combined organic phases were washed with 300 mL of 10% aqueous hydrochloric acid and 300 mL of saturated aqueous sodium bicarbonate, dried over magnesium sulfate, and concentrated under reduced pressure. Recrystallization from absolute ethanol gave 4.8 g (53 Yo) of a mixture that by G U M S analysis was 90% of the desired 2 : 1 cross-coupling product (51) and 9% of 52, the self-coupling product of the bromoalkyne; self-coupling of bromoalkynes is a common side-reaction and difficult to suppress completely in these cross-couplings. Further recrystallizations produced pure silylated open trimer 51 that was suitable for spectral and physical characterization: mp 111-114°C.
9.6.4 Oxidative Cyclization of a Long-Chain a,o-Diyne (53
4
44, Fig. 9-15) [45]
Under nitrogen, 150 mg (0.33 mmol) of acyclic decayne 53 in 30 mL of dry pyridine was added with a syringe pump to a mixture of 1.48 g (15 mmol) of cuprous chloride and 0.87 g (6.5 mmol) of cupric chloride in 100 mL of dry pyridine over 60 h at room temperature. Stirring was continued for 30 h after completion of the addition. The reaction was quenched by the cautious addition of 225 mL of 25% aqueous hydrochloric acid to the cooled (0°C) reaction mixture. The layers were separated, and the aqueous layer was extracted with 3 x 250 mL of pentane. The combined extracts were washed with 200 mL of 10% aqueous hydrochloric acid, 200 mL of saturated aqueous sodium bicarbonate and 200 mL of water, dried over magnesium sulfate, and concentrated under reduced pressure. Purification by column chromatography on silica gel using 3:2 hexanekoluene as eluent gave 133 mg (89%) of the 25-membered ring decayne 44 as white crystals: mp 170°C (dec).
9.6.5 Coupling a Terminal Acetylene with a Tertiary Propargylic Chloride Example (47 69, Fig. 9-20)
- 2: 1
-+
Under a nitrogen atmosphere, a solution of 1.9 g (10.4 mmol) of 3,3,8,8-tetramethyl-1,4,6,9decatetrayne 47 in 5 mL of dry THF was added dropwise to 11 mL (22 mmol) of 2.0 M ethylmagnesium bromide in THF at 0°C. The mixture was warmed to 50-55 "C for 30 min to ensure complete deprotonation, cooled back to room temperature, and then 50 mg (0.51 mmol) of cuprous chloride was added. Next, a solution of 2.45 g (6.6 mmol) of 3-chloro-3-methyl-l-(trimethylsilyl)-l-butyne 8 in 5 mL of THF was added. The mixture was stirred at 50-60°C for 2.5 hours and at room temperature overnight. The reaction mixture was quenched with 20 mL of 10% aqueous hydrochloric acid, and the layers were separated. The aqueous layer was extracted with 2 x 50 mL of pentane. The combined organic layers were washed with 2 x 75 mL each of saturated sodium bicarbonate solution and 10% aqueous ammonium chloride, dried over magnesium sulfate and concentrated under reduced pressure. Column chromatography on silica gel using 10: 1 hexane/ethyl acetate as eluent followed by recrystallization from absolute ethanol gave 2.93 g (61 070) of the bis(trimethylsily1) derivative of hexayne 69 as white crystals: mp 86-87 "C. Removal of the two trimethylsilyl groups was accomplished as follows. To 15 mL of ice-cold methanol was added 2.67 g (5.82 mmol) of the bis(trimethylsily1)hexayne and 1.5 g (22.7 mmol) of 85 Vo potassium hydroxide. The mixture was warmed to room temperature and stirred overnight. The reaction was quenched by the
References
349
addition of 50 mL of water, and the aqueous mixture was extracted with 3 x 100 mL of pentane. The combined extracts were washed with 100 mL of water, dried over magnesium sulfate and concentrated under reduced pressure. Recrystallization of the crude product from absolute ethanol gave 1.67 g (91 Yo) of desilylated hexayne 69 as pale yellow crystals: 68-70°C.
9.6.6 Conversion of a 2,2-dibromovinyl Compound to a Bromoalkyne Two-Fold Example (75 -+ 74, Fig. 9-21)
-
A solution of 10 g (21 mmol) of tetrabromide 75 in 400 mL of anhydrous THF was treated with 9.45 g (84 mmol) of potassium t-butoxide at - 78 "C. The mixture was stirred for 5 h at the same temperature, followed by hydrolysis with 200 mL of water. The aqueous phase was extracted with 3 x 150 mL of hexane. The combined organic layers were washed with 100 mL of 1 M hydrochloric acid and 100 mL of 5 % aqueous sodium bicarbonate, and dried over magnesium sulfate. The solvent was removed under reduced pressure to give 5.9 g (90%) of bis-bromoalkyne 74 as pale yellow crystals that were suitable for use without further purification - CAUTION [MI!
Acknowledgements Financial support from the National Science Foundation and NATO is gratefully acknowledged.
Abbreviations DMF LDA LUMO NBS THF TMS Ts
dimethylformamide lithium diisopropylamide lowest unoccupied molecular orbital N-bromosuccinimide tetrahydrofuran trimethylsilyl tosyl
References 111 G. J. DeCicco, Ph. D. Dissertation, UCLA, 1977. [2]C. Santiago, K. N. Houk, G. J. DeCicco, L. T. Scott, .lAm. Chem. Soc. 1978, ZOO, 692-696. [3] L. T. Scott, W. R. Brunsvold, J. Am. Chem. Soc. 1978, 100, 4320-4321. [4]L. T. Scott, W. R. Brunsvold, M. A. Kirms, I. Erden, Angew. Chem. 1981, 20, 282-283;Angew. Chem., Int. Ed. Engl. 1981, 20, 214-279. [5] L. T. Scott, W. R. Brunsvold, M. A. Kirms, I. Erden, .l Am. Chem. SOC. 1981, 103, 5216-5220. [6]J. L. Hyun, M. S. Thesis, University of Nevada, Reno, 1982.
350
9 Macrocyclic Homoconjugated Polyacetylenes
[7] L. T. Scott, G. J. DeCicco, J. L. Hyun, G. Reinhardt, J. Am. Chem. SOC.1983, 105, 7760-7761. [8] L. T. Scott, G. J. DeCicco, J. L. Hyun, G. Reinhardt, J. Am. Chem. SOC. 1985, 107, 6546-6555. 191 K. N. Houk, L. T. Scott, N. G. Rondan, D. C. Spellmeyer, G. Reinhardt, J. L. Hyun, G. J. DeCicco, R. Weiss, M. H. M. Chen, L. S. Bass, J. Clardy, F. S. Jmgensen, T. A. Eaton, V. Sarkozi, C. Petit, L. Ng, K. D. Jordan, J. Am. Chem. SOC. 1985, 107, 6556-6562. [lo] L. T. Scott, M. Oda, I. Erden, 1 Am. Chem. SOC. 1985, 107, 7213-7214. Ill] L. T. Scott, Pure Appl. Chem. 1986, 58, 105-110. [12] M. Giinther, H. von Puttkamer, P. Schmitt, H. Giinther, L. T. Scott, M. A. Kirms, Chem. Ber. 1986, 119, 2942-2955. (131 L. T. Scott, M. Oda, Chem. Lett. 1986, 1759-1762. [I41 F. Gerson, J. Knobel, A. Metzger, L. T. Scott, M. A. Kirms, M. Oda, C. A. Sumpter, J. Am. Chem. SOC. 1986, 108, 7920-7926. [IS] M. J. Cooney, M. S. Thesis, University of Nevada, Reno, 1987. [I61 L. T. Scott, M. J. Cooney, D. W. Rogers, K. Dejroongruang, J. Am. Chem. SOC. 1988, 110, 7244-7245. [I71 L. T. Scott, C. A. Sumpter, M. Oda, 1. Erden, Tetrahedron Lett. 1989, 30, 305-308. [18] L. T. Scott, M. J. Cooney, D. Johnels, 1 Am. Chem. SOC. 1990, 112, 4054-4055. [I91 L. T. Scott, M. Unno, 1 Am. Chem. SOC.1990, 112, 7823-7825. I201 A. de Meijere, F. Jaekel, A. Simon, H. Borrmann, J. Kohler, D. Johnels, L. T. Scott, 1 Am. Chem. SOC. 1991, 113, 3935-3941. [21] L. J. Schaad, B. A. Hess, Jr., L. T. Scott, J. Phys. Org. Chem. 1993, 6, 316-318. 1221 M. J. Cooney, Ph. D. Dissertation, University of Nevada, Reno, 1993. [23] R. M. GonzAlez, Ph. D. Dissertation, University of Nevada, Reno, 1993. [24] A. de Meijere, S. Kozhushkov, C. Puls, T. Haumann, R. Boese, M. J. Cooney, L. T. Scott, Angew. Chem. 1994, 106, 934-936; Angew. Chem., Int. Ed. Engl. 1994,33, 869-871. 1251 (a) L. T. Scott, M. J. Cooney, C. Otte, C. Puls, T. Haumann, R. Boese, P. J. Carroll, A. B. Smith, 111, A. de Meijere, J. Am. Chem. SOC. 1994, 116, 10275-10283; (b) A. de Meijere, S. Kozhushkov, T. Haumann, R. Boese, C. Pub, M. J. Cooney, L. T. Scott, Chem. EUKJ. 1995, 1, 124-131. (261 Intramolecular Friedel-Crafts-like acylation of a trimethylsilyl acetylene had previously been used to prepare unstrained large ring alkynes: K. Ultimoto, M. Tanaka, M. Kitai, H. Nozaki, Btrahedron Lett. 1978, 2301 -2304. [27] L. T. Scott, M. J. Cooney, P. J. Carroll, A. B. Smith, 111, unpublished work. [28] M. J. S. Dewar, M. K. Holloway, J. Chem. SOC,Chem. Commun. 1984, 1188-1191. [29] A similar controversy surrounds the claim of neutral homoaromaticity in triquinacene by Paquette, et al., who also employed an analysis based on heats of hydrogenation data: J. F. Liebman, L. A. Paquette, J. R. Peterson, D. W. Rogers, J. Am. Chem. SOC.1986, 108, 8267-8268. For leading references, see: (a) A. Holder, 1 Comput. Chem. 1993, 14,251-255. (b) J. W. Storer, K. N. Houk, J. Am. Chem. SOC.1992, 114, 1165-1168. (c) J. W. Storer, K. N. Houk, ibid. 1992, 114, 5907-5908. [30] Cyclic-oligomers of a,o-diynes prepared by this same strategy were used extensively by F. Sondheimer et al. as synthetic precursors to the annulenes; for a summary of the early work, see F. Sondheimer, Pure Appl. Chem. 1963, 7, 363-388. [31] L. T. Scott and D. Johnels, unpublished work. [32] L. T. Scott, M. J. Cooney, T. Haumann, R. Boese, unpublished work. [33] For a freeze-frame picture of one such explosion, together with the X-raycrystal structure of 63,see the front cover of the April 18th, 1994 issue of Angewandte Chemie. [34] (a) M. G. Voronkov, S. F. Pavlov, Zh. Obshch. Khim. 1973,43, 1408-1409; Chem. Abstr. 1973, 79, 66448 g. (b) M. G. Voronkov, 0. G. Yarosh, L. V. Zhilitskaya, A. I. Albanov, V. Yu. Vitkovskii, Metalloorg. Khim. 1991, 4, 368-372; Chem. Abstr. 1991, 115, 29446 d. [351 H. Sakurai, Y. Eriyama, A. Hosomi, Y. Nakadaira, C. Kabuto, Chem. Lett. 1984, 595-598. [36] R. Bortolin, B. Parbhoo, S. D. Brown, J. Chem. Soc, Chem. Commun. 1988, 1079-1081.
References
351
[37] E. Hengge, A. Baumegger, J Organomet. Chem. 1989, 369, C39-C42. [38] A. Baumegger, E. Hengge, S. Gamper, E. Hardtweck, R. Janoschek, Monatsh. Chem. 1991, 122, 661-671. [39] L. T. Scott, M. Unno, P. J. Carroll, A. B. Smith, 111, unpublished work. [40] L. T. Scott, M. Unno, M. J. Cooney, D. Wege, unpublished work. [41] L. T. Scott, Y. Aso, unpublished work. [42] L. T. Scott, Y. Aso, S. Johnson, unpublished work. [43] (a) E. I. Negishi, A. 0. King, W. L. Klima, W. Patterson, A. Silveira, Jr., J Org. Chem. 1980, 45, 2526-2528. (b) E. I. Negishi, A. 0. King, J. M. Tour in Org. Synth. 1985, 64, 44-49. [44] CAUTION! Copper acetylides and some bromoalkynes are potentially explosive when dry; it is safest never to remove solvent completely. [45] D. O’Krongly, S. R. Denmeade, M. Y. Chiang, R. Breslow, J Am. Chem. SOC.1985, 107, 5544-5545.
This Page Intentionally Left Blank
10 Polyacetylene Eric .l Ginsburg, Christopher B. Gorman, Robert H. Grubbs
10.1 Introduction Few organic molecules with a structure seemingly as simple as polyacetylene, (CH),, have been the subject of so much scientific attention. Since the 1970s, when Shirakawa and coworkers reported a facile route to films of the material [l] and then went on to discover a lO”-fold increase in electrical conductivity upon oxidation [2, 31 many physicists, materials scientists, and chemists from both academia and industry have been drawn to the field of conjugated polymers. Their goals include both understanding the properties of polyacetylene and exploring new routes to the polymer or its derivatives. Early work in the field was driven by a vision of using doped polymers as lightweight, processable “plastic metals”; as a consequence, much work has been focused on obtaining increased electrical conductivity. More recently, a wide range of other potential uses for polyacetylene, its derivatives, and other conjugated polymers has arisen. Examples include nonlinear optical waveguides [4, 51, light-emitting diodes [6, 71, gas separation membranes [8- 101 chiral separation membranes [ll], and cell growth media [12]. Furthermore, the structural and electronic changes undergone by conjugated polymers upon oxidation or reduction make them potentially “smart” or “adaptive” materials [13]. It should be pointed out that, despite this attention, 17 years after Shirakawa’s discovery polyacetylene is not yet a commercial polymer, and many early research efforts at industrial laboratories have been discontinued. In part, progress has been hampered by the fact that polyacetylene, and most other unsubstituted conjugated polymers, can be neither dissolved nor melted. In addition, polyacetylene is unstable in air, complicating its incorporation into products. However, the diversity of potential uses for conjugated polymers, combined with the experimental challenges of preparing more tractable materials, has kept the field vibrant. Although the relationships between polyacetylene structure of properties have become better understood, challenges remain here, too. Many of the chemist’s typical analytical techniques are not readily applied to a material which may only be examined in a distordered solid state. For instance, its insolubility makes even a determination of molecular weight difficult. Before precursor routes were developed (see below), the most reliable measure of polyacetylene molecular weight came from radioactive end-group labeling [14]. The material is partially crystalline, allowing interchain separations to be quantified for both cis and trans polyacetylene in crystalline regions using electron and X-ray diffraction techniques. However, in part because these investigations are frequently performed on stretch-aligned samples to maximize crystallinity, and because techniques for stretching a sample may differ, there are some disparities in the literature [15]. Solid-state ”C-NMR is routinely used to determine the overall “structural purity” of samples by assaying the number of sp’ C-atoms, but sensitivity is limited, and the information is spatially averaged. At a more macroscopic level, scanning electron microscopy is used to readily determine the gross morphology of a sample, e. g., whether it is fibrous, smooth, or globular. While morphology has an effect on the bulk pro-
354
10 Polyacetylene
perties of polyacetylene, designing a synthesis of a polyacetylene film with a desired morphology remains an empirical exercise. In general, the specialized nature of many tools for solid-state analysis is not conducive to gaining a general understanding of polyacetylene. A single research group may be able to apply a particular technique with state-of-the-art equipment, but will be able to obtain polymer samples made using only one synthetic route. Conversely, a synthetic group may be capable of synthesizing polyacetylene using a number of different techniques, thereby systematically varying its properties, but the logistics of getting the different materials carefully analyzed can be limiting. While this problem has led to many fruitful collaborations, the overall picture remains fragmented. The resulting lack of structure-property relationships compound the dilemma of understanding electrical conductivity through a bulk organic solid. As typically synthesized, polyacetylene is insulating (conductivity less than = lo-’ 0 cm-’ ). However, because of its extended n-conjugation, the polymer is readily reduced or oxidized. From an organic chemist’s standpoint, this treatment, termed “doping” by analogy to the process in semiconductors, produces cations or anions on the polymer which act as charge carriers. Common oxidative dopants include iodine vapor and AsFS, while Na/NH, solutions have been used as reductive dopants [14]. Iodine doping of highly oriented polyacetylene (see below) results in a material with a conductivity of los a-’ cm-’, which is comparable with metals such as lead and platinum. Much theoretical and some experimental effort has established that oxidation or reduction of polyacetylene produces a cation or anion that is delocalized over about 15 carbon atoms, and which may move along the polymer chain (Fig. 10-1) [16-201. In a bulk sample, however, no individual chain stretches for a macroscopic distance, so for an electrical current to flow charge must hop from chain to chain (Fig. 10-2). The efficiency of this process will depend on the length of conjugated segments within a chain, the interchain alignment within both crystallites and amorphous regions, and the extent of polymer crystallization (Fig. 10-3). It will also depend on the macroscopic polymer morphology, for instance the size and extent of polymer fibers or grains. Quantifying the relative contributions of these processes is an area of on-going research and debate [21, 221. Many theoretical models of conductivity are constructed around the fit of a plot of conductivity versus temperature. The more successful treat electron transport as a series of discrete interchain and intrachain steps through both ordered and disordered domains [23].
-’
oxidative doping
t Figure 10-1 Zntrachain charge mobility after oxidation.
10.1 Introduction
355
Figure 10-2 Inferchain charge mobility. Although the dopant counterion, cg.,I- (which also exists as 1; and 1); has been omitted from these drawings, it may play a role in interchain transport.
Figure 10-3 A cartoon of ordered crystallites and disordered amorphous domains in polyacetylene.
This chapter will focus on some recent developments in the synthesis of polyacetylene itself, (CH),, and of substituted derivatives, (CR),(CH), [24, 251. While the great variety of other conductive polymers, including poly(pheny1ene vinylenes), polythiophenes and polypyrroles, may be thought of as annulated derivatives of polyacetylene [26-281. (Fig. 10-4), their syntheses and properties differ enough to put them outside the scope of this chapter. The reader is referred to a number of reviews covering both these polymers and polyacetylene [15, 17, 291. As the field of polyacetylene chemistry has developed, some common themes have emerged. For instance, a number of different approaches have been used to understand the relationship between a polymer’s effective conjugation length and its molecular or solid-state properties [20, 30-341. “Effective conjugation length” is a measure of how many double bonds in a row are in conjugation. This quantity (n)is typically defined by the length of the discrete polyene molecule, R-(CH=CH),-R, that has the same absorption maximum as the polymer in question. It has been shown [35-381 that this optical absorption energy ( E ) may be related to the conjugation length by the empirical expression E = a + b/n (Fig. 10-5). The values of a and b depend upon a number of factors, including the solvent and the polyene end-groups. If the optical absorption maxima for polyenes are extrapolated to n = 03, the expression gives a lower energy (a equals 1.8 eV, or 680 nm in terms of wavelength) than the optical absorption
356
I0 Polyacetylene
Polythiophene
Poly(para-phenylene vinylene)
Y
Polypyrrole
Poly (para-phenylene) R
R' Polyacetylenes
Polyaniline
Figure 10-4 Common conjugated polymers.
D
0 u)
0
P)
<
v
0.0
0.1
0.2
0.3
0.4
0.5
l/(Number of double bonds)
Figure 10-5 Lowest-energy UV-visible absorption maxima of all-tmns-t-butyl-cappedpolyenes dissolved in pentane. Only the data for the molecules containing five or more double bonds are included in the linear fit. The data are from Knoll and Schrock 141).
maximum actually observed for polyacetylene (650 nm) 1391. This has been taken to suggest that conformational or other defects limit the conjugation in polyacetylene to an effective conjugation length of approproximately 30 double bonds [351, although care must be taken when extrapolations are made from data on polyenes in solution to polyacetylene in the solid state. The concept of effective conjugation length is useful, since properties such as electrochemical potential [40] and solubility [34, 411 may generally be correlated with it. Conjugation lengths have been systematically varied by making discrete oligomers [16, 32, 41, 421 by making block copolymers with varying ratios of conjugated to nonconjugated segments [31, 43, 441;and by varying the steric requirements of side-groups to influence the backbone conformation of the polymer [45] (Fig. 10-6).
10.1 Introduction
357
Me
Me
Figure 10-6 Three ways of systematicallyvarying the effective conjugationlength of a polyene: (top) synthesis of short oligomers of known structure [41, 421 ; (middle) interspersing nonconjugated segments along the polymer chain [46, 471; and (bottom) introducing twist-inducing substituentsat intervals along the chain [45].
The quest for a more tractable material is another thrust of today’s polyacetylene research. %o general solutions to the problem of insolubility have been developed. First, a number of
groups have synthesized substituted polyacetylenes. To a certain extent, though, the intractability of the unsubstituted polymer seems to be the price paid for its exceptional properties. Side-groups, while conferring solubility, also perturb the planar conformation of the polymer backbone. Furthermore, substitution (especially irregular substitution) causes the polymer to be less crystalline in the solid state. The resulting decreases in both intrachain and interchain order will give rise to lower electrical conductivities, although the relative contributions of the two effects are not known. Similar tradeoffs between tractability and the properties associated with extended conjugation are observed when block or graft copolymers of polyacetylene are synthesized [a]. In general, the longer the polyene portion of a copolymer, the greater the tendency to aggregate or precipitate from solution. The second approach to a more tractable polyacetylene is to sidestep the compromises described above by developing multistep syntheses which proceed via a soluble “precursor” polymer [31, 48-50]. Although the final product is still insoluble, the solubility of the precursor allows both the determination of properties, such as molecular weight, as well as more facile manipulation of the polymer into a desired macroscopic shape. In addition, these routes also provide polyacetylene films that differ in microscopic morphology from those synthesized directly from acetylene. However, these advantages come with added complexity; not only do these routes require a reaction, frequently carried out in the solid state, to go to completion, but there is also usually the need to remove an eliminated by-product from the solid product.
358
10 Polyacetylene
The sections below provide specific examples of different routes to polyacetylenes. They are organized from a retrosynthetic standpoint. Most address one or both of the themes mentioned above: the effect of changing conjugation length, and the need to circumvent the intractability of (CH),.
10.2 Syntheses and Properties 10.2.1 Routes from Alkynes 10.2.1.1 Acetylene Polymerization Polyacetylene was first observed as an unwanted side-product of attempts to trimerize or tetramerize acetylene using transition metal catalysts [51]. The original synthesis of free-standing, shiny films of polyacetylene, carried out by Shirakawa's group, used an optimized interfacial Ziegler-Natta Ti(O-n-C4H9)4/Al(C2H5)3 polymerization of acetylene gas at the solvent (typically toluene) surface [l]. Material produced via this route has been termed "Shirakawa polyacetylene" . Propagation in Ziegler-Natta catalysis occurs via insertion into a metal-carbon single bond (Fig. 10-7)[52-541.
Figure 10-7 Ziegler-Natta polymerization of alkynes. "P" represents the growing polymer chain.
The cis/trans ratio of Shirakawa polyacetylene is determined by the temperature of polymerization. Polymerizations carried out at - 78 "C yield a material containing a high percentage of cis double bonds, presumably because of the mode of addition during propagation. Heating [55, 561 or doping [57, 581 the solid leads to isomerization to the predominantly tmns isomer. The thermal isomerization is accompanied by a change in the visible absorbance maximum to longer wavelengths [14]. In an all-planar conformation, the cis and trans isomers of a polyene will not differ greatly in the extent of electron delocalization [41], so the observed change in absorbance is attributed to a transition from a helical to a more linear conformation [59]. Ito et al. found the isomerization to be exothermic by approximately 0.9 kcal (mol CH units)-', with an activation energy of 17 kcal mol-* [60],although barriers as low as 11 kcal/mol have been reported [61]. The different values may arise from different degrees of crystallinity in samples of the cis polymer. A recent variation on the Shirakawa techniques has been developed by Naarmann [62, 631. In his procedure, Ti(OBu),/A1Et3 is dissolved in silicone oil. The solution is then heated and n-butyllithium is added. Acetylene is introduced, and the resulting polymer is stretched either before or after washing out the catalyst. Upon doping, this material exhibits conductivities along the stretching direction of up to lo5 P cm-I. This value is the highest yet obtained for an organic material, similar to that of copper, and greater than any metal on a conductivityper-unit-mass (Q - ] cm-' g-') basis.
10.2 Syntheses and Properties
359
The ill-defined nature of this catalyst system makes it difficult to draw any mechanistic conclusions as to why doped “N-polyacetylene” (for “Naarmann polyacetylene” or “New polyacetylene”) is so conducting. Moreover, variants on this technique by different research groups, while still yielding highly conductive polymer, give materials with physical properties, such as density and morphology, that differ from one another [21]. One proposal is that the material is of higher molecular weight, because preheating the catalyst might reduce the number of active catalytic centers. However, systematic experiments carried out by Chien, in which various molecular weights of Shirakawa polymerization were prepared, found that conductivity was independent of molecular weight [64]. Solid-state I3C-NMR does indicate that N-polyacetylene has fewer sp3-carbon “defects” than does Shirakawa polyacetylene. Also, morphology differences are observed between the two polymers, perhaps arising from the different solubilizing properties of silicone oil and toluene [15, 211. Solvation of the catalyst may affect its reactivity, and solvation of the growing polymer will determine at what molecular weight aggregation and crystallization begin to occur [65]. Also, these morphological differences allow Naarmann polyacetylene to be stretched more than Shirakawa polyacetylene, presumably leading to greater crystallinity and interchain contact.
10.2.1.2 Polymerization of Substituted Alkynes
A growing array of different terminal and internal alkynes have been polymerized [8]. Many polyalkynes are air-stable, soluble materials, and not highly conjugated. As new catalysts allow the polymerization of alkynes with an increasing variety of substituents, an exploration of what properties unsaturated polymers have to offer is warranted. In general, substituted polyacetylenes may or may not be colored, and tend to be more rigid than saturated polymers. Selected materials are described below and compiled in Table 10-1. The polymerization of propyne with Ziegler-Natta catalysts, as well as its copolymerization with acetylene, was reported by Chien’s group in 1981 [66]. Poly(propyne) scarcely resembles unsubstituted polyacetylene; it is orange in color, soluble in organic solvents, and cannot be doped to a high electrical conductivity Q cm-’ after exposure to iodine). It is proposed that both chain-twisting due to steric interactions between methyl groups, and the nonequivalence of adjacent carbon atoms, are responsible for the lower conductivity. It is likely that poorer interchain contact also plays a role. When acetylene-propyne copolymers were synthesized, material containing as little as 15% acetylene was no longer soluble, and electrical conductivity increased by three orders of magnitude as the acetylene content was increased to 33 Yo. The copolymers probably form tapered blocks, assuming acetylene polymerizes faster than propyne, and it is presumably aggregated polyacetylene blocks which are responsible for both the electrical conductivity and the insolubility. In recent years, the use of metathesis catalysts to polymerize alkynes, instead of ZieglerNatta catalysts, has increased. This is in part because they have been found to polymerize a wider range of monomers [8],and because the Schrock group has shown that well-defined metathesis catalysts allow some control of alkyne polymerizations (see below) 132, 67, 681. Metathesis polymerizations differ from Ziegler-Natta polymerizations in that the active species is a metal-carbon double bond, or “alkylidene”. Alkynes add across this bond, in what may be thought of as a [2 + 21 cycloaddition [69] to form metallacycles. These in turn open to form a new alkylidene with the growing polymer chain attached (Fig. 10-8) [70].
360
10 Polyacetylene
Table 10-1 Selected examples of polymers synthesized from substituted acetylenes R, - C C - R,
R,
I,,, (nm)
R2 ~~~~~~
Solubility
Reference
~
H
H
620 (solid)
Insoluble
~41
H
Me
290 (heptane)
Soluble
[661
H
Me3Si
292 (THF)
Soluble
(80,811
H
Ph
<350 (CC14)
Soluble
[72, 741
H
o-(Me,Si)Ph
520 (CC14)
Soluble
[74, 82, 831
Me
SiMe,
273 (cyclohexane)
Soluble
1841
455 (6: 1 methanol/benzene)
Soluble
~761
300-325 (chloroform)
Soluble
[77, 851
R=H: 598 (CH3CN) R=SiMe3: 610 (CH,CN)
Soluble
1731
(Not reported)
Soluble
WI
0
H
H or SiMe,
+o /
\
-
Me
Masuda and co-workers have been among the most active groups working in the area of alkyne polymerizations [S]. They have uncovered the types of classical polymerization catalysts that are best suited for monomers with different substitution patterns. For instance, while Ziegler-Natta catalysts will produce high polymer from sterically undemanding acetylene or n-alkyl terminal alkynes, they will not form soluble high polymer from aryl alkynes, alkynes with tertiary substituents, or disubstituted alkynes. In contrast, Group VI metathesis catalysts, such as WC16/SnPh4, will polymerize more sterically demanding alkynes, such as tert-butylacetylene. They have found that catalysts based on niobium or tantalum pentahalides polymerize even more sterically demanding acetylenes, including ethyl-
10.2 Syntheses and Properties
361
Figure 10-8 Acetylene metathesis polymerization by an alkylidene complex.
phenylacetylene and 1-(triethylsily1)-1-propyne. These Group V catalysts tend to give substituted benzenes (cyclotrimers) from alkynes with less substitution. Table 10-1 contains selected examples of some of the diverse substituted polyacetylenes which have been synthesized. Given the large variety of different alkynes that have been polymerized, few generalizations can be made about their properties. In general, though, placing substituents at every alternate carbon, or at every carbon, of a polyene chain forces the polyene backbone to twist out of conjugation [71]. For instance, the absorption spectrum of poly(trimethylsilylacetylene), which is yellow, has a I,,, of 292 nm (in THF). Its effective conjugation length is thus about four double bonds. The polymer cannot be doped to a conductive state. In contrast, poIy@henylacetylene), because the phenyl group can rotate perpendicular to the polymer chain, is more conjugated, with a I,, of 325-350 nm, and is a semiconducting photoconductor [72]. Poly(phenylacety1enes) with ortho substituents on the phenyl ring and poly(2-ethynyl-N-methylpyridium) derivatives [73] form an interesting subclass of substituted polyacetylenes. It has been found that the absorption maxima of the cis isomers of the phenyl polymers shift to longer wavelength as the size of the substituent is increased. For example, poly(o-methylphenylacetylene) absorbs at 440 nm, and poly(o-trimethylsilylphenylacetylene)absorbs at 520 nm [74]. Evidently, the steric requirements, of the ortho substituents impose a planar conformation on the backbone (Fig. 10-9).
Figure 10-9 A possible conformation of the cis isomer of an ortho-substituted poly(phenylacety1ene). The steric requirements of the substituents enforce a planar conformation.
The drive to make an organic magnet [75] has motivated efforts to attach free-radical-containing side-groups to polyacetylene chains in the hope that the polyene chain will act as a ferromagnetic coupling unit between the spins [76]. To date, although the materials do contain many free spins, they are not ferromagnetically coupled. Nishide et al. made the attempt by placing nitroxide radicals on phenyl substituents (Table 10-1) [76]. It is possible that the nitroxides are too decoupled from one another because the phenyl rings are twisted out of conjuga-
362
I0 Polyacetylene
tion with the polymer chain, and that a more rigid framework would be required for this approach to work. Also shown in Table 10-1 is an (alkylcyclohexylary1oxy)-substituted polyacetylene [77]. Polymers of this general structure have been found to display liquid-crystalline behavior. In contrast to vinyl-based liquid-crystalline polymers, the geometric isomerism of the main-chain double bonds plays a role in determining the type of phase that is found. Advincula et al. have examined Langmuir films of polyacetylenes at the air-water interface [78]. Polyacetylene derivatives are unusual in that the polymer backbone itself acts as a chromophore; therefore, in studies such as these, UV-visible spectroscopy can be a sensitive probe of polymer conformation. An example of an enantiomerically pure polymer is also shown [ll]. Aoki et al. showed that films of a polyacetylene substituted with a (-)-D-pinene derivative formed an effective membrane for chromatographic resolutions of racemic mixtures. ( f)-2-Butanol was resolved to 29.8 Vo e.e. and dl-tryptophan to 86.1 Vo e.e. There is increasing interest in unsaturated polymers for both liquid-phase and gas-phase separation applications [8, 9, 791. It has been suggested that the rigidity and irregularity of the highly substituted polyacetylene chain, combined with the presence of aliphatic substituents which reduce interchain interactions, are important for the polymers’ transport properties [lo]. Most polymerizations of acetylene or substituted acetylenes have been carried out using “classical” metathesis catalyst systems [8, 86, 871. As with Ziegler-Natta polymerizations, the active organometallic species in a classical metathesis system is formed in situ, for example, by mixing WCl, with Me& [88]. Some of the difficulty of characterizing polyacetylene stems from the ill-defined nature of its synthesis. An unknown and potentially variable fraction of the metal centers react to form catalytically active species. In addition, given the Lewisacidic nature of the components, some species that are formed could cause unwanted sidereactions, such as Friedel- Crafts alkylations or cross-linking. To circumvent these problems, and to shed some light on the mechanistic details of these reactions, the Grubbs and Schrock groups have synthesized well-defined, stable compounds containing metal-carbon double bonds or triple bonds [89-921. These do, Group V and VI, complexes not only are metathetically active, but in some cases react to form observable propagating metallacyclobutenes [68, 931. A further advantage of using these well-defined catalysts is that they are more likely to produce “living” polymerizations [32, 94, 951. A living polymerization is one in which initiation of the polymerization is fast relative to propagation, so the catalytic centers are all polymerizing simultaneously, and in which there are no chain transfer or termination steps. If these conditions are satisfied, then the resulting polymer will be monodisperse, i. e., the chains will have a narrow range of molecular weights. Wallace et al., for example, found that monodisperse poly(2-butyne) and poly(diphenylacety1ene) could be prepared using a well-defined trisaryloxy tantalum alkylidene. The addition of pyridine was necessary, presumably because upon complexing tantalum, it increased the rate of initiation relative to propagation [68]. Some polymerizations of sterically hindered alkynes using classical catalyst systems do produce relatively monodisperse polymer, although this is the exception [74, 83, 961. Another type of alkyne polymerization, and one to which well-defined catalysts have been applied [67, 971, is the ring-closing polymerization of 1,6-diynes (Fig. 10-10). The resulting polymers are frequently soluble, and can be more highly conjugated than acyclic analogs because adjacent bonds are held closer to coplanarity by the ring. They are not expected to
10.2 Syntheses and Properties
363
be highly conductive, however, as a relatively large volume of the polymer solid is taken up by organic substituents which are placed in such a way that they would be expected to prevent interchain contact. For example, Choi and co-workers [86]recently reported h,,, of up to 480 nm, and conductivities of approximately 2 ! cm-I for polymers (prepared using a classical catalyst) in which X = (n-C,H&N+ (Fig. 10-10). Schrock’s group [67] has extended the use of Group VI alkylidenes to the polymerization of a variety of esters of dipropargyl malonate [X = (R02C),C]. The system indicates the capability of a well-controlled catalyst. For instance, using the molybdenum alkylidene complex shown in Fig. 10-10 diethyl dipropargylmalonate was reacted to form a series of relatively monodisperse polymers of increasing molecular weight. Although the polymers were found to contain an approximately 1 : 1 mixture of five-membered and six-membered rings, resulting from either tail-tail or head-tail sequential additions to the alkylidene, a plot of h,, versus the reciprocal of the number of bonds in the polymer was reasonably linear.
-’
Figure 10-10 Polymerization of 1,6-diynes using a molybdenum alkylidene catalyst [REis (CF,),CH,C] [67]. The lb-diyne monomer is drawn in two different exaggerated conformations to illustrate that “head-tail” polymerization leads to six-membered rings, and “tail-tail” polymerization leads to fivemembered rings. See Fig. 10-8 for a more mechanistic diagram of acetylene metathesis.
The living behavior of this 1,ddiyne system was also demonstrated by the synthesis of “push-pull” polyenes. This was accomplished by employing a catalyst with a preformed, donor-substituted @afu-Me,N-C,H,) alkylidene, which ends up at one end of the polymer. When all of the monomer has been polymerized, the molybdenum alkylidene at the chain end is replaced with an electron-accepting group, via a Wittig-type reaction with pum-cyanobenzaldehyde. At lower molecular weights, the “push-pull” polymers tend to have lower energy absorption maxima than do analogously prepared “push-push” or “pull-pull” polymers. At higher molecular weights, the I,, for all three types of polymers approaches 550 nm (in THF).
10.2.2 Routes from Alkene Precursors 10.2.2.1 Nonmetathetic Routes When one considers the polymerization of acetylene, it becomes apparent that no matter which catalyst is used, the properties of the polymer will be to some extent determined by a balance between the polymerization rate, the rate at which the product starts to aggregate or
364
10 Polyacetylene
precipitate, and the rate of polymerization in the solid state. One possible way to avoid this dilemma, which is inherent in directly forming an insoluble polymer, is to proceed via a multistep route to unsubstituted polyacetylene. Table 10-2 contains a number of schemes that involve first synthesizing a more tractable “precursor” polymer. Table 10-2 A summary of precursor routes to unsubstituted polyacetylene Monomer
h,,
Precursor polymer
Vinyl chloride
Free-radical initiators
Poly(viny1 chloride)
Isoprene
Anionic
Polyisoprene
Benzvalene
ROMP catalysts
Polybenzvalene (see Fig. 10-19)
Feast monomer
ROMP catalysts
See Fig. 10-18
7-oxanorbornadiene
Pd(OAc),
See Fig. 10-16
Heat
549
[SO]
See Fig. 10-12
Heat
n.r.(d)
[31, 1041
2,4,6-0ctatriene
Schrock metathesis catalyst (ADMET)
426
[lo51
Phenyl vinyl sulfoxide Anionic
This is the longest wavelength of the fine structure. value for unoriented material. ( d ) Not reported. (a)
(b)
Conversion reagents
(nm)
Heat
S40(‘)
Ref.
Polymerization reagents
[48, 1031
Not observed in the undoped form.
(‘)
The
An early example is the use of base to eliminate HCI from poly(viny1 chloride) (Fig. 10-11) [98- 1001. More recently, h u n g and Tan have developed a route in which the extent of elimina-
tion can be carefully controlled (Fig. 10-12) [31]. Poly(pheny1vinyl sulfoxide) was prepared via a normal anionic polymerization. This polymer is unstable, and at 150°C readily eliminates HSOPh to give polyacetylene [104]. However, if the sulfoxides are oxidized to sulfones, they do not eliminate. Therefore, by treating the polymer with varying amounts of rn-chloroperbenzoic acid before heating, a series of copolymers of varying conjugation length were prepared. The logarithm of the conductivity of iodine-doped samples was found to be roughly linear with respect to the percentage of uneliminated “defects” in the polymer chain (Fig. 10-13). One feature of reactions such as these is that the eliminations do not necessarily proceed randomly along the chain. Once a given monomer unit has been eliminated, adjacent units become more reactive due to resonance stabilization from the eliminated unit (Fig. 10-14) [SO, 1041. In those cases where incomplete reaction occurs, it may be because long polyene segments forming early in the reaction cause the polymer to precipitate prematurely, and further elimination is slowed in the solid state.
Figure 10-11 Elimination of HCI from poly(viny1 chloride).
365
10.2 Syntheses and Properties
0
-
anionic
polymerization Ph
an -wn 150 "C
m-CPBA (s toichiometry
varied)
SOPh
0 SOPh
- d i .
Ph
's
Ph
\p
0
150°C
Figure 10-12 A precursor method for obtaining varying conjugation lengths [31]. -1-
A - 2-
A
h
%
c .-
.->
-3-
%A
= I
0 C 0
0, -
4 4-
0,
-0
- 5-
A
- 64 10
A
I
I
I
I
15
20
25
30
I 35
I
40
"Percentage of Defects" (%) Figure 10-13 Conductivity data for partially oxidized and eliminated poly(pheny1 vinyl sulfoxide) from Table V in (311. The conductivity was measured at 11.7 Hz and is in S cm-'. The "percentage of defects" is measured by IR spectroscopy.
\\A&\/ - \\A/+- \!mi/ x
x
slower
X
faster
Figure 10-14 The acceleration of elimination at sites adjacent to sites which have already reacted.
In 1988, the report that 1,Cpolyisoprene showed appreciable electrical conductivity after iodine-doping created a stir in the conductive polymer community [loll. However, initial suggestions that this meant replacing the common paradigm, in which extensive backbone
366
I0 Polyacetylene
delocalization is necessary for significant electrical conductivity, may have been premature. Subsequent experimental studies by Dai and White [lo61 and by Set0 et al. [lo71 as well as a theoretical treatment [108], indicate that the iodine is probably adding across double bonds. HI is then eliminated to produce conjugated segments (Fig. 10-15). Therefore, polyisoprene may better be thought of as a precursor to a partially methyl-substituted polyacetylene.
Figure 10-15 Addition of iodine to l,Cpolyisoprene, followed by elimination of HI to give an unsaturated polymer.
More recently, a Pd(1I) salt was shown to catalyze the 1,2-insertion polymerization of a 7-oxanorbornadiene derivative (Fig. 10-16) [50]. The resulting saturated polymer, when heated, gives polyacetylene via a retro-Diels- Alder reaction. (This reaction is reminiscent of the Durham route to polyacetylene discussed below). One advantage of this technique over other routes is that it employs a late transition metal polymerization catalyst. Catalysts using later transition metals tend to be less oxophilic than the do early transition metal complexes typically used for alkene and alkyne polymerizations [109, 1101. Whereas tungsten alkylidene catalysts must be handled under dry anaerobic conditions, the Pd(I1)-catalyzed reaction of water-insoluble monomers may be run as an aqueous emulson polymerization.
100 O C
EtOpC
CopEt Et02C
C02Et ‘C02Et
Figure 10-16 A precursor route to polyacetylene which proceeds via an insertion polymerization followed by a retro-Diels-Alder reaction [50].
10.2.2.2 Routes Using Olefin Metathesis Since ring-opening olefin metathesis polymerizations (ROMPS) [lll, 1121, yield polymers which contain at least one double bond in the backbone per monomer unit (Fig. 10-17), it is
10.2 Syntheses and Properties
L,M- m
-
C
H
367
R
f
\ L,M=~~zCHR
4
/
gcHR
LnM
Q
Figure 10-17 Ring-opening metathesis polymerization (ROMP) [Ill, 1121. represents a cyclic olefin or an alkyne. Any of the steps may be reversible, depending in part on the relative stabilities of the metallacycle and metallacarbene, and on the ring-strain of the ring in the monomer which is opened.
not surprising that a number of precursor routes to polyacetylene use ROMP.The earliest example of such a route was developed at Durham in 1980 [48]. The tricyclic molecule obtained from the addition of hexafluoro-2-butyne to cyclooctatetraene, and frequently referred to as “Feast monomer”, was polymerized using WCI,/Ph,Sn or TiCI,/AIEt, (Fig. 10-18) [113]. The resulting soluble precursor polymer was found to spontaneously undergo retro-Diels- Alder reactions in solution at room temperature to give o-trifluroromethylbenzene and “Durham polyacetylene”. This scheme relies on the selectivity of the catalyst for opening the strained, unsubstituted bond [lll, 1121. Similar monomers were also investigated, and it was found that precursor polymers which eliminated naphthalene derivatives, rather than a benzene, required higher temperatures for the reaction to occur [114].
WC16/Ph,Sn ___t
toluene
Figure 10-18 The Durham route to polyacetylene.
Durham polyacetylene was the first processable polyacetylene, and its use has been taken up by a number of groups [41, 1151. One physical property which distinguishes it from
368
10 Polyacetylene
Shirakawa polyacetylene is its decreased crystallinity [103, 1161. This is manifested not only by its diffractive properties, but also by a slightly higher energy absorption maximum in the solid state. It was found, though, that stretching the precursor polymer before elimination induced crystallinity and yielded a more “Shirakawa-like” polyacetylene. Knoll and Schrock used the Feast monomer to prepare polyenes containing up to 15 double bonds [41]. They discovered that a living oligomer of Feast monomer prepared from a tungsten alkylidene underwent a retro-Diels- Alder reaction smoothly to give a tungsten-capped polyene oligomer. The metal could be removed from the oligomer in a Wittig-like reaction to give tert-butyl-capped mixtures using either pivalaldehyde or trans-4,4-dimethyl-2-pentenal of oligomers containing either an odd or an even number of double bonds. Careful chromatography allowed the separation and characterization of oligomers by both length and doublebond geometry. Previously, there were no good routes to polyenes of this length. Swager and Grubbs reported an unusual precursor route to polyacetylene which also used a tungsten metathesis catalyst 1491. They polymerized the highly strained benzene isomer, benzvalene, to give highly strained polybenzvalene. There is a potential advantage over the Feast route to a method in which the precursor polymer is an isomer of polyacetylene, because its conversion does not result in any loss of mass or the need to wash out an elimination product. Unfortunately, polybenzvalene, is, like the monomer, prone to detonation. However, if handled carefully, treatment with HgCI, causes the bicyclobutane rings to rearrange to butadienes, yielding polyacetylene (Fig. 10-19). When iodine-doped, the conductivity was approximately 1 0 cm-’. This moderate value probably stems from the relatively high number of saturated carbons present (10- 19%). Presumably these arise from crosslinking of the bicyclobutane units in the precursor [102].
-’
Figure 10-19 The synthesis of polybenzvalene, and its conversion to polyacetylene [49].
The most recent application of olefin metathesis to the synthesis of polyenes has been described by Tao and Wagener [IOS, 1171. They use a molybdenum alkylidene catalyst to carry out acyclic diene metathesis (ADMET) (Fig. 10-20) on either 2,4-hexadiene or 2,4,6-octatriene. The Wagener group had earlier demonstrated that, for a number of nonconjugated dienes [118- 1201, these polymerizations can be driven to high polymer by removal of the volatile product (e. g., 2-butene). To date, insolubility limits the extent of polymerization of unsaturated monomers to polyenes containing 10 to 20 double bonds. However, this route has some potential for the synthesis of new substituted polyacetylenes. Since most of the monomer unit is “preformed” before polymerization, it is possible that substitution patterns which cannot be incorporated into an alkyne or a cyclic olefin can be built into an ADMET monomer.
10.2.3 Ring-Opening of Cyclooctatetraene The application of metathesis catalysis to polyene synthesis may be viewed as following a progressive path : from polymerizing acetylene itself, a “two-membered ring”, to ring-opening a
10.2 Syntheses and Properties
-
369
RCH=CHR
+
Figure 10-20 The early stages of acyclic diene metathesis (ADMET). The reaction is driven by pumping off volatile products. When X is a double bond and R* is methyl (i. e., 2,4,6-octatetraene), oligomeric polyenes are formed using a molybdenum alkylidene catalyst [105, 1171.
masked four-membered cyclobtadiene ring [48], or using ADMET to connect an equally unsaturated butadiene [105], to ring-opening strained six-membered rings (benzvalene) (Fig. 10-21). The next logical step in the series is the polymerization of the eight-carbon polyacetylene isomer, cyclooctatetraene (COT). COT was polymerized by Hocker and coworkers using a classical tungsten catalyst [121], and by Klavetter and Grubbs using an active tungsten alkylidene [46, 471. An active catalyst is necessary, since cyclooctatetraene is a relatively unstrained carbocycle (2.5 kcal mol-' ring strain) [122]. 2:
H-H
Figure 10-21 Two-, four-, six-, and eight-carbon equivalents used in the synthesis of polyacetylene via metathesis catalysts.
370
10 Polyacetylene
When well-defined, less Lewis-acidic metathesis polymerization catalysts are used to polymerize COT, a lower level of detectable sp3 defects are formed. Also, although the polyacetylene produced is still insoluble, the reaction proceeds slowly enough to allow manipulation of the liquid reaction solution before hardening. In this way, one can obtain films in a desired shape and location, e.g., on a semiconductor [123]. This procedure was found to result in better electrical contact than can be obtained when a free-standing film prepared via the Shirakawa route is simply pressed against an electrode. Once the principles of COT polymerization had been established, the extension of this technique to the syntheses of substituted polyacetylenes was examined [34, 45, 124, 1251. This route may be thought of as an alternative method for the preparation of a copolymer of acetylene and a substituted acetylene (Fig. 10-22). Polymerizing substituted cyclooctatetraenes offers more control, in principle, since side-groups may be spaced along the chain in a more regular manner. Also, given the variety of substituted cyclooctatetraenes that are synthetically accessible [126, 1271 the method is versatile, being primarily limited by what types of functionality the tungsten alkylidene catalyst can withstand. This route thus offers the possibiliy of tuning the properties of the polymer by perturbing the polyene chain in a controlled fashion and thereby systematically exploring the interplay between solubility and the properties associated with the extent of conjugation.
Q AorB
1
hv or A
R
Figure 10-22 The synthesis of substituted polycyclooctatetraenes [34].
A number of different high-molecular-weight (M, of lo4- lo’) substituted polycyclooctatetraenes (poly(RCOT)s) have been prepared (see Table 10-3), and they may be coarsely divided into two sets: those which are soluble in organic solvents in the tmns configuration, and those which are not. Polymers containing n-alkyl side-chains are mostly soluble in the cis form, but insoluble in the trans form (except for a very small fraction of the material). In contrast, poly(trimethylsilylcyc1ooctatetraene) and poly(sec-butylcyclooctatetraene) are completely soluble in both the cis and trans forms. Moreover, these two soluble polymers are
10.2 Syntheses and Properties
371
Table 10-3 Optical absorbance (in THF) and solubility data for cis and trans poly(RC(TT)s [34]. %nu
R
cis
(nm) trans
Solubility Insoluble Insoluble
H methyl n-Butyl n-Octyl n-Octadecyl Neopentyl
614 632 630 634
(a)
616
(a)
496 522 360
594 620 556
(a)
366 365
556 550 586 432 540
462 480 538 412
2-Ethylhmyl Methoxy tert-Butoxy Phenyl sec-Butyl
(a) (a)
(3
Insoluble
Isopropyl Cyclopentyl Cyclopropyl tert-Butyl Trimethylsilyl
302 380
(a)
Soluble Soluble Soluble (a)
Soluble Soluble
These polymers are soluble in the cis form, but aggregate in the trans form. The,,A reported for the trans isomers of these polymers are for visually homogeneous, dilute ( M) solutions, although they presumably contain aggregated material since filtration results in filtrates which absorb at higher energies (kmU= 580 10 nm) and contain low-molecular-weight polymer (M,,< 5000).
*
highly conjugated, as evidenced by their color (A,,,= = 540 nm, and 556 nm, respectively, in THF), and the trans isomers can be doped to appreciable conductivities (1 to 10 SZ cm-'). An examination of the polymers' solubility suggests that it is the steric bulk of the sidechain at the position adjacent (a) to the double bond which determines solubility. Thus, poly(neopenty1COT) and poly(2-ethylhexylCOT) (R, = - CH2-), much like the polyn-alkylCOT polymers (R, = -CH2- also), are soluble in the cis form but not in the tram. Similar behavior is observed for alkoxy- (R, = -0 -) and phenyl-substituted derivatives. Polymers containing a secondary (R, = -CHR'R) or tertiary (R, = - C R R ' R ' ) substituent adjacent to the double bond are soluble in both the predominantly cis and predominantly tram forms. All of the substituted poly(cyc1ooctatetraenes) display intense absorptions (E = lo3 cm-' M-I) in the visible spectrum. The optical absorption maximum is a measure of the effective conjugation length of the double bonds in the main chain. The bulky side group in tram-poly(tert-butylCOT)severely limits conjugation to the equivalent of ten double bonds, since its spectrum is similar to the most intense transition of an all-tram polyene with ten double bonds (Fig. 10-5) [41]. If the polyene data in Fig. 10-5 are used to extrapolate to longer wavelengths, those polyRCoTs with methylene units adjacent to the main chain have an effective conjugation length of 25-30 double bonds. Absorption spectra of thin films of these polymers are similar to that of polyacetylene, indicating a similar effective conjugation length. The soluble trans polymers with more bulky substituents adjacent to the main chain (e.g., R = isopropyl, sec-butyl, cyclopentyl, and trimethylsilyl) have slightly blue-shifted
-'
372
10 Polyacetylene
h,,,, indicating a slightly reduced effective conjugation length of 20-25 double bonds. In fact, trans-poly(sec-butylCOT) displays the highest observed effective conjugation length for a soIubIe polyacetylene. The optical absorption and solubility data on the trans-poly-RCOTs suggest that a compromise must be made between conjugation length and solubility in polyacetylene derivatives. More conjugated polymers tend to be less soluble. The simplest explanation for these data is that attractive noncovalent interactions are sufficient to aggregate long polyenes, but that sidegroups which cause the polymer chain to twist render the polymer soluble, albeit with a loss of conjugation. The insolubility of oligomeric all-trans polyenes has been attributed to covalent crosslinking (411. Crosslinks are not thought to be present in the trans-polyRCOTs, however, as their absorption maxima are comparable with that of unsubstituted polyacetylene. It should be noted that poly-RCOTs behave differently from substituted polythiophenes [128, 1291. Attachment of short n-alkyl side-groups (e. g., n-hexyl or n-octyl) serves to solubilize a polythiophene chain, but even the attachment of an n-octadecyl chain at approximately every eighth carbon of trans-polyacetylene is not enough to solubilize it [130]. By copolymerizing two monosubstituted COT derivatives, one which gives a soluble homopolymer and one which gives an insoluble homopolymer (both in the trans form), it is possible to “tailor” the effective conjugation length and solubility of the resulting polymer by adjusting the ratio of the two monomers [34]. The copolymerization of n-octylCOT with trimethylsilylCOT (a monomer in the former category) provides a family of polymers in which the effective conjugation length of the resulting copolymer increases monotonically with the amount of n-octylCOT added to the reaction. As the n-octylCOT/Me,SiCOT ratio is increased, the absorption maximum of the copolymer increases. When k,,, reaches 580 nm, at approximately 50 070 n-octylCOT, the polymer becomes insoluble. The solubility of a copolymer containing an n-alkyl group implies that the insolubility of trans homopolymers of poly(n-alky1COT)s is not caused by crosslinking at the potentially reactivity allylic sites of the side-chain. Data from a variety of polyacetylene block and graft copolymers [44], in which a nonconjugated block acts as a solubilizing tail, also imply that there is a direct tradeoff bevalues as high as 580 nm have been tween conjugation length and solubility. Although &,, reported for solutions of these polymers, there is evidence that the polyacetylene segments in these solutions are aggregated or crosslinked [43,1311. Overall, these copolymer data, combined with the poly(RCOT) data for a variety of substituents, support the contention that there is a maximum soluble conjugation length possible for polyacetylene derivatives. In order to examine possible conformational differences between the different classes of poly(RCOT)s, both force-field (MM2) I1321 and semiempirical quantum mechanical (AM1) [133] calculations on model polyene oligomers were employed (Fig. 10-23) [34]. Using either routine, it is observed that the single bonds adjacent to the trisubstituted double bond (0, and 0,) both strongly deviate from planarity in the models of the soluble polymers. In contrast, in the model of trans- poly (tert-butoxyCOT), an insoluble polymer, 0,is large, but @, is not. The good correlation between polymer solubility and calculated chain twist is indicated R
0, 0,
R
R
Figure 10-23 The twist angles, 0,and tions [34].
a,, in the model polyene used for molecular mechanics calcula-
10.2 Syntheses and Properties
373
by the line that may be drawn across the bar graph of twist angles, which cleanly divides the soluble and insoluble polymers (Fig. 10-24). This model is useful for the design of new poly(RCOT)s, in that if a polymer is modeled computationally before synthesis and both 0, and O2 are found to be sufficiently large, the polymer is predicted to be soluble. cyclopropyl
'
neopentyl
TMS
R
t-BuO
hlleo t-Bu S-BU n-Bu 0
10
20
30
40
50
60
Twist Angle (") Figure 10-24 Twist angles about the single bonds adjacent to the substituted bond of a poly(RC0T) oligomer, modeled using molecular mechanics [34]. Those polymers with twist angles greater than the position of the dashed line are soluble in the trans form.
The versatility of the poly(RC0T) approach is demonstrated by the synthesis of soluble, chiral, highly conjugated polyacetylenes [134]. The hypothesis that any poly(RC0T) with secbutyl substitution would provide enough of a twist for solubility but not so large a twist as to severely lower the conjugation length, made 2-siloxy-3-(cyclooctatetraenyl)butane derivatives, prepared from chiral epoxides, an obvious choice (Fig. 10-25).The backbone x + n* transition of the enantiomerically pure polymers show substantial circular dichroism (CD). The magnitude of the CD for the polymers is characteristic of a dissymmetric chromophore, indicating that the chiral side-groups twist the main chain in predominantly one chiral sense, rather than just inductively perturbing the chromophore. The CD is similar in magnitude to that observed for previously prepared substituted polyacetylenes having chiral substituents at every alternate carbon, although it is substantially shifted to lower energies due to the increased conjugation length [67,135, 1361.
-
Figure 10-25 Synthesis of a chiral COT monomer TBS = 'BuMqSi (1341.
The soluble poly(RC0T)s undergo a photochemical cis/rruns isomerization in solution - a reaction that is not observed for unsubstituted polyacetylene in the solid state. The photochemical cis/truns isomerization of poly(trimethylsilylC0T) can be followed by visible spectroscopy (Fig. 10-26)[124].Here, the chromophore is a collection of double bonds with an ef-
374
I0 Polyacetylene
fective conjugation length of 15-20 double bonds. An isosbestic point is observed during the isomerization, meaning that either all of the double bonds comprising this chromophore isomerize simultaneously without detectable intermediates, or that no observable change in chromophore absorption occurs until some threshold number of individual double bonds have isomerized. Tanaka et al. observed a similar isosbestic point in the visible absorption spectrum upon thermal cis/trans isomerization of unsubstituted polyacetylene in the solid state [61]. 0.41
W
-
0
z u rn U
0 u)
m 4
0.0
I
400
1
I
600
I
I
800
WAVELENGTH (nm)
Figure 10-26 UV/visible spectra of poly(Me,SiCOT) in carbon tetrachloride eight periods of photolysis (10 s each) with a 350 W mercury lamp.
M) obtained between
The rate of the thermal isomerization of poly(sec-butylCOT) in solution was measured by following the absorbance at 560 nm [34]. First-order kinetics were observed, with the molarity considered to be moles of “chromophores”. The activation energy was found to be 21.3 f 0.4 kcal mol-’ with A = 2.4 x 10”. An Eyring plot gave activation parameters of A H * = 20.6 f 2.1 kcal mol-’ and AS* = -13.3 & 4.5 cal mol-* K-’. Isomerization rates were similar in benzene and THF, suggesting a nonpolar transition state. This rate may be compared with that of other conjugated unsaturated systems, for example : stilbenes (AG* = 55.1 kcal mol-’ at 723 K) [137]; a six-carbon cumulene (AG* = 20 kcal mol-’ at 393 K) [137]; and a semi-rigid heptaene (AH* = 27.5 kcal mol-’, AS* = 4.4 cal mol-’ Ky-’) [138]. In comparison, ethylene (CHD=CHD) has a 65 kcal mol-’ barrier for isomerization at 723 K [137]. Evidently, the transition state for poly(sec-butylCOT) isomerization is extremely stabilized due to extended delocalization. More generally, the ability to perform such experiments is an indication of the usefulness of soluble, highly conjugated polyacetylenes. The electrochemical behavior of poly(RC0T)s has also been examined [40]. As expected from the electrochemical properties of unsubstituted polyacetylene, films of poly(RCOT)s coated on an electrode and immersed in an acetonitrile electrolyte solution (in which the polymers are not soluble) are found to undergo reversible oxidative and reductive doping. Unlike unsubstituted polyacetylene, these films may be prepared readily by casting from solution, or, in the case of poly(sec-butylCOT), by electrodecomposition from a THF solution. In contrast to the voltammetry of polymer films, cyclic voltammograms of methylene chloride
10.2 Syntheses and Properties
315
solutions of poly(sec-butylCOT) or poly(Me,SiCOT) were poorly defined, as expected for a redox system with many degrees of freedom and highly coupled redox sites. A significant difference between poly(RC0T)s and unsubstituted polyacetylene is that the former bear substituents which may perturb their electronic properties. The optical absorbances (Table 10-3) probe this to some extent, but electrochemical data are more sensitive. For example, whereas poly(sec-butylCOT) and poly(Me3SiCOT)have similar absorption spectra, both the formal reduction and oxidation potentials of the silyl-substituted polymer are shifted positive of the alkyl-substituted polymer (Fig. 10-27). This is expected, based on the more electropositive nature of the silyl substituent [139]. Also, while the effects are not large, the substitution of either an electron-donating (para-methoxyphenyl) or an electron-withdrawing (para-trifluoromethylphenyl) substituent do perturb the polymer's electronic properties, with the latter material being harder to oxidize and easier to reduce than the former.
-1.5 I
I
t-BuO pCF3Ph
I
I
I
Ph pMeOPh
PBU
I
M,si
I
sec-Bu
I
t-Bu'
R Figure 10-27 The formal oxidation and reduction potentials of poly(RC(TT)s. The potentials are vs. SCE and were determined by averaging the anodic and cathodic peak potentials from voltammograms of polymer films on a glassy carbon electrode immersed in a 0.1 M Me4NBF4 acetonitrile electrolyte [40]. * The oxidation of the lert-butyl-substituted polymer was irreversible
Poly(Me,SiCOT) has been used to illustrate the advantages that a conducting polymer can bring to an electronic device [140]. A thin film of the polymer was cast onto n-doped silicon, then oxidized with iodine to form a semiconductorkonducting-polymersolar cell. When light passed through the polymer onto the silicon, a photocurrent was measured passing through a circuit linking the two materials. The smooth morphology of the solution-cast polymer ensured a uniform contact with the silicon surface, in contrast to press-contacted films of fibrillar, insoluble polymers which have also been examined in electronic devices. Moreover, it was shown that the silicon/conducting-polymer solar cell behaved, in a strict sense, better than a siliconhetal device. It has been found that the photovoltages produced at silicon/ metal interfaces do not vary as expected from theory when silicon samples of different conductivity are used (1411. This has been attributed to the electronic properties of an unwanted metal silicide layer which forms at the interface during the gas-phase metal deposition process. In contrast, higher photovoltages were observed in the silicon/poly(Me3SiCOT) device, indicating that there was no limiting interfacial layer of impurities. The photovoltages observed were those predicted by theory, assuming the silicon/polymer interface behaved in an ideal fashion, and that it was the properties of the bulk silicon which were limiting the efficiency of the device. This behavior has not been observed for siliconhetal solar cells.
376
I0 Polyacetylene
10.3 Conclusions The work mentioned immediately above, in which a conducting polymer behaved more ideally than does a metal, gave the type of result that provides impetus for exploring new conjugated polymers. Polyacetylene research will certainly continue along a variety of different paths. Although syntheses using classical metathesis catalysts are more difficult to control than those using well-defined catalysts, the fact that the highest-quality unsubstituted polacetylene known is made using Ti(OR),/AIRj, plus the ready availability of classical catalysts, will ensure their continued use. More work is needed to understand both the mechanisms by which the active organometallic compounds are formed, and the structure of the propagating species [74, 88, 1421. The examination of conjugated, soluble, substituted polyacetylenes remains in its early stages. There are still a number of interesting synthetic targets which have not been approached. One example is fluorinated polyacetylene. Theoretical reports indicate that the material should behave very differently from normal polyacetylene [143 - 1471. For example, a recent report suggests that poly(difluoroacety1ene)will be nonplanar, and that poly(fluor0acetylene), if polymerized head-to-tail, will be most stable in the cis configuration [148]. Stereoregular substituted polythiophenes have been prepared recently, and the ordering of the side-chains has been found to increase main-chain conjugation [149- 1511. To date the only similar observation for polyacetylenes is that of ortho-substituted derivatives of poly(pheny1acetylene). It would be of interest to prepare more highly conjugated substituted polyacetylenes of well-defined structure, perhaps by the ring-opening metathesis of disubstituted cyclooctatetraenes, or via ADMET, to observe whether a better solubility-conjugation length tradeoff could be made. Finally, theoretically insight into the mechanisms of charge transport should continue to improve. This will happen in concert with the development of better experimental probes of the solid state. Better understanding of the structural origins of the exceptional properties of N-polyacetylene is needed, and such work may point the way toward the syntheses of new families of materials.
10.4 Experimental Procedures A detailed account of the synthesis of Shirakawa polyacetylene may be found in [152].
10.4.1 Synthesis of Substituted Polycyclooctatetraenes The metathesis catalysts, W ( C H ( ~ ~ ~ ~ - B U ) ) ( N - ( ~ , ~ - ( ~ - P ~ ) ~ C ~ H , ) ( 1 O [153], C M ~ (and CF~)~)~, W(CH(o-MeOPh))(NPh)(OCMe(CF3)J2THF, 2 [91], were prepared using literature methods. Polymerizations and subsequent handling of polymer films and preparation of polymer solutions were conducted in a nitrogen-filled Vacuum Atmospheres drybox. In a typical polymerization, 2-3 mg of catalyst (1 or 2) was weighed into a vial and dissolved in a minimum of pentane (2-3 drops). One drop of THF was added to catalyst 1 to slow the rate of polymerization. Lewis bases will reversibly bind to the catalyst, slowing the rate of propaga-
-
10.4 Experimental Procedures
317
tion considerably [46].Catalyst 2 already has a molecule of THF precoordinated to the metal center. To this mixture, 150 10 molar equivalents of monomer (in all cases described here, a yellow liquid) was added, and the contents of the vial were mixed. The mixture typically began to turn a darker orange-brown color within 15 s. Over the next minute or so, the mixture could be transferred by pipette to a glass slide or other non-interacting substrate such as a KBr die. Here, the mixture hardened into a dark film which could then be removed from the substrate with a razor blade. To specifically terminate the polymerization, the polymer was dissolved and 50 equivalents of benzaldehyde or isovaleraldehyde was added immediately. Samples could then be filtered for study by NMR or GPC. Alternatively, films were rinsed repeatedly at 0 "C with dry pentane and methanol under argon in order to remove soluble components such as residual monomer, catalyst termination products, and substituted benzene produced by back-biting during the polymerization. After rinsing, the films were subjected to dynamic vacuum to remove solvent until a vacuum of < 1 mTorr was achieved. By rinsing at low temperature and protecting the films from light, the highest possible cis content was ensured. _+
10.4.2 Cidtruns Isomerization of Soluble Polycyclooctatetraenes Isomerization of the polymer from a predominantly cis configuration to a predominantly trans configuration could be accomplished either thermally or photochemically. Thermal isomerization was accomplished by heating the sample in benzene or THF at 60-80"C in a tube sealed with a Teflon Kontes screw-top until the visible absorption spectrum showed no change. Photochemical isomerization was accomplished at 0 "C (monitored by thermocouple and external meter in the bath) by exposure of the sample dissolved in THF or benzene to light from a Pyrex-filtered, 350 W, medium-pressure mercury Hanovia lamp (approximately 6-12 h for a sample concentration of 1 mg mL-'). Overexposure resulted in a decrease in color indicating decomposition of the material in solution. THF, toluene and benzene were suitable solvents for this experiment. Chlorinated solvents sometimes lead to photobleaching.
10.4.3 A Precursor Route to Polyacetylene 10.4.3.1 Synthesis of Poly(diethy1 l-oxabicyclo[2.2.l]hepta-2,5-diene-2,3-di~rbo~late) To 120 WL (0.00218mmol) of a stirred 1.82 x M P ~ ( O A C solution )~ in THF in a l-dram/(3.7-mL) glass vial with a plastic screw-cap was added 110 mg (0.463 mmol) of diethyl 7-oxabicyclo[2.2.l]hepta-2,5-diene-2,3-dicarboxylate[154]. The reaction mixture was stirred for 12 h (until it gelled) and then allowed to sit for 36 h, at which point the solid orange mixture appeared dry. A solution of the mixture was made with 5 mL of THF, and after addition of 1 mL of concentrated HCl the solution was stirred for 1 h to deactivate the catalytic endgroup by protonation. Although acid-catalyzed opening of the 0x0 bridge might be expected [155], it was not observed spectroscopically. Addition of 15 mL of methanol caused precipitaProvided by A. Safir and Prof. Bruce Novak [50].
378
10 Polyacetylene
tion of a solid that was collected by centrifuge and washed with methanol (3 x 15 mL) to yield 81.6 mg (74070) of poly(diethyl7-oxabicyclo[2.2.l]hepta-2,5-diene-2,3-dicarboxylate)as a white powder. M,: 54603; M,: 29100; PDI: 1.88 IR (KBr): 1724, 1645 cm-'. 'H NMR (400 MHz): 6 1.27 (br, 6), 2.62 (br, 1.5), 3.93 (br, OS), 4.20 (br, 4), 5.21 (br, I), 5.60 (br, 1). 13C-NMR (100MHz): 6 13.7 (br), 37.7 (br), 44.6 (br), 61.5 (br), 80.8 (br), 82.0 (br), 139.2 (br), 153.2 (br), 160.0 (br), 164.8 (br). Analysis: Calcd. for C,,H405: C, 60.50; H, 5.92. Found: C, 60.21;H, 5.98%.
10.4.3.2 Solid-state Production of Polyacetylene from Poly(diethy1 7-oxabicyclo[2.2.1]hepta-2,5-diene-2,3-dicarboxylate) Single layers of a solution of 5.0 mg of poly(diethyl7-oxabicyclo[2.2.l]hepta-2,5-diene-2,3-dicarboxylate) in 1.0 mL of CH,Cl, were placed on either glass slides (for resonance Raman spectroscopy and conductivity measurements) or NaCl plates (for IR spectroscopy) via syringe. Slow evaporation of CH,Cl, yielded thin films which were placed in a vacuum oven. Evacuation to 100 mTorr for 1 h at room temperature followed by heating at 120°C for 2 h caused the polymer to change from colorless to yellow, then to orange, red, maroon, and to finally leave a purplelblack film of polyacetylene. The samples were allowed to cool to room temperature while remaining under vacuum, and were then rapidly transferred into the dry box and stored under N,. IR (NaCl): 1008 cm-' (lit. 1015 cm-I) [156].Resonance Raman: 1086, 1484 cm-' (lit. 1064, 1460 cm-') [157-1593.
10.4.3.3 Solution Production of Polyacetylene from Poly(diethy1 7-0~ab~cy~lo[2.2.llhepta-2~-diene-2,3-di~rboxy~ate) To a 100-mL flame-dried Schlenk flask was added 200 mg of poly(diethy1 7-oxabicyclo[2.2.1]hepta-2,5-diene-2,3-dicarboxylate).The flask was subsequently placed under vacuum at below 1 mTorr for 12 h. At this time, 25 mL of dry/degassed o-dichlorobenzene was transferred via cannulation into the Schlenk flask, giving a clear solution which was kept under a constant flow of argon. The Schlenk flask was then placed in an oil bath at 120°C.Heating the stirred solution at this temperature for 2 h resulted in a slow color change from clear to purple/black. The solution was allowed to cool to room temperature, and was then transferred via cannulation into a 500-mLsolvent pot under argon containing 250 mL of dryldegassed diethyl ether. This procedure caused aggregation of the purple/black solid. The suspension was transferred via cannulation into centrifuge tubes sealed under nitrogen, centrifuged, and washed with dry/degassed diethyl ether (3 x 25 mL). Drying of the collected product under vacuum yielded polyacetylene as a purple/black solid. IR and resonance Raman spectra were identical to those for the polyacetylene samples obtained in the solid state.
Acknowledgements We thank Shokyoku Kanaoka for the helpful translation of an article, K. B. Wagener for sending us a preprint, and Bruce Novak and Adam Safir for supplying experimental details of their work.
References
319
Abbreviations ADMET AM1 CD COT CPBA MM2 poly( RCOT) ROMP TBS THF TMS
acyclic diene metathesis Austin Model 1 circular dichroism cyclooctatetraene chloroperbenzoic acid molecular mechanics 2 high-molecular-weight polycyclooctatetraenes ring-opening metathesis polymerization tert-butyldimethylsilyl tetrahydrofuran trimethylsilyl
References [l] T. Itoh, H. Shirakawa, S. Ikeda, J. Polym. Sci., Polym. Chem. Ed. 1974, 12, 11-20. [2] H. Shirakawa, E. J. Louis, A. G. MacDiarmid, C. K. Chiang, A. J. Heeger, J. Chem. SOC,Chem. Comm. 1977, 578-580. [3] C. K. Chiang, C. R. Fincher, Jr., Y. W. Park, A. J. Heeger, H. Shirakawa, E. J. Louis, S. C. Gau, A. G. MacDiarmid, Phys. Rev. Lett. 1977, 39, 1098-1101. [4] J. Wilson, J. F. B. Hawkes, Optoelectronics: An Introduction, 2nd ed., Prentice Hall, New York, 1989. [5] D. S. Chemla, J. Zyss, Nonlinear Optical Properties of Organic Materials and Crystals, Vol. 2, Academic, Orlando, FL, 1987. [6] J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks, K. Mackay, R. H. Friend, P. L. Burn, A. L. Holmes, Nature (London), 1990, 347, 539-541. [7] Polyacetylene itself has not found application in light-emitting diodes. trans-Polyacetylene has a very low quantum yield for emission, being instead an efficient photoconductor. See Chapter 6 in [141. [81 T. Masuda, T. Higashimura, Adv. Polym. Sci. 1987, 81, 121-165. [9] M. R. Anderson, B. R. Mattes, H. Reiss, R. B. Kaner, Science 1991, 252, 1412-1415. [lo] A. C. Savoca, A. D. Surnamer, C. Tien, Macromolecules 1993, 26, 6211-6216. [ll] T. Aoki, K. Shinohara, E. Oikawa, Makromol. Chem, Rapid Commun. 1992, 13, 565-570. [12] J. Y. Wong, R. Langer, D. E. Ingber, Proc. Natl. Acad. Sci. USA 1994, 91, 3201-3204. [I31 For a discussion of smart materials, see: R. E. Newnham, Muter. Res. SOC.Bull. 1993, 18, 24-27, and articles that follow. [14] J. C. W. Chien, Pobacetykne: Chemistry, Physics, and Material Science, Academic, Orlando, FL, 1984. [15] N. C. Billingham, P. D. Calvert, Adv. Polym. Sci. 1989, 90, 1-104. [16] C. W. Spangler, R. A. Rathunde, J. Chem. Soc., Chem. Commun. 1989, 26-27. [17] T. A. Skotheim, Handbook of Conducting Polymers, Vol. 2, Marcel Dekker, New York, 1986. 1181 T. A. Skotheim, Handbook of Conducting Polymers, Vol. 1, Marcel Dekker, New York, 1986. 1191 A. J. Heeger, S. Kivelson, J. R. Schrieffer, W.-P. Su, Rev. Mod. Phys. 1988, 60, 781-850. [20] L. M. Tolbert, M. E. Ogle, .l Am. Chem. SOC.1990, 112, 9519-9527. [21] J. Tsukamoto, Adv. Phys. 1992, 41, 509-546. 1221 L. Cruz, P. Phillips, Phys. Rev. B. 1994, 49, 5149-5156. [23] A. B. Kaiser, S. C. Graham, Synth. Met. 1990, 36, 367-380.
380
10 Poiyacetylene
[24] For an example of an interesting (CR), isomer that is not polyacetylene, see: 0. T. Visscher, P. A. Bianconi, J. Am. Chem. Soc. 1994, 116, 1805-1811. 1251 Note that older literature sometimes refers to polyynes as “polyacetylenes”. See for example: P. Cadiot, W. Chodkiewicz in Chemistry ofAcetylenes (Ed.: H. G. Viehe), Marcel Dekker, New York, 1969, pp. 597-647. [26] Y. Lee, M. Kertesz, A Chem. Phys. 1988,88, 2609-2617. (271 J. P. Lowe, S. A. Kafafi, J. P. LaFemina, J. Phys. Chem. 1986,90, 6602-6610. [28] M. Liigdlund, W. R. Salaneck, F. Meyers, J. L. Bredas, G. A. Arbuckle, R. H. Friend, A. B. Holmes, G. Froyer, Macromolecules 1993, 26, 3815-3820. [29] C. B. Gorman, R. H. Grubbs, in Conjugated Polymers (Eds. : J. L. Bredas, R. Silbey), Kluwer, Dordrecht, 1991, pp. 1-48. [30] F. L. Klavetter, R. H. Grubbs, Polym. Muter. Sci. Eng. 1988, 58, 855-858. [31] L. M. Leung, K. H. Tan, Macromolecules 1993, 26, 4426-4436. [32] R. Schlund, R. R. Schrock, W. R. Crowe, A Am. Chem. Soc. 1989, 111, 8004-8006. [33] H. Kuzmany, J. Kiirti, Synth. Met. 1987,21, 95-102. [34] C. B. Gorman, E. J. Ginsburg, R. H. Grubbs, A Am. Chem. SOC. 1993,115, 1397-1409. [35] H. E. Schaffer, R. R. Chance, K. Knoll, R. R. Schrock, R. Silbey in Conjugated Polymeric Materials: Opportunities in Electronics, Optoelectronics, and Molecular Electronics (Eds. : J. L. Bredas, R. R. Chance), Kluwer Academic, Dordrecht, The Netherlands, 1990,pp. 365-376. [36] B. E. Kohler, J. J. A. Pescatore, in Conjugated Polymeric Maferials: Opportunities in Electronics, Optoelectronics, and Molecular Electronics (Eds. : J. L. Bredas, R. R. Chance), Kluwer Academic, Dordrecht, The Netherlands, 1990, pp. 353-364. [37] B. E. Kohler, in Electronic Properties of Polymers and Related Compounds (Eds.: H. Kuzmany, M. Mehring, S. Roth), Springer-Verlag, New York, 1985,pp. 100-106. [38] B. Hudson, B. E. Kohler, Synth. Met. 1984, 9, 241-253. 1391 A. 0. Patil, A. J. Heeger, F. Wudl, Chem. Rev. 1988, 88, 183-200. [40] T. H. Jozefiak, E. J. Ginsburg, C. B. Gorman, R. H. Grubbs, N. S. Lewis, J. Am. Chem. Soc. 1993, 115, 4705-4713. [41] K. Knoll, R. R. Schrock, J. Am. Chem. Soc. 1989,I l l , 7989-8004. [42] K. Knoll, S. A. Krouse, R. R. Schrock, J. Am. Chem. Soc. 1988, 110, 4424-4425. (431 S. A. Krouse, R. R. Schrock, Macromolecules 1988, 21, 1885-1888. [44] For a review see: J. A. Stowell, A. J. Amass, M. S. Beevers, T. R. Farren, Polymer 1989,30,195-201. [45] E. J. Ginsburg, C. B. Gorman, R. H. Grubbs, F. L. Klavetter, N. S. Lewis, S. R. Marder, J. W. Perry, M. J. Sailor in Conjugated Polymeric Materials: Opportunities in Electronics, Optoelectronics, and Molecular Electronics (Eds.: J. L. Bredas, R. R. Chance), Kluwer Academic, Dordrecht, The Netherlands, 1990,pp. 65-81. [46] F. L. Klavetter, Ph. D. Dissertation, California Institute of Technology, 1989. (471 F. L. Klavetter, R. H. Grubbs, X Am. Chem. Soc. 1988, 110, 7807-7813. [48] J. H. Edwards, W. J. Feast, Polymer 1980, 21, 595-596. [49] T. M. Swager, D. A. Dougherty, R. H. Grubbs, X Am. Chem. Soc. 1988, 110, 2973-2974. [50] A. L. Safir, B. M. Novak, Macromolecules 1993,26, 4072-4073. [51] For a review, see: R. Fuks, H. G. Viehe in Chemistry of Acetylenes (Ed.: H. G. Viehe), Marcel Dekker, New York, 1969, pp. 425-495. [52] Y.V. Kissin, in Handbook of Polymer Science and Engineering (Ed.: N. P. Cheremisinoff), Marcel Dekker, New York, 1989, pp. 103-131. [53] T. J. Katz, S. M. Hacker, R. D. Kendrick, C. S. Yannoni, A Am. Chem. Soc. 1985,107, 2182-2183. [54] T. C. Clarke, C. S. Yannoni, T. J. Katz, X Am. Chem. Soc. 1983, 105, 7787-7789. [55] B. E. Kohler, X Chem. Phys. 1988, 88, 2788-2792. [56] P. Robin, J. P. Pouget, R. Comes, H. W. Gibson, A. J. Epstein, A Phys. 1983, 44, 77-81. [57] B. Francois, M. Bernhard, J. J. Andre, J. Chem. Phys, 1981, 75, 4142-4152.
References
381
[58] D. M. Hoffman, H. W. Gibson, A. J. Epstein, D. B. Tanner, Phys. Rev. B. 1983, 27, 1454-1457. [59] B. K. Rao, J. A. Darsey, N. R. Kestner, Phys. Rev. B 1985, 31, 1187-1190. [60] T. Itoh, H. Shirakawa, S. Ikeda, J. Polym. Sci., Polym. Chem. Ed. 1975, I3, 1943-1950. [61] M. Tanaka, H. Yasuda, J. Tanaka, Bull. Chem. SOC. Jpn. 1982, 55, 3639-3640. [62] H. Naarmann, N. Theophilou, Synfh. Met. 1987, 22, 1-8. [63] N. Basescu, Z. X. Liu, D. Moses, A. J. Heeger, H. Naarmann, N. Theophilou, Nature (London) 1987, 327, 403-405. [64] M. A. Schen, F. E. Karasz, J. C. W. Chien, Makromol. Chem., Rapid Commun. 1984, 5, 217-220. [65] B. Wunderlich, Macromolecular Physics, Academic Press, New York, 1976. [66] J. C. W. Chien, G. E. Wnek, F. E. Karasz, J. A. Hirsch, Macromolecules 1981, 14, 479-485. [67] H. H. Fox, M. 0. Wolf, R. O’Dell, B. L. Lin, R. R. Schrock, M. S. Wrighton, J. Am. Chem. Soc. 1994, 116, 2827-2843. [68] K. C. Wallace, A. H. Liu, J. C. Dewan, R. R. Schrock, Organometallics 1989, 108, 644-654. [69] T. H. Upton, A. K. Rapp6, J. Am. Chem. SOC.1985, 107, 1206-1218. [70] Both alkylidenes and Fischer carbenes have been found to polymerize acetylene. See J. P. Collman, L. S. Hegedus, J. R. Norton, R. G. Finke, Principles and Applications of OrganotransitionMetal Chemistry, University Science Books, Mill Valley, CA, 1987, p. 610. [71] M. Leclerc, R. E. Prud’homme, J. Polym. Sci, Polym. Phys. Ed. 1985, 23, 2021-2030. [72] P. Ehrlich, W. A. Anderson in Handbook of Conducting Polymers, Vol. 1 (Ed.: T. A. Skotheim, Marcel Dekker, New York, 1986, pp. 441-488. [73] S. Subramanyam, M. S. Chetan, A. Blumstein, Macromolecules 1993, 26, 3212-3217. [74] J. Kunzler, V. Percec, J. Polym. Sci, Polym. Chem. Ed. 1990, 28, 1221-1236. [75] D. A. Dougherty, Arc. Chem. Res. 1991, 24, 88-94. [76] H. Nishide, T.Kaneko, N. Yoshioka, H. Akiyama, M. Igarashi, E. Tsuchida, Macromolecules 1993, 26, 4567-4571. [77] S. Oh, K. Akagi, H. Shirakawa, K. Araya, Macromolecules 1993, 26, 6203-6206. [78] R. C. Advincula, R. S. Duran, J. LeMoigne, A. Hilberer, Macromolecules 1993, 26, 3895-3903. [79] Y. Hayakawa, M. Nishida, T. Aoki, H. Muramatsu, J. Polym. Sci., Polym. Chem. Ed. 1992, 30, 873-877. [80] J. M. Zeigler, Polym. Prepr. 1984, 25, 223-224. [81] J. M. Zeigler, US Patent App. 760433 AO, 1986; Chem. Abstr. 1986, 20, 157042~. [82] T. Masuda, T. Higashimura, Adv. Chem. SeE 1990, 224, 641-661. [83] T. Masuda, J. Fujimori, M. Z. Rahman, T. Higashimura, Polym. 1 1993, 25, 535-539. [84] T. Masuda, E. Isobe, T. Higashimura, Macromolecules 1985, 18, 841-845. [85] S. Oh, R. Ezaki, K. Akagi, H. Shirakawa, 1 Polym. Sci., PoIym. Chem. Ed. 1993, 3I, 2877-2985. [86] K. L. Kang, S. H.Kim, H. N. Cho, K. Y. Choi, S. K. Choi, Macromolecules 1993,26,4539-4543. [87] Y. Nakayama, K. Mashima, A. Nakamura, Macromolecules 1993, 26, 6267-6272. [88] K. J. Ivin, B. D. Milligan, Makromol. Chem., Rapid Commun. 1987, 8, 269-271. [89] J. Feldman, R. R. Schrock, Prog. Inorg. Chem. 1991,39, 1-74. [90] R. H. Grubbs, W. ’hmas, Science 1989, 243, 907-915. Am. Chem. Soc. 1990, II2, 5384-5385. [91] L. K. Johnson, S. C. Virgil, R. H. Grubbs, .l [92] S. T. Nguyen, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc. 1993, II5, 9858-9859. [93] Metallacyclobutenes have also been isolated from the reaction of titanocene metallacyclobutanes and alkynes; however, these products are stable, and do not go on to give polyenes. See: F. N. Tebbe, R. L. Harlow, J. Am. Chem. Soc. 1980, 102, 6149-6151. [94] L. R. Gilliom, R. H. Grubbs, .lAm. Chem. Soc. 1986, 108, 733-742. [95] B. M. Novak, W. Risse, R. H. Grubbs, Adv. Polym. Sci. 1992, 102, 47-72. [96] J. Kunzler, V. Percec, Polym. Prepr. 1988, 29, 80-82. [97] N. Zhang, R. Wu, Q. Li, K. Pakbaz, C. 0. Yoon, F. Wudl, Chem. Mater. 1993, 5 , 1598-1599. [98] H. Kise, J. Polym. Sci., Polym. Chem. Ed. 1982, 20, 3189-3197.
382
I0 Polyacetylene
[99] Y. Shindo, T. Hirai, Makromol. Chem. 1972, 155, 1 - 16. (1001 A. Baruya, D. L. Gerrard, W. F. Maddams, Macromolecules 1983, 16, 578-580. [loll M. Thakur, Macromolecules 1988, 21, 661-664. [lo21 T. M. Swager, Ph. D. Dissertation, California Institute of Technology, 1988. [lo31 C. A. Jones, R. A. Lawrence, J. Martens, R. H. Friend, D. Parker, W. J. Feast, M. Ltigdlund, W. R. Salaneck, Polymer 1991, 32, 1200- 1209. [lo41 R. S. Kanga, T. E. Hogen-Esch, E. Randrianalimanana, A. Soum, M. Fontanille, Macromolecules 1990, 23, 4241 -4246. [lo51 D. Tao, K. B. Wagener, Macomolecules 1994, 27, 1281-1283. [lo61 L. Dai, J. W. White, Polymer 1991, 32, 2120-2127. [lo71 M. Seto, Y. Maeda, T. Matsuyama, H. Yamaoka, H. Sakai, Nucl. Instrum. Methods Phys. Res., Sect. B 1993, I-4, 287-289. [lo81 Z. Shuai, J. L. BrCdas, Macromolecules 1991, 24, 3723-3724. [lo91 For example, acetylene may be polymerized in ethanol using a NaBH,/Ni(II) system: L. B. Luttinger, J. Org. Chem. 1962, 27, 1591-1596. [110] G. Lieser, G. Wegner, W. Miiller, V. Enkelmann, W. H. Meyer, Makromol. Chem., Rapid Commun. 1980, I, 627-632. [lll] K. J. Ivin, Olefin Metathesis, Academic Press London, 1983. (1121 R. H. Grubbs in Comprehensive Organometallic Chemistry (Ed.: G. Wilkinson), Pergamon, New York, 1982, pp. 499-551. [I131 Note that TiCI,/AIEt, is acting here as a classical metathesis catalyst, in contrast to its behavior as a Ziegler-Natta catalyst with acetylene. [114] J. H. Edwards, W. J. Feast, D. C. Bott, Polymer 1984, 25, 395-398. [I151 J. H. Burroughes, C. A. Jones, R. H. Friend, Nature (London) 1988, 335, 137-141. [116] D. C. Bott, C. S. Brown, C. K. Chai, N. S. Walker, W. J. Feast, P. J. S. Foot, P. D. Calvert, N. C. Billingham, R. H. Friend, Synth. Met. 1986, 14, 245-269. 11171 D. Tao, K. B. Wagener, Polym. P r e p 1994, 35, 522-523. [118] K. B. Wagener, J. T. Patton, Macromolecules 1993, 26, 249-253. [119] K. B. Wagener, J. M. Boncella, J. G. Nel, Macromolecules 1991, 24, 2649-2657. [I201 J. E. O’Gara, J. D. Portmess, K. B. Wagener, Mucmmolecules 1993, 26, 2837-2841. I1211 Y. V. Korshak, V. V. Korshak, G. Kanischka, H. Hacker, Makromol. Chem., Rapid Commun. 1985, 6, 685-692. 11221 P. v. R. Schleyer, J. E. Williams, K. R. Blanchard, J. Am. Chem. Soc. 1970, 92, 2377-2386. 11231 M. J. Sailor, F. L. Klavetter, R. H. Grubbs, N. S. Lewis, Nature (London) 1990, 346, 155-157. [I241 E. J. Ginsburg, C. B. Gorman, S. R. Marder, R. H. Grubbs, J. Am. Chem. Soc. 1989, III, 7621-7622. [125] C. B. Gorman, E. J. Ginsburg, S. R. Marder, R. H. Grubbs, Angew. Chem. Adv. Muter. 1989, IOI, 1603- 1606. 11261 G. I. Fray, R. G. Saxton, The Chemistry of Cyclooctatetraene and its Derivatives, Cambridge University Press, Cambridge, 1978. [I271 G. Schroder, Cycloocfutetruen,Verlag Chemie, Weinheim, 1965. [128] K. Y. Jen, R. L. Oboodi, R. L. Elsenbaumer, Polym. Muter. Sci. Eng. 1985, 53, 79-83. [129] S. D. D. V. Rughooputh, S. Hotta, M. Nowak, A. J. Heeger, F. Wudl, Synth. Met. 1987, 21, 41-50. [130] For a discussion of the role of flexible side chains in solubilizing rigid polymers, see: M. Ballauff, Angew. Chem., Int. Ed. Engl. 1989, 28, 253-267. [131] F. L. VanNice, F. S. Bates, F. S. Baker, P. J. Carroll, G. D. Patterson, Macromolecules 1984, 17, 2626-2629. [132] J. Kao, N. L. Allinger, J. Am. Chem. SOC. 1977, 99, 975-986. 11331 M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J. Am. Chem. Soc. 1985, 107, 3902-3909.
References
383
[134] J. S. Moore, C. B. Gorman, Grubbs, R. H., J. Am. Chem. SOC. 1991, 113, 1704-1712. [135] F. Ciardelli, S. Lanzillo, 0. Pieroni, Macromolecules 1974, 7, 174-179. [136] B.-2. Tang, N. Kotera, Macromolecules 1989, 22, 4388-4390. [137] M. Oki, Applications of Dynamic NMR Spectroscopy to Organic Chemistry, VCH, New York, 1985.
[138] W. v. E. Doering, L. Birladeanu, X. Cheng, T. Kitagawa, K. Sarrna, J. Am. Chem. SOC. 1991, 113, 4558-4563. [139] L. A. Paquette, C. D. Wright, 111, S. G. Traynor, D. L. Taggart, G. D. Ewing, Tetrahedron 1976, 32, 1885-1891. [140] M. J. Sailor, E. J. Ginsburg, C. B. Gorman, A. Kurnar, R. H. Grubbs, N. S. Lewis, Science 1990, 249, 1146-1149. [141] R. Z. Bachrach, in Metal-Semiconductor Schottky Barrier Junctions and Their Applications (Ed. : B. L. Sharma), Plenum, New York, 1984, pp. 61-112. [142] H. Shirakawa, Kobunshi 1993, 42, 33-34. [I431 T. Yamabe, K. Tanaka, E.-H. Terama, K. Fukui, H. Shirakawa, S. Ideda, Synfh. Met. 1979/80, I , 321-327. [144] S. M. Abdelaty, M. Sasai, H. Fukutorne, Prog. Theor. Phys. 1985, 74, 195-210. [I451 B. Furlan, U. Osredkar, A. Sebenik, VeSn. Slov. Kem. Drus. 1987, 34, 393-400. [I461 M. M. Ahmad, A. B. Alimuniar, J. H. Edwards, W. J. Feast, I. S. Millichamp, C. Spanomanolis, Polym. Prepr. 1984, 25, 217-218. [147] K. C. Metzger, W. J. Welsh, Polym. Prepr. 1984, 25, 268-269. [148] D. A. Dixon, B. E. Smart in Selective Fluorination in Organic and Bioorgcmic Chemistry (Ed. : J. T. Welsh), American Chemical Society, Washington, DC, 1991, pp. 18-35. 11491 R. D. McCullough, S. P. Williams, 1 Am. Chem. SOC.1993, 115, 11608-11609. [150] W. P. Hu, K. Levon, K. S . Ho, A. S. Myerson, T.K. Kwei, Macromolecules 1993, 26, 1318-1323. [151] R. D. McCullough, S. Tristram-Nagle, S. P. Williams, R. D. Lowe, M. Jayaraman, J. Am. Chem. SOC. 1993, 115, 4910-4911. [152] H. W. Gibson, J. M. Pochan, in Encyclopedia of Polymer Science and Engineering, John Wiley, New York, 1985, pp. 87-130. [153] R. R. Schrock, R. T. DePue, J. Feldrnan, C. J. Schaverien, J. C. Dewan, A. H. Liu, 1 Am. Chem. SOC.1988, 110, 1423-1435. [154] A. W. McCulloch, B. Stanovnik, D. G. Smith, A. G. McInnes, Can. 1 Chem. 1969, 47, 4319-4326. [155] A. Etienne, A. Spire, E. Toromanoff, Bull. SOC.Chim. Fr. 1952, 750-756. [156] H. Shirakawa, S. Ikeda, Polymer 1971, 2, 231-244. 11571 L. Lichtmann, E. Imhoff, A. Sarhangi, D. Fitchen, 1 Chem. Phys. 1984, 81, 168-184. [I581 E. Ehrenfreund, Z. Vardeny, 0. Brafman, B. Horovitz, Phys. Rev. B. 1987, 36, 1535-1553. [159] P. W. Carter, J. D. Porter, Phys. Rev. B. 1991, 43, 14478-14487.
This Page Intentionally Left Blank
11 Acetylenic Compounds as Building Blocks for HighSpin Molecules and Molecular Assemblies Hiizu Iwamura, Kenji Matsuda
11.1 Introduction The lowest triplet state of acetylene lies at ca. 90 kcal/mol above its singlet ground state [l] and appears to be short-lived. The presence of a .C,H radical in the 193-nm photolysis of acetylene has been proven by time-of-flight mass spectrometry and other spectrometric methods [2].A propargyl radical is detected when methylacetylene is irradiated at 77 K [3]. Since these alkynyl derivatives are all reactive intermediates, it is difficult to consolidate them into high-spin organic molecules and molecular assemblies. On the other hand, quite a few acetylenic compounds carrying authentic unpaired electron centers or precursors thereto as substituents are documented. Some of them have been studied variously for the purpose of assembling their electron spins in molecular solids or in oligomer/polymer molecules. There are also other alkynyl compounds usable as key chemicals for the synthesis of the main skeleton for super-high-spin organic molecules. In this chapter, the authors wish to discuss, in the light of the modern theory of electron spin alignment in molecular systems, how acetylenic compounds can be used as versatile building blocks for the design and construction of high-spin species.
11.2 Alkynyl Compounds Carrying Unpaired Electrons Alkynes carrying unpaired electron centers are classified into three groups depending on in which part of the molecule the unpaired electron resides: in a substituent remote from the alkyne moiety, in a transition metal ion ligated with the alkyne, or a to the C =C triple bond.
11.2.1 Alkynyl Compounds Carrying Unpaired Electrons in Remote Substituents Phenylacetylenes (Table 11-1), l-phenyl-1,3-butadiynes, and a l-phenyl-1,3,5-hexatriyne (Table 11-2) having substituents bearing unpaired electrons shed light on the degree of spin delocalization over the x-conjugated molecular framework. While these compounds are mostly paramagnetic, studies are being carried out actively to explore the long-range exchange coupling that would enable alignment of all the electron spins in parallel in their molecular solid and in polymer molecules. In Table 11-1 are also included alkynyl compounds having precursor groups for generating the radical substituents. In a pioneering study, Rieker et al. prepared a series of 4-( 8-substituted ethynyl)-2,6-di-tertbutylphenols (1 and the analogs), oxidized them to the corresponding phenoxyls 2-5 with alkaline K, [Fe (CN),] solution, PbO, or 2,6-di-tert-butyl-4-cyanophenoxyl under a nitrogen
386
I1 Building Blocks for High-Spin Molecules and Molecular Assemblies
Table 11-1 Phenylacetylenescarrying substituents containing unpaired electrons or precursors thereto on the benzene ring of
1-13
No. R 1 2
3 4 5 6
7 8 9 10
11 12 13
H H CH,
R:,
ESR
Color
4-HO-3,5-(tert-Bu), 4-(-0)-3,5-(terr-Bu), 4-(.0)-3,5-(tert-Bu),
aH = 2.79 G (2H)
[41 Red -violet [41 Yellowish red-brown [4]
a, = 5.65 G (3H), 1.81 G (2H); g = 2.0050 n-Pr 4-(.0)-3,5-(tert-B~), a, = 5.35 G (2H), 1.77 G (2H) a, = 1.79 G (2H); g = 2.0050 tert-Bu 4-(.0)-3,S-(tert-Bu), aN= 11.0 G, aH= 2.6 G (2H), H 4-N(O.)tert-Bu 1.3 G (3H) a, = 8.0 G (2N); H 3-(Ullman’s nitronyl nitroxide)“ g = 2.0066 (in CH2C12), a, = 7.6 G (2N); H C(Ullman’s nitronyl g = 2.0070 (in CH2C1J nitroxide)@) a, = 6.0 G (4N) H 3-Verda~ylyl~) H 4-[p-(n-C,gH~s)C,HdCN2 H 4-Go-(n-C,,H,,)C6H,]C. JD/hcJ= 0.389 ~ m - ’ , JE/hcJ= 0.019 cm-’ (in MTHF) H 3-[p-(n-C,,H,,)C,H3CN, H 3-[p-(n-C,,H3,)C,H,)C- JD/hc(= 0.412 cm-I, IE/hcl = 0.020 cm-’ (in MTHF)
Ref.
Yellowish red-brown [4] Green 141
PI Blue
161
Deep blue
[61
Green Red
[71 [Sl
PI Red
PI I81
atmosphere, and studied to what extent the electron spins would delocalize to the alkynyl moieties 141. The hyperfine coupling constants of the phenoxyls given in Table 11-1 show that a reasonable fraction of the spin is delocalized to the alkyne part. None of them gave the isolable monomer radicals in the solid state. In the case of 4-(3,3-dimethylbutynyl) derivative 5, several dimers were obtained. While the phenoxyl oxygen atoms are sterically protected, the delocalized spins at the p-carbons of the p-alkynyl groups appear to be too reactive toward bimolecular reactions. Nitroxide radical 6 is similar in that it is stable in deoxygenated solution but cannot be isolated as free-radical solid. The n-spin density at the terminal ethynyl carbon atom is estimated from the a, value of 1.3 G to be 0.05 [5]. A pair of Ullman’s nitronyl nitroxide radicals carrying m- and p-ethynylphenyl groups at position 2 of the imidazoline ring (7 and 8) are quite stable at room temperature under air and the unpaired electrons withstand the catalyzed polymerization conditions (see Section 11.4.3) [6]. The terminal ethynyl groups are not susceptible to PbO, in benzene/methano1 at room temperature, which is used for the oxidation of the bis(hydroxy1amine)s to the nitronyl nitroxides [6, 71. Fremy’s salt gives the best result for the oxidation of the hydroxylamine 14 carrying a terminal 1,3-butadiyne moiety [9].
11.2 Alkynyl Compounds Carrying Unpaired Electrons
387
Table 11-2 lAryl-1,3-butadiynes carrying substituents containing unpaired electrons or precursors thereto on the benzene ring of
14-22
No. 14
15
16 17
18 19 20 21 22
R
R:,
ESR
Color
Ref.
u, = 13.8 G ; g = 2.0066 (in hexane) uN = 13.8 G, uH = 3.1 G (lH), 2.1 G (lH), 1.0 G (1H); g = 2.0066
Red
[91 [91
Orange red
JD/hcJ= 0.4666 cm-I, ( ~ / h c=l 0.0210 cm-' Red
(D/hcl = 0.4116 cm-', 1E/hcI = 0.0220 cm-'
ID/hcl = 0.3795 cm-', IE/hcJ = 0.0181 cm-' a, = 13.9 G; g = 2.0059 (in hexane)
[lo] [lo1
PI [81
Red
[81 [81
Orange
1341
Oxidation of the precursor hydrazones to the corresponding diazo compounds 10 and 12 can be carried out with active MnOz but not with yellow HgO [8]. When photolyzed in 2-methyltetrahydrofuran (MTHF) glass at 10 K, they show EPR fine structures characteristic of unoriented triplet species. The zero-field splitting (zfs) parameter ID/hc I of a carbene is governed by the dipolar interaction between the unpaired electrons in the o- and n-orbitals at the carbenic center. The smaller 1 D/hcI value of t h e p isomer 11 than that of the m isomer 13 shows the extended delocalization of the n-spin in the former. The corresponding value diminishes further in the p-substituted 1,3-butadiyne 21 181. Hydrazine hydrate reacts with 1,3-butadiyne moieties in competition with the carbonyl groups. Therefore, tosylhydrazine is employed for the selective reaction with the latter. The sodium salts of the resulting tosylhydrazones are heated to give the diazo compounds 16, 18, and 20 [8, lo]. As some of the 1J-butadiyne derivatives in Table 11-2 undergo spontaneous polymerization in the solid state (see Section 11.4.1), they have to be kept in solution. 1,3-butadiyne units are also known to react with diazo groups [ll] and therefore the diazo compounds such as 16, 18, and 20 must be kept refrigerated.
11.2.2 Alkynes Bonded to Paramagnetic hnsition Metals Paramagnetic organometallic compounds with q '-acetylide ligands continue to be the subject of interest since they were initially reported in the late 1960s. V(C=CPh)Cp, (Cp = cyclo-
388
I1 Building Blocks for High-Spin Molecules and Molecular Assemblies
pentadienyl) is obtained as a black crystalline solid, mp 25°C (dec.), when NaCECPh and V2C12Cp2 are mixed in an equimolar ratio [12]. The similarly prepared olive green is air-sensitive but melts at 83-84°C. An X-ray V(C = C(C6H,Me3-2,4,6)](q5-C5Me4Et), structural study at 243 K showed a linear V - C - C - C(mesity1) skeleton symmetrically disposed between the canted cyclopentadienide rings. The well-resolved wide-line 'H-NMR spectrum also characterized the complex and revealed extensive spin delocalization onto the q'(a)-acetylide ligand [13]. From the reaction in a 2: 1 ratio, V ( C E C P ~ ) ~isCobtained ~~ as a red crystalline solid which shows an effective magnetic moment value of 1.54 pB at 96-274 K (calculated spin only 1.73 pB; see Section 11.7.1) and v(C=C) = 2060 cm-' [12]. The new additions to these organometallic compounds are d7 low-spin q (a)-alkynyl iron(1) complexes of the general formula (PP3)Fe(C= CR) where PP3 is P (CH,CH,PPh,), and R = Ph, SiMe3, n-C3H7, n-C5HII,and C(CH,), (23) [14a]. [(PP3)Fe(H)(N2)]BPh4and excess alkynes H C = CR in THF give [(PP3)Fe(C= CR)]BPh4 as orange crystals, which are then reduced electrolytically or chemically with cobaltocene to give dark red crystals of 23. All iron(I1) alkynyls obtained as the intermediates are air-stable in both the solid state and solution, where they behave as 1 : 1 electrolytes. All compounds are paramagnetic with peff values ranging from 3.28 to 3.42 p B at room temperature. Such values correspond to two unpaired spins and, therefore, indicate a low-spin trigonal-bipyramidal d6 electronic configuration. All iron(1) alkynyls 23 are stable in the solid state and in solution where they behave as nonelectrolytes. They are all paramagnetic with bulk magnetic moments ranging from 1.90 to 2.15 pB. These values are consistent with one unpaired electron (d7 low-spin configuration of the metal).
23
Gas-phase He1 and He11 photoelectron spectroscopy has been used to disclose the bonding interactions in q '-acetylides [146]. The predominant x-interactions between the acetylide ligands and the metal are filled/filled interactions between the occupied acetylide n-bonds and the occupied metal d n orbitals. Such interactions become more extensive in the C=CPh ligand and explain the observed electronic communication along the metal-carbon-carbon atom chain. The acetylide ligands resemble n-donor halide ligands rather than a primarily adonor CH3 ligand and a weak n*-acceptor C = N ligand. A variety of q2(n)-acetylides are also known. Room-temperature reactions of VCp, with alkynes R,C = CR, give V(R,C=CR2)Cp2 [15b]. The structure characteristic of the oxidative addition of an acetylene to vanadocene is well established 1161. A dark green adduct = (-)43.4 G) at g = 1.9968 R1 = R, = C02CH3 shows an eight-line EPR spectrum in solution. The low value of the hyperfine coupling constant, compared with those (I I = 60-75 G)of VCp,C12 and similar complexes of simple anionic ligands, reflects a greater delocalization of the unpaired electron density onto the alkyne ligand. X-ray structural studies confirmed the metallacyclopropene structure [16].
11.2 Alkynyl Compounds Carrying Unpaired Electrons
3 89
Another type of q2(n)-acetylene complex is obtained when Group 11 metal atoms are deposited into a mixture of inert hydrocarbon matrices, e. g., adamantane-containing alkynes. Cu, Ag, and Au atoms (M) react with mono- and disubstituted acetylenes (RC=CH and RC=CR) to give a variety of mono- and diligated complexes, M[RC=CH], M[RC=CR], M[RC=CH],, and M[RC=CR],, and organometallic vinyls, MCH=CR. and MCR=CR. [17]. In a typical example of the reaction of Cu with CH,C=CH, a complex EPR spectrum consists of isotropic lines from the monoalkyne n complex Cu(CH,C = CH) (ucu-63= 1338 G, g = 2.0056), the bis(a1kyne) n complex Cu(CH,C = CH), [AlI (Cu) = 0 G, A I (Cu) = 65 G, All (2H) = A I (2H) = 30 G, 811 = g, = 2.0001, propargyl radical (see Section 11.2.3), and a copper-substituted vinyl radical CuCH=C.(CH,) for which only the sum + a, of the hyperfine coupling constants is estimated to be 535 G [17]. While unusual structural features, catalytic activities [15], and synthetic applications of the paramagnetic organometallic compounds are being studied, their assemblage for potential magnetic materials remains to be explored. Studies of the thermal electron transfer rates and the exchange interaction in diferrocene compounds 24 and 25 revealed that, in the mixed valence states of monocation radicals, valence is relatively localized with thermal electron transfer rates being less than the reciprocal of the "Fe Mtissbauer time scale, i. e., lo-' s in 24. Valence is not localized, with the transfer rates greater than ~ O " S - in ~ 25 [18]. The dication of 24 is diamagnetic.
24
25
11.2.3 2-Propynylidenes A propargyl radical HC=CCH2- is formed from propyne under various conditions: UV
irradiation at 77 K [3], by action of vaporized Group 11 metal atoms [17], etc. It is characterized by EPR hyperfine coupling parameters of uH = 18.5 G (2H), uH = 14.8 G (lH), and g = 2.0025. Ethynylcarbenes are key reactive intermediates in various reactions, such as topochemically controlled solid-state polymerization of some 1,3-butadiyne derivatives (see Section 11.4.1) and formation of a variety of bicyclic ring compounds containing the ethynyl group [19]. The parent acetylenic carbene, 2-propynylidene, was suggested by early EPR experiments to have a linear carbene structure, 26s [20]. Early theoretical studies at the STO-3G level suggested an allenic diradical structure (26 b, C,) [21], but later studies incorporating Merller-Plesset treatment of electron correlation favored a bent carbenic structure, 26c, with a very low barrier to "bond shift" interconversion close to the vibrational energy of the ground state [22]. Recently, McMahon and co-workers performed careful I3C-labeling experiments to settle the controversy [23]. They prepared a pair of labeled compounds, 3-dia~o-l-[l-'~C]and -[3-'3C]propyne. When photolyzed (1 > 472 nm) at 10-15 K in an argon matrix, either diazo compound gave a single common species (Scheme 11-1). Its IR absorptions at 3273w, 3257m,
390
I1 Building Blocks for High-Spin Molecules and Molecular Assemblies
H-EGLH
26a
. c". H
Y'C,
c'
26b
H-EG6 'H 26C
1612w, 547w, 400w cm-' were in excellent agreement with those computed for 26b at the very sophisticated level, and did not exhibit good agreement with the IR spectrum computed for a 1 : 1 mixture of bent carbenic structure 26c labeled at two different sites. The I3C-hyperfine coupling constant observed in the triplet EPR spectrum (D/hc = 0.640 and E/hc = 0.000977 cm-') was determined to be 38 G by simulation. The bond angle at the terminal carbon atoms is estimated to be 155'. The bent allenic diradical structure was corroborated by a sophisticated computational treatment of configuration interaction (QCISD/6-3lG*) [24].
Scheme 11-1 Formation of a single common species in the photolysis of 3-dia~o-l-[-'~C]-and -[3-"~]propynes.
EPR spectra obtained by photolysis of 1- and 3-phenyl-3-diazopropynes (27d and 28d, respectively) in MTHF matrices at 9 K consist of a mixture of two sets of fine structures due to two triplet carbenes a ((D/hcl= 0.543 and JE/hcJ= 0.003 cm-I) and b ((D / hc(= 0.526 and IE/hcI = 0,010 cm-I) [25]. Relative amounts of carbenes a and b carried the memory of the starting diazo compounds; a and b were produced mainly from 27d and 28d, respectively. Carbene b isomerized to a at 70-90 K in MTHF at 44-68 K in isopentane, indicating that a is thermodynamically more stable than b on the triplet ground state potential energy surface, Iwamura and co-workers assigned structures 27c and 28c to a and b, respectively (Scheme 11-2) [25]. Carbene 29c (ID/hc( = 0.482 and IE/hcJ = 0.007 cm-') shows a similar EPR spectral change when annealed at 70-80 K (Scheme 11-3). The more linearized carbene 29c' (ID/hcI = 0.497 and IE/hcI z:O.OOO1 cm-') is considered to be an end-product [26]. In the case of carbene 30, the spectral change on annealing accompanies the decrease of the (D/hc( value [27]. Since the bond-shift isomerization is degenerate, the change is ascribed to the conformational change of the phenyl groups.
11.3 Molecular Crystals of Organic Free Radicals that Carry Alkynyl Substituents
A H
P h 27d
hv MTHF,SK
391
~
a = 27c A (70-90 K)
2.. Ph
28d
I hv
MTHF,9K
/A 'H
' 7 Ph
b=2 8 ~
Scheme 11-2 Formation of 1- and 3-phenyl-2-propynylidenesfrom 1- and 3-phenyl-3-diazo-1-propynes.
30 Scheme 11-3 Formation of carbenes from diphenyl substituted diazoalkynes.
11.3 Molecular Crystals of Organic Free Radicals that Carry Alkynyl Substituents 11.3.1 What Makes Acetylenic Compounds Unique in Assembling Their Molecules? The sp-hybridized carbon atoms of alkynes are more electronegative than those of alkanes and alkenes. The hydrogen of terminal ethynyl groups is therefore acidic enough to form a hydrogen bond with the n-electrons of the adjacent acetylenic molecules as well as the lone pairs of electrons of oxygen and nitrogen atoms. Since the two bonds extending from the ethynylene unit are in opposite directions, alkynes have no strong permanent dipole moment. They do have a permanent quadrupole moment which makes an important contribution to the intermolecular forces. Since the local symmetry of the molecules is linear and the sphybridized carbons are dicoordinated, acetylenic compounds are sterically less demanding for the operation of these attractive forces. As a result, the triple bonds have a higher tendency to stick together. These hydrogen bonds and van der Wads interactions are most dramatically demonstrated by the dimer of acetylene in the gas phase and by the crystals of 1,3-butadiynes in which the packing of the molecules is dictated so that the n-conjugated diacetylene moieties may overlap closely and often in parallel. Attempts have been made to take advantage of the latter opportunity for arranging the free-radical molecules in a unique orientation necessary for the development of ferromagnetic interaction in molecular crystals.
392
11 Building Blocks for High-Spin Molecules and Molecular Assemblies
In a number of cases represented by 31, [28b] however, the molecules pile up uniquely but in the wrong direction or so that adjacent radical centers are still too far apart; the crystalline samples behave just as paramagnetic materials. Any interaction between free-radical molecules is weakly antiferromagnetic; in other words, the electron spins tend to align antiparallel to each other and the temperature dependence of the paramagnetic susceptibility is described by the Curie-Weiss law in which 0 = -1-0 K (see Section 11.7.1) [28,29]. The interaction can be made ferromagnetic in a well-designed case to be described in Section 11.3.3.
31
In the crystal of 1,1,6,6-tetraphenyl-2,4-hexadiyne-l,6-diol and its analogs, solvents or third molecules are included in the vacant holes or channels made by the stacking host molecules [30]. These molecular spaces may include organic free radicals or open-shell organometallics in a specific orientation desirable for the ferromagnetic interaction. This possibility seems to be worthy of further exploration.
11.3.2 A Guiding Principle on Aligning Electron Spins in Parallel between Two Neighboring Molecules It is the rule rather than the exception that the electron spin of an organic free-radical molecule couples antiferromagnetically with those of the neighboring molecules in the solid state. Such an antiferromagnetic coupling arises from the Heitler-London exchange between the electron spins with the opposite sign, and may be understood readily by considering the potential energy surfaces made by the approach of two hydrogen atoms giving a hydrogen molecule. As two hydrogen atoms approach each other from infinity to within ca. 10 A, the Coulombic repulsion between the two unpaired electrons becomes significant, lifts the zeroorder degeneracy, and gives rise to singlet and triplet states of different total energy. It is the singlet pair with antiparallel spins that forms the covalent bond and consequently a hydrogen molecule (Fig. 11-1a). The energy of the triplet pair with parallel spins increases as the distance between the two hydrogen atoms decreases; the triplet state is an excited state of the molecule. The same is true for the two overlapping 2p orbitals containing unpaired electrons of organic free-radical molecules (Fig. 11-1b). Thus it is exceptional that organic free radicals show ferromagnetic interactions in molecular solids. Only a few of such exceptions undergo transition to ferromagnets at or below 1 K (311. Orthogonality in space of 2p atomic orbitals bearing unpaired electrons is one of the conditions for the ferromagnetic interaction (Fig. 11-1c). According to another guiding principle presented by McConnell in 1963 [32], the electron spins of two adjacent radical molecules will be allowed to align in parallel when the delocalized fractional spins satisfy the Heitler-London exchange of the spins of opposite sign at the sites of the contact between the two molecules, as shown in Fig. 11-1 d. A direct contact, as in Fig. 11-1b', is of course antiferromagnetic. The theory has been verified experimentally by a set of prototype molecules in which the
11.3 Molecular Crystals of Organic Free Radicals that Carry Alkynyl Substituents
393
Figure 11-1 Schematic drawings of the spin alignments in two atoms and molecules in contact: (a) two hydrogen atoms; (b) two unpaired electrons in two overlapping 2p atomic orbitals; (b‘) interaction between two 2p orbitals in x-conjugated molecules; (c) two unpaired electrons in two orthogonal 2p atomic orbitals; and (d) the McConnell mechanism for the ferromagnetic coupling of the electron spins in x-conjugated molecules. The shapes of the 2p orbitals are not shown for simplicity in (b’) and (d).
orientation of the two triplet diphenylcarbene molecules is fixed by incorporation into a rigid [2.2]paracyclophane skeleton. It is the pseudoortho and pseudoparu isomers (320 and 32p, respectively) in which two S = 1 spins align in parallel to make the quintet ground states (S = 2). The two S = 1 spins cancel each other out in the pseudometu isomer 32m [33].
P h 4
Ph-f
320
32m
32P
11.3.3 Crystals of Antiferromagnetic l$-Butadiyne and Ferromagnetic 1,3,5-Hexatriyne Both Carrying Stable 4-Chloro-3-(N-tert-butyl-N-oxyamino)phenyl Radical Crystalline samples of 1,3-butadiyne 14 [9] and 1,3,5-hexatriyne22 [34] carrying a 4-chloro3-(N-tert-butyl-N-oxyamino)phenyl substituent at the terminal positions of the conjugated
394
\ C
I1 Building Blocks for High-Spin Molecules and Molecular Assemblies
-.
5
9-611
'
14s
22s
triple bonds, show quite contrasting magnetic properties, i. e., antiferro- and ferromagnetic intermolecular interactions at cryogenic temperatures, respectively, and provide additional support for the guiding principle of how to align electron spins in parallel in molecular assemblies. In Fig. 11-2 are shown the temperature dependences of their effective magnetic moments peff [34]. The observed magnetization curve for 22 has been analyzed in terms of the Brillouin function (see Section 11.7.1) to give the average spin quantum number values S of 0.9 at 2.0 K, 0.7 at 4.2 K, and 0.6 at 10 K; the neighboring unpaired electrons of 22 in crystals start to align in parallel at 10 K and form, on average, a parallel pair at 2 K.
%
1.4-
1.2-
Since dimer-like structures are found in one-dimensional arrangements of radical 14 in the single crystal (Fig. 11-3b), the temperature dependence of its peffwas analyzed in terms of a linearly arranged dimer model to give the optimized exchange coupling parameter J/kB= -1.2(3) K. The observed stacking pattern 14s is responsible for the negative (antiferromagnetic) J value. A one-dimensional Heisenberg model [5] was applied with success to
11.4 Spin Alignments in Poly(phenylacety1ene) and Poly(I,3-butadiyne)s
395
analyzed the observed susceptibility data for 22 in the whole temperature range studied. A positive (ferromagnetic) exchange coupling parameter of J/k, = + 1.4(8) K is derived and ascribed to the stacking pattern 22s [34]. Thus a subtle change in the relative orientation of the adjacent molecules in the one-dimensional stacking in the crystals, probably due to the linearly extruding 1,3,5-hexatriyne unit relative to the 1,3-butadiyne unit, leads to quite an opposite effect. While the transition to a ferromagnet has not been observed for 22, this strategy appears to be a promising method for exploring the ordering of spins in molecular solids in general and alkynyl radicals in particular [35].
+ = -
(a)
(b)
Figure 11-3 X-ray crystal structures of 14 viewed (a) along the b axis and (b) along the a/c diagonal.
11.4 Spin Alignments in Poly(phenylacety1ene)s and Poly(l,3-butadiyne)s 11.4.1 Natural Spins Detected during the Solid-state Polymerization of Id-Butadiynes Some 1,3-butadiynes undergo smooth topochemical polymerization in their crystalline states by ultraviolet or high-energy irradiation or by heat. Each molecule is bonded with two adjacent neighbors along a particular crystallographic direction by a sequence of 1,Caddition reactions [36]. Since the crystalline polymers obtained have the extended x-conjugation that is closely responsible for the high third-order nonlinear optical properties, photoconductivities, and other physiochemical properties of great interest, the reactions drawn in Scheme 11-4 are well documented [37]. The crystal of a monomer 1,3-butadiyne acts as a preformed lattice for the polymer crystal. Thus the monomer molecules should stack with the repeating unit of
I1 Building Blocks for High-Spin Molecules and Molecular Assemblies
396
33
Scheme 11-4 Topochemically controlled solid-state polymerization pattern of 1,3-butadiynes.
A to form a one-dimensional chain in which the bond-forming carbons in the adjacent molecules are situated in proximity (s = 3.4-4.0 A). The molecular axis must be slanted by y = 45 from the direction of the one-dimensional chain. Takeda and Wegner suggested a carbene-like structure for the active species of the topochemical reaction [38]. A strong EPR spectrum with a conspicuous fine structure is observed for the partially polymerized samples. This fine structure is due to the magnetic dipole interaction of two or more unpaired electron spins. It is strongly dependent on the direction of the external magnetic field. By the analysis of this anisotropy, various triplet carbenes and quintet dicarbenes are unambiguously identified. The most detailed studies have been made on 2,4-hexadiyne-l,6-diylbis(g-toluenesulfonate) (33, R = CH20Ts in Scheme 11.4) [37]. When the monomer crystal of 33 is irradiated with UV light / A 2 310 nm) for a short period of time at 4.2 K, a large number of different species are detected by absorption and EPR spectroscopic methods. They are stable at liquid helium temperature. Subsequent annealing in the dark produces further intermediate reaction products from which more and different reaction products are produced by further irradiation with visible light. Six1 classified the different series of intermediate products into three categories and found that the diradical and dicarbene series, DR2-D&, DC,-DC,, ..., are genuine intermediates from the monomer to the polymer [39]. When the monomer crystal is irradiated with a UV flash, the dimer DR2 is formed from the monomer by a photoreaction which is followed by a series of thermally activated monomer addition reactions leading via higher DRs and DCs to the polymer (Fig. 11-4). The DRs have been characterized mostly by laser flash photolysis. A single UV laser flash at 4.2 K produces DR, which shows a characteristic 0-0 transition at 422 nm and is stable at this temperature. Annealing the sample in the dark produces DR3, DR,, DR5, and DR6 by addition of one monomer per step. Each intermediate product is detected at the maximum of its optical absorption: DR3 at 514, DR4 at 578, DR, at 664 nm. The whole reaction runs in a 10 ms time scale at room temperature. From the temperature dependence of the rate constants for the decays, the thermal processes are found to have activation energy values of 5-7 kcal mol-'. The DCs have been characterized by their EPR fine structures. When a crystal of perdeuterated 33 was irradiated with UV light (A = 313 nm) at 4.2 K for about 1000 s, an EPR spectrum containing various triplet and quintet species was obtained. The resonance fields for fixed microwave frequency are strongly anisotropic and this different angular dependence of the resonance fields was used to assign the signals to DC,-DCI3. The temperature dependence of these EPR signals shows one common feature. The EPR intensities vanish toward absolute zero temperature, indicating that the quintet dicarbenes have singlet ground states. Since at higher temperature the intensities also vanish in accordance with the Curie law, they 4.9-5.1
O
397
11.4 Spin Alignmenfs in Poly(phenylacety1ene) and Poly(l,3-butadiyne)s
kT
F
c,
.c:R
Figure 11-4 Schematic drawing of the formation of the polymer via a series of DRs and DCs from the photoreaction of 1,3-butadiyne 33 in crystals.
exhibit maximum intensities for which the temperatures are different for different DCs. From these EPR fine structures and their temperature dependence, dicarbenes DC,-DC,, have been fully characterized, as summarized in Table 11-3 [40]. Table 11-3 EPR parameters for dicarbenes DC9-CDI2[40b] Dicarbene
D/hc/cm
E/hc/cm
DC9
0.296(6) 0.293(1) 0.292(1) 0.290(1)
-0.0080(6)
DC,O DCII DCl2
-I
- 0.0071(1) -0.0072(2) -0.0069(1)
RI2/A
AEsQ/cm-'
MSQ/crn-'
> 12 > 12 > 12
0.52(3) 1.55(5) 4.5 (1) 15 (2)
2.2(8) 4.7(4) 14.0(1)
> 12
Iwamura and co-workers tried to generate the dicarbene 35 corresponding to DC2 independently [25a]. When a solution of the corresponding didiazo compound 35d in MTHF was irradiated (A > 540 nm) at 4.2 K, absorptions due to the starting material disappeared rapidly with instantaneous grow-in of absorptions at ca. 260 nm with vibrational fine structure, suggesting efficient formation of a terminal alkyne. l-Phenyl-1,3-butadiyne was actually isolated in ca. 70% yield after annealing and usual work-up (Scheme 11-5). When the photolysis was monitored by EPR (9.443 GHz) spectroscopy, a pair of broad plus/minus signals in a derivative mode with a separation of ca. 110 G grew in at g = 2. A randomly oriented triplet diradical, e. g., 34, corresponding to structure DR, with a small zfs parameter D is suggested to be responsible for the signals. These signals disappeared on continued irradiation or by an-
398
I1 Building Blocks for High-SpinMolecules and Molecular Assemblies
N2
35d
34
Scheme 11-5 Formation of 1,3-butadiyne by the photolysis of didiazo compound 35d.
nealing the sample solution in the dark at 40 K. Neither intermediate triplet monocarbene nor quintet dicarbene corresponding to DC2 was detected. In the IR spectrum obtained by irradiation at 15 K of a PVC film doped with the didiazo precursor, an absorption at 3300 cm-' was observed characteristic of terminal acetylenic C - H stretching as soon as the irradiation was started. All the spectroscopic evidence at cryogenic temperature suggests that the authentically generated dicarbene 35, similarly to DC,, should be an excited state not thermally accessible at cryogenic temperature, that the diradical 34 may be persistent at temperatures below 40 K, and that the center bond would be readily cleaved thermally as well as photolytically. The retrodimerization reaction is fairly reasonable in terms of the thermochemistry of the dimer formation. The enthalpy of formation of the two carbene centers is positive and cannot be compensated for by the formation of one C = C bond. It is only after the polymerization proceeds to a certain degree and the x-conjugation is extended, that the enthalpy of the reaction becomes negative. Thus, although the carbenes are fully analyzed, the concentration of the spins is far lower than the stoichiometric amount. Therefore they are meaningful mechanistically but less important from the point of view of using them to assemble unpaired electrons.
11.4.2 Topological Control of the High-Spin vs. Low-Spin Ground States of n-Conjugated Diradicals and Dicarbenes In the previous section the two carbene centers in DCs interact weakly to produce the singlet ground states; the thermally accessible quintet states are observable. According to MO and valence-bond theory on the electron spin polarization in alternant hydrocarbons [41], the electron spins of n-conjugated radicals are polarized so that up-spins and down-spins appear alternately on the carbon centers containing the x-electrons. Therefore, when a second radical center is placed in phase with this spin polarization, the triplet state with the two spins aligned in parallel will be the ground state. The electron spin of a second radical center placed out of phase with the spin polarization due to the first radical center should align antiparallel to that of the latter producing the singlet ground state (Fig. 11-5). A valence-bond theoretical expression is summarized in Eq. (1):
s = (n* - n ) / 2
11.4 Spin Alignments in Pory(pheny/acetylene)and Poly(l,3-butadiyne)s
399
Figure 11-5 Schematicpresentationof the spin alignments in n-conjugated diradical molecules: (a) triplet and (b) singlet diradicals.
where n* and n stand for the numbers of the starred and unstarred carbons carrying the nelectrons, and S is the ground-state spin, In all the DCs in Section 11.4.1,n* = n and, therefore, the singlet ground states had been predicted as observed. One of the most classical cases of triplet vs. singlet diradicals due to n* - n = 2 vs. 0 is found in Schlenk (36)and Thiele (37) hydrocarbons that have S = 1 and S = 0 ground states, respectively. The radical centers may be replaced with triplet carbene centers to produce m- and p-phenylenedicarbenes 38 and 39 [42]. Since the o-spin is added in parallel with the n-spin at each carbene center, they have quintet and singlet ground states, respectively.
Q Q 36
38
37
$ 39 I
In order to demonstrate the topological control of the high-spin vs. low-spin ground states, several tolane (40)and 1,4-diphenyl-1,3-butadiyne (41) derivatives have been studied. When the interactions of two carbene or nitrene centers, one on each phenyl ring, were studied [43,441 a general rule was obtained showing that o,m'- and p,m'-regiochemistry produces quintet ground states that are well separated from the excited singlet states. The 0 , ~ ' and p,p'-parity leads to singlet ground states. In the m,m' topology, the singlet is again a ground state but the quintet state is readily observable by thermal population [41,431. The om' and p,m' topology is important in designing and constructing super-high-spin organic molecules. Pentanitrene 42 was demonstrated to have an S value of nearly 5 [45].
400
I1 Building Blocks for High-Spin Molecules and Molecular Assemblies
. .
40 (X = Ph9-, *N-)
41 (X = Phk-, &)
42
11.4.3 Attempts at Introducing Stoichiometric Amounts of Spins in Poly(phenylacety1ene)s and Poly(phenyldiacety1ene)s In 1986 the scientific community was excited at the news from Ovchinnikov and co-workers that black polymeric products obtained by the thermal, photochemical, and glow-discharge treatment of bis(nitroxide) diradical 31 contained a fraction which exhibited field-dependent magnetization corresponding to a small amount (0.1 070) of an organic-based ferromagnet [28]. The poly(l,3-butadiyne) skeleton formed by the polymerization might have mediated the exchange coupling of the nitroxide radical centers in the side-chains as in 42. Cao and COworkers reported in 1988 a supporting observation of a 0.7% ferromagnetic contribution for the thermally treated butadiyne 31 [46]. Later, Miller and co-workers tried to reproduce the reports without success; the zero-field magnetization at 90 K was extrapolated to 0.002 emu g-’ which would correspond to about 5 ppm by weight of iron contamination, which is consistent with X-ray fluorescence analysis on samples prior to thermal treatment ~91. Thus, while the production of purely organic ferromagnets was not substantiated, the problematic reports encouraged the design and construction of organic magnetic materials on the basis of the x-conjugated polymers that have pendant radical centers positioned stoichiometrically and in the right topology.
11.4.3.1 Poly@henylacetylene)s The ethynyl groups of ethynylbenzenes undergo polymerization by the action of Group 6, 7, and 9 transition metal catalysts, often used for olefin metathesis, to give the corresponding polyenes which are frequently called “poly(pheny1acetylene)s” (43) [15, 471. A pair of Ullman’s nitronyl nitroxide radicals bearing m- and p-ethynylphenyl groups at position 2 of the imidazoline ring (7 and 8 ) undergo polymerization in the presence of
11.4 Spin Alignments in Poly(phenylacety1ene) and Poly(l,3-butadiyne)s
43
401
I .
44
Rh(COD)(NH3)Cl (COD = 1,5-cyclooctadiene) in ethanol and ethanol/benzene, respectively, at room temperature [6].The greenish homopolymer 43 ( R = m-nitronyl nitroxide) from the m-isomer is soluble in ordinary organic solvents and has an approximate molecular weight [vs. standard polystyrene by gel permeation chromatography (GPC)] on the order of 150000. The dark blue homopolymer 43 (R = p-nitronyl nitroxide) is insoluble. EPR spectra of both of them show a single line of width 7-8 G. The nitronyl nitroxide radicals in the side-chains are estimated from elemental analyses, EPR signal intensities, and Curie constants to be present more than 95% intact. The reciprocal vs. T plots give straight lines characteristic of paramagnetic species and are analyzed in terms of the Curie-Weiss law to reveal very weak antiferromagnetic couplings (0 = - 1.7 and - 1.5 K for the m- and p-polymers, respectively). The magnetization vs. magnetic field strength data on the two isomeric samples at 1.8 K deviated slightly downward from the Brillouin functions with S = 112 (see Section 11.7.1), confirming the presence of antiferromagnetic coupling between the S = 112 spins. The expected ferromagnetic coupling among the radical centers in the side-chains through the conjugated main chain is concluded not to be operative in these polymers. UV-vis absorptions of the polymers agreed nicely with the sum of those of poly(phenylacety1ene) and the nitronyl nitroxide. The main chain is considered to consist of the conjugated double bonds with the s-transoid cis configuration. Torsion around the bonds connecting the phenyl groups with the main chain must be considerable, owing to steric hindrance. The exchange interaction between the contiguous radical centers may not be as strong as designed, although the radical centers are predicted to couple ferromagnetically on the basis of topological symmetry considerations. Phenol 1 is polymerized in the presence of transition metal complex catalysts, e. g., WCI,, Mo(CO), [48]. While polymer samples with reasonable molecular weight are obtained, all attempts to oxidize the phenols to the phenoxyls are far from satisfactory; EPR studies revealed that the concentration of the unpaired electrons in the chemical oxidation products of the polymer is never higher than 30%. In the polymer, the local concentration of the phenoxyls must be quite high at the beginning. Just as 2-5 cannot be obtained in the solid state, both intra- and interchain free-radical coupling reactions must be very extensive. Attempts at avoiding this difficulty by lowering the local concentration of the unpaired electrons by copolymerization of 1 with phenylacetylene followed by oxidation proved not to be satisfactory, probably because the neighboring radical centers are too far apart to communicate with each other. With the idea of avoiding the potential bimolecular coupling reaction of the radical centers in the solution-phase chemical oxidation reactions, a photochemical approach was adopted. Diazo compounds 10 and 12 were treated with the Rh catalyst under basic conditions to give the poly(acety1ene)s 43 [R = p - and m-[p-(n-C,9H39)C6H4]C(N2)] of M, 200000 [8]. While photolysis of the diazo groups proceeded smoothly on neat films at 2 K and broad EPR si-
x
402
11 Building Blocks for High-Spin Molecules and Molecular Assemblies
gnals were observed in the g = 2 region, magnetization curves obtained were fitted to a set of Brillouin functions with S = 2.0-2.5. The alignment of the spins is limited to no more than two or three neighbors. Even in the glassy state at cryogenic temperature, recombination of the two triplet carbene centers created next to each other may not be prohibited.
11.4.3.2 Poly(phenyldiacety1ene)s
l-Phenyl-1,3-butadiynes and 1,3,5-hexatriynes can be polymerized to poly(phenyldiacety1ene)s (44) in the solid state. As shown in the crystal structure (Fig. 11-3), 14 appears as if the packing
of the molecules satisfied the conditions for polymerization described in Section 11.4.1. Unfortunately, however, the distance between the two acetylenic carbon atoms defined as s in Scheme 11-4 is 4.75(8) A in 14 and longer than the required length. As a result, the crystal is quite stable both thermally and under irradiation with UV and y-rays. The crystals of the corresponding hydroxylamine are readily polymerized at 120 "Cin 24 h [9]. The hydrogen bond between the neighboring molecules seems to be effective in shortening the distance s. Mixed crystals of the hydroxylamine and 14 (70: 30 to 50 :50) did undergo polymerization at 140 "C in 20 h to give a black, insoluble solid. The magnetic susceptibility of the polymer samples showed that ca. 90% of the nitroxide radical centers were lost during the polymerization. Most of the remaining spins are S = 1/2, and the rest are in a segment where S > 112. The latter spins were found to be quenched by an anomalous phase transition at ca. 250 K when the samples were annealed from cryogenic temperature in the magnetic field (91. Solid-state polymerization of diazo compounds 18 and 20 was monitored by IR absorptions at 66 and 35 "C,respectively. While an absorption at 3295 cm-' disappeared smoothly, more than 87% of an absorption at 2035 cm-' remained intact. Here again, the magnetization curves obtained did not obey the Brillouin function with any single S value but with a combination of S values in the range 2-2.5. The alignment of the spins is limited to a short range. It is noted that there are at least two polymerization patterns in asymmetrically substituted diacetylenes. It is only 44, but not 45, that has the right topology for the alignment of the spins on substituent Rs. Otherwise, the unsatisfactory results so far obtained are ascribed either to the low-spin polarization of the n-bonds connecting the contiguous spin centers or to the poor sample preparations that have lower spin concentrations that the stoichiometric amounts. In principle, the strategy of constructing super-high-spin molecules on the basis of n-conjugated polymers carrying radical centers as pendants appears to be justified.
45
n
11.5 Cyclotrirnerization Reaction of Benzoylacetylenes
403
11.5 Cyclotrimerization Reaction of Benzoylacetylenes in the Presence of a Secondary Amine The one-dimensionally aligned spins discussed above are said to be unstable thermodynamically as well as chemically. Since the enthalpy of the ferromagnetic interaction between the neighboring spins is not much greater than the thermal energy &, the fully spin-aligned system is unstable in terms of entropy. Therefore it should be difficult to realize the ordered spins at a finite temperature other than absolute zero. Once an unpaired electron fails to be generated or is destroyed chemically, the mechanism for the ferromagnetic spin interaction will be interrupted at that site. The chance of such chemical events will increase as the chain length increases [49]. For these reasons, it is highly preferable to pursue two-dimensional and threedimensional spin alignment rather than one-dimensional. From these considerations, a twodimensional honeycomb-shaped sheet structure 46 is proposed as a strong candidate for purely organic ferromagnets [50]. With this structure as a long-range goal, several attempts have been made to synthesize the key partial structures involved in structure 46. The two-dimensional structure is characterized by the presence of a 1,3,5-benzenetriyl unit, another robust ferromagnetic coupling unit in addition to the rn-phenylene unit [51]. I
I
I
I
I
46
When a monosubstituted acetylene R C i C H is heated with nickel or cobalt carbonyls, it gives the 1,2,6trisubstituted benzene as major product, the 1,3,5-benzene as minor product, and little if any of the 1,2,3-isomer. Coordination of the metal with the triple bond has been considered to be involved [52]. Thus, in the transition-metal-catalyzed cyclotrimerization, the 1,3,5-benzenetriyl unit strictly required for the ferromagnetic coupler is not easy to make. Various benzoylacetylenes were more recently found to be trimerized to give the corresponding 1,3,5-tribenzoylbenzeneswhen heated with a catalytic amount of diethylamine. In 1967 Sasaki and Suzuki observed that 1-(2-furyl)-2-propyn-l-o1was oxidized by active manganese dioxide to give 1,3,5-tri(2-furoyl)benzene in a higher yield than the expected 2-propynoylfuran [53]. The first example of this kind of trimerization in a preparative method is the formation of some 1,3,5-triaroylbenzenesby heating benzoylacetylenes in N,N-dimethylformamide (DMF)
404
I I Building Blocks for High-Spin Molecules and Molecular Assemblies
reported by Balasubramanian et al. in 1980 [54]. In view of the potential versatility of these reactions in the design of branched and dendritic structures for super-high-spin polyradicals and polycarbenes, Matsuda and Iwamura have studied these reactions in more detail and found that a trace of diethylamine decreased the amount of by-products. A large amount of diethylamine gives y-(diethy1amino)propenoylbenzene at the expense of 1,3,5-tribenzoylbenzene, verifying a Michael addition/trimerization/elimination-typemechanism for the formation of the benzene ring. A variety of aryl ethynyl ketones undergo trimerization in the presence of a trace amount of diethylamine (Table 11-4) [53-551. Moreover, not only homotrimerization but also crosstrimerization can be performed (Table 11-5) [55]. The cross-trimerized products always contain a mixture of four kinds of trimers when starting with two kinds of ethynyl ketones, but these products can be separated by GPC. Table 11-4 Trimerization reaction of Ar -CO -C = CH catalyzed by diethylamine
Ar
Reaction time/h
Yield/%
Ref.
Table 11-5 Cross-trimerization of Ar, - CO - C t CH and Ar2- CO - C = CH
Ph (2 equiv.), 3-(Me3Si-C = C - CO)C& 4-terf-Bu-C6H4(2 equiv.), 3-(Me3Si- C = C - CO)C6H,
Reaction time/h
Yield/%@)
Ref.
24 12
31 12
[551 PSI
The yield is for the main unsymmetrical product carrying one Me,Si-CrC group out of the four possible triketone products. (a)
The hexa- and nonaketones thus prepared were converted to the corresponding polydiazo compounds via oxidation of their hydrazones [SOa]. The corresponding polycarbenes 47 and 48 were produced by photolyzing the polydiazo compounds in MTHF solid solutions at cryogenic temperature in a sample basket suspended in a Faraday magnetic balance. The magnetization measurements were carried out in situ to obtain the curves which obeyed the Brillouin functions in which S = 6 [SOa, 561 and 9, [Sob], respectively (Fig. 11-6), the highest
11.5 Cyclotrimerization Reaction of Benzoylacetylenes
405
47: n = 6 48: n = 9
s=9
m
5
v
S=6
C 0
,-
. I -
.-3 10 a)
s=5
c1
s=4
K
3
E
s=3 S = 512
1.o
2.0
3.0
H / T (Tesla / K)
Figure 11-6 Dependence of the magnetization of various paramagnetic species on the external magnetic field strength (H) divided by temperature ( T ) : at 2.1 (o), 4.8 ( 0 ) and 10.0 K (A). The increase in the magnetization with H/T is at first linear (region l), then slows down (region 2), and finally levels off (region 3). See text for explanation, The top two curves representing S = 9 and 6 are for polycarbenes 48 and 47, respectively.
spins of organic molecules reported to date. These polycarbenes are still paramagnetic but show a sign of restriction of the rotation of the magnetic moments at cryogenic temperature. Furthermore, since the ethynyl group can be protected by silylation, the resulting trimerization products may be used as second starting materials after deprotection. By using this strategy, construction of the “Starburst” dodecaketone 49, which is the precursor of dodecacarbene contained in the network polymer 46, was successfully performed (Scheme 11-6).
406
I1 Building Blocks for High-Spin Molecules and Molecular Assemblies
Two kinds of ethynyl aryl ketones were chosen as starting units. One was ethynyl ketone 50 and the other was rn-propynoylphenyl trimethylsilylethynyl ketone 51. Cross-trimerization of these units afforded a mixture of four kinds of trimers: trimer 53 from 3 mol of 50, 2: 1-trimer 52, 1:2-trimer 54, and trimer 55 from 3 mol of 51. These trimers could be separated by GPC. The trimer 52 was subjected to the second trimerization reaction after deprotection to afford dodecaketone 49 [55 b]. h
/
50
toluene (31 %)
52
TMS 51
3:O trimer
1:2 trimer
54
0
KF
*
MeOH (78 %)
toluene
0
0
do 56
*
EQNH (3!j %)
49
Scheme 11-6 A synthetic route employed for dodecaketone 49.
0:3 trimer
55
11.5 Cyclotrimerization Reaction of Benzoylacetylenes
407
Another cross-trimerization reaction was employed to corroborate mechanism A in which one ketoenamine and two ethynyl ketone molecules form a six-membered ring from which the diethylamino group is intercepted by another molecule of ethynyl ketone (Scheme 11-7). There are two more mechanisms, B and C that are conceivable (Schemes 11-8 and 11-9, respectively). The ketoenamine is regenerated and plays the role of the catalyst in mechanism B, and the ketoenamine is in equilibrium with the ethynyl ketone in mechanism C.
0
NEt2
\
o
A r 1 w A r 3
&Ar2
Scheme 11-7 Mechanism A for the cyclotrimerization of ethynyl ketone in the presence of diethylamine
I 1 &
Ar
- ArE
Scheme 11-8 Mechanism B for the cyclotrimerization of ethynyl ketone in the presence of diethylamine.
408
I I Building Blocks for High-Spin Molecules nnd Molecular Assemblies
Scheme 11-9 Mechanism C for the cyclotrimerization of ethynyl ketone in the presence of diethylamine.
When refluxed in toluene, y-(diethy1amino)propenoylbenzene 57 (0.44 mmol) and 4-ferfbutylbenzoylacetylene 58 (0.90 mmol) gave ketoenamines 57 and 59 in 40.5 and 48.6% yields (based on starting 57), and triketones 60 and 61 in 25.1 and 62.9% yields (based on starting 58) respectively (Scheme 11-10) [55 b]. No triketone carrying two or three unsubstituted phenyl groups originating from 57 was found in the trimer fraction. These results are consistent with mechanism A. Only 60, a trimer of 58, should have been formed in mechanism B, and the products would have been a mixture of four kinds of the trimers of all combinations of benzoylacetylene and p-tert-butylbenzoylacetylenein mechanism C . A more quantitative analysis
Toluene reflux
NEQ 58
\
59 61
Scheme 11-10 The cross-trimerization reactions employed for differentiating mechanisms A-C.
11.7 Experimental Procedures
409
has been made by using the recurrency formula on the assumption that the rate-determining step is the trimerization process rather than the trapping of the diethylamino group of the dihydrobenzene by ethynyl ketone, and that the reaction rates of 57 and 59 with 58 similar. The computed product ratios of 0.95 for 59/57 and 2.7 for 61/60 in mechanism A are in good agreement with observations. Since the reactions are ionic, the desired 1,3,5-regioselectivity is ideally kept.
11.6 Conclusions The uniqueness of the C = C bonds in the molecular assembly of alkynes lies in their attractive molecular forces. One can take advantage of this opportunity of stacking the radical molecules in the designed orientation in the solid state. A limited number of alkynyl compounds bearing unpaired electrons have been tested to show ferromagnetic intermolecular interaction in crystals. Since a considerable number of paramagnetic alkynyl organometallic compounds containing vanadium(IV), iron(I), etc., are now known, their magnetic properties should be studied systematically. One of the advantages of q'-alkynyl ligands is the lower steric demand of the locally linear structures which will enable us to obtain novel paramagnetic alkynyl organometallic compounds which might not be possible with the alkyl or aryl ligands. It is also of interest to prepare high-spin polyorganometallic compounds for which there is almost no precedent. Structures like 46 will not be possible without using the rich chemistry of alkynes. In view of the current great interest in molecule-based magnetic materials [57], acetylenic compounds will continue to play their important roles.
11.7 Experimental Procedures 11.7.1 Characterization of Magnetic Properties The magnetization curve in which the magnetic moment per ion or molecule, I = M/N, is plotted against the external magnetic field strength H divided by the absolute temperature T is quite useful in characterizing paramagnetic species [58]. In the absence of an external magnetic field, electron spins are randomly oriented to show no magnetization (the origin in Fig. 11-6). As H i s increased at a given temperature, the spins start to align in proportion to the field strength (region 1). Since the spins undergo thermal vibration at a finite temperature, their direction also fluctuates and is dictated by the Boltzmann distribution (region 2); the magnetization curve is described by the Brillouin function. It is only under a strong applied field and at low temperature that all the spins align parallel to the field (region 3). In region 1, where magnetization is linearly proportional to H/T, the paramagnetic susceptibility x defined by I/H is inversely proportional to T This relation is called Curie's law, and the proportionality constant C is a Curie constant that should be 0.37 and 1.0 K emu mol-' for S = 1/2 monoradicals and triplet diradicals, respectively. The product X T and its function called effective magnetic moment peff= 2.828 I/XT should be independent of T at constant H. When weak magnetic interactions among the paramagnetic ions or molecules develop at cryogenic temperature against thermal fluctuation, the xT vs. T or peffvs. T plots
410
I 1 Building Blocks for High-Spin Molecules and Molecular Assemblies
may become temperature-dependent (Fig. 11-2); the increase and decrease of the XT or p,ff values as the temperature is decreased toward absolute zero correspond to the occurrence of ferro- and antiferromagnetic interactions, respectively. As a first approximation, these interactions are treated with the Weiss field and the susceptibility is now given by a Curie-Weiss equation: x = C/(T - 0). In region 3, where all the spins are aligned in the direction of the applied magnetic field, the height of the saturated value (I,) is proportional to the size of the spins (S)in each ion or molecule: I, = N M = NgSp,. A Faraday magnetic balance system which is installed with a light guide for photolysis from outside and in-situ magnetic measurement at cryogenic temperature are documented [59].
11.7.2 Synthesis of Dendritic "Starburst" Dodecaketone 49 [55 b] 11.7.2.1 l-(3,5-Dibenzoylbenzoyl)-3-(3-trimethylsilyl-2-propynoyl)benzene (52) To a stirred solution of ethynyl phenyl ketone 50 (600 mg, 4.6 mmol) and 1-(2-propynoyl)3-(3-trimethylsilyl-2-propynoyl)benzene 51 (590 mg, 2.3 mmol) in toluene (5 mL) was added one drop of diethylamine, and the mixture was heated at 120 "C for 24 h. Evaporation of the solvent followed by chromatography on silica gel eluted with n-hexane/dichloromethane (4 : 1, v/v) afforded a mixture of four kinds of the trimerized ketones. The third fraction out of four peaks obtained in recycled GPC (Japan Analytical Industry Co. LC-08 Chromatograph System; column, JAIGEL l H + 2H; solvent, chloroform) was collected to give tetraketone 52 (370 mg, 31%) as a yellow oil; IR (KBr) v 2150, 1670 cm-'; 'H-NMR (CDCI,, 270 MHz) S 0.29 (s, 9H), 7.52 (t, J = 8 Hz, 4H), 7.64 (t, J = 7 Hz, 2H), 7.69 (t, J = 8 Hz, lH), 7.86 (d, J = 8 Hz, 4H), 8.12 (dt, J = 8, 1.8 Hz, lH), 8.39 (d, J = 1.8 Hz, 2H), 8.41 (t, J = 1.8 Hz, lH), 8.47 (t, J = 1.8 Hz, lH), 8.58 (t, J = 1.8 Hz, 1H).
11.7.2.2 1-(3,5-Dibenzoylbenzoyl)-3-(2-propynoyl)ben~ne(56) To a solution of tetraketone 52 (600 mg, 1.2 mmol) in methanol (120 mL) was added potassium fluoride (350 mg, 6 mmol) at -20°C and the mixture was stirred for 10 min. The reaction mixture was poured onto ice water. Extraction with benzene, washing with brine, drying with magnesium sulfate, and evaporation afforded deprotected tetraketone 56 (400 mg, 78%) as a yellow oil; IR (KBr) v 2090, 1660 cm-'; 'H-NMR (CDCl,, 270 MHz) 6 3.55 (s, 9H), 7.51 (t, J = 7 Hz, 4H), 7.62 (t, J = 7 Hz, 2H), 7.68 (t, J = 8 Hz, lH), 7.85 (dt, J = 7, 1.5 Hz, 4H), 8.12 (dt, J = 8, 1.8 Hz, lH), 8.38 (d, J = 1.8 Hz, 2H), 8.41 (t, J = 1.8 Hz, lH), 8.45 (t, J = 1.8 Hz, lH), 8.58 (t, J = 1.8 Hz, 1H).
11.7.2.3 1,3,5-~is[3-(3,5-dibenzoylbenzoyl]benzene (49) To a solution of deprotected tetraketone 56 (400 mg, 0.9 mmol) in toluene (2 mL) was added one drop of diethylamine and the mixture was heated at 120°C for 26 h. Evaporation of the solvent followed by chromatography on silica gel eluted with hexane/dichloromethane (1 : 1, v/v), and purification by means of GPC, afforded dodecaketone 49 (140 mg 35 070) as a yellow
References
41 1
oil; IR (KBr) v 3060, 1660, 1600 cm-l; 'H-NMR (CDCl,, 270 MHz) 6 7.49 (t, J = 8 Hz, 12H), 7.61 (t, J = 8 Hz, 6H), 7.69 (t, J = 8 Hz, 3H), 7.83 (d, J = 7 Hz, 12H), 8.06-8.12 (m, 6H), 8.35 (bs, 3H), 8.40 (s, 9H), 8.47 (s, 3H); I3C-NMR (CDCl,, 67.8 MHz) S 128.6, 129.2, 130.0, 131.0, 133.2, 133.9, 134.12, 134.17, 134.25, 134.5, 136.2, 136.7, 137.1, 137.3, 137.9, 138.3, 193.3, 193.6, 194.5.
Acknowledgement This work was supported by a Grant-in-Aid for Specially Promoted Research (No. 03102003) from the Ministry of Education, Science and Culture, Japan.
Abbreviations and Symbols CP CI D/hc, E/hc DMF AESQ EPR GPC J/kB MTHF
cyclopentadienyl configuration interaction zero-field splitting parameters N,N-dimethylformamide energy gap between singlet and quintet states electron paramagnetic resonance gel permeation chromatography exchange coupling parameter
R12
distance between the carbene centers spin quantum number tosyl zero field splitting
S Ts zfs
2-methyltetrahydro furan
References [l] D. F. Dance, I. C . Walker, Chem. Phys. k t t . 1973, 18, 601-603; H.Lischka, A. Karpfen, Chem. Phys. 1986,102, 77-89; P. Duprk R. Jost, M. Lombardi, P. G. Green, E. Abramson, R. W. Field, ibid. 1991, 152, 293-318;Y. Yamaguchi, G.Vacek, H. F. Schaefer, 111, Theor. Chim. Actu 1993,86, 97-113. [2]A. M.Wodtke, Y. T. Lee, J. Phys. Chem. 1985,89,4744-4751;T. R. Fletcher, S. R. Leone,J. Chem. Php. 1989,90, 871-879. [3]C. P. Poole, Jr., R. S. Anderson, J. Chem. Phys. 1959, 31, 346-354. [4]S. Hauff, P. KrauO, A. Rieker, Chem. Ber. 1972, 105, 1446-1455. [S]K. Inoue, Ph. D. Thesis of the Graduate School of the University of Tokyo, 1992. [6]A. Fujii, T. Ishida, N. Koga, H. Iwamura, Macromolecules 1991, 24, 1077-1082. (71 A. Fujii, M. Sc. Thesis of the Graduate School of the University of Tokyo, 1989. [8] N. Tanabe, M. Sc. Thesis of the Graduate School of the University of Tokyo, 1993.N. Tanabe, N. Sasagawa, N. Koga, H. Iwamura, to be published. [9]K. Inoue, N. Koga, H. Iwamura, J, Am. Chem. SOC. 1991, 113, 9803-9810.
412
I 1 Building Blocks for High-Spin Molecules and Molecular Assemblies
N. Koga, K. Inoue, N. Sasagawa, H. Iwamura, Mater. Res. SOC. Symp. Proc. 1990, 173, 39-49. M. Frank-Neumann, P. Geofferuy, J. J. Lohmann, Tetrahedron Lett. 1983, 24, 1775-1778. J. H. Teuben, H. J. de Liefde Meijer, J. Organomet. Chem. 1969, 17, 87-93. F. H. Kohler, W. Prossdorf, U. Schubert, D. Neugebauer, Angew. Chem. 1978,90, 912-913; Angew. Chem., Znt. Ed. Engl. 1978, 17, 850-851. [14] (a) C. Bianchini, F. Laschi, D. Masi, F. M. Ottaviani, A. Pastor, M. Peruzzini, P. Zanello, F. Zanobini, 1 Am. Chem. SOC. 1993, 115, 2723-2730; (b) D. L. Lichtenberger, S. K. Renshaw, R. M. Bullock, J. Am. Chem. SOC. 1993, 115, 3276-3285. [15] (a) R. Tsumura, N. Hagihara, Bull. Chem. SOC. Jpn. 1964, 37, 1889-1890; (b) idem, ibid. 1965, 38, 861-862. [16] J. L. Petersen, L. Griffith, Znorg. Chem. 1980, 19, 1852-1858. [I71 J. A. Howard, R. Sutcliffe, J. S. Tse, Organometallics 1984, 3, 859-866. [18] (a) C. LeVanda, D. 0. Cowan, C. Leitch, K. Bechgaard, J. Am. Chem. SOC.1974, 96, 6788-6789; (b) I. Motoyama, M. Watanabe, H. Sano, Chem. Lett. 1978, 513-516; (c) J. A. Kramer, F. H. Herbstein, D. N. Hendrickson, J. Am. Chem. SOC. 1980, 102, 2293-2301. [19] (a) R. Selvarajan, J. H. Boyer, J. Org. Chem. 1971, 36, 1679-1682; (b) A. Padwa, Y. Gareau, S. L. Xu, Tetrahedron Lett. 1991, 32, 983-986. [20] (a) R. A. Bernheim, R. J. Kempf, J. V. Gramas, P. S. Skell, X Chem. Phys. 1965, 43, 196-200; (b) R. A. Bernheim, R. J. Kempf, E. F. Reichenbecher, .l Magn. Reson. 1970, 3, 5-9. [21] W. J. Hehre, J. A. Pople, W. A. Lathan, L. Radom, E. Wasserman, Z. R. Wasserman, J. Am. Chem. SOC. 1976, 98, 4378-4383. [22] (a) D. J. Defrees, A. D. McLean, Astrophys. J. 1986,308, L3I-L35; (b) G. Meier, H. P. Reisenauer, W. Schwab, P. Carsky, V. Spirko, B. A. Hess, Jr., L. J. Schaad, .lChem. Phys. 1989,91,4763-4773; (c) P. Carsky, V. Spirko, B. A. Hess, Jr., L. J. Schaad, J. Phys. Chem. 1990, 94, 5493-5496; (d) V. Jonas, M. Bohme, G. Frenking, J. Phys. Chem. 1992,96, 1640-1648. [23] R. A. Seburg, J. T. DePinto, E. V. Patterson, R. J. McMahon, J. Am. Chem. SOC.1995, 117, 835-836. [24] R. Herges, Angew. Chem. 1994, 106, 261-283; Angew. Chem., Znt. Ed. Engl. 1994, 33, 255-276. [25] (a) N. Koga, M. Matsumura, M. Noro, H. Iwamura, Chem. Lett. 1991, 1357-1360; (b) M. Noro, T. Masuda, A. S. Ishimura, N. Koga, H. Iwamura, X Am. Chem. SOC.1994, 116, 6179-6190. [26] M. Noro, N. Koga, H. Iwamura, J. Am. Chem. SOC. 1993, 115, 4916. [27] J. T. DePinto, R. J. McMahon, J. Am. Chem. SOC. 1993, 115, 12573-12574. [28] (a) Yu. V. Korshak, A. A. Ovchinnikov, T. V. Medvedeva, V. N. Spector, Pisma Zh. Eh p . Teor. Fk. 1986, 43, 309; (b) Yu.V. Korshak, T. V. Medvedeva, A. A. Ovchinikov, V. N. Spector, Nature (London) 1987, 326, 370-373. [29] (a) J. Zhang, A. J. Epstein, J. S. Miller, C. J. O'Connor, Mol. Cryst. Liq. Cryst. 1989, 176, 271-276; (b) D. W. Wiley, J. C. Calabrese, J. S. Miller, Mol. Cryst. Liq. Cryst. 1989, 176, 277-288. [30] F. Toda, Top. Curr. Chem. 1988, 914, 212-238. [31] (a) M. Kinoshita, P. 'hrek, M. Tamura, Y. Nozawa, D. Shimi, Y. Nakazawa, Y. Ishikawa, M. Takahashi, K. Awaga, T. Inabe, Y. Maruyama, Chem. Lett. 1991, 1225-1228; (b) R. Chiarelli, M. A. Novak, A. Rassat, J. L. Tholence, Nature (London) 1993,363, 147-149; (c) K. Tomioka, S. Mitsubori, T. Ishida, T. Nogami, H. Iwamura, Chem. Lett. 1993, 1239-1242. [32] H. M. McConnell, X Chem. Phys. 1963, 93, 1910. [33] A. Izuoka, S. Murata, T. Sugawara, H. Iwamura, J. Am. Chem. SOC. 1985, 107, 1786-1787; idem, ibid. 1987, I09, 2631-2639. [34] K. Inoue, H. Iwamura, Adv. Mater: 1992, 4, 801-804; idem, Mol. Cvst. Liq. Crysf. 1993, 232, 89-98. [35] (a) K. Inoue, H. Iwamura, Chem. Phys. Lett. 1993,207, 551-554; (b) K. Inoue, H. Iwamura, J. Am. Chem. SOC. 1994, 116, 3173-3174.
[lo] [ll] [12] [13]
References
413
[36] G. Wegner, Z. Naturforsch., Teil B 1969, 24, 824-832; idem, Makromol. Chem. 1971, 145, 85-94; idem, ibid. 1972, 154, 35-48. [37] M. Schwoerer, H. Niederwald, Makromol. Chem., Suppl. (Polym. Specific Prop.), 1985, 12, 61 -82. [38] K. Takeda, G. Wegner, Makromol. Chem., 1972, 160, 349-353. [39] (a) W. Neumann, H. Sixl, Chem. Phys. 1980,50, 273-280; (b) C. Bubeck, W. Hersel, W. Neumann, H. Sixl, J. Waldmann, ibid. 1980, 51, 1-8. [40] (a) R. Huber, M. Schwoerer, Chem. Phys. Lett. 1978,53, 35-38; (b) R. A. Huber, M. Schwoerer, H. Benk, H. Sixl, ibid. 1981, 78, 416-420. [41] (a) H. C. Longuet-Higgins, J. Chem. Phys. 1950, 18, 265-274; (b) W. T. Borden, E. R. Davidson, J. Am. Chem. SOC. 1977, 99, 4587-4594; (c) A. A. Ovchinnikov, Theor. Chim. Acta 1978, 47, 297-304; (d) W. T. Borden, H. Iwamura, J. A. Berson, Acc. Chem. Res. 1994, 27, 109-116. [42] (a) A. M. Trozzolo, R. W. Murray, G. Smolinsky, W. A. Yager, E. Wasserman, J. Am. Chem. SOC. 1963,85,2526-2527; (b) K. Itoh, Chem. Phys. Lett. 1967, I , 235-238; (c) E. Wasserman, R.W. Murray, W. A. Yager, A. M. Trozzolo, G. Smolinsky, J. Am. Chem. SOC.1%7,89,5076-5078; (d) K. Itoh, Pure Appl. Chem. 1978,50, 1251-1259; (e) H. Sixl, R. Mathes, A. Schaupp, K. Ulrich, Chem. Phys. 1986, 107, 105-121. [43] (a) S. Murata, T.Sugawara, H. Iwamura, X Am. Chem. Soc. 1987,109, 1266-1269; (b) H. Iwamura, S. Murata, Mol. Cryst. Liq. Cryst. 1989,176, 33-48; (c) S. Murata, H. Iwamura, J. Am. Chem. SOC. 1991, 113, 5547-5556. [44] (a) P. M. Lahti, A. S. Ichimura, J. Org. Chem 1991, 56, 3030-3042; (b) A. S. Ichimura, N. Koga, H. Iwamura, J. Phys. Org. Chem. 1994, 7, 207-217. [45] S. Sasaki, H. Iwamura, Chem. Lett. 1992, 1759-1762. [46] Y. Cao, P. Wang, Z. Hu, S. Ki, L. Zhang, J. Zhao, Solid State Comrnun. 1988, 68, 817-820; idem, Syn. Met. 1988, 27, B625-B630. [47] (a) T. Masuda, T. Higashimura, Adv. Polym. Sci. 1986, 81, 121-165; (b) T. Masuda, T. Highashimura, Acc. Chem. Res. 1984, 17, 51-56; (c) A. Furlani, C. Napoletano, M. V. Russo, W. J. Feast, Polym. Bull. 1986, 16, 311-317; (d) M. Tabata, K. Yokota, Jpn. Kokai Tokkyo Koho JP 63 275 614 (Cl.C08F38/00) and JP 63 275 613 (CIC08F38/00), Nov. 14th, 1988. [48] H. Nishide, N. Yoshioka, T. Kaneko, E. Tsuchida, Macromolecules 1990, 23, 4487-4488; N. Yoshioka, H. Nishide, T. Kaneko, H. Yoshiki, E. Tsuchida, ibid. 1992, 25, 3838-3842. [49] N. N. Tyutyulkov, S. C. Karabunarliev, Znt. .l Quantum Chem. 1986, 29, 1325-1337. [SO] (a) H. Iwamura, Pure Appl. Chem. 1986,58, 187-196; (b) N. Nakamura, K. Inoue, H. Iwamura, T. Fujioka, Y. Sawaki, J. Am. Chem. SOC.1992, 114, 1484-1485; (c) N. Nakamura, K. Inoue, H. Iwamura, Angew. Chem. 1993, 105,900-901 ;Angew. Chem., Int. Ed. Eng/., 1993,32, 872-874; (d) H. Iwamura, Pure Appl. Chem. 1993, 65, 57-64. [51] (a) G. Schmauss, H. Baumgartel, H. Zimmermann, Angew. Chem. 1965, 77, 619-620; Angew. Chem., Int. Ed. Engl. 1%5,4, 596; (b) E. Wasserman, K. Schueller, W. A. Yager, Chem. Phys. Lett. 1968, 2, 259-260; (c) T. Takui, K. Itoh, ibid. 1973, 19, 120-124; (d) H. Iwamura, Adv. Phys. OR. Chem. 1990,26, 179-253. (e) F. Kanno, K. Inoue, N. Koga, H. Iwamura, 1 Phys. 0%.Chem. 1993, 97, 13267-13272; (f) K. Yoshizawa, M. Hatanaka, Y. Matsuzaki, K. Tanaka, T. Yamabe, J. Chem. Phys. 1994, 100, 4453-4458. [52] W. Reppe, N. von Kutepow, and A. Magin, Angew. Chem. 1%9, 81, 717-723; Angew. Chem., Int. Ed. Engl. 1969, 8, 727-733. [53] T. Sasaki, Y. Suzuki, Tetmhedron Lett. 1967, 32, 3137-3140. [54] K. Balasubramanian, S. Selvaraj, P. S. Venkataramani, Synthesis 1980, 29-32. [55] (a) K. Matsuda, K. Inoue, N. Koga, N. Nakamura, H. Iwamura, Mol. Cryst. Liq. Cryst., 1994,253, 33-40. (b) K. Matsuda, N. Nakamura, H. Iwamura, Chem. Lett. 1994, 1765-1768. [56] K. Furukawa, T. Matsumura, Y. Teki, T. Kinoshita, T. Takui, K. Itoh, Mol. Cryst. Liq. Crysf. 1993, 232, 251-260.
414
11 Building Blocks for High-Spin Molecules and Molecular Assemblies
[57] (a) D. Gatteschi, 0. Kahn, J. S. Miller, F. Palacio, Magnetic Molecular Materials (Eds.) Dordrecht, Kluwer Academic Publishers, 1991; (b) H. Iwamura, J. S. Miller (Eds.), Mol. Cryst. Liq. Cryst., Special issues: Chemistry and Physics of Molecular Based Magnetic Materials 1993,232, 1-360 and 233, 1-366; (c) J. S. Miller, A. J. Epstein, Angew. Chem. 1994, 106, 399-432; Angew. Chem., Int. Ed. Engl. 1994,33, 385-415; (d) A. Rajca, Chem. Rev. 1994,94, 871-893; (e) D. A. Dougherty, Acc. Chem. Res. 1991, 24, 88-94. [58] (a) A. Weiss, H. Whitte, Magnetochemie, Verlag Chemie, Weinheim, 1973; (b) R. L. Carlin, Magnefochernistry, Springer, New York, 1986; (c) C. J. O’Connor (Ed.), Research Frontiers in Magnetochemistry, World Scientific, Singapore, 1993. 1591 (a) H. Iwamura, in [58c], p. 303; @) H. Iwamura, N. Koga, Acc. Chem. Res. 1993, 26, 346-351.
12 Acetylenes in Nanostructures James K. Young, Jeffrey S. Moore
12.1 Introduction Within the past two decades, chemistry has begun to embrace a broader context of the term “synthesis” that now encompasses not only small-molecule and macromolecular constructions based on covalent bonds, but also complex systems involving weaker bonding interactions. With this new connotation, possible targets now include a variety of architectures having unique and well-defined shapes, as well as periodic networks that extend indefinitely in one, two, or three dimensions held together by either supramolecular or covalent bonds. The impetus behind these efforts lies in the potential to fine-tune and optimize material properties through rational means, and to discover new materials with unprecedented characteristics. In this light, major challenges are the invention of target systems and the design of building blocks that can be used to achieve these targets. Without question, the most impressive molecular machinery and most advanced materials are those found in nature. A glimpse at the chemistry of biomolecular systems reveals powerful principles that could serve as a working model to guide the design of target systems and corresponding building blocks. Nature uses structure-controlled macromolecules for both function and supramolecular assembly. All of this chemistry is dominated by only three backbone structures : polypeptides, polynucleic acids, and polysaccharides. A special feature common among these macromolecules is well-defined and essentially invariant shape (shape-persistence). Shape-persistence together with macromolecular dimension provides large-area molecular surfaces with distinctive topographic features and functional group dispositions. These surfaces are responsible for dictating the recognition events between the macromolecule and its surrounding environment leading to controlled supramolecular chemistry. Because the dimensions of these molecules are on the nanoscale and because they possess well-defined and characteristic shapes they are considered here as nanoscale objects or simply “nanostructures” . Another important point about biological macromolecules is that they are “poly” “mers” i. e., their molecular structures consist of basic repeating units. While apparently simple and obvious, this construction approach offers synthetic economy and yet allows enormous structural variation through pendant group modification. It is astounding that all of the diversity in structure and function exhibited by biological macromolecules is realized from just three backbone constitutions! With this model as a guide, a pertinent question is then, “What is a suitable non-natural platform on which to base a new building block set?” This chapter describes synthetic methods that allow for the versatile and efficient construction of oligo- and macromolecular architectures from phenylacetylene monomers. Recognizing the advantages of stepwise, repetitive syntheses that follow geometric progressive growth, we developed versatile chemistry [l] leading to both straight-chain [2] and branched [3], monodisperse, sequence-specific oligomers and polymers. Acetylene-based macromolecules share features common to the best-known natural macromolecular systems, such as nucleic
416
12 Acetylenes in Nanostructures
acids and proteins, in that they have a high level of shape persistence and many architectural variations can be realized by the ordered catenation of a small set of monomers. The examples in this chapter were selected to illustrate that wide structural diversity can be achieved from a small set of constructing reactions (Scheme 12-1). It will also be shown how the solid-phase approach pioneered by Merrifield [4] serves as a model guiding the development of the synthetic methods. Pd(0) Catalyzed Cross-Coupling Reactions
X = Br, I Selective Coupling
0" '4 -
Br
- --
Pd(dba)p Et3N/50'C I Ph3P I Cul
Br
Deprotection Reactions
Scheme 12-1 The set of constructing reactions for phenylacetylene macromolecules.
12.1.1 Structural Parameters of Phenylacetylenes The use of the carbon-carbon triple bond as a linking unit between aromatic rings allows access to relatively rigid and simple scaffoldings that can be further elaborated by incorporating a wide range of pendant functionalities on the aromatic rings. Some of the unique structural features of the phenylacetylene unit that make it a suitable building block for constructing molecular modules are shown in Fig. 12-1. "Valence angles" of 60",120°, or 180" are attained from repeat units with ortho, meta, or para arene connectivity, respectively. This means that any fragment of any trigonal lattice is a potential nanostructural target realized by sequencespecific combinations of appropriate phenylacetylene monomers. Throughout this chapter the simplified dot- and -line convention seen in Fig. 12-1 will be used to represent the phenylacetylene repeat units in nanostructures. While the number of torsionally derived conformations available to macromolecules composed solely of phenylacetylene subunits is limited, the valence angle deformation 6 seen in Fig. 12-1 is significantly more pliable than that found in other carbon-based frameworks. The CHARMm force field constant, F ( 4 , for this deformation is less than half that of an all-sp3 carbon framework 151. Steric interactions between aromatic rings are negligible, so electronic factors are expected to dominate the potential governing torsion about the sp-sp2 carbon-carbon single bond. Delocalization could involve overlap between the aromatic p-orbitals of both rings either with a common pair of alkyne p-orbitals, or with orthogonal pairs of alkyne p-orbitals (favoring the planar or perpendicular geometry, respectively). Experimental data and theoretical
12.1 Introduction
417
Geometric Considerations
02'
"valence Phenylacetylene angles"
1
D 1 8 0 '
t I
I
Structure key
1/
Paulin'g bond
number = 1.18
Conformational Considerations t
0.71 I
c-
2
0.5-
h
ca.1.95A
z
w'
0 3 0.1-
; 1 . 0 -90
-45 0 45 Torsional angle
90
Figure 12-1 Fundamental characteristics of the phenylacetylene monomer unit. In this chapter the three basic phenylacetylene valence angles will be represented with a dot (*) and line ( -) convention, as shown in the structure key. The phenylacetylenelinkageis characterizedby a flexiblebending force constant, F(d) [5], and an sp2-sp bond order of 1.18. The indicated bond distances are taken from the crystal structure of tolane [6].The bottom plot shows the torsional potential energy curve in tolane as determine by supersonic jet spectroscopy (adapted from [8]). The outer edges of van der WaaIs spheres for the two ortho hydrogens on one side of tolane are separated by nearly 2 A in the planar conformation. The result is a virtually barrierless rotation.
calculations are in contradiction as to which of these is the preferred geometry, although both theory and experiment conclude that the extent of overlap and hence the barrier to rotation is small. The Pauling bond number of the single bond in tolane is 1.18 (i. e., 18'70double-bond character) [6],implying a relatively low degree of conjugation between aromatic rings. The majority of single-crystal X-ray structures of simple diarylalkynes in the Cambridge Structural Database [7] have a near-planar geometry. Supersonic jet electronic spectral data [8] and dipole moment measurements [9] for tolane reveal a planar geometry with D2,, symmetry. In contrast, semi-empirical CNDO and INDO calculations [6,91 predict a slightly more
418
12 Acetylenes in Nanosfrucfures
stable (0.65 kcal/mol) perpendicular form. Experimentally the activation energy for torsional rotation is found to be less than 0.60 kcal/mol (Fig. 12-1) [S]. To summarize, the phenylacetylene repeat unit has limited degrees of conformational freedom. Nonetheless, a high degree of shape-persistence in nanostructures may still require introduction of additional steric or topological constraints through nonbonded interactions or cyclization, respectively.
12.2 Phenylacetylene Dendrimers Cascade (dendritic) macromolecules are discrete, highly branched, monodisperse polymers that possess branching patterns described by a nonlinear mathematical progression. It has been demonstrated that although dendrimers have well-defined constitutions, their size in solution may vary drastically with conditions such as pH [lo]. Therefore, shape-persistent
a
b
C
d
e
position
Figure 12-2 Chemical structure and conceptual design of a convergent and directional light-energy conversion device. The bottom diagram shows the variation of excited-state energy as a function of position within the dendrimer hierarchy.
12.2 Phenylacetylene Dendrimers
419
members of this unique class of well-defined nanostructures represent ideal models for studying how physical properties depend on molecular size and architecture. An illustrative example is given below, before the key concepts of phenylacetylene dendrimer synthesis are reviewed. The unique hierarchical nature dendrimers in general and the shape-peristence of phenylacetylene dendrimers in particular can be exploited to control electronic interactions between spatially defined features. For example, Fig. 12-2 depicts a convergent and directional lightenergy conversion device [ll]. Directional transduction of energy along a convergent path is the essential requirement for an efficient antenna. Keeping this criterion in mind, consider the design of 1. The length of the linear segment between triconnected junctures decreases by one unit proceeding from the core to the rim. The 1,3,5-trisubstituted arene branch junctures interrupt delocalization of the dendrimer’s electronic wavefunctions, and the excited state of the fluorescent focal point represents an energy sink relative to the dendrimeric portion of this molecular antenna. Therefore, phenylacetylene monodendron 1 has a smooth gradient in electronic energy from the periphery to the luminophoric perylene focal point. The number of chromophores of a given type decreases by a factor of two at each step from the rim to the core. The abundance of collection sites (“u” in Fig. 12-2) on the periphery of 1 are all coupled to the perylene luminophore by a convergent pathway that is consistent with the anticipated direction of energy migration. Overall, the net effect should be a focusing of light energy collected by the numerous peripheral chromophores down to a single, efficient fluorescent emitter. This is one example that illustrates how the precise positional control afforded by shapeperistent phenylacetylene dendrimers can be exploited.
12.2.1 Synthetic Considerations for Phenylacetylene Dendrimer Construction Since dendrimers possess a high level of chemical regularity, it is possible to use nonlinear growth strategies for their synthesis [12]. The efficiencies of four different repetitive synthetic strategies are compared in graphical form in Fig. 12-3, which plots the degree of polymerization versus the cumulative number of steps. The slope of these curves at any point represents the number of monomers incorporated per synthetic step at that particular point in the synthesis and is an indicator of synthetic efficiency. Curve (a) represents conventional linear synthesis, which is well suited for preparing oligomers having no periodicity in the sequence order of monomers. Curve (b) represents nonlinear growth for straight-chain oligomers having a regular sequence order of monomers. By way of comparison, the efficiency of dendritic, nonlinear growth schemes that employ branched (triconnected) monomers is remarkable (curve (c) = “regular” and curve (d)= “double exponential”). The chemistry used to prepare phenylacetylene dendrimers is based on the well-established and high-yielding palladium-catalyzed cross-coupling reaction of terminal acetylenes and aryl halides [13]. Terminal acetylenic and aryl iodide moieties can be orthogonally masked as trimethylsilylacetylene [14] and l-aryl-3,3-dialkyltria~ene [l] groups, respectively, during phenylacetylene nanostructure synthesis. Selective and facile removal of these protecting groups precedes the Pd(0)-catalyzed cross-coupling reaction. Although dendrimer synthesis typically does not require protection of both coupling partners, it is useful to have dual protecting group schemes available. This availability allows the comparison of two or more synthetic routes to the desired dendrimer. We have found that the advantages of one route versus another are subtle and usually cannot be appreciated until the synthetic attempt is made.
420
12 Acetylenes in Nanostructures
J
I
I
I
I
/
i
L
I
‘il
k a”
40
20
0
0
4
8
12
16
20
Cumulative number of synthetic steps
Figure 12-3 Plots of the degree of polymerization versus the cumulative number of synthetic steps for various repetitive syntheses: (a)conventional linear solid-phase synthesis; (b)nonlinear straight-chain sequence synthesis; (c) dendrimer synthesis (branching multiplicity of three); (d)double exponential dendrimer synthesis (branching multiplicity of three). In all cases, the degree of polymerization is defined as the total number of monomer units per polymer molecule.
12.2.1.1 The Divergent and Convergent Synthetic Approaches
The two major strategies for dendrimer synthesis are the divergent and convergent approaches. The divergent method, developed independently by Newkome [15] and Tomalia [16],involves the repetitive catenation of units around a core. The advantages of this approach are rapid growth of the dendrimer series and an increase in mass of isolated products without significant steric inhibition at early generations. Potential liabilities of the divergent approach are the formation of molecular imperfections at higher generations for some chemistries, and the difficulty in selectively modifying the peripheral functionality. An attempt to synthesize phenylacetylene dendrimers by a divergent approach was severely hampered by solubility problems at an early stage in the synthesis. We chose the aryl iodide coupling partner as the reactive peripheral functionality, because the alternative scheme would involve synthesizing potentially unstable multiterminal acetylenes. Unfortunately, even the first-generation aryl iodide terminated dendrimer showed virtually no solubility in any common solvents. Frkchet and Hawker [I71 and Miller and Neenan [I81 developed a convergent route for preparing dendritic macromolecules, in which the synthetic starting point is the periphery of the dendrimer. The convergent approach begins with the synthesis of “monodendrons” and finishes by the coupling of several of these segments around a multifunctional core. A convergent strategy for phenylacetylene monodendron’ synthesis is shown in Scheme 12-2 1191.
’
Mono-, di-, and tridendrons refer to dendritic architectures radiating outward from a point of single, double, or triple connectivity, respectively.
12.2 Phenylacetylene Dendrimers
421
H
Ar
KzC03 I rt MeOH I CHZCIz
Scheme 12-2 A repetitive, nonlinear growth scheme for the convergent synthesis of phenylacetylene
monodendrons. Entrance into the repetitive cycle is gained by coupling two equivalents of 3,5-di(tertbuty1)phenylacetylene (2: the group that will ultimately be on the dendrimer periphery) to 3’,5’-dibromo-2-phenyl-l-(trimethylsilyl)acetylene(3) (n = 0 at the completion of this step). Removal of the trimethylsilyl protecting group leaves a terminal acetylene at the monodendron focal point. Two equivalents of this alkyne are then coupled with aromatic dibromide 3 to give the first-generation (n= 1) monodendron having four peripheral units. In principle, this process can be continued until the desired generation is achieved. n o characteristics of the convergent approach might be expected to improve solubility. First, the growing mondendrons possess lower symmetry than the dendrimers grown by the divergent approach. Second, it is possible to attach pendant groups onto the periphery of the dendrimer and, unlike the divergent route, these groups are present throughout the entire synthesis.
12.2.1.2 Convergent Synthesis of-Phenylacetylene Dendrimers Vital considerations during convergent phenylacetylene dendrimer synthesis are solubility and ease of purification, which vary with the nature of the focal point group and the functionalities on the periphery, as well as the progressively diminished focal point reactivity of successive generations. The dendrimer series [20] ZCascade: benzene[3- 1,3,5] :(5-ethynyl-1,3phenylene)G:5-ethynyl-1,3-di(ferf-butyl)benzene shown in Fig. 12-4 was constructed using monodendrons prepared by the convergent strategy shown in Scheme 12-2.The fourth-generation dendrimer 8 (the largest member of this series) contains 94 aromatic rings and has an approximate diameter of 55 A. In general, the nature of the terminal group is the main determinant of solubility for phenylacetylene mono-, di-, and tridendrons. However, the high symmetry of tridendrons may result in poor solubility, even if highly soluble monodendrons are used for their construction. For example, (4-ferf-butylpheny1)acetylene-terminatedmonoden-
422
I2 Acetylenes in Nanostructures
7 generation n = 3 8
R=H
R=D generation n = 4
Figure 12-4 Abbreviated structures for generations 1-4 of the phenylacetylene dendrimer series Z-Cascade: benzene[3- 1,3,5] :(5-ethynyl-1,3-phenylene)G:5-ethynyl-1,3-di(tert-butyl)benzene.
drons are very soluble in most common organic solvents, while the corresponding tridendrons are only marginally soluble. In contrast, mono- and tridendrons containing di(fert-butyl) peripheral groups display excellent solubility, even in n-alkanes. The only significant impurity detected during synthesis of the monodendrons was the diacetylene formed by oxidative dimerization of the starting monodendron. The diacetylene by-product can be removed by careful and tedious low-pressure silica gel chromatography. The dendrimers 4-8 (Fig. 12-4) are obtained by Pd(0)-catalyzed cross-coupling between the appropriate monodendron with acetylene at its focal point and tribromo- or triiodobenzene. Because of the progressively diminished focal point reactivity of successive generations, tridendron yields diminish rapidly when using 1,3,5-tribromobenzeneas the core. The fourth-generation dendrimer 8 cannot be prepared in this manner as only the diacetylene by-product of monodendron oxidative dimerization is formed. However, tridendron 8 is obtained in 37 9'0 yield when 1,3,5-triiodobenzene is used as the core. Aryl iodides allow the coupling reaction to be performed at lower temperatures than the corresponding bromides and as a consequence the cross-coupling becomes competitive with the self-coupling reaction manifold. As the molecular weight of the monodendron increases, it becomes necessary to use larger quantities of the catalyst (e. g., 10 mol 9'0 based on the terminal acetylene instead of the typical 2 mol%). When larger quantities of the catalyst are used, it is essential to run the reaction at low temperatures (35-40 "C)for longer periods of time (> 48 h).
12.2 Phenylacetylene Dendrimers
423
12.2.1.3 Effect of Varying Focal Point Functionality on the Convergent Synthesis of Phenylacetylene Dendrimers
We previously mentioned that the nature of the terminal group is the main determinant of solubility for phenylacetylene dendrimers. Similarly, the nature of the focal point group is the main determinant of yield in the synthesis of monodendrons. A plot of isolated yields of monodendrons with different focal point functionality versus generation is shown in Fig. 12-5 [21]. From the second generation on, there is a dramatic improvement when the triazene functional group is at the focal point, as opposed to the methods that we have discussed so far in which the protected acetylene is at the focal point. This is a recent observation that escaped our notice for some time, because as previously mentioned the advantages of one route versus another are subtle and usually cannot be appreciated until the synthetic attempt is made. This example illustrates the utility of dual protecting group schemes.
0
1
3
2
4
Generation
1
I
Focal Point Monomer Etd-N
Ill
Figure 12-5 Plot of isolated yields versus generation of rnonodendrons with different focal point func-
tionality.
12.2.1.4 SYNthesis of Dendrimers by Repetition of Monomer Enlargement (SYNDROME Method)
Since the nonlinear mathematical relationship that forms the basis of dendrimer growth is simple geometric progression, the corresponding molecular structures quickly fill space in three dimensions. The increase in volume per generation, based on added molar mass, changes at a faster rate than the space available, given the monomers’ geometric constraints, leading to a “dense-packed” state [22]. Scheme 12-3 illustrates an alternative nonlinear growth
424
I2 Acetylenes in Nanostructures I
I I M-1 SiMe3
* 111
w-1
M-5
M-3
-
1 4 c-1
M - l Z M - l & SiMe3~ - 3 _ ---t
w-3
w-7
I
C-4
c-7
SiMe3
W-17
w-39
p o
c-1
c-1
G4 1
1
4% D-4 1.7 nm
1
& D-1 0
2.8 nm
D-25 5.1 nm
D-sa 8.5 nm
D-127 12.5 nm
Scheme 12-3 Method for the preparation of phenylacetylenedendrimers that combines a linear increase in monomer core size per generation with the convergent growth process (SYNDROME strategy). Adapted from [23].
strategy that combines a linear increase in monomer size per generation with the convergent approach leading to geometrically expanded, large-size macromolecules after only a few repetitive cycles [23]. The method requires a series of monomers (M-1, M-3, M-5) as well as cores (C-1, C-4, C-7, C-10) that gradually increase in size. The enlargement of the monomers and cores is accomplished by repetitive catenation of para-linked phenylacetylene structural units. To illustrate how rapidly the size of the molecule increases by this method the thirdgeneration SYNDROME tridendron, D-58, containing 58 aromatic rings, elutes by gel permeation chromatography with a polystyrene equivalent hydrodynamic radius that is nearly 1.5 times larger than that of the fourth-generation “regular” tridendron 8 (containing 94 aromatic units).
12.2.1.5 “Double Exponential” Dendrimer Growth
Scheme 12-4 outlines the synthetic strategy for a remarkable example of exponentially accelerated growth [curve (d)in Fig. 12-31 that leads to dendritic macromolecules [24]. For the convergent method with trifunctional monomers, Eq. (1) shows how degree of polymerization, dp, grows as a power of n (n = generation number). In contrast, by the growth process shown in Scheme 12-4, the degree of polymerization depends on 2 raised to the power of 2,
12.2 Phenylacetylene Dendrimers A,
+A
-
A-(Bp BP A+B
A,+Bp
3 A-B
BP
-
425
A p t B p BP
9
Bp +B
BP 10
B
B
11
Scheme 12-4 “Double exponential” growth scheme for phenylacetylene dendrimers. A,,
B, are pro-
tected versions of the functional groups A, B.
which is raised to the generation number n, i.e., double exponential dendrimer growth (DEDG). The general form for the degree of polymerization versus generation follows double exponential growth as shown in Eq. (2).
The concept of double exponential growth developed from a generalization of the nonlinear repetitive method described below, for synthesizing straight-chain oligomenc sequences of “AB” monomers [2, 251. Scheme 12-4 shows the first two generations of DEDG starting from trifunctional monomer 9. The first generation (10) is a trimer obtained by combining the monoprotected derivatives of 9 in a 2 : 1 ratio. Monodendron 10 has identical types of functional groups to 9 except that the number of peripheral B, groups has doubled. Repetition of this process in a 4 : 1 ratio from 10 gives the second-generation product 11, a 1.5-mer with 16 peripheral B, groups. If 11 is subjected to this reaction sequence once more in a 16: 1 ratio, a monodendron having a degree of polymerization of 255 is expected. Fig. 12-6 shows the chemical structures of a starting monomer 12 and the corresponding trimer 13 and 15-mer 14 that have been prepared by double exponential growth. The first two cycles proceed smoothly yielding monodendrons 13 and 14. Complete removal of the 16 peripheral trimethylsilyl groups from monodendron 14 is difficult, due to poor solubility of the resulting product. However, when KOH is used in a solvent mixture of THF and methanol, complete deprotection can be achieved. The resulting monodendron is obtained as a rather unstable,* The solid could be stored at - 50°C for up to three weeks without loss of purity. Decomposition is evident by the appearance of higher-molecular-weight impurities in (SEC) traces.
426
12 Acetylenes in Nanostructures
white, amorphous solid. Fortunately, this solid exhibited reasonable solubility in THF and DMF, and a 'H-NMR spectrum in d,-THF revealed the total absence of trimethylsilyl groups. Because of its instability, the product was not characterized further but used immediately in the next coupling reaction to give a nearly defect-free 255-mer monodendron, as characterized by laser desorption mass spectrometry [24]. SiMea
Me,Si
Q
Ill SiMe3 Me,Si
M e 3 s i y 2
SiMe,
12
Me3Si
SiMe3
SiMes
SiMea
SiMe3
14
Figure 12-6 Chemical structures of a starting monomer 12 and the corresponding trimer 13 and 15-mer 14 that have been prepared by double exponential dendrimer growth (DEDG).
12.3 Phenylacetylene Macrocycles Fig. 12-7 gives an overview of structure-directed assembly using site-specificallyfunctionalized and geometrically controlled phenylacetylene nanostructures as modular building blocks. These shape-persistent skeletons can be used to position and orient steric and electronic features that dictate intermolecular interactions. Just as in the polymeric backbones of proteins, nucleic acids, and polysaccharides, the ordered catenation of a small set of phenylacetylene monomers provides a means for the precise control of functional group placement and chain length of various phenylacetylene sequences (PASS). Upon deprotection of the termini, these PASs can be cyclized to phenylacetylene macrocycles (PAMs) in high yield. Such a versatile and efficient approach to this family of geometrically well-defined macrocycles offers the potential for producing a set of modular building blocks to rationally assemble molecular crystals [26], liquid crystals [27], and monolayer surfaces [28].
421
12.3 Phenylacetylene Macrocycles monomers
H
Oligomeric Sequence
I Cyclization 1 1 L FG
Modular Assembly tt via Directed Organization FG
FG
1
L
Surfaces
Liquid Crystals
Modular Units
I
I
I
I Porous Crystals
Figure 12-7 Modular construction of organic materials based on phenylacetylene nanostructures. Sitespecifically functionalized and geometrically controlled macrocycles are easily made by cyclization of the corresponding sequence-specific oligomeric sequences. The synthesis of the oligomeric sequence encodes the module with its supramolecular information. This information can be expressed in the assembly of organized monolayers, columnar liquid crystal, or crystalline networks. The process of encoding the oligomer as an ordered sequence of monomers in analogous to the preparation of oligopeptides. FG, functional group.
428
12 Acetylenes in Nanostructures
12.3.1 Phenylacetylene Macrocyclic Framework As mentioned earlier, any framework consistent with a trigonal lattice can be assembled by using combinations of ortho-, meta-, and para-phenylacetylene monomers (60°,120°, and 180" angles, respectively). Fig. 12-8 shows a plot of the number of unique monocyclic geometries versus the number of phenylacetylene monomer units, as well as some of the possible unstrained monocyclic geometries comprised of three to eight phenylacetylene subunits. Clearly, a large variety of hydrocarbon frameworks are accessible, not to mention the large number of structures resulting from the attachment of exo- and/or endo-oriented functional groups. The shape and functionalization of these structures can be viewed as the information that makes up the complete set of instructions for all of the subsequent supramolecular chemistry [29].
Number of phenylacetylene monomets 3-monomer macrocycle 0. 4-monomermacrocycle
7-monomer macrocycles
9.
8-monomer macrocycles
5-monomer macrocycle
bR
.ct 6-monomer macrocycles
OWQ# Figure 12-8 Plot of the number of unique cyclic geometries versus the number of phenylacetylene monomer units and structures of the unique cyclic geometries for phenylacetylene macrocycles comprised of between tcree and eight monomer units.
Fig. 12-9 is a collage of some of the trigonal-based macrocyclic architectures (15-20) that have been realized by the cyclization of the appropriate PASS. Complex topologies, such as the macrobicycles 16 or macrotetracycle 17, are available by double cyclization of appropriate PASs as described below.
12.3 Phenylacetylene Macrocycles
429
17
OH
HO
\\
4
10
19
20
Figure 12-9 Examples of some unique cyclic geometries that have been realized using functionalized or branched phenylacetylene monomers.
430
12 Acetylenes in Nanostructures
12.3.2 Synthetic Considerations for Phenylacetylene Macrocycle Construction Recall that conventional linear synthesis is well suited for preparing sequences in which there is no periodicity in the sequence order of monomers, while geometric progressive growth is more efficient for straight-chain sequences with monomer sequence order periodicity (curves (a) and (b) in Fig. 12-3, respectively). Oligomers which have periodic or extended periodic segments can be most efficiently prepared using geometric progressive growth, provided that dual protecting group schemes are available to allow the differentiation of chain ends. This approach has been used for the preparation of oligomeric alkanes [25, 301 and polyurethane segments [31] and is the preferred approach for the preparation of PAMs from the corresponding oligomeric sequences. The cyclization of a PAS to a PAM is carried out under pseudo-high-dilution conditions [32] with the sequence solution being slowly added to the palladium catalyst solution by a syringe pump. Oxygen is the major cause of deactivating the catalyst. Therefore, in order to keep the catalyst active during the whole addition period, which usually lasts for more than 12 h, it is essential to exclude oxygen from the reaction system. An active catalyst is characterized by its bright yellow color in the solution and a deactivated catalyst is dark brown in the solution. Qpically, up to gram scale can be operated in a single run. The cyclization proceeds exceptionally well in terms of yield and purity of the product. The isolated yields of the cylic compounds range from 70 to 80%.
12.3.2.1 The Double Cyclization of Branched Phenylacetylene Oligomers Three-dimensional (3D) phenylacetylene nanoscaffolding can be constructed in a manner that is analogous to PAM synthesis as shown in Scheme 12-5 [3]. Double cyclization of the branched sequence 21 proceeds smoothly to provide macrobicycle 16a in 60% yield after purification by silica gel chromatography. The double-cyclization reaction was carried out under pseudohigh-dilution conditions [32] to favor the intramolecular cyclization. A solution of the branched sequence (0.02 mol/mL) was slowly added to the palladium catalyst solution using a syringe pump (0.05 mmol/h). Although the cyclization involved formation of two covalent bonds from four reactive sites, it proceeded exceptionally well in terms of yield and purity of the pro-
H
'I'
h
fl
Pd(dbah I Cul I PPh3 TEA I PhH I70'C
*
16a
21
Scheme 12-5 Double cyclization to give a phenylacetylene macrobicycle.
12.3 Phenylaceiylene Macrocycles
431
duct. The pseudo-high-dilution conditions and the rigidity of the branched sequence structure presumably facilitate the intramolecular cyclization. Isolation of the product by chromatography was facile, because the oligomers formed by intermolecular reactions have much larger relative size and mass, and as a consequence were immobilized on the silica gel column under the optimized eluting conditions. The resulting macrobicycle 16a is a white powdery solid that is soluble in a wide range of organic solvents and is stable in air.
12.3.2.2 Tandem Bimolecular Coupling Followed by Intramolecular Cyclization to Form a Foldable Phenylacetylene Macrotetracycle
The macrotetracycle 17 described here required a carefully planned sequence design to obtain the desired topology during the cyclization reactions [33]. Scheme 12-6 illustrates this point by showing three retrosynthetic routes leading to two isomeric macrotetracycles from a pair of nine-unit branched sequences. In this scheme, “A” and “B” refer to functional groups capable of forming a new covalent bond while “Ap” and “Bp” are corresponding protected versions of these groups. For all three routes shown in Scheme 12-6, the first cyclization step entails a tandem bimolecular coupling of two nine-unit branched segments and subsequent intramolecular cyclization. The second step involves double cyclization of the resulting 18-unit branched sequence. The three routes differ as to the substitution pattern on the branched sequences. Only one of these routes leads unambiguously to the desired connectivity. In particular, the desired macrotetracycle is the unique isomer obtained by combining one sequence which is symmetrically substituted across the branch juncture with another sequence that is unsymmetrically substituted across the branch juncture.
Scheme 12-6 Retrosynthetic analysis of a unique macrotetracyclic nanostructural target.
The nine-unit branched sequences 22 and 23 of Scheme 12-7 were synthesized without incident in 70% and 46% overall yield, respectively using a combination of branched and unbranched monomers. The tandem cyclization step was the most challenging and crucial reac-
432
12 Acetylenes in Nanostructures
iiMe3
I
[Pd(dba)*] / PPh3 Cul /TEA / 50'C 24 h I0.002M
\
1) CH3l/ 1Oo'C / sealed tube 2) KOH (cat.) / MeOH I CH& 3) [Pd(dbah] / PPh3 / Cul / TEA / 70'C/High Dilution 17
Scheme 12-7 Synthesis of phenylacetylene macrotetracycle 17.
tion of the synthesis. This cyclization involved a bimolecular coupling and a unimolecular cyclization all in one pot. The reaction conditions, in particular the concentration of the starting branched sequences, required careful optimization. At high concentrations bimolecular coupling as well as intermolecular oligomerization were favored, whereas low concentrations M to 1 x favored intramolecular cyclization. Under dilute conditions (from 2 x M), the major product isolated was the self-coupling of the terminal acetylenes. The self-coupling of terminal acetylenes is more severe under dilute conditions and is catalyzed by adventitious molecular oxygen. It was found that the strict exclusion of oxygen, low reaction temperatures, and extended reaction times were optimal. Extensive experimentation established that a 50% molar ratio of catalyst to sequence and 0.002 M sequence concentration in dry triethylamine solution gave the highest yield of the 18-unit tetrafunctional sequence 24. Interestingly, the best mode of addition was to add both sequences 22 and 23 together at the start, rather than attempt a controlled slow addition of one sequence to the other.
12.4 Synthesis of Sequence-Speci3c Phenylacetylene Oligomers
433
The triazene groups of the 18-unit branched sequence 24 were converted to the corresponding iodides before the trimethylsilyl protecting groups were removed (67 Vo yield for the two steps). Double cyclization of this precursor gave macrotetracycle 17 in 50% isolated yield. The macrotetracycle 17 has very poor solubility in common organic solvents, presumably because of its highly symmetric structure. Characterization of macrotetracycle 17 was difficult due to its low solubility. dl-Bromoform was found to be the solvent of choice for NMR studies because of its high boiling point and the adequate solubility of 17 at elevated temperatures. The structure of 17 was confirmed by complete ‘H-NMR assignment as well as high-resolution Fourier transform mass spectrometry (FT-MS).
12.4 Synthesis of Sequence-Specific Phenylacetylene Oligomers and Dendrimers on an Insoluble Solid Support To this point the synthetic routes to precisely defined linear, dendrimeric, and macrocyclic poly(phenylacety1ene)-based nanoarchitectures have been discussed, as has the analogy to the natural “backbone” motifs. Combinatorial strategies for peptide or oligonucleotide syntheses have recently emerged and, in general, are possible for syntheses that employ a variety of monomers in repetitive coupling and protection-deprotection cycles [34]. A multiple parallel synthetic approach toward the construction of phenylacetylene sequences will provide access to small quantities of a diverse pool of the corresponding functionalized macromolecules, which can then be screened for desired properties. This approach may be relevant to problems in supramolecular chemistry by yielding structure-property correlation maps that should provide insights into key molecular parameters and strategies for specific aggregate association. We have taken steps toward this end by demonstrating the feasibility of a polymer-supported Merrifield-like phenylacetylene sequence synthesis [35]. Enhanced yields and greatly simplified purification have been observed and the method is suitable for small-scale (<100 mg) to multigram procedures for either monodendrons or phenylacetylene sequences. One can envision three possible scenarios for adapting this chemistry to an insolublepolymer-supported synthesis : mask the terminal acetylene or the aryl iodide with the polymer, or link the sequence to the support through a pendant functional group. The efforts described here focus on the strategy of using the insoluble polymer as a masked aryl iodide by attaching functionalized aromatic monomers to polystyrene beads via the l-aryl-3,3-dialkyl triazene group as shown in Scheme 12-8. Scheme 12-9 illustrates the general synthetic approach that was taken during these studies. During this nonlinear repetitive growth scheme, a portion of the supported oligomeric sequence is liberated from the polymer and then coupled to the remaining supported sequence that has had the trimethylsilyl protecting group previously removed. The nonlinear repetitive doubling strategy was pursued without incident through the hexadecamer stage. The completeness of each reaction was estimated by infrared analysis of the polymer-bound substrate as shown in Fig. 12-10. Absorptions at 3311 cm-’ (strong) and 2109 cm-’ (weak) are characteristic of the terminal acetylenic carbon-hydrogen and carbon-carbon stretches, respectively and an absorption at 2156 cm-I (strong) is assigned to the carbon-carbon stretch of the trimethylsilyl-protected terminal acetylene [36]. The coupling reaction is accompanied by the disappearance of the 3311 cm-’ and 2109 cm-’ bands and the appearance of the 2156 cm-’
434
12 Acetylenes in Nanostructures
31
29
-Br
Scheme 12-8 Linking functionalized aromatic monomers to polystyrene beads via the l-aryl-3,3-dialkyl triazene functional group. Reagents: (a) dicyclohexylcarboiimide, 1-hydroxybenzotriazole, DMF, 25 "C, 48 h (b) sodium hydride, THF, 70°C,96 h; (c) potassium carbonate, DMF, O'C, 2 h. (d) n-propylamine, THF, 70T, ten days.
r
b
After n couplings
1
14 SiMe3 2"
Scheme 12-9 Nonlinear repetitive growth scheme for the synthesis of linear phenylacetylene oligomers. Reagents :(a) bis(dibenzy1ideneacetone)-palladium (0), cupious iodide, triphenylphosphine, triethylamine, DMF, 65"C,24 h; (b) potassium hydroxide, THF, MeOH, 75"C,1 h; (c) methyl iodide, llO"C, 6 h.
absorption. The trimethylsilyl deprotection step was similarly monitored, in that the 2156 cm-' band disappears, while those at 3311 cm-' and 2109 cm-' appear. The reliability
of this "null to null" infrared monitoring was confirmed by removal and characterization of
12.4 Synthesis of Sequence-Specific Phenylacetylene Oligomers
435
the product at every alternate step (with the trimethylsilyl protecting group in place). The impurities observed upon liberation of the sequence from the polymer are not identifiable as synthetic precursors and are easily removed by filtration through a small amount of SO,. The gel permeation chromatographs of the liberated sequences which have been purified by silica gel chromatography are shown in Fig. 12-10. Each sequence (monomer, dimer, tetramer, octamer, hexadecamer, and 32-mer) was analyzed separately and the chromatographs are overlaid. Infrared Spectra r
r
,
r
,
Gel Permeation Chromatographs
l2
-0.2 I 21
Wavenumber (cm-’)
23
25
27
29
1 31
Retention Time (min)
Figure 12-10 Infrared spectroscopic monitoring of coupling and deprotection reactions on polymerbound substrates and gel permeation chromatograms of the liberated sequences (monomer, dimer, tetramer, octamer, hexadecamer, and 32-mer). In the IR spectra n represents the number of repetitive reaction cycles: for example, n = l and n = 1’ correspond to trimethylsilyl-protected and unprotected dimer, respectively.
Attempts to double oligomer length from 16 to 32 were only partially successful, due largely to solubility problems. Not only was the liberated hexadecamer poorly soluble in media suitable for the Pd(0)-catalyzed cross coupling reaction, but perhaps more critically the swelling behavior of the polymer was diminished. The diminished swelling was initially observed during the post-coupling washing procedure, which involves repetitive shrinking and swelling of the resin. The infrared analysis is also adversely affected to the point where it is no longer an accurate indicator of reaction completeness, presumably because the sample preparation involves swelling the polymer with carbon tetrachloride. These experiments lead us to believe that the synthesis of monodisperse high-molecular-weight linear phenylacetylenes may be possible by lessening the polymer loading and incorporating more solubilizing groups. The multiple parallel synthesis of sequence-specific precursors to phenylacetylene macrocycles is a task that is well within the current range of usefulness of these solid-phase techniques. With the efficient way these polymer-bound reactions can be monitored, the combined time for the Pd(0) cross-coupling, trimethylsilyl deprotection, and liberation has been reduced to the sum of the conventional liquid-phase reaction times (31 h). Since intermediate oligomeric sequences must be isolated, purified, and characterized during conventional liquid-phase synthesis, the solid-phase approach is in practice significantly more efficient.
436
I2 Acetylenes in Nanostructures
12.5 Conclusions While rigid phenylacetylene macromolecules are precisely defined structurally and monodisperse, they generally have limited solubility characteristics and they are in the size regime where the ability to characterize fully by “traditional” means (particularly NMR) may be limited. Indispensable supplemental characterization methods include size exclusion chromatography, silver chemical ionization laser desorption (LDCI) mass spectrometry, and ultraviolet matrix-assisted laser desorption/ionization time-of-flight (MALDITOF) mass spectrometry. All of these techniques require only marginal solubility of the analyte. An example shown in Fig. 12-11 is the partial LDCI Fourier transform m a s s spectrum of the silverattached macrotetracycle 17, for which the anticipated molecular weight was obtained within experimental error [37]. The essential information regarding sample preparation for the mass spectroscopy techniques is given in Section 12-6.
Measured Mass: 2465.9420
I
..
246oM)
, . I . .
246250
“ I . ” ‘ I . . . . I ” ” I
2465.W
246750
2470130
247250
m/z Figure 12-11 Partial LDCI Fourier transform mass spectrum of silver-attached macrotetracycle 17 (calcd. mass for C,,oH,,2Ag0, is 2465.9244 u; found 2465.9420 u).
The examples in this chapter were selected to illustrate how wide structural diversity can be achieved from only a few constructing reactions and monomers. The unique structural features of the phenylacetylene unit make it a suitable component of molecular modules for
12.6 Experimental Procedures
437
supramolecular assembly. The power of acetylene chemistry is in its versatility, making the chemist's imagination the only limit to its use in creating new nanoarchitectures. It is a safe bet that modern acetylene chemistry will continue to play a prominent role in forging the frontiers of synthetic chemistry and materials science.
12.6 Experimental Procedures 12.6.1 48-Cascade:Benzene[3-1,3,51: (5-Ethynyl-1,3-phenylene)G: 5-Ethynyl1,3-di(tert-buty1)benzene (8) A single-neck flask with a side arm was charged with the deprotected 31-mer monodendron (140 mg, 0.0286 mmol), triiodobenzene (3.3 mg, 0.007 mmol), [Pd(dba),] (1.6 mg, 0.003 mmol), triphenylphosphine (6 mg, 0.023 mmol), copper(1) iodide (0.6 mg, 0.003 mmol), and dry triethylamine (3 mL). The flask was stoppered, degassed under vacuum, back-filled with nitrogen and stirred at 40°C for two days. The disappearance of terminal acetylene was monitored by TLC. After completion, the reaction mixture was diluted with dichloromethane, washed with water ( 3 x ) and saturated aqueous sodium chloride, dried over anhydrous magnesium sulfate, and filtered. The crude mixture was separated by gravity column chromatography, eluting with 7 : 1 hexane/CH2C1, increasing to 6 : 1 hexane/CH,Cl, to give the major product as a clear, colorless glass (40 mg).
12.6.2 General Procedure for Double Cyclization A septum-stoppered flask charged with [Pd(dba),] (0.10 g, 0.18 mmol), triphenylphosphine (0.27 g, 1.08 mmol), copper(1) iodide (0.03 g, 0.18 mmol) was evacuated and back-filled with nitrogen. Dry triethylamine (40 mL) was added and the solution was heated to 70°C. The deprotected branch sequence (0.12 mmol) was taken up in a mixture of triethylamine (40 mL) and toluene (10 mL) and added to the catalyst by a syringe pump at a rate of 2.5 mL/h-'. At the end of the addition, the solution was stirred for another 10 h and then diluted with CHzClz. The solution was washed with aqueous 1 M HC1, brine, and water, and dried over sodium sulfate. After the solvent had been removed, the residue was purified by flash chromatography to afford a white solid.
12.6.3 Sample Preparation for Mass Spectrometry For infrared laser desorption Fourier transform mass spectrometry (LD-FTMS), all dendrimer samples were prepared by dissolving ca. 1 mg of sample in CHZClz, followed by deposition upon stainless steel probe tips by the aerosol spray technique described previously [37]. Dendrimer samples 4 and 5 were deposited directly onto a stainless steel probe tip. The sample 6 was prepared by first spraying 50 mL of a saturated silver nitrate/ethanol solution (containing ca. 3 mg of silver nitrate) onto the rotating probe tip, prior to dendrimer deposition. Samples were introduced into the vacuum system and the source cell pressure reduced Torr, before analysis. to 2.2 x lo-' Torr and the analyzer cell pressure to 2.0 x
438
12 Acetylenes in Nanostructures
For ultraviolet MALDI-TOF mass spectrometry of the dendrimers 7 and 8 , trans-retinoic acid was dissolved in CH,Cl, (Fisher, ACS grade) to form a 0.03 M solution which was combined with CH2C12solutions of the dendrimers to produce solutions with a matrix/analyte ratio of 550: 1. Prior to use, the retinoic acid matrix solution was irradiated with fluorescent light for approximately 48 h at room temperature. For analysis of dendrimer 7, an aliquot corresponding to 75 pmol of the sample was transferred to the surface of a stainless steel sample pin and allowed to dry. Following that, a 0.02 M solution of poly(ethy1ene glycol)-1000 (PEG-1000) in CH2C12was aerosol-sprayed onto the pin to form a film on top of the dried 7/matrix mixture. For analysis of dendrimer 8 a 0.9 mL aliquot of the PEG-1000 solution was deposited upon a sample pin and allowed to dry, prior to addition of an aliquot of the Wmatrix mixture corresponding to 74 pmol of dendrimer 8.
12.6.4 Procedures for Solid-Supported Phenylacetylene Chemistry 12.6.4.1 General Procedure A: Pd(0)-Catalyzed Coupling Reactions (Except for 'himethylsilylacetylene) A flask equipped with a heavy-walled side-arm was charged with polymer-supported terminal acetylene (1.0 equiv.), oligomeric aryl iodide (1.1 equiv.), bis(dibenzy1ideneacetone)palladium(0) (22 mequiv.), triphenylphosphine (110 mequiv.), and copper(1) iodide (22 mequiv.). The flask was evacuated and back-filled with dry nitrogen three times, a degassed mixture (4.5mL/g of polymer) of triethylamine (2 parts) and N,N-dimethylformamide (1 part) was added, the flask was sealed, and the suspension was stirred at 65 "C for 24 h. The polymer was transferred to a tared fritted filter using MeOH, washed sequentially (ca. 30 mL/g of polymer) with MeOH, DMF, CH2CI2, MeOH, a solution of sodium diethyldithiocarbamate (377 mg) and diisopropylethylamine (0.33 mL) in DMF (33 mL), DMF, CH2C1,, and MeOH, and dried in vacuo.
12.6.4.2 General Procedure B : Pd(0)-Catalyzed Coupling with Trimethylsilylacetylene A flask equipped with a heavy-walled side-arm was charged with polymer-supported aryl bromide (1.0 equiv.), bis(dibenzylideneacetone)palladium(O) (22 mequiv.), triphenylphosphine (110 mequiv.), and copper(1) iodide (22 mequiv.). The flask was evacuated and back-filled with dry nitrogen three times, a degassed mixture (5.7mL/g of polymer) of triethylamine (2 parts) and N,N-diemthylformamide (1 part) was added, trimethylsilylacetylene (2.0 equiv.) was added, the flask was sealed, and the suspension was stirred at 75°C for 48 h. The polymer was transferred to a tared fritted filter using MeOH, washed sequentially (ca. 30 mL/g of polymer) with MeOH, DMF, CH2C12,MeOH, a solution of sodium diethyldithiocarbamate (377 mg) and diisopropylethylamine (0.33 mL) in DMF (33 mL), DMF, CH2C12and MeOH, and dried in vacuo.
12.6.4.3 General Procedure C: 'Ikimethylsilyl Deprotection Polymer-supported aryl trimethylsilylacetylene (1.00 g) and finely powered KOH (0.10 g) were evacuated to 10 mm Hg and then blanketed with dry N,. A degassed mixture of THF (7 mL)
12.6 Experimental Procedures
439
and MeOH (4.5 mL) was added, the flask was sealed, and the suspension was heated with gentle stirring at 75°C for 1 h. The polymer was transferred to a tared fritted filter using MeOH, washed sequentially (ca. 30 mL) with MeOH, H,O, MeOH, isopropanol, DMF, CH2C12,and MeOH, and dried in vacuo.
12.6.4.4 General Procedure D: Liberation of the Oligomeric Sequence from the Support
The polymer was suspended in the minimum amount of degassed methyl iodide, blanketed in nitrogen, and heated at 110°C in a sealed tube for 6 h. After the methyl iodide had been removed in vacuo, the product was extracted from the resin using hot CH,Cl,, the resulting solution was cooled to 25 "C and filtered through SO,, and the solvent was removed in vacuo to give the product (ca. 95% pure).
12.6.5 Peptide Linkage to Aminornethylated Polystyrene (26) Aminomethylated polystyrene (5.00 g, 0.9 mequiv., 200-400 mesh, 1To divinylbenzene) in DMF (50 mL) was stirred at 25 "C for 1 h, acid 25 (1.48 g, 4.72 mequiv.), and l-hydroxybenzotriazole (0.64 g, 4.77 mequiv.) were added, stirring at 25 "C was continued for 0.5 h, a solution of dicyclohexylcarbodiimide (0.98 g, 4.77 mequiv.) in DMF (25 mL) was added, and the resulting suspension was stirred at 25 'C for an additional 48 h. The polymer was transferred to a tared fritted filter using MeOH, washed sequentially (ca. 30 mL/g) with MeOH, DMF, CH,CI,, DMF, and MeOH, and dried in vacuo. The polymer was stirred at 60°C in a mixture of acetic anhydride (200 mL) and pyridine (5 mL), transferred to a tared fritted filter using MeOH, washed sequentially (ca. 30 mL/g) with MeOH, DMF, CH2Cl,, DMF, and MeOH, dried in vacuo, and analyzed by Kaiser's qualitative ninhydrin test [38].
12.6.6 Ether Linkage to Chlorornethylated Polystyrene (28) [39] Chloromethylated polystyrene (10.0 g, 0.7 mequiv./g, 1070 cross-linked with divinylbenzene, 200-400 mesh), 1-(3-bromophenyl)-3-(3-piperidinemethanol)triazene(27; 7.8 g, 261 mmol), NaH (770 mg, 131 mmol) and dry THF (100 mL) were combined and heated at 70°C in a sealed tube for 96 h. The polymer was transferred to a tared fritted filter using MeOH, washed sequentially (ca. 30 mL/g of polymer) with MeOH, THF, H,O, THF, and MeOH, and dried in vacuo.
12.6.7 Propylaminomethylated Polystyrene (29) A suspension of 10.0 g of chloromethylated polystyrene (0.7 mequiv./g, 1Yo cross-linked with divinylbenzene, 200-400 mesh), n-propylamine (2.3 mL, 28.0 mmol) and THF (120 mL) was heated at 70"C for ten days. The polymer was transferred to a tared fritted filter using MeOH, washed sequentially (100 mL each wash) with MeOH, H,O, 15% NaOH, H,O, THF and
440
12 Acetylenes in Nanostructures
MeOH, and dried in vacuo to give 9.84 g of propylaminomethylated polystyrene 5 (0.7 mequiv./g of nitrogen, 19'0 cross-linked with divinylbenzene, 200-400 mesh): Analysis - found: C, 90.44; H, 7.87; N, 0.95; C1, 0.18%.
12.6.8 Direct Triazene Linkage to Propylaminomethylated Polystyrene (31) 3-Bromobenzenediazonium tetrafluoroborate [40] (30; 3.76 g, 13.8 mmol) was added to a chilled (0 "C) suspension of propylaminomethylated polystyrene 29 (0.7 mequiv/g of nitrogen 1 Vo cross-linked with divinylbenzene, 200-400 mesh) (10.00 g), potassium carbonate (2.10 g, 15 mmol), and DMF (8 mL). The suspension was stirred for 2 h at 0°C. The polymer was transferred to a tared fritted filter using MeOH, washed sequentially (ca. 30 mL of polymer) with MeOH, H,O, MeOH, THF, and MeOH, and dried in vacuo.
Acknowledgement This work was supported by the National Science Foundation (Grants DMR-94-96225 and CHE-93-96298), the NSF Young Investigator Program (Grant CHE-94-96105) and the Office of Naval Research (N00014-94-1-0639). Additional support from the 3M Company, Menicon Co. Ltd., the Petroleum Research Foundation, and the Camille Dreyfus Teacher-Scholar Awards Program is gratefully acknowledged.
Abbreviations CHARMm dba DEDG DMF FTMS F(4 LD-FTMS LDCI MALDI-TOF PAM PAS rt SEC SYNDROME TEA THF
B
molecular mechanics force field bisdibenzylideneacetone double exponential dendrimer growth N,N-dimethylformamide Fourier transform mass spectrometry CHARMm force field constant laser desorption-FTMS chemical ionization laser desorption matrix-assisted laser desorption/ionization time-of-flight phenylacetylene macrocycle phenylacetylene sequence room temperature size exclusion chromatography synthesis of dendrimers by repetition of monomer enlargement triethylamine tetrahydro furan valence angle deformation
References
441
References [l] J. S. Moore, E. J. Weinstein, Z. Wu, Tetrahedron Lett. 1991, 2465-2466. [2] J. Zhang, J. S. Moore, Z. Xu, R. A. Aguirre, J. Am. Chem. SOC. 1992, 114, 2273-2274. [3] Z. Wu, S. Lee, J. S. Moore, J. Am. Chem. SOC. 1992, 114, 8730-8732. [4] R. B. Merrifield, Angew. Chem. Int., Ed. Engl. 1985, 24, 799-892. 151 The force constant was obtained from the QUANTA Version 3.3 Parameter Handbook, Molecular Simulations, Inc. 200 Fifth Avenue, Waltham, MA 02154, USA. [6] A. Mavridis, I. Moustakali-Mavridis, Acta Crystallogr. Sect. B: Struct. Sci. 1977, 33, 3612-3615. [7] Cambridge Structural Database User Manuals, Crystallographic Data Centre, Cambridge, 1991. [8] K. Okuyama, T. Hasegawa, M. Ito, N. Mikami, J. Phys. Chem. 1984, 88, 1711-1716. [9] A. Liberles, B. Matlosz, J. Org. Chem. 1971, 36, 2710-2713. [lo] (a) J. K. Young, G. R. Baker, G. R. Newkome, K. F. Morris, C. S . Johnson, Jr., Macromolecules, 1994,27,3464-3471; (b) G. R. Newkome, J. K. Young, G. R. Baker, R. L. Potter, L. Audoly, D. Copper, C. D. Weis, K . F. Morris, C. S. Johnson, Jr., ibid. 1993, 26, 2394-2396. [ll] Z. Xu, J. S . Moore, Acta Polym. 1994, 45, 83-87. [12] For review articles, see: (a) D. A. Tomalia, H. D. Durst, Top. Cum Chem. 1993, 165, 192-313; (b) H.-B. Mekelburger, W. Jaworek, F. Vogtle, Angew. Chem. Int. Ed. Engl. 1992,31, 1571-1576; (c) C. N. Moorefield, G. R. Newkome, Advances in Dendritic Macromolecules JAI Press, 1994, Vol. 1, pp. 1-67; (d) G. R. Newkome, C. N. Moorefield, G. R. Baker, Aldrichim. Acta 1992, 25, 31-38. (131 (a) K. Sonogashiri, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 4467-4470; (b) L. Cassar, J. Organomet. Chem. 1975, 93, 253-257; (c) H. A. Dieck, R. F. Heck, ibid. 1975, 93, 259-263. [I41 (a) R. Eastmond, D. R. M. Walton, Tetrahedron 1972, 28, 4591-4599; (b) C. Fxiborn, D. R. M. Walton, J. Organomet. Chem. 1965, 4, 217-228. [15] G. R. Newkome, Z.-Q. Yao, G. R. Baker, V. K. Gupta, J. OR. Chem. 1985, 50, 2003-2005; (b) G. R. Newkome, Z.-Q. Yao, G. R. Baker, V. K. Gupta, P. S. Russo, M. J. Saunders, J. Am. Chem. Soc. 1986, 108, 849-850; (c) G. R. Newkome, C. N. Moorefield, G. R. Baker, A. L. Johnson, R. K. Behera, Angew. Chem., Int. Ed. Engl. 1991,30, 1176-1178; (d) G. R. Newkome, X. Lin, J. K. Young, Synlett 1992, I , 53-54; (e) G. R. Newkome, A. Nayak, R. K. Behera, C. N. Moorefield, G. R. Baker, J. Org. Chem. 1992, 57, 358-362. [I61 D. A. Tomalia, H. Baker, J. D d d , M. Hall, G. Kallos, S. Martin, J. Roeck, J. Ryder, P. Smith, Polym. J. (Tokyo) 1985, 17, 117-132; (b) D. A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S. Martin, J. Roeck, J. Ryder, P. Smith, Macromolecules 1986, 19, 2466-2469; (c) A. B. Padius, H. K. Hall, Jr., D. A. Tomalia, J. R. McConnel, J. Org. Chem. 1987, 52, 5305-2312. [17] (a) C. J. Hawker, J. M. J. Frtchet, J. Chem. Soc,, Chem. Commun. 1990, 1010-1013; (b) C. J. Hawker, J. M. J. Frkhet, 1 Am. Chem. SOC.1990,112,7638-7647; (c) K. L. Wooley; C. J. Hawker, J. M. J. Frtchet, J. Chem. SOC.Perkin. 7 h s . I 1991, 1059-1076; (d) K. L. Wooley, C. J. Hawker, J. M. J. Frechet, J. Am. Chem. Soc. 1991, 113, 4252-4261. [18] T. M. Miller, T. X . Neenan, Chem. Muter. 1990,2, 346-348; (b) E. W. Kwock, T. X.Neenan, T. M. Miller, ibid. 1991, 3, 775-777; (c) T. M. Miller, T. X. Neenan, R. Zayas, H. E. Blair, 1 Am. Chem. SOC.1992, 114, 1018-1025. [19] (a) J. S. Moore, Z. Xu, Macromolecules 1991,24,5893-5894; (b) Z . Xu, J. S. Moore, Angew. Chem. Znt. Ed. Engl. 1993,32, 246-2.48; (c) Z. Xu, M. Kahr, K. L. Walker, C. L. Wilkins, J. S. Moore, J Am. Chem. SOC.1994, 116, 4537-4550; (d) Z. Xu, B. Kyan, J. S . Moore, Advances in Dendritic Macromolecules, JAI Press 1994, Vol. 1, pp. 69-104. [20] For a proposed systematic nomenclature scheme for dendrimers, see: (a) G. R. Newkome, G. R. Baker, J. K. Young, J. G. Traynham, J. film. Sci., Pprt A 1993,31, 641-651; (b) G.R. Baker, J. K. Young, Advances in Dendritic Macromolecules, JAI Press 1994, Vol. 1, pp. 169-186. [21] P. Bharathi, U. Patel, T. Kawaguchi, D. Pesak, J. S. Moore, submitted for publication. [22] P. G. de Gennes, H. Hervet, J. Phys. Lett. 1983, 44, L351-L360.
442
12 Acetylenes in Nanostntctures
[23] Z. Xu, J. S. Moore, Angew. Chem., Int. Ed. Engl. 1993, 32, 1354-1357. (24) T. Kawaguchi, K. L. Walker, C. L. Wilkins, J. S. Moore, J. Am. Chem. Soc. 1995, 117, 2159-2165. [25] E. Igner, 0. I. Paynter, D. J. Simmonds, M. C. Whiting, J. Chem. Soc, Perkin Trans. I 1987, 2447-2454. [26] D. Venkataraman, S. Lee, J. Zhang, J. S. Moore, Nature 1994, 371, 591-593. [27] J. Zhang, D. J. Pesak, J. J. Ludwick, J. S. Moore, J. Am. Chem. SOC. 1994, 116, 4227-4239. [28] A. S. Shetty, K. F. Stork, J. S. Moore, P. W. Bohn, unpublished results. [29] J.-M. Lehn, Angew. Chem, Int. Ed. Engl. 1990, 29, 1304-1319. [30] (a) 0. 1. Paynter, D. J. Simmonds, M. C. Whiting, J. Chem. Soc, Chem. Commun. 1982, 1165-1166; (b) I. Bidd, M. C. Whiting, J. Chem. Soc, Chem. Commun. 1985, 543-544. [31] C. D. Eisenbach, H. Hayen, H. Nefzger, Makromol. Chem., Rapid Commun. 1989, 10, 463-475. [32] K. Ziegler, H. Eberle, H. Ohlinger, Justus Liebigs Ann. Chem. 1933, 504, 94-115. [33] 2. Wu, Ph. D. Thesis, The University of Michigan, 1994. [34] G. Jung, A. G. Sickinger, Angew. Chem., Znt. Ed. Engl. 1992, 31, 367-383. 1351 (a) J. K. Young, J. C. Nelson, J. S. Moore, Polym. Prepr. (Am. Chem. Soc, Div. Polym. Chem.) 1994, 35(2), 988-989; (b) J. K. Young, J. C. Nelson, J. S. Moore, J. Am. Chem. Soc. 1994, 116, 10841- 10842. [36] T.B. Grindley, D. F. Johnson, A. R. Katritsky, H. J. Keogh, C. Thirkettle, R. D. Topsom, J. Chem. Soc, Perkin Trans. 2 1974, 3, 282-289. [37] This information was obtained by Mr. Michael S. Kahr in Professor Charles L. Wilkins’ laboratory. For additional information see: M. S. Kahr, C. L. Wilkins, J. Am. Soc. Muss Spectrorn. 1993, 4, 453-460. [38] E. Kaiser, R. L. Colescott, C. D. Bossinger, P. I. Cook, Anal. Biochem. 1970, 34, 595-598. [39] W. M. Mackenzie, D. C. Sherrington, . I Chem. SOC,Chem. Commun.1978, 541-543. I401 M. P. Doyle, W. J. Bryker, J. OR. Chem. 1979, 44, 1572-1574.
13 Oligoacetylenes Francois Diederich
13.1 Introduction Acetylenic scaffolding has been identified as a promising way to prepare novel molecular and polymeric carbon allotropes and carbon-rich nanoarchitecture with unusual structure, electronic, and optical properties [l-31. In 1987, with the arrival of graduate student Yves Rubin in my laboratory at UCLA, we started a research program aimed at the preparation of such compounds, and this account will provide a survey mainly of the synthetic chemistry that was developed since then. In Section 2, the synthetic approaches to the cyclo[n]carbons (cycloC,), n-membered monocyclic rings of sp-hybridized C-atoms, are summarized. The following section describes the synthesis of a variety of differentially protected tetraethynylethene derivatives, some of which are potential precursors to two-dimensional all-carbon networks. Starting from these monomeric tetraethynylethene derivatives, perethynylated dehydroannulenes and expanded radialenes with large carbon cores and unusual electronic properties were prepared on the way to the two-dimensional networks, and these developments are presented in a separate section. Finally, Section 5 introduces the reader to the syntheses of molecular wires with new all-carbon backbones. A brief survey of historic developments is presented at the beginning of each section introducing a new class of compounds, and key experimental procedures are given as well.
13.2 Synthetic Approaches to the Cyclocarbons Cyclo[n]carbons (cyclo-C,) are n-membered monocyclic rings of sp-hybridized C-atoms with unique electronic structures resulting from two perpendicular systems of conjugated norbitals, one in-plane and one out-of-plane [4]. In synthetic approaches aimed at preparing these molecular carbon allotropes from stable, well-characterized precursors, we first targeted cycloc 1 8 for which early theoretical work predicted special Hiickel-aromatic stabilization as a result of two orthogonal (4n + 2) %electron systems [ 5 ] . More recent computational studies made different predictions of the electronic structure of cycIo-Cl8:self-consistent field (SCF) calculations with a 3-21G or larger basis set predicted that the cyclic acetylenic D,, structure 1a with alternating bond lengths represents the ground-state geometry [4]. But optimizations at the Msller-Plesset second-order perturbation theory (MP2) level including valence electron correlations [6] as well as density functional theory calculations [7] favor the cumulenic Dlgh structure 1b as the most stable planar monocyclic geometry. Scheme 13-1 shows the three synthetic routes, via precursor molecules (2-7) with the desired all-carbon macrocyclic frameworks already preformed, that were pursued in the synthesis of C~clo-C,~ and higher homologs. The parent dehydroannulenes 8 and 9, related to macrocycles 2-4, had previously been prepared by Sondheimer and co-workers as highly unstable compounds that exploded upon heating [8, 91.
444
13 Oligoacetylenes
7
n lb
la
n
o
n
6, n = l 7, n = 2
3, n = l 4. n = 2 5; n = 3 f''
..," ./'
P'. CyClO-Ci a
Scheme 13-1 Synthetic routes to c y ~ l o - C , ~ .
H
H
n
H
H 8,n-1
9,n=2
-12 co
- 3 Ph2PCHpPPh2
445
13.2 Synthetic Approaches to the Cyclocarbons
13.2.1 The Retro-Diels-Alder Route to Cyclo-C,, Compound 2 was chosen as a direct precursor to cyclo-C,, since it should lose three anthracene molecules in a retro-Diels-Alder reaction under thermal conditions (Scheme 13-1). The synthesis of 2 (Scheme 13-2) started with the Diels-Alder reaction of anthracene and trans-dichloroethene [lo], followed by dehydrochlorination and subsequently bromination to 10.The latter conversion was best achieved by simply adding elemental bromine to a solution of the vinyl anion formed with n-BuLi [ll]. Palladium-catalyzed alkynylation of 10 with trimethylsilylacetylene in a-butylamine followed by deprotection with aqueous KOH in MeOH gave the diethynyl derivative 11 as very unstable crystals, which in one case exploded spontaneously. 1) E-CICH=CHCI, 215 'C,
90% 2) tBuOK, THF, A, 90% 3) nBuLi.TMEDA, - 78 "C, then Br2, - 60 'C, 86%
10
H
Scheme 13-2 Synthesis of hexadehydro[l8]annulene 2.
The oxidative coupling of 11 under Eglinton-Glaser conditions [12] afforded the cyclic trimer 2 as the only identifiable product, in 25% yield. The absence of dimer, tetramer, or pentamer can be explained by the steric matching between monomer 11 and the macrocyclic oligomer, which is found experimentally to determine largely the product distributions in Eglinton-Glaser couplings of cis-enediynes to form dehydroannulenes [l b, 131. The C = C - C(sp) angle, a, in 11 (Scheme 13-2) is calculated by MM2 (molecular mechanics 2) as 125" and by AM1 (Austin model 1) as 127" - values close to the average value of 123.3 found for the corresponding bond angles in the X-ray crystal structure of 2. The X-ray crystal structure of 2 shows a nearly perfectly planar hexadehydro[lS]annulene ring which is sterically shielded by the six benzene rings and therefore protected with respect to bimolecular and polymerization reactions. Accordingly, 2 is highly stable to air and can be heated in a sealed tube to 250 "C in the solid state without appreciable decomposition. Both 'H-NMR and electronic absorption spectroscopic data suggest that 2 is aromatic. Starting from 2, cycIo-CI8was prepared in the gas phase by laser flash heating and the neutral product, formed by stepwise elimination of three anthracene molecules in retroDiels- Alder reactions, was detected by resonant two-photon ionization time-of-flight mass O
446
13 Oligoacetylenes
spectrometry [4].Attempts to prepare macroscopic quantities of the cyclocarbon by flash vacuum pyrolysis using solvent-assisted sublimation 1141 only afforded anthracene and polymers.
13.2.2 The 3-Cyclobutene-1,2-dione Route to the Cyclocarbons The synthesis of the macrocyclic carbon oxide precursors 3-5 to the cyclocarbons C,8, C,,, and C , started with the preparation of the 3,4-dialkynyl-3-cyclobutene-l,2-diones 12a-g, which were unknown despite the extensive work reported on squaric acid and derivatives [15-171. Using organotin [18]and organocopper 1191 compounds, two methods were developed to prepare these colored compounds in good to high yield (Table 13-1) [14,20, 211. The direct route to the carbon oxides required the oxidative cyclization of the diethynyl derivative 13 (Scheme 13-3). However, all attempts to prepare this compound by deprotection of 12d-f failed, presumably due to the extreme reactivity of 13 as a Michael acceptor. Instead, the diketall4 was formed from dione 12e under unusually forcing conditions, followed by desilylation (Scheme 13-3). Hay coupling [22] of 14 subsequently yielded the three cyclobutenefused dehydroannulenes 15-17 in good overall yield. In contrast to the parent compounds 8 and 9, the three macrocycles 15-17 are stable in moderate temperature ranges and when exposed to air. The extra stabilization of the cyclobutene-fused derivatives is a result of enhanced rigidity of the ring skeleton due to the annelation [lb, 131. Table 13-1 Preparation of 3,4-dialkynyl-3-cyclobutene-1,2-diones
R'
12a-g
R
Product
Yield (To), m.p., appearance
Ph-C=C nPr Me,Si (iPr),Si
12a 12b 12c 12d 12f
70, 114-115"C, orange needles 11, > 184"C, deep red crystals 51, yellow oil 30, 95-97 "C, yellow needles 80, 37-38 "C, yellow needles
(B) Copper acetylide route (M = Cu)@) Me+ tBuMelSi (iPr),Si Me$ -C= C
12d 12e 12f 1%
27
R
(A) Tributyltin route (M = SnBu,)" Ph
68, 57-59"C, orange crystals 59 27, unstable brown oil
Conditions: [Pd(PPh,),], CICH2CH2CI,20 "C. @) Conditions: treatment of alkyne with nBuLi, then Cur; then alkynylation of 3,4-dichloro-3-cyclobutene-1,2-dione in THF, 20°C.
447
13.2 Synthetic Approaches to the Cyclocarbons
H BuMepSi
H 13
SiMe2iBu
H
12e
14
0 3
14
O2/ acetone n
n
' 0 0J
n = 1, 15 (16%) n=2,16(18%) n = 3 , 1 7 ( 7%)
0
0
n = 1 , 3 (99%) n = 2,4 (79%) n = 3 , 5 (90%)
Scheme 13-3 Synthesis of the carbon oxides 3-5.
The removal of the ketal protecting groups in 15-17 to form the corresponding carbon oxides proved to be exceptionally difficult and could only be achieved by dissolution of the compounds in concentrated sulfuric acid [21]. The particular sensitivity of the products formed also required unusual workup conditions. Dilution of the sulfuric acid solutions with water led to instant polymerization of the product. Therefore, the sulfuric acid solutions were extracted with 1,2-dichloroethane and the organic layer neutralized with powdered CaCO, to give 3-5 in high yields. The X-ray crystal structure of orange-yellow 3 showed a perfectly planar annulene perimeter with considerable angle strain in the three butadiynyl moieties [21]. The carbon oxides 3-5 are extremely sensitive to nucleophiles which induce polymerization through initial Michael additon. Also, they undergo rapid Diels-Alder reactions with dienes, for example furan, leading to strained derivatives that subsequently polymerize. They are stable at room temperature but explode above 80 "C. The results of low-temperature matrix isolation studies with 3 are quite consistent with the photochemical formation of cycZo-C18via ketene intermediates and subsequent loss of six CO molecules [lo]. When 3 in a glass of 1,2-dichloroethane at 15 K was irradiated at X > 338 nm, the strong cyclobutenedione C = O band at 1792 cm-I in the FT-IR spectrum disappears quickly and a strong new band at 2115 cm-' appears. This band is assigned to diketene substructures that are photochemically generated from the cyclobutenedione moieties. Irradiation at X > 280 nm led to gradual decrease in intensity of the ketene absorption at 2115 cm-I and to the appearance of a weak band at 2138 cm-' which was assigned to the CO molecules extruded photochemically from the diketene intermediates. Attempts to preparatively isolate cyClo-C,, by flash vacuum pyrolysis of 3 or low-temperature photolysis of 3 in 2-methyltetrahydrofuran in NMR tubes at liquid-nitrogen temperature have not been successful.
448
13 Oligoacetylenes
Results of Fourier-transform mass spectrometry (FT-MS) studies with the carbon oxides 3-5 conclusively demonstrate that one reaction pathway to the formation of fullerenes is the coalescence of large cyclocarbon ions [21, 231. In the negative-ion mass spectra of the carbon oxides, the expected successive losses of CO molecules from the precursor anions to give the cyclocarbon ions C ,, ,C , and C, were observed. In the positive-ion mode, gas-phase coalescence reactions of the cyclocarbon ions produced fullerene ions. Ion-molecule reactions starting from the cyclocarbon cations C& (formed by laser desorption of 3), and C& (formed from 4) led through distinct intermediates C& C& .+ C& C& -+ C& + C2 and C& C& C& -+ C$, + C2, respectively) to fullerene C,o as the major product ion. These reactions are remarkably selective since the formation of the C& ion is not observed. In contrast, the ion-molecule dimerization reaction of C ~ C ~ O(produced -C,~ from carbon oxide 5 ) leads selectively to the buckminsterfullerene cation C&. These experiments, together with others [24], provide strong support for the chain-to-ring-to-sphere mechanism for the Kratschmer-Huffman fullerene production process starting from graphite [25]. +
-+
+
--$
13.2.3 The Transition Metal Complex Route to Cyclo-C,, When an alkyne reacts with [Co,(CO),], the two C=C-C angles in the formed (pacety1ene)dicobalt hexacarbonyl complex are reduced to a value of = 140" [26]; therefore, dicobalt hexacarbonyl fragments have been used as protecting groups to allow geometrically disfavored cyclization reactions by bending an alkyne moiety [27]. Since the alkyne can be
/
H
(Pr)3Si
I
18
1) [cO2(co)$, hexane, then dppm, toluene 75% 2) nBu4N+ F,wet THF
Scheme 13-4 Synthesis of the cobalt complexes of C,* (6) and C, (7).
13.3 Tetraethynylethenes
449
easily liberated from its complex through oxidation [28], alkyne-ligand exchange [29], or flash vacuum pyrolysis [30], the synthesis of cyclo-c,, seemed promising along the pathway outlined in Scheme 13-4 [lo, 311. Reaction of 1,6-bis(triisopropylsilyl)-1,3,5-hexatriyne with [co,(co),] afforded the sterically least crowded, rather unstable dicobalt complex 18. Attempts to remove the Si(iPr)3 groups in 18 with tetrabutylammonium fluoride (TBAF') in wet THF led only to decomposition, presumably due to the lability toward bases and nucleophiles of the free acetylene formed. When the hexatriyne was subjected to a one-pot reaction with [co2(co),] followed by ligand exchange with the bridging bis(dipheny1phosphino)methane (dppm) ligand, a heatand air-stable dicobalt complex was obtained which could be deprotected smoothly in nearquantitative yield to the very stable, dark red cyclization precursor 19. Oxidative coupling under high dilution conditions finally afforded the very stable cobalt complexes of c y ~ 1 0 - C ~ ~ (6) and cyclo-C2, (7), both as shiny black needles. The X-ray crystal structure of 6 showed considerable angle bending of the three butadiyne moieties within the nearly planar C18ring. With a value of 161 some of the C = C - C angles in 6 approach the degree of bending exThe UV spectra of 6 and 7 strongly suggest the presence of partial pected for c~c~o-C,,. cyclic conjugation in the central all-carbon rings [32]. Decomplexation of 6 and 7 has failed so far, due to the particular stability of the complexes provided by the dppm ligands; the preparation of phosphine-free complexes will be required to liberate the cyclocarbons under mild conditions. O,
13.3 Tetraethynylethenes, Fully Cross-Conjugated n-Electron Chromophores, and Other Perethynylated Molecules Tetraethynylethene (20) and its differentially protected derivatives are versatile building blocks for two-dimensional all-carbon networks and carbon-rich nanomaterials [l]. In addition, they attract interest for their fully cross-conjugated n-electron system [33]. The first tetraethynylethene derivative, 21a, was reported in 1969 by Hori and co-workers [34], and the persilylated and peralkylated derivatives 21b-d were prepared in the mid-1970's by Hauptmann [35]. In 1991, Hopf et al. [36] summarized this early synthetic work (Scheme 13-5) and reported the X-ray crystal structure of 21a; the authors also suggested in their paper the potential of substituted tetraethynylethenes as monomers for new polymers. Also in 1991, Rubin et al. [37] reported the first synthesis of the parent compound 20 by a synthetic route, which, after suitable modifications, provided access to tetraethynylethenes with any desired substitution and protection pattern. These transformations are the subject of this Section; the application of these compounds as precursors to two-dimensional all-carbon networks and carbon-rich nanomaterials will be discussed in the following sections.
13.3.1 Synthesis of Tetraethynylethene (20) and Geminally Bisdeprotected Derivatives In a general synthetic route (Scheme 13-6) (371, dialkynylketone 22 was converted [38] into the dibromomethylene derivative 23 and subsequent Pd(0)-catalyzed alkynylation [39] afforded the protected tetraethynylethenes 21d and 24ah. The X-ray crystal structure of 21d showed
450
13 Oligoacetylenes
RyR
R Br route a)
H
i
0
pyrrolidone, 73%
KI,acetone,
route b)
17-30°/o
i
route c)
/
;XR
BarnfordStevens = 4%
R
R
21a 21b 21c 21d
R
R = Ph (routes a,b) R = CH3 (route a) R = C(CH& (routes a,c) R = Si(CH3)3 (routes a,c)
Scheme 13-5 Early routes to tetrathynylethene derivatives. 0
Car4, PPh,
II
MeaSio
S
i
M
e
3 50%
21d R = SiMe3 (66%) 24a R = Si(FVj3 (65%) 24b R = Ph (61%)
eSiMe
Me3Si
nBuNHp, CsHe
20 R=H(67%) 25a R = Si(lPr)s (99%) 25b R = Ph
Scheme 13-6 Synthesis of tetrathynylethene (20) and geminally bisdeprotected derivatives.
a nearly planar molecule with a maximum deviation from the least-squares plane through the tetraethynylethene framework including the four Si-atoms of 0.032(9) A. Removal of the Me@ groups with catalytic amounts of K2C03 in methanol afforded smoothly the free alkynes 20 and 25a/b. The parent compound 20 crystallizes out of pentane at - 10°C as white plates which, at 25 "C, polymerize rapidly even in the absence of oxygen. The disilylated compound 25a decomposes more slowly, whereas the diphenyl derivative 25 b is extremely unstable in pure form: the neat oil polymerizes to a black hard mass in a matter of seconds. All three compounds can be stored for longer periods of time in refrigerated ethereal solution.
13.3 Tetraethynylethenes
451
13.3.2 Synthesis of Monodeprotected Tetraethynylethenes The synthesis of monodeprotected tetraethynylethenes starts from the unsymmetrically protected dibromomethylene derivatives 26a/b that are prepared as shown in Scheme 13-6 for 23 [40-421. Table 13-2 shows the reaction conditions for the palladium-catalyzed ethynylation to 27 a-d and the subsequent monodeprotection to 28 a-d. Remarkable is the high-yielding kinetically controlled removal of a trimethylsilyl in the presence of triethylsilyl and triisopropylsilyl protecting groups in the synthesis of 28 b. Table 13-2 (A) Preparation of tetrasubstituted tetraethynylethenesfrom unsymmetrically protected dibromomethylene derivatives
H,
R2
catalyst, Cul (cat), amine, CsHs Compound
R'
R2
Catalyst
Amine
Product
Yield (To)
26b 26 a 26b 26 a
(iPr),Si Et,Si (iPr),Si Et$i
(iPr),Si (iPr),Si Ph Et,Si
[PdC12(PPh3)2] [PdC12(PPh3)2] [PdCl,(PPh,),] [Pd(PPh,),]
(iPr),NEt (iPr),NEt (iPr),NEt nBuNH,
27a 27 b 27c 27d
41 41 38 44
(B)Synthesis of monodeprotected tetraethynylethenes catalyst, MeOHKHF (1:l) 28a-d Compound
R'
R2
Catalyst
Product
Yield (Vo)
27 a 27b 27 c 27 d
(rPr),Si Et,Si (1Fr),Si EtSi
(iPr),Si (iPr),Si Ph EtlSi
K2CO3 1 M NaOH(')
28s 28b 28 c 28d
98 98 51 89
()'
K2C03
1 M NaOH
Addition of a few drops and stirring for 2-5 min at 2OOC.
13.3.3 Synthesis of trans-Bis(triisopropylsily1)-Protected and trans-Bisdeprotected Tetraethynylethenes tmm-Bis(triisopropyIsily1)-protected tetraethynylethenes, including the bisdeprotected tmmenediyne 30a, are readily prepared in multigram quantities by the general route shown in Scheme 13-7 [43]. The synthetic sequence starts from dimethyl 2,3-dibromofumarate, which
452
I3 Oligoacetylenes Si(Pr)3 (IPT)~S~--S~BU~
1) DIBAL-H, CH2C12,O “C, 95% t
0
OMe [(PPh&PdC12], THF, 9 d, 92%
xsi(~r)3
2) PCC, CH~CIZ,85 %
CBr4, PPh3, Zn,
1 -i 9~
x
~
1) LDA r (6 )equiv),3THF, -78 “C
*
4
CH2CI2,75%
(IPT)~S~
t
//
(~Pr)~si
xR
R
Si(Br)3
2) Electrophile, 20 “C
Br
-
30a H
30b 3OC CHs(H0)HC I 3w (CH&CHCH202C
(IPr)aSi
Br 29
30e
95 23 81 65 79
Ph
31
is converted in four high-yielding steps into tetrabromide 29. Elimination of HBr and metalation with LDA generates a purple dianion which can be quenched with a number of mildly reactive electrophiles to yield compounds 30a-e. All of these derivatives are stable crystalline compounds, and the free diethynyl derivative 30a and diester 30d have been characterized by X-ray crystallography. The stability of 30a is quite remarkable since bisdeprotected tmns-enediynes typically polymerize upon concentration of their solutions [MI. An analysis of the crystal packing shows that the stability of 30a is related to the “insulating effect” of the Si(iPr), groups: the molecules align in a herringbone fashion with the reactive free ethynyl groups pointing directly toward the silyl groups of neighboring molecules. The importance of the Si(iPr), groups for the stability of 30a is clearly revealed in a comparison with the free trans-enediyne 31 which is prepared by a similar route [MI.The diphenyl derivative 31 is only stable in dilute solutions and decomposes instantaneously in the solid state.
13.3.4 Synthesis of cis-Bisdeprotected Tetraethynylethenes The route to the cis-enediyne 32 a starts from cis-2,3-dichloro-2-buten-l ,bdiol, which is obtained by exhaustive reduction of mucochloric acid (Scheme 13-8) (451. The two dibromomethylene residues in 36,the direct precursor to 32a, are introduced sequentially in a synthetic route which includes as the key step the selective reduction of orthoester 33 to the monoprotected diol34. Since compound 35 and all other intermediates with dibromomethylene groups
13.3 Tetraethynylethenes
( ‘ 3 s i x o H
(1?3)~Si
DIBAL-H (3 equiv),
1) PDC, CH2C12,
*
CHpCI2,
*
- 78 ‘C,
94 %
(~Pr)~si
’
1) a o \ B0’ - B r
4 34
O-OMe
2) CBr4, PPh3, Zn, 77%
453
XiOM 4
(~Pr)~si
35
LDA (7 equiv), THF,
65-70%
*
then NH4CI, 90 %
2) PDC, CHzCl2 3) CBr4, PPh3, Zn, 49 % 36
32a, R = S i ( ~ f r ) ~ 32b, R = SiMeg
Scheme 13-8 Synthesis of cis-bisdeprotected tetraethynylethenes.
on the way to 32a underwent facile cis-trans isomerizations in the presence of acid, acidic reaction and workup conditions had to be strictly avoided. Tetrabromide 36 was found to be most sensitive to isomerization: it was obtained as a yellow oil which, upon standing on the laboratory bench, yielded pale yellow crystals which were shown by X-ray crystallography to be the corresponding trans-tetrabromide 29 1451. To form the free cis-enediyne 32a, the deep purple solution of the cis-di(1ithium acetylide) formed by treatment of 36 with LDA was quenched at 0°C with saturated aqueous NH4Cl. Compound 32a was obtained as a yellow oil which is quite stable at room temperature in dilute solution. The neat oil decomposes slowly at room temperature and immediately at 60°C. By the same route, the bis(trimethy1sily1)-protected analog 32b was also obtained as a colorless unstable oil.
13.3.5 Other Perethynylated Compounds as Potential Monomers for Carbon Networks In Section 13-4, it will be shown that tetraethynylethenes are potential monomers for the construction of all-carbon networks [l]. Other perethynylated compounds have been reported, which could also serve as building blocks for infinite two- and three-dimensional carbon nets [46].They all represent fascinating small molecules of substantial structural and electronic interest [l, 2, 471. Vollhardt and co-workers [48] reported in 1986 the synthesis of hexaethynylbenzene 137, Scheme 13-9(a)] which, by oxidative polymerization could lead to a large planar carbon sheet
454
I3 Oligoacetylenes
a)
R Br
U ,.
I
Br
Ill
111
Br
37
38a R = Si(1Pr)~o(l8%)
38b R = BU(36A)
11 R
I
Me3Si
1) LDA, Ph2S2,88% 2) DIBAL-H, 88%
3) MCPBA, 81 % * M 4 ) TrisNHNH,, 90%
~ #'""~
s
~
I
MesSi'
Me3Si
4 Li-C=C-SiMe,, 22 hexane, 20 'C, 90% Me3Si
xH
3;
4
13.3 Tetraethynylethenes
,iL;:;
l)~~&$;c'
2) LDA, then HCI, 59%
/l)
CH3l
/
H BrZ, CH~CIZ, 82% 2) NaNH2, liq. NH3, then 5 N HCI, 67%
1) Br2, CH~CIZ, 41% 2) KNHz, liq. NH3, then 5 N HCI, 18%
Hxl H
42
H
>;; ::yXwR)
R
e)
455
2) CUI'PBU~
R = SiMe3 (23) R = Si(~pr)~
0
< R
,
43a R=SiMe3(55%) R 43b R = .Si(Pr), (42%)
R
H
H
1) Me3Si-SnMe3
[Pdddba)d, Aspha, DMF. 83% 2) K2CO3, CH30H,
79%
44
Scheme 13-9 Synthesis of perethynylated molecules and transition metal complexes as potential precursors to carbon networks.
named graphyne [49]. More recently, they also described the synthesis and structural properties of hexasilylated (38 a) and peralkylated (38b) derivatives of hexabutadiynylbenzene [SO]. Tetraethynylmethane (39), a potential monomer for a three-dimensional superdiamonoid carbon network [l], was elusive for many years [Sl, 521, until its synthesis was accomplished in 1993 by Feldman and co-workers [53]. The key step in the synthesis was the acid-mediated Johnson orthoester variant of the Claisen rearrangement, which provided the central quaternary methane C-atom with suitable functional groups for the ultimate transformation into 39 [Scheme 13-9(b)]. Solid 39, like tetraethynylethene (20), decomposes rapidly at room temperature in either the presence or absence of oxygen. The earlier efforts to prepare tetraethynylmethane had yielded the peralkynylated derivatives 40-42 [Scheme 13-9(c, d)] [Sl, 521. Tetraethynylallene represents another potential precursor for a three-dimensional carbon network [l], but remains elusive; of the perethynylated [n]cumulenes, so far only the silyl-protected [3]cumulenes 43a and 43b [Scheme 13-9 (e)] have been prepared [54]. With 44 [Scheme 13-9 (01, the first transition metal complex of a perethynylated ligand is now available [55].
456
13 Oligoacetylenes
13.4 Perethynylated Dehydroannulenes and Expanded Radialenes: Large Carbon Cores on the Way to All-Carbon Sheets Since tetraethynylethenes represent a repeat unit in more than one two-dimensional carbon network including 45 and 46 (Fig. 13-1) [I] the preparation of a specific network cannot be accomplished by simple oxidative polymerization of 20, but rather requires a more characteristic macrocyclic precursor as starting material. Macrocyclic precursors to extended carbon sheets are perethynylated dehydroannulenes [56] and expanded radialenes, novel carbon-rich materials with interesting and unusual structures and functions.
Figure 13-1 Planar all-carbon networks 45 and 46 derived from tetraethynylethene.
13.4.1 Perethynylated Dehydroannulenes Macrocyclic precursors to the carbon networks 45 and 46 are the perethynylated tetradehydro[l2]annulene 47a/b and hexadehydro[l8]annulenes48 a/b, respectively (Scheme 13-10). They are readily prepared by oxidative Hay coupling of the corresponding cis-bisdeprotected tetraethynylethenes 32a/b [57]. The yields of these coupling reactions are highly concentration-dependent (Scheme 13-10): the dimeric macroring forms preferentially at lower concentrations and the trimeric cycle at higher concentrations. The total yield of dimer and trimer is strongly reduced with increasing concentration of starting cis-enediyne. Higher cyclic oligomers are only observed in the coupling of the Me# derivative 32b; they do not form in the reactions of the (iPr)$i derivative 32a presumably for reasons of steric crowding of the bulky triisopropylsilyl groups in the larger macrocycles.
13.4 Perethynyiated Dehydroannulenes and Expanded Radialenes
457
47a R = (iPr)gSi 47b R=Me3Si
CuCI, TMEDA, II
R 1
48a R = (~Pr)~si 48b R = Me3Si
4 8 ~R = H
oligomers
0.045 0.08
1:4:2 1:3:4
I
R
Scheme 13-10 Preparation of perethynylated dehydroannulene precursors to planar all-carbon networks.
The X-ray crystal structure of trimeric 48b [Fig. 13-2(A)] shows a perfectly planar carbon frame with linear butadiyne fragments in the 18-membered ring [58]. The electronic absorption spectra characterize the bright yellow silyl-protected derivatives 48a and 48b as stable Hiickel-aromatic [18]annuleneswith a large HOMO-LUMO gap of 2.57 eV. The X-ray crystal structure of dimeric 47a shows a highly strained 12-membered ring [Fig. 13-2(B)] [58]. The butadiyne fragments are significantly bent and C = C - C angles as low as 164.5 are observed [59]. According to the electronic absorption spectra, 47a and 47b are antiaromatic [12]annulenes with a low HOMO-LUMO gap of 1.87 eV. The Me@-protected derivative 47 b undergoes slow decomposition in dilute solutions even at -2OOC and cannot be isolated as a stable solid. Dilute solutions of the (iPr),Si derivative 47a are much more stable and the crystalline compound decomposes only at 200OC. Crystals of 47a remain unchanged for months when exposed to light and air at ambient temperature. Interestingly, concentration of dilute solutions initially leads to rapid decomposition of 47 a which only stops with the onset of crystallization. The unusual stability of solid 47a results from the “insulating” effect of the bulky (iPr),Si groups in the crystal, which completely surround and isolate the delicate annulenic cycles [la]. This insulating effect, which has already been described in Section 13.3.3 for compound 30a, has been observed in several X-ray crystal structures of oligoacetylenes and represents a general stabilizing mode for these compounds in the solid state [43, 581. The two Me,Si derivatives 47 b and 48 b were fully deprotected with borax in methanol [58]. The resulting free perethynylated dehydroannulenes 47c and 48c were obtained as highly unstable molecules which can only be handled in dilute solutions for short periods of time without decomposition. Oxidative polymerization of 47 c should lead to the target network 45, and 48c should be the starting material for network 46. Both experimental correlations [l, 131 and AM1 [58] calculations predict that oxidative polymerization of 47c, with its relatively large external C = C -C angles of 122- 124”, should give exclusively the trimeric hexadehydro[l8]annulene rings in 45 and no undesired dimeric tetradehydro[l2]annulenes. O
458
I 3 Oligoacetylenes
(A)
Figure 13-2 X-ray crystal structures of perethynylated hexadehydro[l8]annulene 48b at 100 K (A) and tetradehydro[12]annulene 47 a at room temperature (B).
13.4 Perethynylated Dehydroannulenes and Expanded Radialenes
459
13.4.2 Perethynylated Expanded Radialenes Radialenes (49) are a homologous series of all-exo-methylenecycloalkanesof molecular formula C,H, (Fig. 13-3) [60]. Upon insertion of butadiynyl moieties into the cyclic framework between each pair of vicinal exo-methylene units, the carbon-rich expanded radialenes 50 (C,,H,) are obtained. Starting from suitable protected tetraethynylethenes, the perethynylated expanded radialenes 51-53 were prepared via 54, as shown in Scheme 13-11 [41]. They possess large carbon cores with silyl-protected peripheral valences and can be viewed as CN + 8 (iPr),Si (51), Cs0 + 10 (iPr),Si (52), and Cm + 12 (iPr),Si (53). The diameters of these large carbon-rich molecules are in the nanometer range with values of = 17 A (for 51), = 19 (for 52), and = 22 A (for 53).
A
49; CnHn 50; C3nHn
Figure 13-3 Scheme to increase the C/H ratio by inserting butadiynyl moieties into radialenes (491, thereby producing expanded radialenes (SO).
The expanded radialenes 51-53 are remarkably stable, with melting points above 260"C, and are readily soluble in common aprotic organic solvents. Laser desorption time-of-flight mass spectra, recorded in the negative-ion mode without matrix assistance, were particularly useful in their analytical characterization; they showed strong molecular ions for each of the three compounds as well as a complete absence of any fragmentation or impurity peaks. Despite the high degree of unsaturation, compounds 51-53 are only yellow-colored and show similar end absorptions in the UV spectra at = 500 nm. Careful comparisons show that n-electron delocalization in 51-53 is limited to the longest linearly conjugated fragment (shown in bold for 51 in Scheme 13-11) and that cross-conjugation, i.e., the conjugation between butadiynyl (C = C - C = C) fragments across the sp2-C-atoms in the central ring, is not efficient. The particularly high stability of 51-53 may well be due to inefficient cross-conjugation [61], which cannot compete with linear n-electron conjugation in these systems [l a]. Inefficient cross-conjugation could be a general stabilizing principle of unsaturated carbon-based matter: fullerene Cm is best described as a cross-conjugated molecule with [5]radialene substructures and graphite may also be viewed as cross-conjugated. The high solubility and stability of 51-53 raises hope that much larger carbon surfaces with similar electronic properties can be prepared and handled in the future, as long as the peripheral valences contain stabilizing and solubilizing groups such as the (iPr)$i groups.
460
I3 Oligoacetylenes
1) CU(OAC)~, pyridine/PhH, (l:l), 70%
28b
CNOAC)~,
H
-
pyridine/PhH (3:2)
2) K2C03,MeOH/THF ( l : l ) , R - H
67%
+R
R
R
28b (2 equiv)
+
25a
1) CU(OAC)~, pyridine/PhH, (l:l), 9% 2) K2C03,MeOHmHF (l:l), 60%
H
R
R
R = (t?r),Si
54 (3mM, 1 equiv), CU(OAC)~, pyndinePhH.
R
R
(3:2),15%
Scheme 13-11 Synthesis of the persilylethynylated expanded radialenes 51-53. Shown for 51 in bold is the longest linearly n-conjugated fragment in 51-53.
13.5 Molecular Wires: From Polytriacetylenes to Carbyne
461
13.5 Molecular Wires: From Polytriacetylenes to Carbyne Polyacetylenes (PAS) [62] and polydiacetylenes (PDAs) [63] (Fig. 13-4) are the only linearly conjugated organic polymers with an all-carbon backbone not composed of aromatic rings and have been widely explored for their interesting materials properties. When doped, transpolyacetylene exhibits electrical conductivity similar to that of copper [62, 641. Polydiacetylenes show excellent third-order nonlinear optical properties [65]. An extension of the progression of linearly conjugated nonaromatic all-carbon backbones leads via polytriacetylenes (PTAs) and higher poly-n-acetylenes to carbyne, a carbon rod composed exlusively of spC-atoms. This section describes the synthesis and properties of oligomers with the polytriacetylene and the carbyne backbone.
&
trans-Polyacetylene (PA)
H
R
I
.
trans-Polydiacetylene (PDA) R
W
n
Carbyne
Figure 13-4 Progression of polymeric backbones from polyacetylene to carbyne.
13.5.1 Linear Polyynes: Short Oligomers of Elusive Carbyne An intriguing aspect of carbon chemistry is the preparation of infinite one-dimensional rods constituted of alkyne units. Interest in such compounds dates back to Baeyer, who studied the oxidative Glaser [66]coupling of acetylenes in the search for infinite linear chains of pure carbon [67]. A century later, the preparation and characterization of infinite linear polyynes [ - (C = C), -1 named “carbynes” were actively pursued using a variety of methods [68]. But reports of the preparation and structure of these materials, which are calculated to be onedimensional conductors, remain controversial [69]. Materials closest to the infinite carbyne structure are the extended polyynes reported in 1972 by Walton and co-workers [70], which incorporate up to 16 conjugated C = C units and are stabilized by terminal triethylsilyl protecting groups. These compounds, which form stable solutions in petroleum ether, were constructed by repeated oxidative coupling of monoprotected polyynes under Hay conditions [22] followed by monodeprotection with aqueous methanolic NaOH [e. g., Et,Si - (C = C), - H + Et3Si- (C = C), - SiEt, +Et3Si- (C = C), - H --* Et,Si - (C = C), - SiEt, --t Et,Si - (C= C), - H -P Et3Si- (C= C)16- SiEt,].
462
13 Oligoacetylenes
Since long-chain polyynes were usually constructed in tedious stepwise procedures by multiple Hay [22], Eglinton-Glaser [I21 or Cadiot-Chodkiewicz [71] coupling reactions starting from mono- or diynes, new synthetic entries have been pursued more recently. A general preparative-scale method for the synthesis of symmetrically and unsymmetrically substituted which is transformed into linear polyynes starts from 3,4-dichloro-3-cyclobutene-l,2-dione, 3,4-dialkynyl-3-cyclobutene-l,2-diones (Table 13-1) followed by pyrolysis to the polyynes under extrusion of two CO groups (Table 13-3) [14]. This method also allows the preparation of polyynes with an odd number of C = C bonds, for which no convenient access existed previously. Since the majority of the 3,4-dialkynyl-3-cyclobutene-l,2-diones showed poor volatility and thermal instability, they could not be subjected to conventional flash vacuum pyrolysis (FVP) [72] and an alternative method, called solution-spray flash vacuum pyrolysis (SS-FVP) was applied. In this procedure, benzene solutions of the 3,4-dialkynyl-3-cyclobutene-1,2-diones 12a-i (Table 13-3) were introduced by capillary tubing as aerosols [73] into a hot pyrolysis tube made of quartz glass and filled with quartz rings. The triynes and pentaynes 55a-i were obtained in good to excellent yields. The reaction could even be applied to convert the bis(cyc1obutenedione) 56 into the hexayne 57 (31 Vo yield, Scheme 13-12). As in Fowler’s “solvent-assisted sublimation” method [74],the sublimation of the solid substrate into the gas phase during SS-FVP is assisted by the flash evaporation of benzene. Table 13-3 Preparation of linear polyynes by solution-spray flash vacuum pyrolysis (SS-FVP)
SS-FVP c
R-R‘
650 ‘C 12a-i
55a-i
Dione
R
R’
Product (Yield, To)
12 a 12b 12c 12d 12e 12f 12h 12i
Ph-C=C Ph - C=C - C C nPr-C=C Me,Si - C = C tBuMe,Si - C = C iPr,Si - C = C r?r,Si -C = C - C = C iPr& - C = C
Ph-CSC Ph -C E C - C s C nPr -C= C Me,Si - C = C tBuMe,Si- C E C Pr,Si - C = C iPr,Si - C = C - C = C Me,% - C = C
55a (97)
I
IBuMe2Si
56
I
SiMe2fBu
Scheme 13-12 Preparation of hexayne 56 by SS-FVP.
57 (31%)
55b (59) 55e (78) 55d (99) 55e (99) 55f (95) 55 h (42) 55i (71)
13.5 Molecular Wires: From Polytriacetylenes to Carbyne
463
A photochemical method for the generation of odd-membered linear polyynes occurs by [2+2]cycloreversion of [4.3.2]propellatrienes under extrusion of indane (Scheme 13-13) [75]. Although the cycloreversion step is high-yielding, the preparation of the propellane starting material is much more tedious than the preparation of the 3,4-dialkynyl-3-cyclobutene-1,2diones for SS-FVP. OMOM 1) MOMOCH~CSCCH~OMOM,
hv, CHzClz, 70%
0
1) NaBH,,
2) Me3SiOTf, Et3N, CCI4 3) Pd(OAc)p, CHBCN, 53%
CeCl3.7H20, MeOH
2) C6H3(N0z)zSC19E W , CICHzCHzCI, 84%
0 OMOM 1) aqueous HCI, THF, 68% 2) Mn02, CH2CIz, 78%
3) CC13PO(OEt)2, nBuLi, THF, 64% R
nBuLi, THF, then Me3SiCl (78%)
f & R
or lBuMe2SiCl (47%)
Br-C-C-Si(Pr)3, CuCI, EtNHz, NHzOH.HCI, MeOHTTHF. 95%
hv, quartz tube,
R
=
=
Z
R
low pressure Hg lamp
c
R= R= R= R=
SiMe3 SilBuMezJ TBAF, THF, H 78% c~C-si(Pr)~
55d R = SiMe3 (MYo) 55e R = SiBuMe2 (67%) 55h R = C=C-Si(/Pr)3 (74%)
Scheme 13-13 Preparation of polyynes by photochemical cycloreversion.
13.5.2 Stable Soluble Conjugated Carbon Rods with a Polytriacetylene Backbone By end-capping polymerization of 30a (Scheme 13-14), a series of highly stable conjugated nanometer-sized molecular rods with the polytriacetylene backbone were obtained [43]. They are amazingly stable, high-melting materials that remain unchanged for months on the laboratory bench. X-ray crystal structures of monomeric 58a and dimeric 58b show nearly perfectly planar conjugated carbon frames including the end-capping phenyl rings. In the electronic absorption spectra of highly colored 58a-e, the wavelength of the end absorption increases with the length of the oligomeric chain. Electrochemical studies of 58a-e showed facile one-electron reductions of the oligomers, with the number of reversible one-electron reduction steps corresponding to the number of tetraethynylethene moieties in each rod. As the oligomer length increases, the first reduction potential occurs at increasingly less negative potentials at -1.57 V vs. ferrocene for 58a and -1.07 V for 58e in THF). In contrast, none of the oligomers could be oxidized below +1.0 V vs. ferrocene, which helps to explain their stability in air.
464
13 Oligoacetylenes
lsi'pr3 1 ) CUCI, TMEDA, 0 2 , CHC13, 20 'C, 2 d
H
.9
2) PhCECH (2 equiv)
58a 1 58b 2 58c 3
28 33 16
19.4 26.8 34.3
Scheme 13-14 Preparation of the soluble carbon rods 58a-e which have a polytriacetylene backbone- The
length shown is the distance between the two extreme tips of the end-capping phenyl rings.
The remarkable stability of oligomers such as 58a-e (or larger analogs) could make them useful as molecular wires in molecular electronics [76].To make contact with these wires, any desired substituent can be attached to the ends of the rods by changing the end-capping reagent. The lateral silyl groups can be easily exchanged for other substituents to allow attachment of such molecular wires to substrates such as silicon wafers.
13.6 Conclusions Modern acetylene chemistry plays a critical role in the current world-wide efforts to synthesize new molecular and polymeric carbon allotropes as well as carbon-rich nanomaterials. It is rapidly becoming clear that the preparative challenges in this interdisciplinary research area at the interface between materials science and chemistry are formidable and rival those in the more established synthesis of natural products and biologically active compounds. Modern materials research strongly relies on advanced synthetic methodology, and it is hoped that the field will attract many synthetically oriented chemists into its ranks: their efforts will be rewarded by the development of materials with unique properties and unprecedented applications.
13.7 Experimental Procedures 13.7.1 3,4-Bis[triisopropylsilyl)ethynyl]-3-cyclobutene-l,2-dione (12f)
-
To a solution of (iPr),Si C E CH (10.94g, 60 mmol) in dry degassed THF (300 mL) under Ar at 0 "C was added dropwise a 1.6 M solution of n-BuLi in hexane (37.5mL, 60 mmol) over 15 min. After 10 min, CuI (12.38 g, 65 mmol) was added under a positive pressure of Ar, and
13.7 Experimental Procedures
465
the solution became dark green. After stirring for 15 min at 0 "C, 3,4-dichloro-3-cyclobutene12-dione (7.54 g, 50 mmol) was added at once under a positive pressure of Ar, and the brown solution was stirred at 20°C for 1 h. The solution was evaporated to dryness together with 50 g of SiO,. The solid was loaded on top of a flash silica gel column packed with hexane as eluent. Elution with hexane/ether (9: 1) gave 7.82 g (59%) of 12f as yellow crystals, mp 36.5-37.5 "C.
13.7.2 Oxidative Hay Coupling of 14 to the Cyclobutene-Fused Dehydroannulenes 15-17 A decanted solution of Hay catalyst (50 mL, 18 mmol), prepared from CuCl(l0 g, 0.1 mol) and TMEDA (4.18 g, 36 mmol) in acetone (100 mL) under Ar, was added to a stirred solution of 14 (3.86 g, 17.7 mmol) in acetone (1 L) under oxygen (1 -2 psi), and the yellow-green solution was stirred vigorously at 20°C for 2 h. The solution was filtered through Celite, and the filter cake was washed with CHCl, (5 x 200 mL). The combined filtrates were extracted with H 2 0 (2 x 2 L), then dried, and evaporated at 530°C at water aspirator pressure. The remaining crude solid was purified by flash chromatography with CHCl,/acetone (9 : 1 to 7 :3). The first fractions of pure trimer 15 were collected and evaporated. The fractions containing mostly tetramer 16 together with some trimer 15 and pentamer 17 were combined and evaporated to =15 mL. The poorly soluble tetramer 16 was filtered off, washed with CH,C12 (3 x 5 mL), and dried. The filtrates were evaporated and chromatographed again to give additional 15-17. The fractions containing pentamer 17 from the first chromatography were evaporated and chromatographed a second time with the same solvent mixture to give the pentamer 17, next to some tetramer 16. The total cyclization yield was 41 %. The fractions containing 15 were evaporated and redissolved in CCl,/CH,Cl, and gave, after slow evaporation, 15 (619 mg, 16%) as stable, pale yellow prisms, mp > 100°C (dec., sealed tube). The fractions containing 16 were evaporated, and the residual solid was washed with a small amount of CH2C12, affording 16 (702 mg, 18%) as a stable microcrystalline orange-red powder, mp > 180°C (dec., sealed tube). The fractions containing 17 were evaporated, and slow evaporation from CH,Cl, afforded 17 (254 mg, 6.5 9'0) as stable, bright yellow, fine needles, mp > 100°C (dec., sealed tube).
13.7.3 3-Dibromomethylene-l,5-bis(trimethylsilyl)-l,4-pentadiyne(23) To a solution of 22 (18.92 g, 85 mmol) and CBr, (36.48 g, 0.11 mmol) in benzene) (1 L) under Ar was added PPh, (57.7 g, 0.22 mol). After stirring at 20°C for 12 h, hexane (500 mL) was added to the suspension and the mixture was filtered through Celite. TLC analysis (hexane) showed the presence of a new compound at Rf = 0.44 besides starting material (Rf=0.05). Evaporation of the filtrate gave a pale yellow semi-solid which was triurated with hexane and filtered. After evaporation of the filtrate, flash chromatography of the residual oil on silica gel with hexane gave 23 (16.21 g, 50%) as a colorless oil.
466
13 Oligoacetylena
13.7.4 Q-l,2-Diethynyl-l,2-bis[triisopropylsilyl)ethynyl]ethene (30a) To a solution of 29 (1.0 g, 1.3 mmol) in dry THF (10 mL) at -78 "C was added slowly a solution of LDA in THF [lo mL, 7.8 mmol; prepared by the addition of 4.9 mL of n-BuLi (1.6 M in hexane) to a solution of diisopropylamine (0.8 g) in THF (10 mL) at OOC]. The solution turned bright green, then blue, and finally purple, signifying the complete reaction of the tetrabromide. After 10 min, saturated aqueous NH4CI ( 5 mL) was added, and the solution rapidly turned pale yellow. The THF solution was diluted with hexane and extracted with H,O ( 2 ~ ) The . hexane layer was filtered through a plug of silica gel and concentrated to 5 mL. After 12 h at -2O"C, crystals had formed which were filtered to give 30a (0.54 g, 95oJo); mp 76°C (dec.).
13.7.5 Eglinton-Glaser Coupling of 54 to the Expanded Radialenes 51 and 53 A flask was charged with distilled pyridine/benzene (3 : 1, 8 mL), and anhydrous CU(OAC)~ (35.2 mg, 0.194 mmol). Over 4 h, a solution of 54 (17.9 mg, 0.021 mmol) in benzene (2 mL) was added, After stirring at room temperature for 20 h, the mixture was poured into 30% aqueous CuS04 and was washed several times with 30% aqueous CuSO,, once with H20, and once with saturated aqueous NaCI. The dried, concentrated oil was passed through a flash Si02 plug with hexane. The bright yellow, stable solution was concentrated once again and purified by reverse-phase chromatography (CH3CN/CH,Cl,, 1 :2, then 1 :3) to give four fractions. Fraction 1 (Rf= 0.31, CH3CN/CH2CI2, 1 :2) upon evaporation gave 51 (2.6 mg, 15Vo) as a stable yellow solid, mp > 260°C. Fraction 2 (Rf= 0.19, CH3CN/CH2C12, 1 :2) upon evaporation gave 53 (3.4 mg, 20%) as a stable yellow solid, mp > 260°C. Fraction 3 (R,= 0.31, CH3CN/CH2C12, 1 :3) upon evaporation gave the corresponding octameric persilylethynylated expanded radialene C224H336Si16 (MW 3478.54) as a stable yellow solid (3.0 mg, 17%); laser desorption timeof-flight mass spectrum (LD-TOF) MS m/z 3475 (100%, M - ) . Fraction 4 (Rf=0.21, CH3CN/CH2C12, 1 :3) upon evaporation gave a trace mixture of octameric, decameric [C2soH420Sim,MW 4348.18; LD-TOF MS m/z 4344 (loo%, M - ) ] , and dodecameric [C336HS04Si24, MW 5217.82; LD-TOF MS m/z 5215 (33 '70,M - ) persilylethynylated expanded radialenes.
13.7.6 General Procedure for Solution-Spray Flash Vacuum Pyrolysis (SS-FVP) The apparatus is shown in Fig. 13-5. (1) Flask (a) contains the solution of the starting material. (2) A 1 mm inner diameter (6 mm outer diameter, 20 cm long) capillary glass tube (e) is drawn into a fine capillary in the standard manner and then cut at the fine portion to a length of 5-7 cm. It is then inserted through a conventional thermometer adaptor or a pierced rubber stopper (1) fitting the quartz tube (g). The other end of the capillary tube (e) is attached with connector (f) to the needle adaptor (d) and the teflon needle (b) passing through the punctured septum (c) placed on flask (a).
467
13.7 Experimental Procedures
a Flask with precursor in benzene b. Teflon needle c. Septum d Needle adaptor e Glass capillary tube f Rubber-tubing connector g Quartztube h Quartz ring fillings i High-temperature oven j. Trap k. Side-arm for sampling I Teflon tubing adaptor
N2-gas (1-2 PSI)
Figure 13-5 Schematic diagram of the solution-spray flash vacuum pyrolysis (SS-FVP) setup.
(3) The quartz tube (g), (20 mm outer diameter, 36 cm long), filled with quartz rings (4 mm height by 6 mm length) to two-thirds of its length, is heated to 650°C in the hightemperature oven (i). It is essential to use quartz fillings to increase the contact time of the substrate. Otherwise, a large quantity of starting material is usually recovered. (4) The trap is used in this setup to avoid the condensation of products on the walls of the quartz tube (g) in proximity to the oven. This problem occurs particularly when concentrated solutions (0.1 M) and/or high-molecular-weight compounds are used. The sidearm (k) on the trap is designed for easy sampling of the benzene solution collected in the trap.
u)
To perform the SS-FVP experiment, nitrogen gas (100 mm Hg reading at pressure gauge) is passed into the flask (a) with the teflon needle (b) staying above the liquid level. The pyrolysis apparatus is placed under vacuum by means of a mechanical pump, and the trap G) is filled with liquid nitrogen. The quartz tube (g) is then heated in the oven (i). When the temperature reaches 650 "C (measured with an external iron-constantan thermocouple), the teflon needle (b) is placed below the level of the solution in flask (a) and rapid aspiration of the solution through the capillary occurs. As soon as the liquid reaches the tip of the capillary, a sprayed aerosol forms inside the hot tube. The products of pyrolysis condense together with benzene in the trap (j).Solutions in benzene can be relatively concentrated (up to 0.1 M). However, solutions of unstable and/or oily precursors tend to clog the capillary tube easily if they are >0.02 M. The tip of the capillary tube (e) should not be allowed to reach inside the oven, otherwise charring from the extreme heat and subsequent clogging always occur. At the end of the addition, a few milliliters of fresh solvent are added with a syringe to the flask (a) to complete the transfer of the starting material to the pyrolysis tube. After cooling of the .oven (i), the products (along with frozen benzene) are allowed to thaw under nitrogen, collected, and worked up. Any black deposits inside the quartz tube are most conveniently removed by heating the tube under oxygen at normal pressure (balloon) between 25 and 700 "C.
468
I3 Oligoacetylenes
The polyynes were purified by flash chromatography on Si02 (55a-h and 57) or on Florisil (55i). A mixture of hexane/CH2C12 (1 : 1) was the solvent in the separation of 55b and 55c;pure hexane was used in all other runs. Solid polyynes were further purified by recrystallization.
Acknowledgement This work was supported by the Swiss and the U.S. National Science Foundations.
Abbreviations AM1 CSA dba DIBAGH DME DMF dPPm FT-MS FVP HOMO-LUMO LDA LD-TOF MS MCPBA MM2 MOM MP2 PA PCC PDA PDC PTA SCF SS-FVP TBAF Tf THF TMEDA TrisNHNH,
Austin model 1 canphorsulfonic acid dibenzylideneacetone diisobutylaluminum hydride dimethoxyethane dimethylformamide bis(dipheny1phosphino)methane Fourier transform mass spectrometry flash vacuum pyrolysis highest occupied MO-lowest unoccupied MO lithium diisopropylamide laser desorption time-of-flight mass spectrometry m-chloroperbenzoic acid molecular mechanics 2 methoxymethyl M~rller-Plessetsecond-order perturbation theory polyacetylene pyridinium chlorochromate polydiacetylene pyridinium dichromate polytriacetylene self-consistent field solution-spray N P tetrabutylammonium fluoride triflyl (trifluoromethanesulfonyl) tetrahydrofuran tetramethylenediamine
2,4,6-(Triisopropyl)benzenesulfonhydrazide
References
469
References [l] (a) F. Diederich, Nature (London) 1994, 369, 199-207; (b) F. Diederich, Y. Rubin, Angew. Chem.
1992, 104, 1123-1146; Angew. Chem., Znt. Ed. Engl. 1992, 31, 1101-1123. [2] U. H. F. Bunz, Angew. Chem. 1994, 106, 1127-1131; Angew. Chem., Int. Ed. Engl. 1994, 33, 1073-1076. [3] Z. Wu, S. Lee, J. S. Moore, J. Am. Chem. SOC.1992, 114, 8730-8732. [4] F, Diederich, Y. Rubin, C. B. Knobler, R. L. Whetten, K. E. Schriver, K. N. Houk, Y. Li, Science 1989, 245, 1088-1090. [5] R. Hoffmann, Tetrahedron 1966, 22, 521-538. [6] V. Parasuk, J. Almldf, M. W. Feyereisen, J. Am. Chem. SOC. 1991, 113, 1049-1050. [7] J. Hutter, H. P. Liithi, F. Diederich, J. Am. Chem. SOC. 1994, 116, 750-756. [8] (a) W. H. Okamura, F. Sondheimer, J. Am. Chem. SOC. 1967, 89, 5991-5992; (b) H. P. Figeys, M. Gelbcke, Tetrahedron Lett. 1970, 5139-5142. [9] R. M. McQuilkin, P. J. Garratt, F. Sondheimer, J. Am. Chem. SOC. 1970, 92, 6682-6683. [lo] F. Diederich, Y. Rubin, 0. L. Chapman, N. S. Goroff, Helv. Chim. Acra 1994, 77, 1441-1457. [ll] H. Hart, S. Shamouilian, Y. Takehira, J. Org. Chem. 1981, 46, 4427-4432. [12] 0. M. Bebr, G. Eglinton, A. R. Galbraith, R. A. Raphael, J. Chem. SOC. 1960, 3614-3625. [13] Y. Li, Y. Rubin, F. Diederich, K. N. Houk, J. Am. Chem. SOC. 1990, 112, 1618-1623. [14] Y. Rubin, S. S. Lin, C. B. Knobler, J. Anthony, A. M. Boldi, F. Diederich, J. Am. Chem. SOC.1991, 113, 6943-6949. [15] (a) A. H. Schmidt, Synthesis 1980, 961-994; (b) A. H. Schmidt, W. Ried, ibid. 1978, 1-22; (c) H. Knorr, W. Ried, ibid. 1978, 649-666; (d) A. H. Schmidt, W. Ried, ibid. 1978, 869-880. [16] (a) J. E. Ezcurra, C. Pham, H. W. Moore, J. Org. Chem. 1992,57, 4787-4789; (b) L. M. Gayo, M. P. Winters, H. W. Moore, ibid. 1992, 57, 6897-6899, and references cited therein. [17] (a) L. S. Liebeskind, J. Wang, Tetrahedron Lett. 1990, 31, 4293-4296; (b) L. S. Liebeskind, K. Z. Granberg, J. Zhang, J. Org. Chem. 1992, 57, 4345-4352; (c) J. P. Edwards, D. J. Krysan, L. S. Liebeskind, ibid. 1993, 58, 3942-3952, and references cited therein. [I81 J. K. Stille, Angew. Chem. 1986, 98, 504-519; Angew. Chem., Int. Ed. Engl. 1986, 25, 508-524. [19] M. W. Logue, G. L. Moore, J. Org. Chem. 1975, 40, 131-132. [20] (a) Y. Rubin, F. Diederich, J. Am. Chem. SOC.1989, 111, 6870-6871; Y. Rubin, C. B. Knobler, F. Diederich, ibid. 1990, 112, 1607- 1617. [21] Y. Rubin, M. Kahr, C. B. Knobler, F. Diederich, C. L. Wilkins, . I Am. Chem. SOC. 1991, 113, 495 -500. [22] A. S. Hay, J. 0%.Chem. 1962, 27, 3320-3321. [23] S. W. McElvany, M. M. Ross, N. S. Goroff, F. Diederich, Science 1993, 259, 1594-1596. [24] (a) T. Grosser, A. Hirsch, Angew. Chem. 1993, 105, 1390-1392; Angew. Chem., Inl. Ed. Engl. 1993, 32, 1340-1342; (b) M. T. Bowers, P. R. Kemper, G. von Helden, P. A. M. van Koppen, Science 1993, 260, 1446-1451; (c) J. Hunter, J. Fye, M. F. Jarrold, Science 1993, 260, 784-786; (d) H. Schwarz, Angew. Chem. 1993, 105, 1475-1478; Angew. Chem., Znt. Ed. Engl. 1993, 32, 1412-1415. [25] W. Krtitschmer, L. D. Lamb, K. Fostiropoulos, D. R. Huffman, Nature (London) 1990,347,354-358. [26] R. S. Dickson, P. J. Fraser, Adv. Organomet. Chem. 1974, 12, 323-311. [27] (a) P. Magnus, P. A. Carter, J. Am. Chem. SOC.1988, 110, 1626-1628; (b) S. L. Schreiber, T. Sammakia, W. E. Crowe, ibid. 1986, 106, 3128-3130. [28] (a) H. Petersen, H. Meier, Nouv. J. Chim. 1980, 4, 687-689; (b) Y. Shvo, E. Hazum, J. Chem. Soc, Chem. Commun. 1974, 336-331. [29] G. Cetini, 0. Gambino, R. Rossetti, E. Sappa, J. Organomet. Chem. 1967, 8, 149-154. [30] N. T. Allison, J. R. Fritch, K. P. C. Vollhardt, E. C. Walborsky, J. Am. Chem. SOC.1983, 105, 1384-1386. [31] Y. Rubin, C. B. Knobler, F, Diederich, J. Am. Chem. SOC.1990, 112, 4966-4968.
470
13 Oligoacetylenes
[32] P. Sautet, 0. Eisenstein, E. Canadell, New J Chem. 1987, 21, 797-806. [33] B. Ma, Y. Xie, H. F. Schaefer 111, Chem. Phys. Lett. 1992, 191, 521-526. [34] Y. Hori, N. Noda, S. Kobayashi, H. Taniguchi, Tetrahedron Lett. 1969, 3563-3566. [35] (a) H. Hauptmann, Angew. Chem. 1975, 87, 490-491; Angew. Chem., Int. Ed. Engl. 1975, 14, 498-499; (b) H. Hauptmann, Tetrahedron Lett. 1975, 1931- 1934; (c) H. Hauptmann, Tetrahedron 1976, 32, 1293-1297. (361 H. Hopf, M. Kreutzer, P. G. Jones, Chem. Ber. 1991, 124, 1471-1475. [37] Y. Rubin, C. B. Knobler, F. Diederich, Angew. Chem. 1991, 103, 708-710; Angew. Chem., Int. Ed. Engl. 1991, 30, 698-700. [38] E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 3769-3772. [39] K. P. C. Vollhardt, L. S. Winn, Tetrahedron Lett. 1985, 26, 709-712. [40] A. M. Boldi, J. Anthony, C. B. Knobler, F. Diederich, Angew. Chem. 1992, 104, 1270-1273; Angew. Chem., Int. Ed. Engl. 1992, 31, 1240-1242. [41] A. M. Boldi, F. Diederich, Angew. Chem. 1994, 106, 482-485; Angew. Chem., Znt. Ed. Engl. 1994, 33, 469-471. (421 A. M. Boldi, Ph. D. Thesis, University of California at Los Angeles (UCLA), 1994. [43] J. A. Anthony, C. Boudon, F. Diederich, J.-P. Gisselbrecht, V. Gramlich, M. Gross, M. Hobi, P. Seiler, Angew. Chem. 1994, 106, 794-798; Angew. Chem., Znt. Ed. Engl. 1994, 33, 763-766. [44] J. A. Walker, S. P. Bitler, F. Wudl, J. Org.Chem. 1984, 49, 4733-4734. [45] J. Anthony, C. B. Knobler, F. Diederich, Angew. Chem. 1993,105,437-440; Angew. Chem., Znt. Ed. Engl. 1993, 32, 406-409. [46] K. M. Merz, R. Hoffmann, A. T. Balaban, J. Am. Chem. SOC. 1987, 109, 6742-6751. [47] B. Ma, H. M. Sulzbach, Y. Xie, H. F. Schaefer 111, J. Am. Chem. SOC. 1994, 116, 3529-3538. [48] R. Diercks, J. C. Armstrong, R. Boese, K. P. C. Vollhardt, Angew. Chem. 1986,98,270-271; Angew. Chem., Int. Ed. Engl. 1986, 25, 268-269. [49] R. H. Baughman, H. Eckardt, M. Kerstesz, J. Chem. Phys. 1987, 87, 6687-6699. [50] R. Boese, J. R. Green, J. Mittendorf, D. L. Mohler, K. P. C. Vollhardt, Angew. Chem. 1992, 104, 1643-1645; Angew. Chem., Znt. Ed. Engl. 1992, 31, 1643-1645. [51] A. H. Alberts, H. Wynberg, 1 Chem. SOC, Chem. Commun. 1988, 748-149. [52] U. Bunz, K. P. C. Vollhardt, J. S. Ho, Angew. Chem. 1992,104, 1645-1648; Angew. Chem., Znt. Ed. Engl. 1992, 31, 1648-1651. [53] K. S. Feldman, C. M. Kraebel, M. Parvez, 1 Am. Chem. SOC. 1993, 115, 3846-3847. [54] J.-D. van Loon, P. Seiler, F. Diederich, Angew. Chem. 1993,105, 1235- 1238; Angew. Chem., Znt. Ed. Engl. 1993, 32, 1187-1189. I551 (a) U. H. F. Bunz, V. Enkelmann, Angew. Chem. 1993,105,1712-1714; Angew. Chem., Znt. Ed. Engl. 1993, 32, 1653-1655; (b) U. H. F. Bunz, V. Enkelmann, J. RBder, Organornetullics 1993, 12, 4745-4747. [56] (a) F. Sondheimer, Acc. Chem. Res. 1972, 5, 81-91 ; (b) H. A. Staab, J. Ipaktschi, A. Nissen, Chem. Ber. 1971, 104, 1182-1186; (c) M. Nakagawa, Angew. Chem. 1979,91, 215-226; Angew. Chem., Int. Ed. Engl. 1979, 18, 202-214. [57] J. Anthony, C. B. Knobler, F. Diederich, Angew. Chem. 1993,105, 437-440; Angew. Chem., Int. Ed. Engl. 1993, 32, 406-409. [58] J. Anthony, Ph. D. Thesis, University of California at Los Angeles, 1993. [59] Q. Zhou, P. J. Carroll, T. M. Swager, J. Oe. Chem. 1994, 59, 1294-1301. [60] H. Hopf, G. Maas, Angew. Chem. 1992, 104, 953-977; Angew. Chem, Int. Ed. Engl. 1992, 31, 931-954. [61] (a) N. F. Phelan, M. Orchin, J. Chem. Educ. 1968, 45, 633-637; (b) J. Klein, Tetrahedron 1983,39, 2733-2759; (c) M. Traetteberg, P. Bakken, A. Almenningen, W. Liittke, J. Janssen, J Mol. Struct. 1982, 81, 87-103. [62] A. J. Heeger, A. G. McDiarmid in The Physics and Chemistry of Low-Dimensional Solids (Ed. : L. Alcacer), Reidel, Dordrecht, 1980, pp. 353-391.
References
411
[63] G. Wegner, Pure Appl. Chem. 1977,49, 443-454. [64] T. A. Skotheim (Ed.), Handbook of Conducting Polymers, Vols., 12, Dekker, New York, 1986. [65] H. S. Nalwa, Adv. Muter. 1993, 5 , 341-358. 1661 C. Glaser, Ann. Chem. 1870, 154, 137-171. 1671 A. Baeyer, Ber. dt. Chem. Ges. 1985, 18, 674-681. 1681 (a) V. V. Korshak, Y. P. Kudryavtsev, Y. V. Korshak, S. E. Evsyvkov, V. V. Khvostov, V. G. Babaev, M. B. Guseva, Makromol. Chem., Rapid Commun. 1988,9, 135-140; (b) Y. P. Kudryavtsev, S. E. Evsyvkov, V. G. Babaev, M. B. Guseva, V. V. Khvostov, L. M. Krechko, Carbon 1992,30, 213-221; (c) A. G. Whittaker, Nature (London) 1978,276, 695-696; (d) I. V. Stankevich, M. V. Nikerov, D. A. Bochvar, R w . Chem. Rev. (Engl. i%msl.) 1984,53, 640-655. 1691 P. P. K. Smith, P. R. Buseck, Science 1982,216, 984-986. 1701 R. Eastmond, T. R. Johnson, D. R. M. Walton, Tetrahedron 1972, 28, 4601-4616. 1711 (a) W. Chodkievicz, Ann. Chim. (Paris) 1957,2, 819-869; (b) F. Bohlmann, P. Herbst, H. Gleinig, Chem. Ber. 1961,94, 948-957; (c) G. Eglinton, W. McCrae, Adv. Org. Chem. 1963,4, 225-328. [72] (a) U. E. Wiersum, Red. Trav. Chim. Pays-Bas 1982, 101, 317-332, 365-381; (b) R. F. C. Brown, Pyrohtic Methods in Organic Chemistry, Organic Chemistry Monographs, Vol. 41, Academic Press, New York, 1980. 1731 (a) M. G. Clancy, D. G. Hawkins, M. M. Hesabi, 0. Meth-Cohn, S. Rhouati, J. Chem. Res. Synop. 1982, 78; (b) M. G. Clancy, M. M. Hesabi, 0. Meth-Cohn, J. Chem. Soc., Perkin Trans. I 1984, 429-434. [74] J. Magrath, F. W. Fowler, Tetrahedron Lett. 1988, 29, 2171 -2174. [75] Y. Tobe, T. Fujii, K. Naemura, J. Org. Chem. 1994, 59, 1236-1237. [76] (a) E L. Carter, Molecular Electronics Devices ZI, Dekker, New York, 1987;(b) Nanostructures Based on Molecular Materials (Eds.: W. Gepel, Ch. Ziegler), Weinheim, VCH, 1992;(c) J. M. Tour, R. Wu, J. S. Schumm, J. Am. Chem. Soc. 1991,113, 7064-7066.
This Page Intentionally Left Blank
Index
ab-initio calculations 2 ab-initio STO-3G calculations 14 acetylene anion radical 6 acetylene cycloaddition chemistry 139 ff (p-acetylene)dicobalt hexacarbonyl complex 448 acetylene polymerization 358 f acetylene-propynecopolymers 359 acetylene scaffolding 33, 443 acetylenes 1 ff - bridged 4 - Pauson-Khand reaction 154 - terminal, methyl ketones 345 f acetylenic substitution, nucleophilic 77 acetylide-aldehyde addition, intramolecular 245 acetylide-aldehyde ring closure 227 acetylide anion-aldehydering closure 226 ql-acetylide ligands 387 0-acetylide transition metal complexes 88 q*(n)-acetylides 388 q'-acetylides, photoelectron spectroscopy 388 actinoxanthin 221 f action mechanism, enedyne antibiotics 207 activation - dynernicinA 271 - enediyne 240 - NCS chromophore 210 activation energy, cyclization 9 activation enthalpy 225 acyclic diene metathesis (ADMET) 368 f acyclic diethynyl ketones 330 acyclic dimerltrimer, of dimethyl-1,4-pentadiene 332 acyclic homoconjugatedpolyacetylenes 325 acyloin shift, diasteroselective 242 acylphosphanes 174 adaptive materials 353 addition - electrophilic 108 - M-C 110 f - M-H, hydrogenation 109 f - Michael 154,447 - nucleophilic 107 - tandem Dotz/nucleophilic aromatic 146 addition reactions, cyclic alkynes 308 f
addition-elimination process 67, 72 addition-elimination-rearrangementpathway 77 ADMET, acyclic diene metathesis 368 f aglycone 214 - precursors coupling 259 Al-acetylene complex 12 aldol-based construction, calicheamicin 245 aligning electron spin, parallel 392 f alkoxyacetylenes 81,118 alkoxycarbene complexes 107 alkoxycarbonylation 80 sec-alkylacetylenes 121 alkylation, trimethylsilyl acetylene 324 alkylative cyclizations 125 alkylcyclohexylaryloxy-substitutedpolyacetylene 362 alkylidene 359 alkylidene complexes, acetylene metathesis polymerization 361 alkylidene-1,6-diphosphorane 38 alkylidyne complexes 129 alkynation, palladium-catalyzed 445 alkyne/allene isomerization 118 alkyne/Fischer carbene cycloaddition 150 alkyne-metal carbonyl reactions 99 alkyne polymerization 360 alkyne-vinylidene isomerization 116 alkyne-vinylidene rearrangement 117,128 alkynes - cyclic 285 ff - - ring-closure reactions 286 f - - structure 296 f - - synthesis 286f - novel complexes 104 ff - pericyclic reactions 9 - scission/metathesis/polymerization 114 f alkynoic esters 80 alkynyl(ary1)iodonium salts 68 alkynyl benzoates 82 alkynyl carboxylate 81 alkynyl carboxylate esters 82 alkynyl compounds, unpaired electrons 385 alkynyl dialkyl phosphate esters 82 alkynyl dialkylphosphates 82 17-alkynylestradiols 128
474
Index
alkynyl lactone 143 alkynyl(pheny1)iodoniumsalts 67 alkynyl(pheny1)iodoniumtriflates 68 alkynyl phosphates 81 alkynyl phosphorodithioates 83 alkynyl sulfonate 81 alkynyl thiocarboxylates 83 alkynyl thiocyanates 83 alkynyl thiotosylates 83 alkynylation, Pd(0)-catalyzed 449 alkynyldiazonium salts 67 alkynyliodonium carboxylates 82 alkynyliodonium salts 67 ff, 73 f, 83,93 - formation mechanism 71 - molecular structures 75 alkynyliodonium sulfonates 68 f, 81 alkynyliodonium tetrafluoroborates 69 f alkynyliodonium tosylates 80 alkynyliodonium triflates 89 alkynylphosphonium salts 86 o-alkynylplatinum(I1) complex 88 alkynylstannanes 69 all-carbon molecules 15 all-carbon networks 449,453 - two-dimensional 443 all-carbon sheets 456 f all-exo-methylenecycloalkynes 459 allene-ene-yne type intermediate 228 allenes 118 - naturally occurring 59 o-allenyl derivatives 116 ally1 metals 126 allylic trisulfide trigger 243 AM1 (Austin model 1) 445 - calculations 457 - potential energy surface 11 j3-aminoethynyl-iodoniumtriflates 84 f angle bending 449 angle strain 447 angular dependence, resonance fields 396 angularly fused triquinanes 162 f annealed cyclopropenones, photolysis 293 annulation, Dotz reaction 142 annulene synthesis, Sontheimer 308 anomalous phase transitions 402 antarafacial migration 102 f, 124 anthracene 42 anthracycline antibiotics 217 anthracycline derivatives 146 anthraquinol 219 anthraquinone 217
antiaromatic [12]annulenes 457 antibiotic deoxyfrenolicin 145 antibiotics - broad-spectrum 216 - chromoprotein enedyne 221 f, 273 - DNA minor groove 249 - enediyne anticancer 125 - enediyne 203ff - naturally occurring enediyne 240 - j3-lactams 126 anticancer agents 273 antiferromagnetic coupling 392, 401 antiferromagnetics 392 antineoplastic activity 223 antitumor activity, in-vivo 224 antitumor agents 216 antitumor antibiotics 307 apical sites, alkynyliodonium salts 75 apoprotein 208, 221 Arbuzov-type process 87 aromatic systems, synthesis 206 f aromaticity 24, 286 aromatics, carbon nucleophiles coupling 122 arylaminocarbene 144 l-aryl-1,3-butadiynes 387 l-aryl-3,3-dialkyltriazene 419 l-aryl-3,3-dialkyltriazenegroup 433 arylethynyliodonium tosylates 91 aryl ethynyl ketones 404 o-aryl-metal complexes, thermolysis 104 aryl-tin bonds 144 aryloligosaccharide 214 arylsulfinate salts 84 asymmetric allylboration reaction 245 asymmetric crotylboration 126 asymmetric Pauson-Khand reactions, alkoxyacetylene 118 auromycin 224 automerization 22 5-aza-l-chroma-1,3,5-hexatrieneintermediate 154 2-azabutadiene, Diels-Alder reactions 8 azaphospha-Dewar-benzene 179 azaphosphinine 179 1,4-azaphosphinine 179 (Z)$-azidovinyl, iodonium salts 80 p-azidovinyliodonium tosylate 81 -azidovinylsilane 81 azophosphabenzvalenes 179 azophosphaprismanes 179
Index n back-donation 101 bacterial phosphotriesterase 82 banana bonds 1 barrelene 40 barrierless rotation 417 Basidiomycete fungi 59 basketene derivatives 44 bending, triple bonds 285 trans-bending 11 bending force constant 417 bent allenic diradical structure 390 bent bond model 1 f bent triple bonds 286 cis-bent triplet acetylene 5 trans-bent triplet acetylene 5 benzannulation reaction 144 benzene-1,4-diyl radicals 9 1,3,5-benzenetriyl 403 1,4-benzenoiddiradical 203,206,208,219 benzofuran angelicin 145 benzofuranophane 42 benzofurans 82,144 benzoylacetylenes,cyclotrimerizationreaction 403 f benzyne 105 benzyne complexes 104 benzyne ligand 104 Bergman cyclization 106,240,307 - enediynes 9 Bergman cycIization/pinacol rearrangement cascade, triggering 263 Bergman cycloaromatization 207,238 f Bergman reactions 10,215 f bicyclic cycloheptadienes 152 bicyclic cyclopropane product, chemoselectivity 151 bicyclic enediyne 241 f bicyclic enediyne core 216 bicyclic enediyne core triggering, calicheamicin 238 bicyclic enone 305 bicycle[ 3.3.0]oct-l-en-3-ones 155,161 bicyclo[3.3.0.]octenones 127 bicycle[4.3.0Jnon-1(9)-en-8-ones 161 3,3-bicyclo[5.3.0]decanes, sigmatropic rearrangement 152 Bieri method, electrical discharge 49 bimetallic v-alkyne derivatives 101 bimolecular coupling 431 bimolecular coupling reaction, radical centers 401
475
binuclear metal-alkyne complexes 101 bioreduction, triggering, natural products 264 bioreductive processes 240 biphosphonium diynes 86 bis(acy1oxy)iodobenzene 82 bisalkynyl benzoates 82 bisalkynyl tosylates 81 bis(a1kynyliodonium)tosylates 81 bis-cyclopenteneformation, Michael addition 79 cis-bisdeprotectedtetraethynylethenes 452 f 1,l-bis-(dialky1amino)ethenes 54 bis(dipheny1phosphino)methane 449 bis-iodonium diyne bistriflates 73 bis-iodonium diynes, Michael addition 79 bis-iodonium ethyne 72, 91 bis-iodonium(p-phenylene) bistriflates 71 bis[phenyl(iodonium)] diyne triflates 88 bis[phenyl(iodonium)]ethy ne 82 bis(pheny1thio)alkynes 86 bis-propargylic sulfones 233 - photolytic ring construction 236 bis-tin-alkynes 73 1,6-bis(triisopropylsilyl)-1,3,5-hexatriyne 449 trans-bis(triisopropylsily1)-protected tetraethynylethenes 451 f trans-bis-protected tetraethynylethenes 451 f blastmycinone 126 block copolymer 357 Boltzmann distribution 409 bond angles, cyclic monoalkynes 296 bond dissociation energy 5 bond formation, transannular, C-C 305 bond length variation 15 bond lengths - cyclic diacetylenes 299 - cyclic monoalkynes 296 o -n bond model 2 bond-shift isomerization 390 n-bonded terminal alkynes 117 bonding, metal-alkyne complexes 99 f boron enolates - chiral, Schreiber-Evans methodology 124 - Evans-type homochiral 124 boron-mediated aldol condensation 232 bound dissociation energy 14 bovolide 153 bridged acetylene 4 bridged cage compounds 313 bridging bimetallic complexes, conjugated 89 Brillouin function 394,401 f, 409
476
Index
broad-spectrum antibiotics 216 1-bromoacetylene 50 bromoacetylenes 52 1-bromo-2-iodoacetylene 50 3-bromo-2-propyn-1-01 57 bromoalkynes, terminal alkynes 347 buckminsterfullerene 448 bulky (iPr),Si groups, insulating effects 457 1,3-butadiene crystal, antiferromagnetic 393 butadienes, Diels- Alder reactions 8 1,3-butadiynes 13 f - solid-statepolymerization 395 f butadiynes, monohalogenated 50 butadiynyltributylatannanes 72 butenolides 153 (E/Z)-l-buten-3-yne-1,4-dicarbonitrile 46 t-butylethynyliodonium tosylate 87 t-butylethynyl(pheny1)iodonium triflates 70 2-tert-butyl-k3-phosphinine 177 tert-butylphosphaacetylene 174 tert-butylphosphaacetylenecyclotetramers 192 I3Cchemical shift, sp center 297 C NMR spectra, alkynyliodonium salts 75 C2 18 C, 18 Cq 19 cs 19 Ca 20 C, 19 C8 20 Cg 19 Clo 20,23 c 1 1 22 c11-c17 22 C18 22 Clg molecules 106 cyclo-clg 443 ff C18/C24,cobalt complexes 448 c, 15 cyclo-c, 443 C-1027 221f C-1027 chromophore 204,223 f C(4') hydrogen atom abstraction, deoxyribose 223 C- C bond dissociation 4 C-C triple bond 33 C- H-activated cyclopentenones 150 Cadiot-Chodkieviczreaction 288 Cadiot-Chodkieviczcoupling 57, 462 13
cage compounds 173
- bridged 313 calicheamicin 207,212 - theoretical/synthetic studies 238 f calicheamiciny: 203, 204 calicheamicin aglycone 241 f - Magnus approach 244 - total sythesis 244, 246 f calicheamicin-antibody immunoconjugates 273 calicheamicin diradical 216 calicheamicin/dynemicinhybrid, synthesis 259 calicheamicin/esperamicin carbohydrate fragments, synthesis 249 f calicheamicin/esperamicincascade 245 calicheamicin/neocarzinostatin field 287 calicheamicin oligosaccharide 249 f calicheamicin-typeanti-tumor antibiotics 294 calicheamicinone 214,245 - cycloaromatization 247 Cambridge structural database 417 cancer, enedyne antibiotics 203 carbacycline analog 162 carbapenem 124 carbene-alkyne complexes 114 carbene insertion, carbon nucleophiles 79 carbene-tethered alkynes 146 carbocation stabilization, B-Si substitution 72 carbocyclic five-membered ring annulation 149 carbohydrate fragments 245 - calicheamicin/esperamicin 249 f carbon allotropes 13, 443 carbon chains, even-numbered 18 carbon clusters - ground-stategeometries 17 - isomers 20 carbon cores 456 f carbon network, superdiamonoid 455 carbon nucleophiles 78 f carbon oxides 446 f carbon-rich nanoarchitecture 443 carbon-rich nanomaterials 449 carbon rods 47 - stable conjugated 463 carbonylation, oxidative 107 carbynes 13,461 cascade macromolecules 418 f CASSCF 9 CASSCF calculation, full valence 18
Index
CASSCF correlation energy 2 catalysis, intramolecular Pauson-Khand reaction 166 catalyst, general-base 216 catalytic Pauson-Khand reaction 155 cationic alkyne complexes 128 cationic cyclizations 85 cell death, programmed 270 cell growth media 353 cell lines 269 cell-type selectivity 270 chain-to-ring-to-spheremechanism 448 charge mobility, interchain 354 f charge-transfer interactions 45 charge transport mechanism 376 CHARMm force field 416 chelate effect 305 chemoselectivity - bicyclic cyclopropane product 151 - Dotz reaction 142 chiral cluster cations 120 chiral COT monomer 373 chiral separation membranes 353 chirality transfer 120 chloroacetylenes 48,52 1-chloroacetylenes 49,55 1-chloro-2-fluoroacetylene 50 1-chloro-1H-phosphirenes 192 4-chloro-3-(N-tert-butyl-N-oxyamino)phenyl, free-radical substituent 393 chromacyclobutene 141 chromacyclohexadiene 141,147 chromatin, proteolytic mechanism 223 chromium alkylmethoxycarbene 150 chromium arylalkoxycarbene 140 chromium arylaminocarbene 147 chromium-complexed4-alkoxynaphthol 141 chromium-complexed4-methoxy-1-naphthol 139 chromium-complexed naphthol 140 chromium furanalkoxycarbene 144 chromium vinylalkoxycarbene 150 f chromophore, cytotoxicity 209 chromophore/apoproteininteractions, C-1027 224 chromoprotein enediyne antibiotics 221 f, 273 CI 9 CI (QCISD/6-31G*) 390 circular dichroism (CD) 373 cis-fused tricycling ring systems 122 cis-semihydrogenation 109
477
cis-trans isomerization 453 ClSD calculations 21 CISD(+Q)/DZPlevel 6 Claisen rearrangement, Johnson orthoester variant 455 m-chloroperbenzoic acid (MCPBA) 54 closed-shell cumulene 15 cluster substitution 128 cluster substitution/expansion 115 clusters 18 - even-membered 20 - platinum, nanoscale 128 CO, Pauson-Khand reaction 154 coalescence, large cyclocarbon ions 448 cobalt-complexedpropargyl alcohols, dehydration 120 cobalt complexes - C&224 448 - cyclo-cls 449 - dynemicin A model systems 264 - N-propargylation 122 cobalt-mediated propargylation 124 cobalt methoxycarbenes 153 cobalt-propargyl radicals 127 cobalt-stabilized propargyl cation chemistry 163 f cobaltacyclobutenes 115 cobaltacyclopentadienes 111 cognates, heterocyclic, pericyclynes 340 f collection sites 419 collision-induced dissociation reactions 47 1,3-complexed dynes 117 complexed terminal alkynes, base 117 computational aspects, acetylene chemistry 1 ff concerted mechanism 7 concerted transition structure 8 conductivity - electrical 461 - models 354 conductors - one-dimensional 461 - organic, polyacetylenes 114 conjugated bridging bimetallic complexes 89 conjugated diradical molecules, spin palignement 399 p-conjugated diradicals, spin ground states 398 f conjugation effects, polyacetylenes 13 conjugation length, effective 354,356,361, 371 f convergent route 420
418
Index
convergent synthesis, phenylacetylene dendrimers 421 f convergent synthetic approaches 420 f coordinated alkynes, bent geometry 104 p-q2,q3-coordination 102 copolymerization 359, 372 copolymers 357 copper acetylides 288 copper-mediated coupling reaction 325 cores, carbon 456 f coriolin 162 correlation, electron 11 correlation effects, electron 2 coupled cluster methology 4,18, 21 coupled-electron-pair approximation (CEPA) 13 coupling - Eglinton-Glaser 445, 466 - ferromagnetic 361 - oxidative 445, 449, 461 coupling constants, hyperfine 386, 388 coupling reactions 110 Cp,Zr-phosphaalkyne dimer complexes 186 crinipellin B 164 cross-conjugated n-electron system 449 cross-conjugation 459 cross-coupling, 1,3-diyne 347 cross-coupling reactions 416 cross-trimerization 404 f crossed (2+2)-cycloaddition 179 crossed aldol reactions 125 crotylboration, asymmetric 126 crystal structure, tetraphosphacubane 185 crystallinity 359 cubane 173 (2)cumulene-ene-yne, cycloaromatization 224 cumulenes, perethynylated 455 cumulenes intermediate 209 cumulenic structure 19 Curie law 396, 409 Curie-Weiss law 392, 401, 410 cyanoacetylenes 46 1-cyanoacetylenes 33 ff 2- cyano-l,3-butadiene 46 cyanocarbons 47 I-cyanocyclobutene 46 cyanodiacetylene 35 cyanoethynyl radical 46 I-cyanohexatriyne 35 I-cyano-4-iodobutadiynes 53
cyano(pheny1)iodoniumtriflates 68 f cyclic alkynes 285 ff - elimination reactions 292 f - ring-closure reactions 286 f - structure 296 f cyclic allene/cyclic acetylene mixtures 303 cyclic C, 15 cyclic carbon clusters, isomers 21 cyclic diacetylenes 297 cyclic dialkynes 295 cyclic diallenes 304 cyclic dienediyne systems 230 cyclic dienediynes 286 cyclic diynones 295 cyclic electron delocalization, N-phospha[N]pericyclynes 345 cyclic enediynes 286,294 cyclic homoconjugation 321,324, 334 cyclic sulfides 289 cyclic thiadiynes 289 cyclic voltammograms 374 cyclization - cationic 85 - intramolecular 430 f - oxidative 446 cycloaddition 39 f - crossed, (2+2) 179 - Fischer carbenes 139 - Pauson-Khand reaction 154 cycloaddition reactions 67,90 f, 309 f - organometallic 139 ff (2+1)-cycloaddition reactions 309 (3+2)-cycloaddition reactions 309 (2+1)-cycloaddition 189 (2+2)-cycloaddition 41,141,189 - formal intramolecular 230 - metallacycles 359 (2+2+2)-cycloaddition 182,190, 321 - acetylene 7 (3+3)-cycloaddition 184 (4+1)-cycloaddition 189 (4+2)-cycloaddition 191 (8+2)-cycloaddition 182 (2+4)-cycloaddition 40 f, 54 (4+2+1-2)-cycloaddition 150 1,s-cycloalkadiynes 105 cycloalkyne/cycloallene equilibrum 304 cycloalkynes complexes 104 cycloaromatization 204, 209, 224 - 1,4-benzenoid diradical 219 - calicheamicinone 247
Index - dynemicin 262
- ketone 233 - polyenyne systems 205 - trigger 220 cyclobutadienecomplexes 110, 312
- tricyclic 311 cyclobutadienedicarbonitrile 41 cyclobutadiene-AK31,complex 313 cyclobutene-fuseddehydroannulenes 446 cyclobutenedione 447 3-cyclo-butene-1,2-dioneroute, cyclocarbons 446 f cyclobutenones 147,149 f cyclobutyne 104 cyclocarbons 443 f, 446 cyclo(n)carbons 15, 24, 443 ff cyclocolorenone 124 cyclocoupling, three-component 113 1,6-cyclodecadiyne 301 f, 311 5-cyclodecynone 305 1,7-cyclododecadiyne 311 cycloelimination reactions 293 f cycloheptadienes - bicyclic 152 - tricyclic 152 cycloheptadienones 151 cycloheptyne 104,293,309 cyclohexadienes 110 1,4-cyclohexadienes 146 cyclohexadienones 141 5-cyclo-hexyl-2-phenyl-2-cyclopentenones 158 cyclohexyne 104,293,311 cyclomerization, transition-metal-catalyzed 403 cyclonona-1,4,7-triyne 321 cyclononyn-3-one 295 1,lO-cyclooctadecadiyne 311 1,5-cyclooctadiyne 292,297 cyclooctatetraenes,ring-opening 368 f cyclooctatetraenes(COT) 110, 369 cyclooctene-diynederivatives 106 cyclooctenynes 104,294 cyclooctyne 104 cyclooligomer chemistry, phosphaalkynes 195 cyclooligomerization 128 - Lewis acid-mediated 189 cyclopentadienes 149,158 cyclopentadienonecomplexes 110 cyclopentadienones 150 cyclopenteneannulations, Michael addition 79 cyclopentenes 158
479
cyclopentenonesynthesis 139 cyclopentenones 85,110,155 cyclopentenyl sulfones 84 cyclopentine 104 cyclophane 345 - ortho, para- 42 cyclophane chemistry 294 cyclopropanation 152 cyclopropanes 151 cyclopropenes 158 cyclopropenyliumcation 11 cyclopropyl-ally1 rearrangement 182 cyclopropyldihydro-furans 152 (2+2)-cycloreversion 463 1,7-cyclotetradecadiyne 304 1,8-cyclotetradecadiyne 311 cyclotetramerization, thermal 184 - two-step 176 cyclotrimerization 7 - alkynes 110 - ethynyl ketone 408 - Pd(I1) catalysts 111f cyclotrimerization reaction, benzoylacetylenes 403 f cylindrical triple bonds 286 cytotoxicity 209,269 - designed enediynes 270 - kedarcidin 223 d low-spin $(u)-alkynyl iron(II) complexes 388 damsin 124 daunomycin 143 daunomycinone 143 decamethyl[5]pericyclyne 322,325 decomplexation,oxidative 289 defects, polymer chain 364 dehydration, cobalt-complexedpropargyl alcohols 120 1,6-dehydro[lO]annulene 290,307 dehydroannulenes 286 f, 443 - perethynylated 443,456 f 1,4-dehydrobutadiene 308 1,5-dehydronaphthalene 307 delivery systems 273 demetalation 115 f dendrimer growth, double exponential 424 f dendrimer synthesis 420 - monomer enlargement repetition 423 f
480
lndex
dendritic macromolecules 424 dendritic structures 404 dense-packed state, dendrimer synthesis 423 density functional study 18 density functional theory 4, 24 - calculations 443 deoxydynemicin A 217 deoxyribose, C(4') hydrogen atom abstraction 223 designed enediynes 262, 270, 273 - triggering, tempered reactivity 268 Dewar benzenes, doubly bridged 313 Dewar-Chatt-Duncansonmodel 99 f 2-Dewar-phosphinines, intramolecular (2+2)cycloadditions 178 DFT calculations 22 f diacetylenes 13 f - cyclic 297 - electronically excited 14 dialkyl alkynylphosphonates 87 gem-dialkyl effect 161 dialkynes, cyclic 295 f 3,4-dialkynyl-3-cyclobutene-1,2-diones446, 462 dialkynyliodonium triflates 72 f diallenes, cyclic 304 diastereoselectiveacyloin shift 242 syn-diastereoselectivity 125 f diatetryne 37 diazacyclodeca-3,8-diyne 297 3H-1,2,4-diazaphospholes 176 diazo group photolysis 401 3-diazo-l-(l-"C)-pne, photolysis 390 3-dia~o-l-(3-'~C)-propyne, photolysis 390 dibenzoisobulhalene 44 dibenzolumivalene 44 dibenzotriquinacene 44 dibromoacetylene 48 l,l-dibromo-2,2-difluoroethene 49
3',5'-dibromo-2-phenyl-l-(trimethylsilyl) acetylene 421 dicarbene-cumulene 16 dicarbene-cumulenestructure 15 dicarbenes 397 f - spin ground states 398 f P-dicarbonyl compounds 78 f 0-dicarbonyl enolates 78 dicationic propargylium complexes 105 1,2-dichloroacetylene 50,53 dichloroacetylene-diethyl ether complex 54 1,2-dichloroacetylenes 55
dichlorobutadiynecomplex 51 cis-2,3-dichloro-2-butene-1,4-diol452 3,4-dichloro-3-cyclobutene-1,2-dione462 dichlorovinylation 55 dicobalt hexacarbonyl fragments 448 dicyanoacetylenes 34 f, 44 1,2-dicyanobenzene 46 2,3-dicyano-1,3-butadiene 46 1,2-dicyanocyclobutene 46 dicyanodiacetylenes 34 f 4,5-dicyano-1,3-dithiol-2-one 35 dicyanopolyacetylenes 35 dicyanopolyynes 13 1,4-dicyclopropylbutadiyne 334 di-t-diphospha[4]pericyclyne 341 Diels-Alder addition 43,54 Diels- Alder adduct 40 Diels-Alder approach, enediynes 271 (2+4)Diels-Alder cycloaddition reactions 90 f Diels-Alder process 44 (4+2)Diels-Alder process 111 Diels-Alder reactions 7, 39 f, 146, 178, 184 - hetero 187 - homo 177,182 - intramolecular, esperamicin 242 - lithium-perchlorate-accelerated 45 - Livinghouse's Rh-catalyzed 113 - phosphaalkynes 177 - phosphacyclohexane 180 - retro 22,158,161,191, 366 f Diels-Alder route, retro 445 f Diels-Alder tandem reactions 145 dienediyne, palladium-mediatedconstruction 234 dienediyne core 237 dienediynes cyclic 286 diferrocene compounds 389 1,2-difluoroacetylene 50 1,2-dihalogenoacetylenes 56 - preparative uses 53 f dihalogenodiacetylenes 51 1,2-dihalogenoethynes 48 f 1,6-dihetero-3,8-cyclodecadiynes 303 1,6-dihydro[lO]annulene 226 dihydrobenzonaphthalenophane 43 dihydrofurans 158 dihydrogen exchange 10 dihydrogen transfer reactions 10 dihydrogenltriplebond addition 308 dihydrojasmone 124
Index
dihydrothionaphthoquinone 145 diiodoacetylene 49 1,2-diiodoacetylenes 51 diiodobutadiyne 51 4,9-diisopropylidene-1,6-cyclodecadiyne 286, 302 f diketene substructures 447 h e r , acyclic, dimethyl-1,4-pentadiene 332 dimer complexes, phosphaalkynes 185 dimerization, strained cycloalkynes 310 dirnetallacyclobutene 101 2,4-dimethyl-cyclohexen-3-one 55 di-0-methyl dynemicin esters 271 4,9-dimethylene-l,6-~yclodecadiyne302 f gem-dimethyl groups 322 3,5-dimethyl-3-hepten-2-one 121 3,3-dimethylpenta-174-diyne 331 f dimethylsilylene,photochemical extrusion 294 dinuclear propargylium complexes 118 f dinyliodonium triflates 72 f dinyl(pheny1)iodonium triflates 72 diodes, light-emitting 353 1,6-dioxa-3,8-cyclodecadiyne 287 3,4-dioxyl-1,5-enynes 126 diphenoxyacetylene 82 diphenyl substituted diazoalkynes 391 diphosphacyclopentadiene 191 diphosphatetracyclodecenes 176 f diphosphatetracycloundecadienones 182 diphosphatetrahedranes 187 diphosphatricyclooctenes 180 f diphosphirenes 183 1 AS, 3 X’ diphospholium ions 87 1,fdipolar cycloaddition reactions 8, 245 1,fdipolar cycloadditions 39, 91 dipole moment measurements 417 diradical formation/DNA cleavage 235 diradical polyyne 15 f diselenacyclodecadiyne 288 1,6-diselena-3,8-cyclodecadiyne 303 disila[6]pericyclyne 341 disodium acetylide 55 disordered amorphous domains, polyacetylenes 355 dissociation reactions, collision-induced 47 distance, transannular 297,299, 302 distance rule 239 distortion, DNA double helix 219 4,5-disubstituted cyclopentenones 158 3,5-di(tert-butyl)phenylacetylene 421 dithia[4]pericyclyne 341 f
481
dithia[6]pericyclyne 341 f divergent synthetic approaches 420 f cis-divinylcyclopropane 152 diyne cyclizations 307 diyne dithiocyanates 83 1,3-diyne, cross-coupling 347 @)-1,5-diyn-3-ene 203 1,6-diynes, ring-closing polymerization 362 diynones, cyclic 295 diynyliodonium salts 74 DNA, enedyne antibiotics 203 DNA affinity/selectivity 273 DNA alkylation pathway 229 DNA alkylation/DNA cleavage 235 DNA cleavage 209 - C-1027 223f - calicheamicin 215 - dynemicin 220 DNA cleaving - chromatin 223 - kedarcidin chromophore 222 DNA cleaving ability, esperamicins 216 DNA cleaving action, calichearnicinl esperacimin 239 DNA cleaving agents 214,219,239 DNAdamage 269 - NCSchromophore 211 DNA damaging properties 204 - kedarcidin 222 DNA double helix 208 DNA footprint experiments, oligosaccharide 215 DNA recognition sequences 222 DNA strand breakage 270 DNA strand cleavage, hydrogen atom abstraction 211 f dodecacarbene 405 dodecaketone 406 - starbust 405 dodecamethyl[6]pericyclyne 325 dopants 354 Dotz mechanism, metallacyclobutenes, intermediate 153 Dotz reaction 140 f double cyclization 437 double exponential dendrimer growth (DEDG) 424 f double exponential dendrimer synthesis 420 double Michael addition-carbene insertion 79 double stereodifferentiatingprocess 124 double-strandedlesions 211
482
Index
double-stranded supercoiled DNA, cleavage of phage ax174 240 doubly bridged Dewar benzenes 313 dry-state absorption methodologies 164 dry-state adsorption techniques 142 Durham polyacetylenes 367 Durham route, polyacetylenes 366 dynemicin 125,208,217 f - DNAcleavage 220 - synthetic studies 261 f dynemicin A 203 f dynemicin cascade 265 edge-opened tetraphosphaprismane 183 effective conjugation length 354 f, 361,371 f effective magnetic moments 394 Eglinton coupling 288 Eglinton-Glaser conditions 445 Eglinton-Glaser coupling 445,462,466 EHMO calculations 102 eight-membered cyclic diynes 297 electrical conductivity 461 - polyacetylenes 353 f electrical discharge 37,50 - Bieri method 49 electrochemical behavior, poly(RC0T)s 374 electrocyclic ring-opening 42 6 x-electrocyclization 154 electron back-donation 100 electron centers, unpaired, alkynyl compounds 385 n-electron conjugation, linear 459 electron correlation 11 - M~ller-Plessettreatment 389 electron correlation effects 2 electron correlation interactions 12 18-electron count, alkyne complexes 108 electron diffraction measurements, cyclic monoalkynes 296 electron impact ionization 46 electron localization 24 electron microscopy 353 electron-rich aromatic nucleophiles 122 electron spin, parallel, aligning 392 f electron spin polarization 398 electron transfer process 77 electron transmission spectroscopy 325 electronic device 375 electronic interactions, transannular 308 electronic structures 1 f
electronically excited diacetylenes 14 electrophilic acetylene equivalents 92 electrophilic addition 108 electrophilic reactions 10 f element-element bonds 173 1,2-elimination 292 f elimination reactions, cyclic alkynes 292 enantiomerically enriched 4-substituted cyclopentenone 161 enantioselectivesynthesis - calicheamicin 245 - calicheamicin aglycone 258 end-capping polymerization 463 end-effect, bond length variation 16 ene reactions 39,181 - NCSchromophore 236 - thermal intramolecular 232 ene reactivity, phosphaalkynes 180 enediyne anti-cancer antibiotics 125 enediyne antibiotics 57,105, 203 ff - medical applications 273 f enediyne cycloaromatization reaction 205 enediyne-containing dynemicin model compounds 264 enediynes - Bergman cyclization 9 - cyclic 286, 294 - model studies 240 cis-enediynes 445 trans-enediynes 451 f ene-yne-allene phosphine oxides 235 ene-ynes - a-methoxyp-acylation 121 - Pd-catalyzed cycloisomerization 113 - Zr-promoted bicyclization 113 energy, isomerization 3 energy surface, potential 5 enol derivatives 123 - intramolecular alkylation 125 enol silane propargylation 124 enone, dicyclic 305 enyne[3]cumulene,cycloaromatization 234 enynes, reactivity, Fischer carbene 151 1,Fenynes 126 a,o-enynes 127 enzymic catalysis 305 epoxidation, asymmetric 226 epoxide opening 219 f a,fl-epoxyacetylene complexes, opening 119 Ma, 17a-epoxycorticosterone 121 (+)-epoxydictymene 164
Index
epoxy dienediyne, NCS chromophore 227 EPR signals, temperature dependence 396 equatorial position, alkynyliodoniumsalts 75 equilibrium, cycloalkyne/cycloallene 304 Eschenmoser fragmentation 293 esperamicin 207, 216 f - theoreticalhynthetic studies 238 f esperamicin A , 204 esperamicin A , trisaccharide 255 f esperamicin/calicheamicin aglycones 58 espramicin/calicheamkin 125 17P-estradiol 128 16a-substituted-l7~-estradiols 123 estrone 112 ethidium bromide 269 4-ethylamino sugar (ring E), calicheamicin 257 ethylphenylacetylene 360 ethynologs 50, 56 ethynylcarbenes 389 ethynyldiazonium ions 67 2-ethynyl-3-hydroxytetrahadrofurans 119 ethynyliodonium salts 74,78 ethynyliodonium trif lates, P-functionalized 90 f ethynyl ketone, cyclotrimerization 408 ethynylmalonates 78 ethynyl(pheny1)iodonium triflates 68, 75 Evans-type homochiral boron eneolates 124 even-membered clusters 18 f even-numbered carbon chains 18 evolution, enedyne antibiotics 203 exchange coupling parameter, ferromagnetic 395 exo-face selectivity 160 exo/endo product ratio 166 expanded radialenes 443, 456 ff, 466 expansion, cluster 115,128 exploded [3]pericyclynes 333 exploded [5]pericyclynes 332 f exploded [4]pericyclynes 334 f exploded [6]pericyclynes 334 f exploded [8]pericyclynes 334 f exploded pericyclynes 321,330 f extended polyynes 461 Eyring plot 374 Faraday magnetic balance 410 fast bombardment (FAB) mass spectra, alkynyliodonium salts 74 f
483
"Fe Mossbauer time scale 389 Feast monomer 367 fenestrane derivatives 127 fermentation, microbial 212 ferromagnetic coupling 361, 393, 401 f ferromagnetic interactions 392 - molecular crystals 391 ferromagnetic intermolecular interactions, crystals 409 ferromagnetic spin 403 ferromagnetism, organic 6 films - Langmuir 362 - polycetylenes 354,357 Firestone's biradical mechanism 8 Fischer carbene/alkyne reactions 153 Fischer carbene complexes 152 Fischer carbene-tethered alkynes 151 Fischer carbene-tethered enynes 151 Fischer carbenes 139,144,151 - cycloaddition 139 flash evaporation 462 flash vacuum pyrolysis 53,177, 446 f, 462 fluoroacetylenes 52 1-fluoroacetylenes 53 fluorocyanoacetylene 37 1-fluoro-2-iodoacetylene 50 fluxional behavior, propargylium-metal complexes 102 focal point functionality, phenylacetylene dendrimer synthesis 423 focal point group 423 formal oxidation, poly(RC0T)s 375 formation, metallacycle 111 formation heat 19 formation mechanism, alkynyliodonium salts 71 four-electron donor, metal-alkyne complexes 101 Fourier transform mass spectrometry, (FT-MS)433,448 Fourier transform mass spectrum, silverattached macrotetracycle 436 fourth-generationdendrimer 422 fredericamycin A precursor 143 free-radical coupling reactions 401 free-radical molecules 392 Fremy's salt 386 frequencies,imaginary 22 Friedel-Crafts alkylation 362
484
Index
Friedel-Crafts-like alkylation, trimethylsilyl acetylene 324 Fritsch-Buttenberg- Wiechell rearrangement 295
frontier orbital calculations 182 FTIR markers 128 full valence CASSCF calculations 18 fullerene C60 459 fullerenes 448 - Kratschmer-Huffman 448 fully homoconjugatedcages, tricyclic cyclophanes 346 functional group placement 426 functionalized acetylenes 33 ff a,p, C-functionalizedacetylenes 123 P-functionalized ethynyliodonium trif lates 90 f
furan synthesis 153 furanones, propargylated 123 furans 80,147 furochromone khellin 144 tram-fused bicyclo(3.3.0)octane ring system 164
fusicoccin sesquiterpenoid skeleton 127
G2 theory 4 gas-phase electron affinities 325 gas-phase pyrolysis 35 gas-phase separation applications 362 gas separation membranes 353 gel permeation chromatography 401,424,435 geminally bisdeprotected tetraethynylethene derivatives 450 geometric progressive growth 430 germene 181 gilvocarcin V aglycone 143 Glaser coupling 288 - oxidative 461 glycals, glycosidation reactions 255 Glyphos 160 golfomycin A 228 f graft copolymer 357 Gram-negative bacteria 214 Gram-positive bacteria 212,224 graphite 459 graphitic structure 17 graphyne 455 Grignard derivatives,acetylenes 287 Gob-type fragmentations 157 ground-state geometries, carbon clusters 17
group VI vinylcarbene complexes 148 growth - progressive 430 - repetitive, nonlinear 434 guaianolide 157 GVB calculations 8 (1, 5)-Hshift
182
'H N M R spectra, alkynyl(pheny1)iodonium species 74 f haloacetylenes 67 P-haloalkylidenecarbene 87 haloalkynes 67 halocyanoacetylenes 37 halocyclopentenes 87 halogen nucleophiles 87 f (1.3)-halogen shifts 175 halogenoacetylenes 48 ff I-halogenoacetylenes 33 ff, 48 ff, 56 ff - derivatives 52 f - preparative uses 53 f halogenobutadiynes 51 1-halogenoethynes 48 f (Z)$-halovinyliodonium halides 87 Hammond postulate 4 hard nucleophiles 77 Hartree-Fock level 22 Hartree-Fock molecular orbitals 2 Hay conditions 461 Hay coupling 288,446,465 - oxidative 456 f HCN-acetylene complex 12 head-tail polymerization 363 heat - formation 19 - hydrogenation 7 - reaction 9 Heck reaction 287 Heisenberg model, one-dimensional 394 Heitler-London exchange 392 hetero Diels-Alder reaction 187 hetero-cyclocouplings 112 heteroatom derivatives, pericyclynes 340,343 heterobimetallic Mo-Co complexes 102 heterocyclicalkynyliodoniumspecies 70 f heterocyclic cognates, pericyclynes 340 f heterocyclic ring systems 153 heteronuclear addition reactions 309 hexabutadiynylbenzene 455 hexacarbonyldicobalt-alkynecomplex 155
Index
hexadehydro[l8]annulenes 445, 456 f hexaethynylbenzene 453 hexafluoro-2-butyne 39 hexamethylbenzene 43 hexaphosphapentaprismane 192 f 1,3,5-hexatriyne crystal, ferromagnetic 393 HF-SCF-MO methods 13 high dilution conditions 449 high-frequency Meller-Plesset energies, singledeterminant 24 high-spin ground state 399 high-spin molecules, acetylenic compounds 385 ff high-spin polyorganometallic compounds 409 high-temperature pyrolysis 37 highly orientated polyacetylenes, iodine doping 354 homo Diels-Alder reactions 177,182 f HOMO-HOMO interactions 9 HOMO-LUMO gap 101,457 HOMO-LUMO interactions 2 HOMO(dienediyne)-LUMO(disu1fide)interaction 209 homoaromatic stabilization 327 homochiral y -alkoxyallylboranes 126 homochiral propargyl ether complexes 124 homoconjugated cage structure 345 homoconjugated macorocycles 337 f homoconjugation 286,302 - cyclic 321 homoconjugative interaction 302 homogeneous semihydrogenation 109 homonuclear addition reactions 308 f homotetraphosphapentaprismane 177 Horner-Emmons condensation, intramolecular 245 Horner-Emmons Wittig olefination 243 host-guest chemistry 294 Hiickel-aromatic [18]annulenes 457 Hiickel-aromatic stabilization 443 Huckel’s rule 337 f Hiickel theory, extended 15 Hiickel-type theories 14 hydration, Pt(I1)-catalyzed 108 hydride addition, acetylene 11 hydrocarbon, Nenitcescu 40 hydrogen atom abstraction 211 - deoxyribose 223 hydrogen cyanide 34 hydrogen migration 4 1,5-hydrogen shifts 42
485
hydrogen transfer 4 hydrogenation 109 - cyclic dienediynes 308 hydrogenation heat 7 - decamethyl[5]pericyclyne 325 f hydrophobic pocket 209,224 hydroquinone-quinone redox process 240 f hydroxylamine linkages 214 [hydroxy(phosphoryloxy)iodo]benzene 82 [hydroxyl(tosyloxy)iodo]benzene 68 hydrozirconation, Schwartz’ 110 hyperfine coupling constants 386 f
imaginary frequencies 22 iminium ion salts 126 iminocarbene complexes 154 immunoconjugates, calicheamicin-antibody 273 imperfections, molecular 420 in-plane n-MOs 301 in-plane orbitals, triple bonds 285 in-vivo antitumor activity 224 indane 144 indanones 147 f indenes 144,147 INDO calculations 417 infrared analysis 435 infrared spectroscopic monitoring 435 inhibitors, PQQ 92 insect pheromone 5-(Z)-tetradecenyl acetate 127 1,2-insertion polymerization 366 insulating effect - bulky (iPr)3 Si groups 457 - Si(iPr)3 group 452 x l o -interaction 286 interactions - HOMO-LUMO 2 - transannular 301 intercalation 219 interchain charge mobility 354 f interchain contact 359, 363 interchain interactions 362 interchain order 357 intermediates, interstellar 46 f intermolecular cyclocoupling reactions 110 intermolecular oligomerization 432 intermolecular Pauson-Khand reaction 157 f interstellar chemistry 35 interstellar intermediates 46 f
486
Index
interstellar organic molecules 47 interstellar space 35 intracellular receptor, enediynes 270 intrachain order 357 intramolecular (2+2)-cycloadditions, 2-Dewarphosphinines 178 intramolecular (4+2)-cycloaddition 181 intramolecular 1,3-dipolar cycloaddition reaction 245 intramolecular acetylide-aldehyde addition 245 intramolecular acetylide-aldehyde ring closure 228 intramolecular aldol-based construction, calicheamicin 245 intramolecular alkylation, enol derivatives 125 intramolecular alkyne cyclocoupling 113 intramolecular aromatic alkylations 122 intramolecular cyclization 430, 432 intramolecular Diels-Alder approach - enediynes 271 - esperamicin enediyne core 243 intramolecular Diels-Alder reaction, esperamicin 242 intramolecular Dotz reaction 145 intramolecular heterocyclicalkynyliodonium salts 71 intramolecular Horner-Emmons condensation 245 intramolecular macrocyclization 334 intramolecular Nicholas reaction 289 intramolecular nucleophilic trapping 105 intramolecular Pauson-Khand cyclization 127 intramolecular Pauson-Khand reaction 161 f intramolecular radical recombination 290 L3-iodane 71 iodine doping, highly orientated polyacetylenes 354 iodine species, polyvalent 67 1-iodoacetylenes 53 (2)-1-iodo-1-akenes 56 iodof luoroacetylene 49 iodonium tetrafluoroborates 70 iodonium transfer agent 69 iodonium triflates 75 iodonium-substitutedalkynes 77 1-iodopropargylalcohol 57 iodosobenzene 68 iodosyl triflates 72 ion spectrum, metastable 47 ion-moleculedimerization reaction 448
ion-molecule reactions, cyclocarbon cations 448 ionic bonding, alkynyliodonium salts 76 ionization, electron impact 46 ionization potentials 15 ionized dicyanodiacetylene 47 iron(I) alkynyls 388 iron@) alkynyls 388 isocyclooclorenone 124 isodesmic transformations 13 isolaurepinacin 121 isolobal metal fragment substitution 115 isomerization - alkyne-vinylidene 116 - alkyne/allene 118 - vinylidene 3 1,3-isomerization, alkyne complexes 118 isomerization energy 3 isomers 20 isometric cage compounds, phosphaalkynes 186 isosbestic point 374
2J(P,P)coupling constant 178 Jahn-Teller distortion 101 Johnson orthoester variant, Claisen rearrangement 455 kedarcidin 208,221 f kedarcidin chromophore 204,221 kedarosamine synthesis 273 Kende synthsis, calicheamicin bicyclic core 242 ketal protecting groups 447 ketene intermediates 447 ketone propargylation 123 kinetics, cycloaromatization 225 Kratschmer-Huffman conditions 35 Kratschmer-Huffman fullerene production 448 lachrimators 49 lactams 153 p-lactams 124 p-lactams antibiotics 126 y-lactams 85 Lalezari reaction 294 Langmuir films 362
Index lanthanide-alkynen-complex 100 laser desorption time-of-flight mass spectra 459 laser flash heating 445 laser flash photolysis 396 Lepidoptera 56 lesions, double-stranded 211 leukemia cell lines 269 leukemias 214 leukotriene-E4 121 Lewis acid catalysis 8, 40 Lewis acid-mediated, cyclooligomerization 189 Lewis-acid-promotedalkylations - silyl ketene acetals 125 - silylenol ethers 125 Lewis-acidic metathesis polymerization 370 ligand-to-metal charge tranfer 88 ligand-to-metal electron donation 100 ligands, perethynylated, transition metal complex 455 light-emitting diodes 353 light-energy conversion device 418 f Lindlar catalysts 109 linear carbon clusters, isomers 21 linear solid-phase synthesis 420 liquid crystals 426 - alkynyliodonium salts 88 liquid-crystallinepolymers 362 lithium chloroacetylene 56 lithium-perchlorate-acceleratedDiels-Alder reaction 45 living polymerizations 362 Livinghouse’s Rh-catalyzed Diels- Alder reaction 113 long-chain a,o-diyne, oxidative cyclization 348 long-chain polyynes 462 long-range exchange coupling 385 f low-spin $(a)-alkynyl iron(I1) complexes 388 low-spin ground state 399 low-temperature matrix isolation studies 447 low-temperature photolysis 447 lumibullvalene 44 M-C addition 110 M-C insertions 128 M-H additionlhydrogenation 109 f macrocycles
487
- homoconjugated mixed polyalkynesldiynes
337 f - phenylacetylene
426 f macrocyclic homoconjugatedpolyacetylenes 321 ff macrocyclization, intramolecular 334 macromomycinlauromomycin 221 f macroscopic polymer morphology 354 macrotetracycle 431 maduropeptin 208,221 f magnetic moments, effective 394 magnetic properties, high-spin molecules 409 magnetization curves 394, 402 magnetization measurements 404 Magnus cyclization 289 manicone 120 marasin 59 Markovnikov selectivity, metal-alkyne complexes reactions 108 mass spectra, laser desorption time-of-flight 459 mass spectrometry - resonant two-photon ionization time-of flight 445 - sample preparation 437 - neutralization-reionization 46 matrix isolation, cycloalkynes 293 MB49 murine bladder carcinoma cells 229 MCSCFl4-31G 9 medical applications,enediyne antibiotics 273 f medium-sized cycloalkynes 126 melanoma 214,224 melt-thermolysis 114 f membranes 353 Merrifield polymer 157 Merrifield-likephenylacetylene sequence synthesis 433 metal acetylides, nucleophilic substitution 286 f metal-acetylene reactions 12 metal-alkyne complexes 99 ff metal-alkyne complexe reactions 107 ff metal--bonded alkyne comlexes 99 metal carbene cycloaddition 139 metal-catalyzed alkyne polymerization 110 metal-catalyzed oligomerization 37 metal cluster chemistry 128 metal-complexed-stabilizedNicholas cations 290 metal-dependent chemoselectivity 148
488
Index
metal-promoted additions, alkynes 108 metal-promoted coupling reactions, alkynes 110 metal-to-ligand charge transfer 88 metal-vinyl complexes 108 f metallacyclopropene 100 metallacycle formation 111 metallacycles, (2+2)-cycloaddition 359 metallacyclobutadienes 114 metallacyclobutane 151 metallacyclobutenes 141,362 - intermediate 153 metallacycloheptatrienes 111 rnetallacyclopentadienes 110 metallapyran 113 metallcyclopropenestructure 388 metastable ion spectrum 47 metathesis 128 - alkynes 114f metathesis catalyst system 362 metathesis catalysts 359 f metathesis polymerization 359 a-methoxy 9-acetylation, ene-ynes 121 Ip-methylcarbapenemprecursors 121 1-0-methyldefucogilvocarvin 143 methyl glycoside, calicheamicin oligosaccharide 254 methyl ketones, terminal acetylenes 345 f 12-0-methyl royleanone 144 methyl thioglycolate 230 10-methylthioisoborneolmoiety 160 mgnetic susceptibility 402 Michael acceptor 55,446 Michael addition 78,87,154,447 microbial fermentation 212 microcanonical variational transition-state analysis 11 microwave spectra, halogenbutadiynes 51 migration 103 migration, antarafacial 102,124 migration, hydrogen 4 migratory aptitude, ethynyliodonium salt substitution 78 MIND02 study 16 minor groove 207 minor groove, double-helical DNA 214 Misumi cyclization 288 MM2 (molecular mechanics 2) 445 MNDO calculations 16 MO calculations, tetraphosphacubanes 185 MO theory 398 f Mo-Co complexes, heterobimetallic 102
modular constructions 427 moieties, valency isoelectronic phosphorus(111) 173 molecular antenna 419 molecular assemblies, acetylenic compounds 385 ff molecular complexes 12 f molecular crystals 426 - organic free radicals 391 f molecular dynamics simulations 17 molecular electronics 463 molecular-engineering 33 molecular imperfections 420 molecular interactions 7 ff molecular machinery 415 molecular mechanics calculations, model polyenes 372 molecular orbital theory 1,15 molecular orbitals, Hartree-Fock 2 molecular wires 461 f molecular wires, all-carbon backbones 443 molecules, perethynylated 455 Maller-Plesset energies, single-determinant high-frequency 24 Msller-Plesset pertubation levels 3 Maller-Plesset pertubation theory 17 Maller-Plesset second-order pertubation theory 443 Msller-Plesset treatment, electron correlation 389 MOLT-4 leukemia cell line 269 molybdenum alkylidene catalyst 368 molybdenum alkylalkoxycarbene 152 molybdenum arylcarbene complexes 148 molybdenum carbene complexes 151 f mono-thia[S]pericyclyne 340 monoalkynes, cyclic 296 f monocyclic Clo 23 monocyclic rings 17 monocyclic structures 20 monodendrons 420 monodeprotectedtetraethynylethenes 451 f monodisperse high-molecular-weight linear phenylacetylenes 435 monodisperse polymers 362 f monohalogenatedbutadiynes 50 monolayer surfaces 426 monomer enlargement repetition, dendrimer synthesis 423 mononuclear q3-propargyl/a1leny1complexes 102
Index mononuclear q3-propargylium-M complexes, nucleophile addition 116 n MOs - in-plane 301 - triple bonds 285 MP2 level 8 MP2/6-31G* 10 MP2/6-31G* 21 MP2/6-31G** 12 MP4/6-31G* 20 MRCI modified coupled-pair functional (MCPF) 21 mucochloric acid 452 multiconfiguration SCF calculations 9 multiconfiguration wavefunction 2 multifunctional core 420 multiple-drug-resistant TCAF-DAX cell line 269 multireference configuration interaction (MRCI) 19 murine tumors 214 Myers cycloaromatization 228 Myers-type cycloaromatization 230 Naarman-polyacetylene 359 nanomer-sized molecular rods 463 nanoscaffolding, phenylacetylene 430 nanoscale platinum cluster 128 nanostructural target, macrotetracyclic 431 nanostructures, acetylenes 415 ff naphthalene derivatives 145 2-naphthylcarbenes 143 naphtols 140 f natural enediyne antibiotics 275 naturally occurring enediyne antibiotics 240 NCS apoprotein 209 - chromophore interactions 209 NCS chromophore - DNAdamage 211 - epoxydienediyne 227 NCS chromophore activation 210 NCS chromophore core structure, Wittig rearrangement strategy 236 negative hyperconjugation, alkynyliodonium salts 76 Nenitcescu hydrocarbon 40 neocarzinostatin 208 f, 287 neocarzinostatin chromophore 125,204,233 - modelsystems 224f neocarzinostatin-type diradicals 225
489
neutralization-ionization mass spectroscopy 47 neutralization-reionization mass spectroscopy 46 Nicholas reaction, intramolecular 289 Nicholas cobalt-stabilized propargyl cation chemistry 164 Nicholas-type reaction 242 nickel-catalyzed oligomerization 99 Nicolaou distance rule 239 nine-membered ring - NCSchromophore 227 - neocarzinostatin chromophore 231 nitrogen nucleophiles 80 f nitronyl nitroxide radicals 401 nitroxide radical centers 400 f nitroxide radicals 361, 386 noncarbon nucleophils 121 noncovalent interactions 372 nonlinear growth scheme 419 nonlinear optical materials 53 nonlinear optical properties, third-order 395 f nonlinear optical waveguides 353 nonlinear optics, alkynyliodonium salts 88 nonlinear repetitive growth 434 nonlinear straight-chain sequence synthesis 420 nonlocal density functional formalism 17 nonracemic intermolecular Pauson-Khand chemistry 160 norcaradiene valence isomers 42 normal cell lines 269 normanicone 120 Nozaki cyclization 290 Nozaki reaction 58 Nozaki-type coupling 240 nucleophile addition, mononuclear q3 propargylium-M complexes 116 nucleophiles - alkynes 77f - electron-rich aromatic 122 nucleophilic acetylenic substitutions 67 nucleophilic additions 11 - metal-alkyne complexes reactions 107 nucleophilic attack, propargylic centers 289 nucleophilic substitution 67 - metal acetylides 286 f nucleophilic trapping, intramolecular 105 nucleophils, noncarbon 121
5-octamethylpericyclyne 329 octamethyltetrasilacyclohexyne 294 odd-membered clusters 18 odd-membered linear polyynes 463 oleate-crepenylate pathway, polyacet ylenes 212 olefin metathesis 151, 366 olefin polymerization catalists, Ziegler-Natta 114 olefins, Pauson-Khand reaction 154 oligoacetylenes 286, 443 ff oligobenzocyclynes 306 oligomerization - intermolecular 432 - nickel-catalyzed 99 - strained cycloalkynes 310 oligomers, one-electron reduction 463 oligometric sequence liberation, from support 439 oligonucleotide synthesis, combinatorial strategies 433 oligopeptides 427 oligosaccharides - calicheamicin 249 f - DNA footprint experiments 215 oligosilanes, cyclic 294 one-configuration approximation 2 one-electron reduction, oligomers 463 opening, a,P-epoxyacetylene complexes 119 optical materials, nonlinear 53 optical properties - acetylide metal complex 88 - nonlinear, third-order 395 f, 461 optical waveguides, nonlinear 353 x-orbital, out-of plane 301 orbital interactions, through space 327 ordered crystallites, polyacetylenes 355 organic conductors, alkynyliodonium salts 88 organic ferromagnetism 6 organic ferromagnets 403 organic free radicals 392 - molecular crystals 391 f organic magnet 361 organic molecules, interstellar 47 organic-based ferromagnet 400 organometallic compounds - paramagnetic 387 - paramagnetic alkynyl 409 organometallic cycloaddition reaction, actylenes 139 ff
organometallic radicals 127 organometallic species, alkynyliodonium salts 88 f organophosphorus compounds 173 ff organotransition metal chemistry, alkynes 99 f orthoester variant, Johnson, Claisen rearrangement 455 out-of plane x-orbital 301 ovarian cancer 273 oxabicyclo(7.2.l)enediyne 58 oxadiphosphapentacyclononadecapentaenones 182 7-oxanorbornadiene 366 oxaphospholes 153 oxepinoparacyclophane 45 oxidation, formal, poly(RC0T)s 375 oxidative carbonylation 107 oxidative coupling 128, 445, 449, 461 oxidative cyclization 446 - long-chain a,w-diyne 348 oxidative cyclotrimerization, C 18 106 oxidative decomplexation 289 oxidative demetalation 115 f oxidative dopants 354 oxidative Glaser coupling 461 oxidative Hay coupling 456 oxidative polymerization 456 f 1-oxido-alkydene-chromiumcomplex 152 p-0x0-A'-iodane 68 2-oxoazetidinylmalonates 78 oxygen nucleophiles 81 f oxygen-tethered acetylenic carbenes 150 ozonolysis 9 P-functionalization, tetraphosphacubane 188 "P-NMR spectroscopy 174 palladium-catalyzed alkynation 445 palladium-catalyzed cross-coupling reactions 419 palladium-mediated dienediyne construction DC 234 paracyclophane 42 [2.2]paracyclophane 40,393 paramagnetic alkynyl organometallic compounds 409 paramagnetic transition metals 387 Pauli bond number 417 Pauling's resonance theory 2
Index
Pauson-Khand chemistry, nonracemic intermolecular 160 Pauson-Khand reactions 110,142,154 f - asymmetric, alkoxyacetylene 118 - intramolecular 161 f - tandem nucleophilic coupling 127 Pauson-Khand substrates 155,158 Pd(0)-catalyzed alkynylation 449 Pd(0)-catalyzed cross-coupling 422 Pd(0)-catalyzed intramolecular coupling 230 Pd(0)-mediated C-C bond-forming process 230 Pd-catalyzed CO insertion 80 Pd-catalyzed cylcloisomerization, ene-ynes 113 Pd-mediated coupling reactions 287 PE spectroscopy 301 f pentacarbonyl(methoxypheny1carbene) chromium 139 1,2,3,4,5,-pentafluorostyrene 51 pentalenene 163 pentalenic acid 163 pentanitrene 399
pentaphosphatetracyclononene-W(CO)s complex 184 pentaprismane 173 peptide synthesis, combinatorial strategies 433 perethynylated (n)cumulenes 455 perethynylated dehydroannulenes 443,456 f perethynylated tetradehydro[l2]annulenes 456 perfluoro-3-methyl-1-butyne 53 perfluorodialkyl-1,2,3-triazine 53 pericyclic reactions 7 f pericyclic reactions, alkynes 9 pericyclynes 321 ff - higher 324 [3]pericyclynes 321, 327, 341 [4]pericyclynes 326 f - heteroatom derivatives 340, 343 [SJpericyclynes 328 f [5-8]pericyclynes 324 [8]pericyclynes 326 - heteroatom derivatives 340,343 [lO]pericyclynes quinone 328 f [Nlpericyclynes 321, 341 f pericyclynones 329 f periodic network 415 permethylated pericyclynes 287,322 perspirocyclopropanated 14-membered ring pentayne 338
491
perspirocyclopropanated pericyclenes, exploded 334 f pertubation levels, Meller-Plesset 3 pertubation theory, Meller-Plesset 443 perylene luminophore 419 Peterson olefination 231 phage DX174 double-stranded supercoiled DNA 240 phase I clinical trials 273 phase transitions, anomalous 402 phenanthrenes 143 phenoxyls 385 f l-phenyl-1,3,5-hexatriyne 385 f l-phenyl-1,3-butadienes 385 f, 397 1-phenyl-2-propynylidenes 391 3-phenyl-2-propynylidenes 391 rn-phenyl-enedicarbene 399 p-phenyl-enedicarbene 399 phenyl-rings, end-capping 463 phenylacetylene dendrimer growth, double exponential 425 phenylacetylene dendrimers 418 f - convergent synthesis 421 f phenylacetylene macrobicycle, double cyclization 430 phenylacetylene macrocycle framework 428 f phenylacetylene macrocycles 426 f phenylacetylene macromolecules 416 phenylacetylene macrotetracycle, foldable 431 phenylacetylene monodendron synthesis 420 phenylacetylene monomers 428 phenylacetylene sequences (PASS) 426 phenylacetylene valence angles 417 phenylacetylenes 385 f - monodisperse high-molecular-weight linear 435 phenylvinylium 11 pheromone synthesis 56 phophaalkyne tetramers 190 phophabenzvalenes 178 f phophabicyclobutane 179 phophahydroquinones 153 phospha(N)pericyclynes 341 f (N)phospha(N)pericyclynes 344 - cyclic electron delocalization 345 2-phospha-Dewar-benzenes 178 phosphaalkyne cyclomerization 194 phosphaalkyne dimer complexes 185 f phosphaalkynes 173 ff - synthesis 174f phosphabarrelene 176
492
Index
phosphabenzenes 178 f 2-phosphabicyclooctadiene 177 phosphacubane 176 phosphacubane salts 86 phosphacyclohexene, Diels- Alder reaction 180 phosphaprismanes 178 f phosphavinylcarbenes 175 L3-phosphinines 176,178 1-phosphirenes 175 phosphorus nucleophiles 86 f phosphorus-carbon cage compounds 173 ff phosphorus-carbon cage compounds, synthesis 177 f phosphorus-carbon-aluminum cage compounds 194 f phosphotriesterase 82 photoadditions 46 photochemical cis/trans isomerization 373 photochemical cycloreversion 463 photochemical extrusion, dimethylsilylene 294 photochemical formation, cyclo-C 447 photochemical ring-closure 44 photochemically triggered systems 271 photoelectron spectra, cyclic diacetylenes 301 f photoelectron spectroscopy 325 - tf-acetylides 388 photoinduced simulation, dynemicin A 268 photoiodination 56 photolysis - annealed cyclopropenones 293 - diazogroups 401 - laser flash 396 - low-temperature 441 photolytic ring construction, bis-propargylic sulfone 236 photosensitive triggering 269 photosensitive triggering devices 263 photovoltages, silicon/poly(Me&COT) device 375 pinacol rearrangement - dynemicin 262 - Tsuchihashi pinacol 242 planar bridged acetylene 4 plastic metals 353 polyacetylene films 354, 357 polyacetylenes 13 ff, 38,286,353 ff, 461 - macrocyclic homoconjugated 321 ff - naturally occuring 59
- organic conductors 114 N-polyacetylenes 359, 376 polyalkynes/diynes macorocycles, homoconjugated mixed 337 f polybenzavalene 368 poly(l,3-butadiynes),spin-alignements 395 f poly(sec-butylCOT), thermal isomerization 374 poly(sec-butylcyclooctatetraene) 370 polycarbenes, super-high spin 404 polycarbon nitride radicals 47 polycoordinated iodine(I1I) chemistry 76 polycyclic phosphanes 173 polycyclooctatetraenes (poly(RC0T)s) 370 polydiacetylenes 461 polydiazo compounds 404 poly(difluoroacety1ene) 376 poly(diphenylacety1ene) 362 polyene backbone 361 polyenyne systems cycloaromatization 205 poly(2-ethylhexylCOT) 371 poly(2-ethynyl-N-methylpyridinium) 361 poly(f luoroacetylene) 376 1,4-polyisopropene 365 polymer backbone 357 polymer chain defects 364 polymer conformation 362 polymer crystallization 354 polymer morphology, macroscopic 354 polymer thin films 371 polymerization 128 - end-capping 463 - oxidative 456 f polymerization alkynes 114 f polymerization degree 420 polymerization rate 363 polymerized acetylene 38 poly(o-methylphenylacetylene) 361 poly(neopentylC0T) 371 polynuclear alkyne complexes 115 polynucleic acids 415 polypeptides 415 poly(phenylacety1enes) 361 - spin-alignments 395 f - spin amounts 400 f poly(phenyldiacetylene), spin amounts 400 f poly(pheny1ene vinylenes) 355 poly(pheny1 vinyl sulfoxide) 364 poly(propyne) 359 polypyrroles 355 polyradicals, super-high spin 404
Index poly(RCOT)s, electrochemical behavior 374 polysaccharides 415 polystyrene beads 434 trans-poly(tert-butoxyCOT) 372 trans-poly(tert-butylCOT) 371 polythiophenes 355 - stereoregular substituted polythiophenes 376 polytriacetylene backbones 463 polytriacetylenes 461 poly(trimethylsilylacetylene) 361 poly(trimethylsilylC0T) 373 poly(trimethylsilylcyc1ooctatetraene) 370 poly(o-trimethylsilylphenylacetylene) 361 polyvalent iodine species 67 poly(viny1 chloride) 364 polyyne carbon 13 polyyne dinitriles 46 polyyne structure 15 polyynes 15 - linear 461 - linear infinite 461 potential energy surfaces 3 f, 11, 392 - AM1 11 PQQ inhibitors 92 PRDDO-CI 9 precursor polymer 357 precursor routes 353, 363 f prismanes 173, 313 propargyl anions 288 propargyl cations 11,288 propargyl alcohol 39 q3-propargyl/alleny1Pt complex 88 propargyl-cobalt coupling reactions 124 (propargyl)Co,(CO), radicals 127 propargyVhydrocarby1coupling 127 o-propargyl derivatives 116 propargyl radicals 288,385,389 propargyl synthons 118 propargylated furanones 123 propargylic-allylic anion 290 propargylic anions 290 propargylic a-carbon 118 f propargylic radicals 290 propargylium-cobalt complexes 105 (propargylium)Co2(CO): complexes 123 (propargylium)Co2(CO)),Lcomplexes 118 q’-propargylium complexes 128 propargylium-metal complexes 101 f (4.3.2)propellatrienes 463
493
propynes, polymerization, Ziegler-Natta catalysts 359 2-propynylidenes 389 f 13-cis-prostaglandin 57 prostaglandin analogs 123 prostaglandin synthesis 57 protease inhibitors, alkynyl benzoates 82 proteolytic activity 223 proton affinity 13 proton loss/elimination 120 protonation reaction, acetylene 10 f proximity 305 pseudo-high-dilution conditions 430 pseudodiaxial steric interaction 161 pseudoguaianolide 157 Pt( 11)-catalyzed hydration 108 push-pull polymers 363 push-pull polyynes 53, 363 push-pull ynamides 80 push-push polymers 363 trans-pyramidalization 11 pyrazolopyridine quinones 153 pyridines 110 f, 153 pyrolysis, flash vacuum 53,177, 446 f, 462 pyrolysis, gas-phase 35 pyrolysis, high-temperature 37 pyrolysis, vacuum 175 a-pyrone 8 pyrones 112,153 pyrroles 153 f pyrromethanenones 124 quadrupole moment 391 quantum chemistry 1 quinone methide intermediate 268 quinter ground state 399 quintet states, thermally accessible 398 [Slradialene substructures 459 radialenes 291, 459 f - expanded 443,456 f, 466 radical additions 11f radical cation 6 radical ions 6 f radical recombination, intramolecular 290 radicals 127 radioactive end-group labeling 353 Ramberg-Biicklund reaction 238,294
494
Index
Ramberg-Backlund ring-contraction reaction 289 reaction heat 9 reactivities 7 ff - acetylene 286f - phosphaalkynes 175 rearrangement - cyclic alkynes 303 f - Fritsch-Buttenberg-Wiechell 295 receptor, intracellular, enediynes 270 redox activation, enediyne systems, Myer’s approach 241 reduction potentials, poly(RC0T)s 375 reductive dopants 354 regioselectivity, intermolecular Pauson-Khand reaction 165 rehybridization, metal-alkyne complexes 101 Renner-Teller molecule 6 repetitive growth, nonlinear 434 resonance energy 14 resonance fields, angular dependence 396 resonant two-photon ionization time-of-flight mass spectrometry 445 retinoid chemistry 57 retro Diels-Alder reaction 22,157,161,191, 366 retro-Diels- Alder process 44 retro-Diels-Alder route 445 f retrodimerization reaction 398 retrosynthetic analysis, calicheamicin oligosaccharide 249 f rhenacycles 116 RHF ab-initio calculations 15 RHF theory 13 RHFI6-31G 7 RHFIUHF instabilities 24 D-(+)-ribonolactone-derived substrate 162 ring annulation 206 - five-membered, carbocyclic 149 ring contraction 294 f ring pentadiynes, 14-membered 339 ring strain 286 ring-closing polymerization, 1,6-diynes 362 ring-closure - acetylide anion-aldehyde 226 - acetylide-aldehyde 227 - photochemical 44 ring-closure reactions, cyclic alkynes 286 f ring-enlargement, thermal 44 ring-enlargement reactions, cyclic acetylenes 295
ring-opening, cyclooctatetraene 368 f ring-opening olefin metathesis polymerization (ROMPSs) 366 Ritter reaction 122 ROMPS, ring-opening olefin metathesis polymerization 366 RRKM 11 sample preparation, mass spectrometry 437 sandwich complexes 177 scaffolding, acetylenes 443 SCF MP2 calculations 17 Schmidt trichloroacetimidate methodology 261 Schreiber-Evans chiral boron enolate methodology 124 Schwartz’s hydrozirconation, alkynes 110 scission, alkynes 114 f selenadiazoles 294 selenoxide elimination chemistry 243 selenoxocarbenes 176 self-consistent field calculations, (SCF) 443 self-consisting fields 2 self-coupling, terminal acetyienes 432 semibullvalenophane 40 semiconductor/conducting-polymersolar cell 375 semiempirical CNDO calculations 417 semiempirical MO theory 15 semiempirical study (AM1) 17 semihydrogenation, homogeneous 109 sequence specificity, calicheamicins 215 sequence-specific double-strand cuts 214 sequence-specific phenylacetylene oligomers, solid support 433 f sequence-specific single-stranded cuts 222 seven-membered trans-cycloalkene 309 shape-persistence 415 sharpless asymmetric epoxidation 226 Shirakawa polyacetylene 358,368,376 Si(iPr), group, insulating effects 452 Si substitution, P-carbocation stabilization 72 sigmatropic rearrangement, calicheamicin oligosaccharide 249 f [3,2]-sigmatropic rearrangement 235 [3,3]-sigmatropic rearrangement, bicyclo(5.3.0)decanes 152 sila-acetylene 68 sila-alkynes 69, 72 sila-ethynyliodonium triflate 78
Index
silacyclodecadiyne 290 silicon, pericyclynes 344 silicon/conducting-polymersolar cell 375 silicon/poly(Me3SCOT) device, photovoltages 375 silver chemical ionization laser desorption mass spectroscopy 436 silver-attached macrotetracycle, Fourier transform mass spectrum 436 silyl ethynyliodonium tetrafluoroborates 69 silyl ketene acetals, Lewis-acid-promoted alkylations 125 1,2-silyl migration 117 silyl-protected [3]cumulenes 455 silyl-protected peripheral valences 459 p-silyl vinylcarbenes 146 silylated terminal alkynes 287 silylenol ethers, Lewis-acid-promoted alkylations 125 single crystal X-ray molecular structure, alkynyliodonium salts 75 single-determinant ab-initio molecular orbital theory 13 single-determinant HF calculations 16 single-determinant high-frequency Meller-Plesset energies 23 single-stranded DNA 209 single-stranded DNA bulges, thiol-independent cleavage mode 211 single-stranded DNA cuts 216 singlet diradicals 399 singlet ground state, acetylene 385 singlet ground states 15 singlet-triplet coupling 6 singlet vinylidene 5 size exclusion chromatography 436 small-ring pericyclynes 344 smart materials 353 Sn substitution, carbocation P-stabilization 72 soft nucleophiles 77 solar cell - semiconductor/conducting-polymer 375 - silicon/conducting-polymer 375 solid support, sequence-specific phenylacetylene oligomers 433 f solid-phase synthesis, linear 420 solid-phase techniques 435 solid-state polymerization 402 - 1,3-butadiynes 395 f - topochemically controlled 389
495
solid-supported phenylacetylene chemistry 438 soluble carbon rods 463 soluble substituted polyacetylenes 376 solution-spray flash vacuum pyrolysis 462, 466 solvent-assisted sublimation 446 - Fowler 462 solvent-induced rearrangements 46 Sondheimer annulene synthesis 308 sp center, "C chemical shift 297 spectroscopic properties - alkynyliodonium salts 74 f - cyclic alkynes 296 f spin alignment 393 - poly(phenylacetylenes)/poly(1,3-butadiynes) 395 spin-coupled wavefunction 2 x-spin density 386 spin ground states 399 spin localization 385 f spin-orbital effects 75 spirocyclic phosphorus derivatives 189 spirocyclopropanes 339 spirocyclotrimerization 189 squaric acid 446 stabilization - carbocations, p-Si substitution 72 - Hiickel-aromatic 443 stable free-radical substituent, antiferromagnetic crystals 393 stacking, one-dimensional 395 stannyl acetylene 144 Staphylococcusaureus 59 starbust dodecaketone 405 Staudinger reactions 189 stemodin 124 step-wise mechanisms 7 stereocontroling interaction, intramolecular Pauson-Khand reaction 162 stereodifferentiatingprocess, double 124 stereoregular substituted polythiophenes 376 stereoretentivecleavage 109 stereoselective synthesis 56 E-stereoselectivity 119 steroid ring system 146 STO-3G level 389 STO-3G minimal basis set 13 straight-chain sequence synthesis, nonlinear 420 strain energies 239
496
Index
strained cycloalkynes 292 dimerization/oligomerization 310 strained triple bonds 286 stranined cycloalkynes 104 Straws reaction 52 structural properties, cyclic alkynes 296 f structure - cumulenic 19 - dendritic 404 - electronic 1 - graphitic 17 - metal-alkyne complexes 99 f - monocyclic 20 - tetraphosphacubane 185 - transition 4 structure data, alkynyliodonium salts 76 structure-directed assembly 426 structure-property correlation maps 433 structure-property relationship 354 sublimation, solvent-assisted 446 substituted alkynes, polymerization 359 f 4-(P-substituted ethynyl)-2,6-di-fertbutylphenols 385 f substitution, cluster 115 substrates, Pauson-Khand 155 sugar fragment, esperamicin A, 254 sulfides, cyclic 289 b-sulfone triggers 270 sulfones, bis-propargylic 233 sulfonylvinyliodonium salts 85 sulfur-eliminating ring-contraction procedures 294 sulfur nucleophiles 83 f super-high spin polycarbenes, dendritic structure 404 super-high-spin molecules 402 super-high-spin organic molecules 385, 399 super-high-spin polyradicals, dendritic structure 404 superdiamonoid carbon network 455 superphanes 311 supersonic jet electronic spectral data 417 suprafacial migration 103 supramolecular chemistry 415,433 susceptibility, magnetic 402 SYNDROME method 423 synthesis - aromatic systems 206 f - cyclic alkynes 286 f synthetic precursors, permethylated pericyclynes 323
-
Ta-alkyne complexes 113 tail-tail polymerization 363 tandem bimolecular coupling 431 tandem Claisen-Bergman rearrangement 207 tandem cyclization 431 tandem Dotz/nucleophilic aromatic addition reactions 146 tandem Ireland-ClaisedBergman rearrangement 206 tandem mass spectroscopy methods 47 tandem Michael addition-carbene insertion reaction 79 tandem nucleophilic coupling, Pausen-Khand reaction 127 tandem reactions, Diels-Alder 145 target sequence recognition, carbohydrate fragment 245 taxamycin 207 taxane 207 taxol 207 temperature dependence, EPR signals 396 tempered reactivity, designed enediynes triggering 268 ten-membered 1,6-diynes 297 ten-membered enediyne 244 ten-membered enediyne ring 216,238,243, 270 ten-membered ring, diacetylenes 298 terminal acetylenes - methyl ketones 345 f - self-coupling 432 terminal alkynes, base 117 tertiary propargylic chloride/terminal acetylene coupling 348 tetra@-buty1)tetraphosphacubane 86 tetracyanocyclooctatetraenes 46 tetracyclic anthracyclinone 143 tetradehydro[12]annulenes, perethynylated 456 f tetradehydrobenzenoidmolecules 105 5,6,11,12-tetradehydrodibenzo[a,e]cyclooctene 297 tetraethynylallene 455 tetraethynylethene derivatives 443 tetraethynylethenes 449 f tetrahomocyclooctatetraene 308 tetrahydronaphthalene synthesis 207 tetrahydrophenanthridine 261 2,5,7,10-tetraisopropylidene-1,6-ditha-3,8cyclodecadiyne 291
Index
1,1,6,6-tetraphenyl-2,4-hexadiyne-1,6-diol 392 tetraphospha[4]pericyclyne 344 tetraphospha[8]pericyclyne 341 tetraphosphacubane 176,184 ff tetraphosphacuneane 184,188 tetraphosphasemibullvalene 191 tetraphosphatetracyclooctene 184, 187 tetrapyrroles 124 tetraradical-polyyne 16 tetrasilacyclohexyne 286 tetrasubstituted tetraethynylethenes 451 tetrathiafulvalene 128 tetrathiafulvalene systems 310 theoretical aspects, acetylene chemistry 1ff thermal cycloaromatization, enedyne antibiotics 203 thermal cyclotetramerization 184 thermal intramolecular ene reaction 232 thermal isomerization, poly(sec-butylCOT) 374 thermal ring-enlargement 44 thermolysis, o-aryl-metal complexes 104 thia[4$ericyclyne 341 thiacycle 294 thiadiynes, cyclic 289 thienamycin 124 thin films, polymer 371 thioacetate triggering device 273 thiol-independent cleavage mode, singlestranded DNA bulges 211 thiophenophanes 312 third-order nonlinear optical properties, acetylide metal complex 88 three-component cycloaddition, acetylene 154 three-component cyclocoupling 113 three-dimensional phenylacetylene nanoscaffolding 430 through-bond interactions 301,334 through-space interactions 301 f, 334 through-space orbital interactions 327 tin-acetylene 68 tin substitution, carbocation P-stabilization 72 tolane 417 topochemically controlled solid-state polymerization 389, 396 topologigal control, diradicals/dicarbenes 398 f transannular addition, HzO,cyclic diyne 305 transannular C-C bond formation 305 f
497
transannular distance 291 f, 302 transannular electronic interactions 308 transannular interactions 301 transannular reactions, cyclic alkynes 305 f transfer, hydrogen 4 transition metal, perethynylated 455 transition metal catalyzed construction, neocarzinostatin chromophore 231 transition metal catalyzed cyclomerization 403 transition metal complex route 448 transition metal complexes, cr-acetylide 88 transition metal polymerization catalyst, late 366 transition metals - paramagnetic 387 f - perethynylated 455 transition structure 4 transport properties 362 trapping, intramolecular nucleophilic 105 trialkylaluminum reagents 194 triaryloxytantalum alkylidene 362 triazene functional group 423 triazene linkage, propylaminomethylated polystyrene 440 tribenzocyclyne 106 1-(tributylstanny1)-1-pentyne 144 tri-fert-butylazete 179 tri-t-butyl-triphospha[6]pericyclyne 341 trichloroacetimidate methodology, Schmidt 261 1,2,3-tricyanobenzene 41 1,2,4-tricyanobenzene 40,46 1,3,5-tricyanobenzene 46 tricyclic cyclobutadiene complexes 311 tricyclic cycloheptadienes 152 tricyclic cyclophanes, fully homoconjugated cages 346 tricycl0(5.3.0.0?~~)decenone ring system 157 tricyclopropabenzene 321 tridendron 422 - SYNDROME 424 1-(triethylsily1)-1-propyne 361 1,3,5-trifluorobenzene 50 1,3,5-tri-(2-furoyl)benzene 403 trigger 215, 224 - p-sulfone 270 - allylic trisulfide 243 - cycloaromatization 220 - trisulfide 261
498
Index
triggering - Bergman cyclization/pinacol rearrangement cascade 263 - bicyclic enediyne core, calicheamicin 238 - natural products, bioreduction 264 trigonal-based macrocyclic architectures 428 triiodoethylene 56 trimer, acyclic, dimethyl-l,4-pentadiene 332 tri-0-methyl dynemicin esters 271 (1,3)-trimethylsilyl shift 174 trimethylsilylacetylene 419 trimethylsilylacetylene, Friedel-Crafts-like alkylation 324 triphospha-Dewar-benzenes 190 f triphosphametallahomobenzvalene 194 f triphosphole 184 triple bonds 285 f triple-bond migration 304 triple-bond protection 128 triple-bond protecting groups 128 triplet diradicals 399 triplet state, acetylene 385 triplet vinylidene 5 triquinacene derivatives 163 trisaccharide, esperamicin A1 255 f triscacharide fragment, calicheamicinl esperamicin core 258 trishomobenzene 321 tris-iodonium salt 73 tris(trimethylsily1)phosphane 174 trisulfide trigger 261 trithienocyclotriyne 106 tropone 182 tropone reactions, phosphaalkynes 182 tropylidenophanes 42 Tsuchihashi pinacol rearrangement 242 tumor cell lines 269 tungsten alkylidene 366 tungsten arylcarbene complexes 148 tungsten metathesis catalyst 368 tungsten pentacarbonyl complexes 183 tungsten pentacarbonyl-phosphinide complex 184 tungsten-capped polyene oligomer 368 twist angle, model polyene 372 two-alkyne annulation reaction 146 two-dimensional Huckel molecular orbital theory (HMO) 14 two-step cyclotetramerization 176
Ullman’s nitronyl nitroxide radicals 386,400 UMP2/6-31G level 6 unpaired electron centers, alkynyl compounds 385 unsubstituted polyacetylenes, precursor routes 364 urethane anion chemistry 257 UV matrix-assisted laser desorption/ionization time-of-flight 436 vacuum pyrolysis conditions 175 valence, full, CASSCF calculation 18 valence angle deformation 416 valence angles, phenylacetylene 417 valence-bond theory 2,398 f valence electron correlations 443 valency isoelectronic phosphorus(111) moieties 173 van der Waals complexes 12 vanadocene 388 Vaska’s complex 88 vibrational frequencies 15 vinblastine 122 vincristine 122 vinyl cation 71 o-vinyl 128 vinyl-metal derivatives 107 a-vinyl-substituted propargylium complexes 119 vinylcarbenes 141,146 vinylcopper reagents 80 vinylethynyl carbinol derivatives 119 vinylidene 3 vinylidene-acetylene isomerization energy 3 vinylidene-acetylene rearrangement 3 vinylidene carbene 295 vinylidene complexes 128 vinyliodonium salt 77 vinylketene 141 vinylketene intermediate 149 voltammograms, cyclic 374 wavefunction, multiconfiguration 2 wavefunction, spin-coupled 2 Weiss field 410 wires, molecular 461 f - all-carbon backbones 443 Wittig olefination 245 Wittig rearrangement 233
Index Wittig rearrangement strategy, NCS chromophore core structure 236 Whig-type reaction 363 X-ray crystallography 193 X-ray measurements, cyclic rnonoalkynes 296 p-xylene 43 ynamides, push-pull 80 ynediamines 54
499
Z-allene-ene-yne 225 Z-cascade:benzene(3-1,3,5):(5-ethynyl-1,3phenylene)G:5-ethynyl-l,3-di(terf-butyl) benzene 421 zero-field magnetization 400 zero-field splitting (zfs) parameter 387 zero-point energy correction 4 zethrene 307 Ziegler-Natta catalysts 358 ff - propynes, polymerization 359 Ziegler-Natta olefin polymerization catalysts 114 Ziegler-Natta polymerization 358 - alkynes 358 zipper reaction 306 zirconadiphosphacyclopentadiene 186 Zr-promoted bicyclization, eneynes 113
This Page Intentionally Left Blank
Author Index
Adamowitz L. 18 Adams, R.D. 104 Almlof, J. 20,22 Bach, R.D. 7,8 Baeyer, A. 13,461 Baker, J. 6 Balasubramanian, K. 404 Bartlett, R.J. 18, 21, 23 Barton, T.J. 286 Bauschlicher, C.W. 18 Beau, J.-M. 257 Becker, G. 174 Bennett, M.A. 104 Bergman, R.G. 9,205,207,225,240 Bernardi, E 9 Bernholc, J. 16 Bernhold, D.E. 19 Bianchini, C. 107,109,111 Binkley, J.S. 4,17,20 Boger, D.L. 273,274 Borden, W. 7,11 Botschwina, P 13 Brandsma, L. 34,53 Briickner, R. 230 Cadiot, F! 57, 462 Calkins, TL. 206 Chen,Y 3 Chien, J.W.C. 359 Chin, D.-H. 209,211 Chodkiewicz, W. 57, 462 Choi,KX 363 Ciganek, E. 38 Clementi, E. 15 Cooney, M.J. 287 Cotton, FA. 101 Coxon, J.M. 7 Cremer, D. 9 Dai,L. 366 Dai, W.-M. 219 Dale, J. 304
Danishefsky, S.J. 244-247 Deakyne, C.A. 14 Delavarenne, SY 48 Diederich, E 15, 22,106 Dotz, K.H. 139 Dragovich, ES. 241 Dykstra, C.E. 4 Edo,K. 208 Eglington, G. 445,462 Ervin, K.M. 3 Eschenmoser, A. 293 Ewing, D.W. 19 Fallis, A.G. 207 Fan, Q. 15 Feldman, K.S. 455 Feyereisen, M. 17 Fortt, S.M. 271 Fowler, EW. 462 Franck-Neumann, E 34,35 Frechet, J.M.J. 420 Ganesh, P 126 Geoffroy, G.L. 114 Glaser, C. 445, 461, 462 Goldberg, LH. 209,211 Gonzales, J. 8 Gready, J.E. 14 Grisson, J.W. 206 Grove, D.G. 122 Grubbs, R.H. 362,368,369 Grunwell, J.R. 305 Hafelinger, G. 13 Halvick, I! 4 Hamilton, T.P 6 Hanoaka, M. 125 Harcourt, R.D. 8 Hauptmann, H. 449 Hawker, C.J. 420 Hehre, W.J. 13
502
Author Index
Hirama, M. 208,230-234,273,274 Hoffmann, D.M. 101 Hoffmann, R. 7,14,15,22,117 Hofmann, E 141 Hopf,H. 449 Hori, Y. 449 Houk, K.N. 7 , l l Hubel, W. 99 Hutter, J. 18, 20, 23 Ishida, N. 208 Isobe,M. 271 Itoh, T 358 Iwamura, H. 390,397,404 Jacobi, PA. 124 Jaouen, G. 122,128 Kadakov, EB. 2 Kadow, J.E 243 Kaneda,E 308 Kende, AS. 242 Khand, 1. 85,139 Khane, D. 255,258 Kim, K.D. 206,207 Klavetter, EL. 369 Klemperer, W. 5 Kloster-Jensen, E. 49,59 Knoll, K. 356,368 Kollmar, H. 14 Konishi, M. 216,217 Kraemer, W.l? 18 Kraka, E. 9 Krebs, A. 231,309 Kurita, N. 17 Kurtz, J. 18 Lalezari, I. 294 Langhoff, S.R. 18 Lee, M.D. 212 Liang, C. 18, 24 Lineberger, W.C. 6 Lisy, J.M. 5 Little, R.D. 8 Lown, J.W. 235 Luthi. H.E 20
Magnus, P 232,235,239,242,244,271 Magnus, R. 122,125 Magriotis, PA. 206, 207 Maier M.E. 243 Marshall, A. 126 Martin, J.M.L. 17,19 Mash, E.A. 257 Masuda, T 360 Matsuda, K. 404 Mayer, J. 205 Mayrhofer, R.C. 4 McConnell, H.M. 392 McDouall, J.J.W. 9 McKee, M.L. 10 McMahon, R.J. 389 Meier, H. 294,297 de Meijere, A. 334 Melikyan, G.G. 105,127 Merrifield, R.B. 416 Mikami, K. 232 Miller, J.S. 400 Miller, T.M. 420 Misumi, S. 308 Moffat, J.B. 14 Moore, W.R. 303 Muetterties, E.L. 109 Myers, A.G. 209,224,225,226,235,241 Naarman H. 358 Nazarov, LN. 85 Neenan, T.X. 420 Negeshi, E.-I. 113 Newkome, G.R. 420 Nicholas, K.M. 105,126,127, 242 Nicolaides, A. 7 , l l Nicolaou, K.C. 203-283,289,294, 307 Nishide, H. 361 Nixon, J.E 177 Nozaki,H. 58 Nuss, J.M. 230, 234 Ogawa,T 308 Ovchinnikov, A.A.
00 f
Parasuk, V. 19,20, 22, 23 Patai, S. 33 Pauson, EL. 85,139 Petasis, N.A. 230,234
Author Index
Pettit, R. 128 Pfeiffer, G.V. 15, 19 Phillips, J.C. 16 Pitterna, T. 232,235 Pitzer, K.S. 15 Politzer, I! 14 Pople, J.A. 4,13 Raghavachari, K. 17,20 Ramasesha, S. 6 Ranganathan, S. 14 Raphael, R.A. 33 Rees, D.C. 209 Regen, S.L. 289 Regitz, M. 176 Rieker, A. 385 f Roth, K.-D. 126 Roush, W.R. 126 Rubin, Y 443,449 Rutledge, TE 33 Sakurai, H. 344 SasakiT 403 Schaefer, H.E 18,24 Schafer, H.E 4,6 Scharf, H.-D. 257 Schlenk, W. 399 Schleyer, h.R. 11 Schmidt, R.R. 261 Schreiber, S.L. 126,215,242,243,271 Schrock, R.R. 356,359,362,363,368 Scott, LX 287 Scuseria, G.E. 18 Semmelhack, M.E 241 Seto, M. 366 Seyferth, D. 128 Shibuya, M. 235 Shirakawa, H. 353,358 Sibert, E.L. 4 Sinha, B. 6 Sixl, H. 396 Slanina, Z. 16,19 Smith, C.A. 242 Snyder, J.P 9,239 Sondheimer, E 205,206,308,443 Stanbury, D.M. 10 Stuart, J.G. 127 Suffert, J. 230,231
Suzuki,Y 403 Swager TM. 368 Takahashi, T 233 Takeda, K. 396 Tanaka, M. 374 Tao, D. 368 Tatsuata, K. 233 Teets, K.A. 230,234 Terashima, S. 228 Thiele, G I 399 Tomalia, D.A. 420 Tomioka, K. 243 Toshima, K. 233,235,213,274 Townsend, C. A. 216 Trost, B.M. 113 Van Zee, R.J. 15 Viehe, H.G. 48 Voionkov, M.G. 344 Vollhardt, K.W. 112,124, 453 von Helden, G. 20 Wagener, K.B. 368 Wallace, K.C. 362 Walton, D.R.M. 13, 461 Wang,H. 11 Ward, H.R. 303 Watts, J.D. 18,21,23 Wegner, G. 396 Weltner, W.Jr. 15 Wender, PA. 231,233,235,236,270,271 Werner, H. 117 White, J.W. 366 Wigley, D.E. 111 Wojcicki, A. 115 Wong, H.N.C. 206 Woodward. R.B. I Yamazaki, H.
111
Zahradnik, R. 16 Zein, N. 221f Zenneck, U. 177 Zhou, J. 273
503
This Page Intentionally Left Blank
Experimental Procedures
B-alkylethynyl(pheny1)iodonium triflates 93 alkynyl(pheny1)iodonium salts 94 aminomethylated polystyrene 439 48-cascade: benzene(3-1,3,5):(5-ethynyl-l, 3-phenylene)G:5-ethynyl-l,3- di(tert-buty1)benzene 437 bicyc~o(6.1.0)non-1(8)-en-9-one 315 bis(~'-cyclopentadienyl)(2,4-di-tert-butyl-1,3-diphospha- bicyclo[l.l.0]butan-2,4-diyl)zirconium 197 bis( phenyl( [(trifluoromethyl)suIfonyl]oxy 1iodo)ethyne 94 bisiodonium diyne bis-triflates 93 bromoalkyne from terminal alkynes 347 tert-butylphosphaacetylene 196 chloroacetylene 61 chloromethylated polystyrene 439 (cyano([(trif luorornethyl)sulfonylloxy)iodo)benzene 92 cyanoacetylene 59 cyclobutene-fused dehydroannulenes, oxidative Hay coupling 465 1,7-cyclododecadiyne 314 cyclonon-2-ynone 315 cyclopentenones 94 1,8-cyclotetradecadiyne 314 dendritic starbust dodecaketone 410 Dewar benzenes 315 dialkynes, cyclic 314 1-(3,5-dibenzoylbenzoy1)-3-(2-propynoyl)benzene 410 l-(3,5-dibenzoylbenzoyl)-3-(3-trirnethylsilyl-2-propynoyl)benzene 410 dibromoacetylene 61 3-dibromornethylene-l,5-bis(trirnethylsilyl)-l,4-pentadiyne 465 2,2-dibromovinyl to brornoalkyne conversion 349 dichloroacetylene 61 dicyanoacetylene 60 dicyanodiacetylene 60 (E)-1,2-diethynyl-l,2-bis[(triisopropylsilyl)ethynyl]ethene 466 diiodoacetylene 62 dimethyl tetracyclo[7.5.2.0.02'8]hexadeca-2,15-diene-15,16-d~carboxylate 315 dimethyl tetracyclo[l2.2.0.0 '17.08~'4]hexadeca-7,15-diene-15,16-dicarboxylate 315 2,2-dimethyl-l-(trirnethylsiloxy)propylidene(trimethylsilyl)phosphane 196 (2,2-dimethylpropylidyne)phosphane 196 ~-[(~2,~2-dl-3,4-~phe~yl)-l,5-cyclooctadiyne]-bis-hexacarbony~dicob~t 131 1,3-diyne, cross-coupling 347 a,o-diyne, long-chain, oxidative cyclization 348 expanded radialenes, Eglinton-Glaser coupling 466 P-functionalized ethynyl(pheny1)iodonium trif lates 93 cidtrans isomerization, soluble polycyclooctatetraenes 377 y-lactams 94 2-(1-methyl-2-propynyl)cyclohexanone 131
506
Experimental Pocedures
~-[(~2,~2-1-methyl-2-propynylium)dicobalthexacarbonyl] tetrafluoroborate 130 oligomeric sequence liberation from support 439 Pd(0)-catalyzed coupling reaction, trimethylsilylacetylene 438 p-phenylethynyl(pheny1)iodonium triflates 93 polyacetylene from poly(diethyl7-oxabicyclo[2.2.l]hepta-2,5-diene-2,3-dicarboxylate) - solid-state production 378 - solution production 378 poly(diethyl7-oxabicyclo[2.2.l]hepta-2,S-diene-2,3-dicarboxylate) 377 precursor route to polyacetylene 377 propylaminomethylatedpolystyrene 439 Shirakawa polyacetylene 376 solution-spray flash vacuum pyrolysis (SS-FVP) 446 f substituted polycyclooctatetraenes 376 terminal acetylenes from methylketone 345 tertiary propargylic chloride/terminal acetylene coupling 348 2,5,6,8-tetra-tert-butyl-l,3,4,7-tetraphosphatetracyclo[3.3,0.02~4.0'~6~oct-7-ene198 2,4,6,8-tetra-tert-butyl-1,3,5,7tetraphosphapentacyclo[4.2.O.O2~s.O3~~.O4~7loctane 197 2,5,7,9-tetra-tert-butyl-3,3,4-triethyl-4-aluminato-3,6,8-triphospha-lphosphoniatetracyclo[4.2.1.0'~s.04~9]~~tane 198 2,4,6-tri-tert-butyl-1,5-diphospha-3phosphoniaspiro[3.4]hexa-1,4- diene-6-trichloroaluminate 197 1,4,6-tri-tert-butyl-2,5,7,7,8,8-hexaethyl-5,8-dialuminato-3-phospha2,7-diphosphoniatetracyclo[3.3.O.Oz~4.O3~6]octane198 triazene linkage to propylaminomethylatedpolystyrene 440 3,4-bis-[(triisopropyIsilyl)ethynyl]-3-cyclobutene-l,2-dione 464 trimethylsilyl deprotection 438 1,3,5-tris[3-(3,5-dibenzoyl)]benzene 410
