Stereoselective Synthesis: A Practical Approach

Mihiily N6griidi

Stereoselective Synthesis

OVCH Verlagsgesellschaft mbH, D-69451 Weinheim (Federal Republic of Germany), 1995 Distribution: VCH, P.O. Box 10 2 1 61, D-6945 1 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 CBI 1HZ (England) USA and Canada: VCH, 220 East 23rd Street, New York, NY 10010-4606 (USA) Japan: VCH, Eikow Building, 10-9 Hongo I-chome, Bunkyo-ku, Tokyo 113 (Japan) ISBN 3-521-29243-8

Mihaly N6grAdi

Stereoselective Synthesis A Practical Approach Foreword by A. I. Meyers Second, Thoroughly Revised and Updated Edition

VCH

-

Weinheim New York Base1 Cambridge Tokyo

Prof. Dr. Mihaly Nogradi Institute of Organic Chemistry Technical University XI., Gellert ter 4 H-1521 Budapest

This book was carefully produced. Nevertheless, author and publisher do not warrant the information contained therein to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

1st edition 1987 2nd edition 1995

Published jointly by VCH Verlagsgesellschaft mbH, Weinheim (Federal Republic of Germany) VCH Publishers, Inc., New York, NY (USA)

Editorial Directors: Dr. Ute Anton, Dr. Thomas Mager Assistant Editor: Eva Schweikart Production Manager: Dip1.-Ing. (FH) Hans Jorg Maier Library of Congress Card No. applied for.

British Library Cataloguing-in-Publication Data: A catalogue record for this book is available from the British Library.

Deutsche Bibliothek Cataloguing-in-Publication Data: Nogradi, Mihaly: Stereoselective synthesis: a practical approach / Mihaly Nogradi. 2., rev. and enl. ed. - Weinheim; New York: VCH, 1995 ISBN 3-527-29243-8 (Weinheim) brosch. ISBN 3-527-29242-X (Weinheim) Pp. ISBN 1-56081-895-6 (New York) Pp.

-

0 VCH 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 in 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 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: K+V Fotosatz GmbH, Beerfelden. Printing: betz druck gmbh, D-64291 Darmstadt. Bookbinding: Wilh. Osswald + Co., Gronbuchbinderei, D-67433 Neustadt. Printed in the Federal Republic of Germany.

Foreword

The quest for selectivity in synthesis still continues as the most important focal area for chemists, and the drive for stereoselectivity is among the most coveted goals. For the past fifty years, organic chemists have pursued the understanding and control of stereochemical behavior, and in 1956 an outstanding monograph appeared entitled “Steric Effects in Organic Chemistry”, edited by M. S. Newman. This work placed stereochemistry, both qualitatively and quantitatively, in its proper position of importance in organic chemistry. Following this, in 1969, Hassel and Barton shared a Nobel Prize for the concepts of molecular conformation, further propelling stereochemistry into the limelight. Again, in 1975, the Nobel Prize was awarded to Cornforth and Prelog for stereochemical reactions, and further intensified efforts to enhance both relative and absolute stereoselectivity. Thus, a large number of papers appeared between 1975 and 1980 dealing with asymmetric syntheses of diastereomers and enantiomers. Many, if not most of these, are based on stereoelectronic effects (e.g., chelation) as the major component to control selectivity in molecular construction. In 1984, the now famous five-volume series, “Asymmetric Synthesis”, was published. This series chronicled the effort in what was then the most intense research area in organic chemistry. What was a mechanistic curiosity in the 196Os, has today become one of the most valuable techniques of organic chemistry - the ability to perform enantioselective syntheses in the laboratory. This long sought-after goal of organic chemists has now reached the level where organic compounds of virtually every type can be obtained in complete enantioselective or diastereoselective form. How did this come about in the short period of 20-25 years? The contributions of many brilliant investigators from all parts of the world were key to these successes. The clearer understanding of reaction mechanism, kinetics, and solution and solvent effects all contributed to the phenomenal leaps forward in the rational planning of stereoselective syntheses. Thus, it would be unjust to celebrate the advances in stereochemical synthesis without paying due homage to those investigators who helped build a sound foundation in reaction theory. For without this knowledge, many rational routes to absolute and relative stereochemistry would not have been possible. Furthermore, one must not ignore the multitude of analytical tools (e.g., HPLC, NMR) which were also invaluable in allowing researchers to assess the level of their stereochemical control.

VI

Fore word

The many synthetic reactions that Professor Nogradi has laid out, in this and the previous edition of this book, are truly a tribute to the excellent work of the modern organic chemist. The road has now been cleared for future generations to address more complex problems in biology and materials science by using the tools developed in these recent synthetic studies. Fort Collins, Colorado, July 1994

A.I. Meyers Professor of Chemistry Colorado State University

Preface to the Second Edition

The favorable acceptance of the first edition of this book and the unabated flow of important publications on stereoselective synthesis prompted us to publish a thoroughly revised second edition. Over the past years, developments in stereoselective synthesis have been impressive. Just in the past ten years, the number of publications devoted to this topic has much surpassed the total number of all papers written from Emil Fischer’s time until 1984. In addition, two new journals devoted in part to our topic, Tetrahedron Asymmetry (Pergamon Press) and Asymmetry (Plenum Press), have been launched. Also some important qualitative changes can be perceived: (i) catalytic variants have been developed for almost all types of stereoselective reactions and (ii) the number of serious attempts to clarify the mechanistic aspects of stereoselectivity increased, although the share of such studies is still negligible, compared with the total number of publications. The dramatic development of stereoselective methods is well illustrated by the following statistics about this book: Abstracts were prepared from 1400 papers; from these 831 were selected for citation in the new edition. In turn, 350 references were omitted from the first edition, giving a total of 1874 references for the second one. Literature in this edition is covered up to December 1992. Rewriting the book was not easy, due not only to the formidable amount of information to be covered, but also due to our firm resolution to keep the price of the new edition within the reach of individual buyers. This required a rigorous adherence to the standards of selection explained in the Preface to the First Edition, and, in addition, sizeable parts of the first edition had to be deleted. This was a very painful exercise, and I felt guilty each time I had to cut out a method, which was pioneering in its own time, but has since been superseded by more selective ones. Now some technicalities should be mentioned: (i) In order to avoid repetition of chiral moieties within the same equation, they are separated from the part of the molecule undergoing transformation by a wavy line and symbolized later by R *. (ii) To save space, usually only the major product is depicted, even when more than one minor product is also formed. (iii) In order to be able to compare competitive methods, when enantiomerically impure substrates or chiral auxiliaries were used, ee values were recalculated for 100% enantiomeric purity. (iv) For methods that have been optimized, the best conditions are quoted. (v) When the optical purity of a product was upgraded by crystallization or chromatography, selectivity values for the crude product are quoted. (vii) When both a preliminary and a full paper cover the same subject, only the latter is quoted.

VIII

Preface to the Second Edition

Finally we call attention to the section General References, a rather arbitrary list of important reviews, which does not fit into any single chapter. The author is indebted to Dr. Thomas Mager, Dr. Ute Anton, Eva Schweikart and Dip1.-Ing. Hans-Jorg Maier for bearing all the nuisance of welding together old and new, to Dr. Gy. M. Keseru for drawing the new formulas, and to Mrs. I. Berenyi for assistance in editing. Budapest, September I994

M. Nogradi

Preface to the First Edition

The present work is an attempt to review practical methods of stereoselective synthesis with emphasis on recent advances. It embraces a wide variety of subjects, such as hydrogenations over chiral catalysts, reductions with chiral hydride donors, stereoselective epoxidations, pericyclic reactions and the rapidly expanding field of “acyclic stereoselection”. This is a very broad topic and therefore several restrictions had to be imposed on the subjects to be covered. First of all, enzymatic transformations were omitted because these have been extensively reviewed [I]. Neither are we going to discuss stereochemical aspects of reaction mechanisms in general. There is an almost endless number of examples in the literature for syntheses starting from an optically active compound, which are carried through a number of more or less selective transformations, to end up with a product in which the original stereogenic element is retained. No coverage will be given to such syntheses since this would have required the inclusion of such immense fields as transformations of steroids, terpenoids, carbohydrates and the like. Although there is no specific date from which the literature has been processed, earlier methods giving poor stereochemical yields will not be discussed, unless they served as a starting point for more efficient processes. Most of these methods have been amply described in earlier works [2, 31. The hectic activity in the field of stereoselective synthesis precludes a comprehensive treatment of even the literature of the last 15 years. Therefore, methods with low stereochemical yields and the application of efficient methods to molecules of high complexity will generally be omitted. Again for reasons of space, this book is somewhat biased in favor of methods falling under the rather ill-defined term “asymmetric synthesis”. Just by their sheer number diastereoselective methods could not have been covered with any claim to comprehensiveness. On the other hand, it would have been rather controversial not to deal at all with diastereoselectivity, since, as will be apparent from the introductory chapter, the underlying phenomenon of enantioselectivity is in fact diastereoselectivity. In summary, we wish to serve the practically minded, synthetic organic chemist rather than the theoretician. The author was induced to write this book, not only by the extremely rapid advancement and fascination of this field, but also by his conviction that the development of stereoselective synthetic methods has reached a turning point from which

x

Preface to the First Edifion

applications to practical problems have become a realistic proposition. Synthesis of natural amino acids, of non-racemic pharmaceuticals with fewer side-effects, of prostanoids and steroids, of insect hormones and pheromones are only the most rewarding fields in which such methods are of key importance. The literature has been reviewed up to December 1984. The author is indebted to Dr. C. Dyllick-Brenzinger for carefully revising the manuscript and to Mr. T. Goschi for his help in preparing the figures. Budapest, August 1986

M. Nogradi

[I] J. Rktey and J. A. Robinson, Stereospecgicity in Organic Chemistry and Enzymology. Verlag Chemie, Weinheim, 1982 and refs. 16 through 20 in Chapter 1. [2] J. D. Morrison and H. S. Mosher, Asymmetric Organic Reactions. Prentice-Hall, Englewood Cliffs, N. J., 1971. [3] Y. Izumi and A. Tai, Stereodifferentiating Reactions. Academic Press, New York, 1917.

Contents

List of Symbols and Abbreviations Introduction XVII

XV

1

General Concepts of Stereoselective Synthesis

1.1 1.2 1.3 I .4 1.4.1 1.4.1.1 1.4.1.2 1.4.2 I .4.2.1 1.4.2.2 1.4.2.3 1.4.2.4 1.4.3 1.4.3.1 1.4.3.2 1.5

Principles of Differentiating Molecules 1 Characterization of Stereoisomers. Conformation and Configuration 4 Intramolecular Symmetry. Topicity and Prochirality 7 Selectivity in Chemistry 11 Substrate Selectivity 11 Substrate Diastereoselectivity (Diastereomer Selectivity) 12 Substrate Enantioselectivity (Enantiomer Selectivity). Kinetic Resolution 13 Product Selectivity 17 Formation of Stereoisomers 18 Conditions Necessary for Stereoselectivity 23 Concept of Stereodifferentiation 27 Methods for Inducing Stereoselectivity 27 Stereoselective Synthetic Strategies 36 Enantioconvergent Synthesis 36 Selective Preparation of Both Enantiomers from a Single Substrate 39 Kinetics and Thermodynamics of Stereoselective Reactions 41

2

Stereoselective Catalytic Reductions

2.1

Stereoselective Homogeneous Hydrogenations with Rhodium-, Ruthenium- and Iridium-Phosphine Catalysts 45 Hydrogenation of Olefinic Bonds 48 Chiral Phosphorus Containing Ligands 48 Enantioselective Catalytic Hydrogenation of Ketones and Imines 66 Diastereoselective Hydrogenations in Homogeneous Phase 68 Catalytic Hydrosilylation 70

2.1.1 2.1.1.1 2.1.2 2.1.3 2.2

1

45

2.3 2.3.1 2.3.2

Heterogeneous Stereoselective Catalytic Hydrogenations 76 Enantioselective Heterogeneous Catalytic Hydrogenations 77 Diastereoselective Heterogeneous Catalytic Hydrogenations 78

3

Stereoselective Non-Catalytic Reductions

3.1 3.1.1 3.1.1.1 3.1.1.2 3.1.2 3.1.3 3.1.4 3.1.4.1 3.1.4.2 3.1.4.3 3.1.4.4 3.1.5 3.2 3.2.1 3.2.2

Enantioselective Reductions 82 Chiral Lithium Aluminum Hydrides 81 Reduction of Carbonyl Groups 81 Reduction of C = N and C = C Bonds 88 Chirally Modified Hydridoborates 88 Chiral Boranes and Boronates 90 Enantioselective Reductions with Hydride Transfer from Carbon 92 Chiral Trialkylboranes 92 Chiral Metal Alkyls 94 Chiral Metal Alkoxides 95 Chiral 1,4-Dihydropyridines 95 Correlation of Substrate Constitution and Enantioselectivity 96 Diastereoselective Reductions of Carbonyl Groups 100 Stereochemistry of Diastereoselective Ketone Reductions 101 Practical Aspects of Diastereoselective Ketone Reductions 110

4

Stereoselective Oxidations

4.1 4.1.1 4.1.2 4.2 4.2.1 4.2.2

Enantioselective Oxidations 123 Epoxidation with Chiral Oxidants 123 Oxidations in the Presence of Chiral Catalysts Diastereoselective Oxidations 130 Diastereoselective Epoxidation 130 Various Diastereoselective Oxidations 133

5

Stereoselective Carbon-Carbon Bond Forming Reactions by Nucleophilic Addition to Carbonyl Groups 135

5.1 5.1.1 5.1.2 5.1.3 5.2 5.2.1 5.2.2

Addition of Simple Nucleophiles to Carbonyl Compounds 136 Enantioselective Additions 136 Diastereoselective Additions 139 Additions to C = N Bonds 154 Additions Involving Allylmetal and Allylboron Compounds 157 General Aspects of Allylmetal Addition 157 Addition of Allylboron Compounds 162

81

123

124

Contents

XI11

5.3.5 5.3.6 5.3.7 5.4

Addition of Allyltitanium Compounds 167 Addition of Allylsilanes 170 Addition of Allylstannanes 173 Addition of Other Allylmetals 175 Stereoselective Aldol Reactions 177 Stereochemistry and Mechanism of the Aldol Reaction 178 Generation of Enolates and their Addition to Achiral Aldehydes and Ketones 184 Addition of Enolates to Ketones (the Cross-Aldol Reaction) 197 Addition of Achiral Enolates to Achiral Aldehydes Under Chiral Catalysis 198 Addition of Chiral Enolates to Achiral Aldehydes 200 Addition of Achiral Enolates to Chiral Aldehydes 205 Stereoselective Aldol Condensations 209 Catalytic Addition of Isonitriles to Aldehydes 210

6

Stereoselective Carbon-Carbon Bond Forming Reactions 213

6.1 6.1.1 6.1.1.1 6.1.1.2 6.1.1.3

Carbon-Carbon Bond Formation Involving Olefins 213 Addition of Carbanions to Activated Olefins 213 The Michael Addition 213 Addition of Metalorganics to a,PUnsaturated Ketones and Esters 21 8 Addition of Metalorganics to cr,PUnsaturated Azomethines and 2-Vinyloxazolines 224 Allylic Alkylation 226 Stereoselective Hydrocarbonylation 229 Stereoselective Alkylations 230 Alkylation of Enolates 230 Alkylation of Enolates with a Removable Chiral Auxiliary Group 231 Diastereoselective Alkylation of Enolates 240 Alkylation of Stabilized Carbanions 241 Alkylation of Enamines and Metallated Azomethines 244 Alkylation of Metallated Hydrazones and Amidines 246 Enantioselective Grignard Cross-Coupling and Other Stereoselective Catalytic Carbon-Carbon Bond Forming Reactions 249 Grignard Cross-Coupling 249 Stereoselective Codimerization of Olefins 25 1 Miscellaneous Stereoselective Carbon-Carbon Bond Forming Reactions 252

5.2.3 5.2.4 5.2.5 5.2.6 5.3 5.3.1 5.3.2 5.3.3 5.3.4

6.1.2 6.1.3 6.2 6.2.1 6.2.1.1 6.2.1.2 6.2.2 6.2.3 6.2.4 6.3 6.3.1 6.3.2 6.4

XIV

Contents

7

Stereoselective Carbon-Carbon Bond Formation by Pericyclic Reactions 257

7.1 7.1.1 7.1.2 7.1.3 7.1.4 7.2 7.2.1 7.2.2 7.2.3

Asymmetric Induction in Cycloadditions 257 Asymmetric Induction in the Diels-Alder Reaction 258 Asymmetric Induction in [3 +2] Cycloadditions 268 Asymmetric Induction in [2 + 21 Cycloadditions 269 Stereoselective Carbene Additions 273 Chirality Transfer in Sigmatropic Rearrangements 277 [3,3] Sigmatropic Rearrangements 278 [2,3] Sigmatropic Rearrangements 283 Ene Reactions 287

8

Stereoselective Formation of Carbon-Heteroatom Bonds 293

8.1 8.2 8.3 8.4 8.5 8.6

Stereoselective Stereoselective Stereoselective Stereoselective Stereoselective Stereoselective

Formation of Carbon-Nitrogen Bonds 293 Formation of Carbon-Phosphorus Bonds 296 Formation of Carbon-Oxygen Bonds 297 Formation of Carbon-Sulfur Bonds 303 Formation of Carbon-Halogen Bonds 305 Protonation and Hydrogen Migration 306

Appendix 311 References 3 15 General References 357 Subject Index 359

List of Symbols and Abbreviations

Groups Ac acac BBN Bn Bu Bz cHex CP cPent Et Hept Hex iBu iPr M Me MEM (-)Ment Oct Pent Ph Pr sBu tBu Tf To1 Tos

acetyl 2,4-pentanedione 9-borabicyclo[3.3. llnonyl benzyl 1-butyl benzoyl cyclohexyl cyclopentadienyl cyclopentyl ethyl 1-heptyl I-hexyl 2-methylpropyl 2-propyl metal methyl (2-methoxyethoxy)methyl (- )menthy1 1-octyl 1-pentyl phenyl 1-propyl 2-butyl 1,l-dimethyl-I-ethyl trifluoromethylsulfonyl p-tolyl p-toluenesul fonyl

XVI

List of Symbols and Abbreviations

Reagents and solvents COD DBU DME DMPU HMDS HMPA LDA LAH MCPBA MEM MS NBD NMO PCC PhMe THF TMEDA

1,5-cyclooctadiene 1,5-diazabicyclo[5.4.01 undec-5-ene 1,2-dimethoxyethane 1,3-dimethyl-2-oxohexahydropyrimidine hexamethyldisilazane hexamethylphosphoric amide lithium diisopropylamide lithium aluminum hydride metachloroperbenzoic acid (2-methoxyethoxy)-methyl molecular sieve norbornadiene N-methylmorpholine oxide pyridinium chlorochromate toluene tetrahydro furan

N,N’-Tetramethylethylenediamine

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

1 General Concepts of Stereoselective Synthesis

Although the present book is primarily oriented towards the practical aspects of stereoselective synthesis, it is necessary to describe briefly the basic principles of stereoisomerism, chemical selectivity in general and stereoselectivity in particular. Also, it is important to define the nomenclature and the system of notation to be used. The following introductory sections, however, should not be regarded as stereochemistry in a nutshell since only aspects important for our topic will be discussed.

1.1 Principles of Differentiating Molecules The two main objectives of chemistry are the analysis and synthesis of molecules. The analysis of molecules is a rather abstract task, since there is no obvious, easily recognizable correlation between the outer appearance of a chemical substance and its internal properties which we generally call chemical structure. Differences in the structures of molecules are manifold, and it is possible to define a hierarchy of characteristics by which molecules can be distinguished. As we go down this ladder of hierarchic characteristics, molecules become more and more similar until we reach complete identity. Molecules which are identical in terms of higher ranking features may be distinguished by lower ranking ones. (i) Molecules can differ in their qualitative composition, i.e. by the nature of elements they contain. Potassium carbonate and sodium carbonate, though both colorless crystalline solids, are of different qualitative composition. (ii) Molecules of identical qualitative composition may differ in their quantitative composition i.e. by the ratio of the different elements they contain. Carbon monoxide and carbon dioxide, e.g., differ in this respect. (iii) Compounds of identical qualitative and quantitative composition may differ in their molecular weight. Acetylene, benzene, and cyclooctatetraene are examples for such a relationship.

2

1.1 Principles of Differentiating Molecules

(iv) Molecules which are found to be identical by criteria (i) - (iii), may be different due to the different connectedness of their atoms. Here we enter the domain of isomerism, and molecules which only differ by the sequence of their atoms are called constitutional isomers. Constitutional isomers, and of course the constitution of a single molecular species, can be fully characterized by enumerating each of their atoms and stating the nature and number of all the atoms connected to each particular atom by chemical bonds. It should be noted that the constitution of a molecule can always be adequately characterized without using words denoting directions such as “under” or “over”, “left” or “right”. Molecules which are found to be identical by criteria (i) - (iv), but are nevertheless distinguishable are sfereoisomers. Stereoisomers occupy two steps on our scale of differentiation. (v) Diastereomers are molecules of identical constitution but which can be differentiated by some scalar property, the most important being internuclear distances of a selected pair of groups (atoms) or in complex cases by the distances of several such pairs. The following examples serve to illustrate how we can characterize diastereomers by internuclear distances. Thus diastereomers of 1,2-dibromoethene differ by the internuclear distance of the two bromo atoms. Conventional prefixes attached to names describing constitution enables one to identify diastereomers without taking recourse to formulas.

’ic=c7

H/x

(Z)-1,2-dibromoethene

Br

\

H

/

c=c

H

‘Br

(E)-1,2-dibromoethene

The prefix Z in the above formula means that the distance between the selected pair of groups in this diastereomer is smaller than that in the E isomer. Rules for assigning the above and other conventional prefixes have been agreed upon by international conventions called the IUPAC Rules of Nomenclature. Most important from our point of view are the “Rules of Stereochemical Nomenclature” [I]. Any pair of diastereomers, however complex their constitution should be, can be adequately characterized by a set of statements referring to internuclear distances. Differences in internuclear distances serve not only for the identification of diastereomers, more importantly they form the basis of differences in their physical and chemical characters. (vi) Enantiomers are pairs of stereoisomers with the highest level of similarity. If their formulas are written down according to the same convention, internuclear distances for any given pair of atoms are identical. Enantiomers can, however, be distinguished by stating the sequence of selected groups following a certain convention. The conventional character of distinguishing enantiomers must be emphasized, because words such as clockwise - anticlockwise or right-handed - left-handed are

1.1 Principles of Diyferentiating Molecules

3

meaningless in themselves and come to life only by a world-wide agreement about their significance. A typical statement describing the difference between dextrorotatory and levorotatory lactic acid is the following: when their formulas are depicted according to the same convention (hydrogen remote from the viewer) the sequence of the groups hydroxy, carboxy and methyl is anticlockwise for the dextrorotatory and clockwise for the levorotatory enantiomer.

(-)-(A)-lactic acid

(+)-(S)-lactic acid

A system of nomenclature for the unambiguous characterization and distinction of enantiomers by pairs of simple prefixes ( R and S, P and M ) has been worked out by R. S. Cahn, C. K. Ingold, and V. Prelog (the S.C.C.I.P. convention) [2]. When represented following the same conventions, formulas of diastereomers are not mirror-images, while those of enantiomers are mirror-images. An object (e.g. a molecule) which is not identical with its mirror image is called chiral, otherwise it is achiral. Molecules forming enantiomers are chiral by definition, while chirality is not a condition for a diastereomeric relationship. Thus the diastereomeric 1,2-dibromoethenes are both achiral. (-)-Tartaric acid and meso-tartaric acid are diastereomers, the former is chiral, the latter is achiral. The last stage in our molecular identity-non-identity hierarchy is complete identity, which we are not interested in. Note that identity is a concept dependent on the depth of our insight. Thus molecules which we regard in our discussions as being identical may differ in their isotopic composition, electronic or nuclear quantum levels etc.

4

1.2 Characterization of Stereoisomers. Conformation and Configurafion

1.2 Characterization of Stereoisomers. Conformation and Configuration Molecules can be characterized by a set of geometrical parameters. These are the van der Waals radii of the individual atoms (relevant to the concept of steric hindrance), equilibrium bond lengths between directly bonded atoms, equilibrium bond angles formed by the bonds of two atoms bonded to a common third atom and, finally, torsional angles describing the spatial relationship of the terminal atoms in a linear chain of four atoms. The complete set of all possible torsional angles of a molecule defines its conformation. Certain well characterized conformations are called conformers. For practical purposes we usually disregard the torsional angle of bonds attached to a double bond (which we take as fixed at 0" and 180°, respectively) and those associated with groups rotating very fast, such as the methyl group. Molecular species having different conformations are, by definition, stereoisomers, since they are different entities with the same constitution, although such stereoisomers are usually inseparable due to their rapid interconversion. Examples for stereoisomers which differ in their conformation are the (P)-synclinal and antiperiplanar conformers of n-butane (inseparable) and ( R)- and (S)-2,2'-diiodobiphenyl-6,6'-dicarboxylicacid [(I?)- and ( S ) ( I ) ] (Fig. 1-1) (separable). The relationship of any two conformers may be either diastereomeric (as that of the two n-butanes) or enantiomeric (as that of the two biphenyls).

' : ;y H'

H

Me

HH $

H '

H Me

(P)-synclinal butane

antiperiplanar butane %e, ~c 180"

-

me,Mo-+m0

1

Fig. 1-1. Characterization of conformations.

1.2 Characterization of Stereoisomers. Conformation and Configuration

5

While (apart from signs) Conformation, i.e. describing stereoisomers by a set of torsional angles, is essentially a quantitative approach, configuration characterizes a molecule in a qualitative way. Configuration has a different role when describing diastereomers and enantiomers, but in both cases it essentially covers a system of conventions. Diastereomers may be compared by their relative configuration, i.e. by differences in the intramolecular relationship of selected groups within each diastereomer. An example for such an internal comparison was given in the preceding section, in which the distinguishing features were epitomized as conventional prefixes such as E and 2. Diastereomers which are interconvertible by rotation around a single bond can be described by the pertinent torsional angles or, more conveniently, by conventional names associated with specific ranges of torsional angles, such as synperiplanar (or syn) for O-t30°, synclinal (or gauche) for +60*30", anticlinal for k 120+30" and antiperiplanar for 180"k30" etc. Diastereomers which arise by different combinations of two or more chiral centers can be conveniently labelled by listing the configurational symbols R and S for each center, e.g. 2R, 3R for dextrorotatory and 2 R, 3 S for meso-tartaric acid. In this book, in conformity with the IUPAC rules, the following nomenclature will be used to describe the relative configuration of diastereomers. For geometrical isomers the 2- E notation will be used with some unavoidable exceptions, when, in order to embrace wider groups of compounds, we will be forced to fall back on the cis-trans notation. The prefixes cis and trans are useful to define the relative arrangement of groups attached to rings, for which purpose 2 and E should not be used. The names recommended by IUPAC (occasionally in their abbreviated form) will be applied to conformers generated by rotation around a single bond. Note that syn and anti are also used to describe the stereochemistry of addition and elimination ~31. Considerable confusion was created some years ago in describing the relative configuration of two chiral centers in a linear molecule. Originally, the relative configuration of two groups which are on the same side in a Fischer projection (as in erythrose) was called erythro, while that of those on opposite sides (as in threose) threo. Apart from occasional difficulties in selecting the main chain, this system served well with two chiral centers. The Fischer projection is undoubtedly an unnatural representation, and when the chain is drawn in the more realistic zig-zag form, erythro substituents end up on opposite sides whereas threo ones are on the same side. In Fig. 1-2 the (2R,3R) and the (2R,3S) diastereomers of 3-hydroxy-2methylbutanoic acid are shown in Fischer projection, in the zig-zag conformation and in a simplified representation of the latter. In order to put substituents back into their "customary" relationship, Heathcock suggested in 1981 [4] an inversion of the nomenclature, thereby causing bewildering confusion in the literature. Thus the (2R, 3R)-compound (a) previously called erythro now became threo, while its diastereomer was renamed erythro. Also, Heathcock's suggestion gained rapid and widespread acceptance. The new problem was soon recognized and several sugges-

6

1.2 Characterization of Stereoisomers. Conformation and Configuration

Me H

(a )

Me

COzH

‘C02H I

Me

H

rhreo by the “aldol” (Heathcock) notation

(2R3)

erythro by the classical notation

COzH

(b)

H+

Me

Me

Me H Me+

~ e

OH

UH

COzH

=

HO H

C02 H

Me+ OH

eryrhro by the “aldol” (Heathcock) notation

(2WS)

thhreo by the classical notation

Fig. 1-2. Characterization of relative configuration in aldol-type compounds.

tions for alternative ways to identify relative configuration in acyclic molecules have been devised [5-71. The most comprehensive and consistent among them is the one by Prelog and Seebach [8] (cf. p. 21). To dissociate ourselves from both the old and new usage of erythro and threo, the prefixes syn and anti will be used in the sense shown in the following examples: *

syn

anti

syn,anti

anti, anti

The steric disposition of groups with a geminal hydrogen atom will be indicated by heavy and broken lines, respectively. In the case of quaternary centers or when it is deemed necessary for a better presentation of the situation, traditional wedges will be used as well. Enantiomers can be identified by quoting the sign of their optical rotation, by constructing or drawing a model representing the molecule or by reference to a certain convention. Fortunately nowadays only one system of convention, that of Cahn, Ingold and Prelog, also called the sequence rules, is in use. Its basic principles and application, at least for simple cases, is well known and need not be discussed here [ 2 ] . Well known optically active compounds containing more than one center of chirality, such as menthol, a-pinene, ephedrine etc. can be conveniently identified by their sign of rotation. Even this can be omitted for some compounds, such as the Strychnos alkaloids and steroids, which in nature only occur in one enantiomeric form.

*

This is not a perfect solution either, since IUPAC recommended these words for the characterization of the mode of approach in addition reactions. Later Heathcock turned to the syn-anti notation too [3].

1.3 Intramolecular Symmetry. Topicity and Prochirality

7

1.3 Intramolecular Symmetry. Topicity and Prochirality Topicity

Analysis of molecular symmetry is of fundamental importance for stereochemistry and therefore also for the thorough understanding of stereoselective reactions. While for the discussion of stereoisomerism it is the global symmetry of a molecule which is relevant, from the point of view of stereoselectivity we also have to consider the symmetry relationships of certain subunits of the molecule, namely, those of groups and faces. As a group we define in our context any subunit of a molecule; this can be as simple as a hydrogen atom or as complicated as a monosaccharide unit. Groups may be classified according to a hierarchical scheme similar to that used for molecules. Thus groups, when regarded in isolation, may differ (i) in qualitative composition (e.g. Br and I), (ii) in quantitative composition (e.g. CHO and CO,H), in constitution (e.g. propyl and isopropyl) and (iii) in stereostructure (e.g. bornyl and isobornyl which are diastereomeric or ( R)- and (S)-2-phenylethyl which are enantiomeric *). Two or more groups in a molecule which are identical by the above criteria may have different relationships to each other. (i) Groups can have the same or different connectedness with the rest of the molecule, in other words their constitutional position may be the same or different. Thus in 2,4,6-trinitrotoluene nitro groups in the 2- and 6-position have the same connectedness, while those in 2- and 4-position are constitutionally different. In discussing stereoselectivity we are only interested in identical groups of the same connectedness. Their relationship can be diastereotopic, enantiotopic or homotopic [lo]. (ii) Diastereotopic are groups which cannot be exchanged by any symmetry operation. Since in asymmetric molecules such as 2 in Fig. 1-3 symmetry elements cannot be present by definition, geminal groups in such molecules (set boldface) are always diastereotopic. Similarly to diastereomers, diastereotopic groups can be readily distinguished by their relationships (near-remote) to a reference group, i.e. in scalar terms. Chiral molecules with rotational symmetry (e.g. 3 ) and achiral molecules (e.g. 4 ) may also contain diastereotopic pairs of groups, but the symmetry element(s) must be unrelated to these groups. Thus geminal hydrogens within each CH2 group (but not those at different carbons) of the cyclopentanone 3 (C2 symmetry) are diastereotopic, one being near to, the other remote from the adjacent methyl group. * Note that although often done for convenience, the assignment of R and S descriptors to groups is ambiguous since these are dependent on the nature of the fourth ligand.

8

1.3 Intramolecular Symmetry. Topicily and Prochirality

Note that although 4 has a plane of symmetry, methyl and carboxyl groups resp. which lie in this plane cannot be exchanged by it. Although for conformationally mobile molecules such as 2 a given near-remote relationship between diastereotopic groups is only valid for a certain conformation, the molecular environment surrounding each member of the group is inherently different for each conformation. Although the magnitude of the difference (expressed as some physical parameter, e.g. magnetization) is dependent on conformational equilibria, the mere fact of this difference cannot be eliminated by fast rotation or ring inversion [ I l l .

C2-t

0

(6)

a ( 7)

Fig. 1-3. Molecules with diastereotopic, enantiotopic and homotopic groups.

(iii) Enantiotopic are groups which can be exchanged by a rotation-reflection axis, which is most often a plane (0= S , ) or a center (Ci = S,) of symmetry. Thus enantiotopic groups can only occur in achiral molecules. Enantiotopic are CH2C02Hgroups in citric acid (5) and subunits thereof (CH,, C02H). (iv) Homotopic are groups which can be exchanged by a symmetry axis. It follows that any achiral or chiral molecule which has an axis of symmetry contains at least one set (usually a pair) of homotopic groups. Compounds which contain a set of two and three homotopic hydrogens, respectively, are dichloromethane and chloromethane (Fig. 1-3). Diastereo- and enantiotopic groups are called heterotopic. The terms dia-, enantio- and homotopic express the relationship of one group to another and may therefore change with the partner. Thus, in 6 atoms HA are homotopic and each has two kinds of enantiotopic relationships to atoms HB, which also form a homotopic set. Neither is connected by any symmetry operation to the homotopic pair of atoms Hc which have a different connectedness. In 7 the HA and HB atoms form two enantiotopic sets, while any HA is diastereotopic to any HB.

1.3 Intramolecular Symmetry. Topicity and Prochirality

9

One of the most frequent synthetic operations is the addition of a group to a tricoordinate center to form a tetracoordinate center. The tricoordinate center is usually a double bonded atom, the three valencies of which constitute a plane with two faces. The topicity notation explained above for groups can be conveniently extended to the symmetry relationships of such faces (Fig. 1-4). (i) Diastereotopic are two faces of any molecular plane which is no plane of symmetry and does not contain a coplanar axis of symmetry. Thus faces in asymmetric molecules (e.g. that of both C = 0 and C = C in 8) are always diastereotopic, no matter how fast bond rotation may be. A plane of symmetry perpendicular to the plane to be qualified is not incompatible with diastereotopicity, as can be seen with compound 7 which contains a carbony1 group with diastereotopic faces. (ii) Enantiotopic are two faces of a molecular plane which is at the same time a molecular plane of symmetry but which does not contain a coplanar axis of symmetry. Enantiotopic are the faces of acetaldehyde and of phenyl-methyl-sulfide. (iii) Homotopic are two faces of a molecular plane which contains a coplanar axis of symmetry. Such faces can be found both in achiral molecules, such as acetone, isobutene, and in chiral ones as, e.g., in 3. For conformationally mobile molecules symmetry relationships usually change with conformation. Here a practical standpoint can be adopted and conformational changes much faster than the process investigated should be disregarded. Thus for a low temperature NMR study the methyl groups and the faces of the carbonyl group of 2,2-dimethylcyclohexanone should be regarded as diastereotopic (9),while for a hydride transfer reaction at room temperature the S.C.statistical symmetry of the molecule, i.e. that of its most symmetrical (possibly non-populated) conformer (10, enantiotopic in the present case), should be invoked.

(8 )

Me

+ e o i - tM : e:: : 4 ,

(9 Fig. 1-4. Molecules with diastereotopic and enantiotopic faces.

Me

(10)

10

1.3 Intramolecular Symmetry. Topicity and Prochiraliiy

Prochirality Since, as will be discussed later, transformation of any one group of an enantiotopic pair of groups into another group or addition of a new group to a center with enantiotopic faces gives rise to a chiral compound, enantiotopic groups and faces are called prochiral [I 21. The same is true for diastereotopic groups which d o not coincide with a plane of symmetry and for diastereotopic faces which have no perpendicular plane of symmetry. By application of the sequence rules, the labels p r o 3 and pro-S are given to geminal prochiral groups and prochiral faces. This is very convenient because it avoids referring to a drawing. According to Prelog and Helmchen [ 131, a prochiral group is called p r o 3 (or briefly Re) when the sequence of ligands in decreasing order of priority (the other group is remote from the observer) is clockwise, and proS (Si),when it is anticlockwise (Fig. 1-5). The procedure for faces is even simpler: regarding the center to be qualified from the pro-R (Re) face the sequence of ligands in decreasing order of priority is clockwise, while anticlockwise from the p r o 3 (Si) face.

X-pro-R

X’-pro-S

(Re)

(Si)

X

Fig, 1-5. Prochirality.

Note that there is no correlation between the prochirality descriptor and the configuration of the product arising from transformation of the group or face concerned, e.g.:

1.4.1 Substrate Selectivity

I1

1.4 Selectivity in Chemistry Two main modes of selectivity can be defined in chemistry, one concerns the substrates and the other the products of a reaction. These two modes are defined by the following schemes: (i) Substrate selectivity: reagent

reagent

A-X;

B - Y

k*x

kBY

~ A X * ~ B Y

(ii) Product selectivity:

A

reagent

* *

X+Y+Z+. . .

[XI [ Y ] [ Z ]. . .

Substrate selective is a reagent which transforms different substrates (e.g. A and B ) under the same conditions at different rates to the products X and Y. Product selective is a reaction or process in which more than one product can be formed but the products are formed in a ratio which differs from the statistically expected one. The latter is the ratio of the number of sites which can react (e.g. 2 :2 : 1 for o-, m-,and p-nitrotoluene in the mononitration of toluene).

1.4.1 Substrate Selectivity Substrate selectivity can be conveniently classified following the hierarchic scheme set up for the differentiation of molecules and described in Section 1.1. Reactivity of constitutional isomers may differ markedly. Thus when a mixture of n-butanol and tert.-butanol is treated with cold cc. hydrochloric acid, the former remains unchanged, while the latter is converted to tert.-butyl chloride. Our real interest lies, however, in substrate selectivity exhibited by stereoisomers.

12

1.4 Selectivity in Chemistry

1.4.1.1 Substrate Diastereoselectivity (Diastereomer Selectivity) Since diastereomers differ in their scalar properties it can be expected that their reactivity may also be different towards any type of reagent. In practice, these differences are the more significant the closer the reaction center is to the stereogenic units of the molecule. Several modes of substrate diustereoselectivity can be distinguished: 1. Diastereomers can lead to the same product at different rates, e.g. k,,/k,, = 3.2 in 8OVo acetic acid for the following reaction [14]:

dOH aH \ HO \

t Bu

tBu

CrOr

2. Diastereomers can give rise to different diastereomers at different rates as exemplified by the debromination of diastereomeric 2,3-dibromobutanes [ 151: Me

Me

3. Diastereomers can even yield products with different constitutions, as shown by the classical example of the Beckmann-rearrangement of (E)- and (2)-oximes:

, ) I Ph

Me

-

PhNHCOMe

(E)

H q N Ph

A (Z )

Me

-

PhCoNHMe

1.4.1 Substrate Selectivity

13

4. Since equilibria are in fact dynamic systems characterized by the ratio of the rates of the forward and reverse reaction ( K = k / k -), equilibria involving diastereomers can be regarded as a special case of substrate diastereoselectivity. The simplest and most important type of such reactions is epimerization. One of the best known examples for the equilibration of diastereomers is the mutarotation in solution of a- or ~-D-glUCOSe,where the reagent is the solvent.

1.4.1.2 Substrate Enantioselectivity (Enantiomer Selectivity). Kinetic Resolution Diastereomers can be relatively easily separated and therefore differences in their reactivity can be explored in separate experiments. With enantiomers the situation is just the opposite. Their separation is rather difficult and therefore conditions are sought under which they react at different rates, whereby ultimately their separation can be effected. Transformation of enantiomers at different rates is called in the present terminology substrate enantioselectivity. Since enantiomers are the mirror images of each other, they can only be distinguished by effects or objects lacking reflectional symmetry. Such an effect is circularly polarized light and such an object is any non-racemic chiral reagent *. This situation can be easily visualized by using two-dimensional chiral and achiral figures as suggested by V. Prelog [13]. A figure (e.g. a triangle) is chiral in two dimensions if it cannot be superimposed onto its mirror image by in-plane rotation and/or translation. Otherwise it is achiral. Molecules are symbolized by triangles, transition states, or interactions in general, by placing one side each of “substrate” and “reagent” in contact with each other. It is apparent from Fig. 1-6 (a) that interaction of enantiomeric chiral substrates [( +)-Su and (-)-Su] with an achiral reagent (Rg)produces “transition states” [( +)-E and (-)-Ts] which are related as mirror images. This remains true for any specific side (“functional group”) of substrate and reagent brought into contact with another side. When the same is done with a chiral reagent (Rg*)**(Fig. 1-6 (b)) the resulting figures ( E l and Ts2) are no longer mirror images but diastereomeric and thus differ in their scalar properties. Since the ground states of enantiomers are of the same free energy and, being diastereomeric, the transition states are of different free energy, it follows that in reactions involving chiral agents or effects enantiomers are transformed at different rates.

* A racemic mixture of chiral molecules behaves as if having reflectional

symmetry. For the sake of brevity throughout this book the term “chiral reagent” denotes a non-racemic assembly of chiral molecules. ** Symbols marked with an asterisk represent chiral groups or non-racemic chiral compounds.

14

1.4 Seleciiviiy in Chemisiry

(-)-TS

Ts

'

Ts

Fig. 1-6.Two-dimensional representation of the reaction of enantiomers with (a) achiral (Rg)and (b) chiral (Rg*)reagents.

From a practical point of view we have to distinguish those cases when (i) the transformation rate of the less reactive enantiomer is virtually zero or (ii) when the two rates are comparable. So far the first case has been restricted to enzymatic transformations and was first exploited by Pasteur. He subjected racemic ammonium tartrate to fermentation with the mold Penicillium glaucum, whereby the unnatural ( - ) - ( 2 S , 3s) enantiomer remained unchanged and could be isolated. In recent years enzyme-catalyzed stereoselective transformations have become a powerful tool in preparative organic chemistry *. In the second case, i.e. when the rates are comparable, one of the enantiomers is transformed faster, and therefore an excess of the less reactive substrate gradually

*

For recent monographs see refs. [16-201.

1.4.1 Substrate Selectivity

0

t

Tim

k'+'>k'-'

15

cx)

Fig. 1-7. Kinetic resolution as a function of time.

builds up. This excess goes through a maximum and disappears on completion of the reaction (Fig. 1-7). If the reaction is interrupted before completion or if less than the necessary amount of reagent is applied, the result is a non-racemic mixture of the starting material and of a product in which an excess of the more reactive enantiomer of the substrate is incorporated. This way of partial separation of enantiomers is called kinetic resolution. The main types of kinetic resolution will be discussed below. The chiral agent may be (i) a catalyst, a medium or a reagent donating or accepting a non-chiral entity (e.g. a hydride ion) or (ii) a reagent which is linked to the substrate in the course of the reaction. In the first case the chirality of the substrate can either be retained or destroyed. If it is retained then the pair of enantiomeric substrates is transformed to a pair of enantiomeric products. When chirality is destroyed both enantiomers give rise to the same achiral product(s). In Fig. 1-8 examples for both situations are given. The simplest chiral agent is circularly polarized light, however, kinetic resolutions by this means are only of theoretical interest. In example (a) it is the medium which is chiral. When the racemic oxaziridine (+)-I1 was decomposed by heating for 1.25 min at 148" in cholesteryl benzoate, the unchanged starting material was enriched to 20% in the (-)-enantiomer [21]. Upon oxidation by the Sharpless method, it is the reagent which is chiral and discrimination between enantiomers of (&)-I -phenyl-2-pyrrolidino-ethanol(b) was exceptionally high [22]. The product may also be chiral and in the simplest case it is the enantiomer of the substrate itself. This process is called deracemization and can be brought about by equilibration in chiral media or using chiral catalysts. Thus on heating racemic phenylglycine with (+)-10-camphorsulfonic acid [( +)-CSA] the salt of the pure R enantiomer (12) crystallizes out [23]. When the association of a racemic substrate with a chiral agent is not simply transient but leads to the formation of a compound in which the stereogenic elements of both are retained, the product is a mixture of two diastereomers. A classical example is the partial esterification of (&)-mandelic acid with (-)-menthol (Fig. 1-9). Actually this was the very reaction which led to the discovery of kinetic resolution by

16

1.4 Selectivity in Chemistry

-<

( R ) , 37% in 95% ee

(-1 - diiso pr o p y It art rate

(b) PhL

N

g

Ti(OiPr)n, tt3uOlH, - 2 0 ° C

(S),59% in 63% ee

w

(c) Ph

Atop

(+)-CSq EtC0,H 7h, 100 O C , 80%

(+)-CSA'

-

4

Ph

(3 Fig. I-8. Kinetic resolution.

)I:

Ph Ph-CH(OH)COzH

COr(-)-Ment

1

Ph

C02H

(*)

+

I

OH

HO

I

Ph*COz(--)-Ment (-) -Me nt -OH

kR7kS

(Sji-)

Fig. 1-9.The experiment of Marckwald and McKenzie.

Marckwald and McKenzie in 1899 [24]. Incomplete reaction leaves an excess of (S)-mandelic acid, while total hydrolysis of the esters gives a mixture enriched in (R)-mandelic acid. Kinetic resolution is an inherently wasteful process for producing optically active compounds and can only compete with conventional resolution (of poor economy

1.4.2 Product Selectivity

17

itself) when rate differences are extreme*. With few exceptions this has so far only been realized with enzymes, a field which lies outside the scope of this book. Since different conformations of a compound can be regarded as unstable stereoisomers, their reactivity is governed by the same principles as those of stable stereoisomers. Differences in the reactivity of stereoisomeric conformers are, however, difficult to verify experimentally due to the fast equilibration of the substrates and, occasionally, also of the products.

1.4.2 Product Selectivity The main subject of our book is product stereoselectivity, i.e. the case when out of two or more possible stereoisomeric products, arising from a single substrate, one is formed preferentially. When the number of products is more than two, selectivity can be best characterized by the percentage distribution of the products. When only two products ( X and Y ) are formed, selectivity may be characterized by giving the product ratio ( r ) , or the percentage of the products or the excess ( e )of the major product **. If [XI > [ Y ] , the following simple relationships hold: [XI - t YI .loo% [ Y ] ’ e = 1x1+ [ YI

r = -[XI

and X %

=

r- 1 100% rs1

VOX-YO Y = -*

100r r+ 1

=-

When the products are enantiomers or diastereomers the specific terms enantio-

meric excess (ee) and diastereomeric excess (de) respectively may be used. Since most often the diastereomeric products are later transformed by the removal of a chiral auxiliary group to a mixture of enantiomers, quoting diastereomeric excess values (de) is preferable. The percentage of the major stereoisomer is often called the stereoselectivity of the reaction. In this case, however, the values should be standardized to

X%+Y%

* **

=

100%

It can be calculated [25] that the enantiomeric purity of unreacted substrate at 50% conversion is 80% when the relative rate is 25, and about 93% when it is 100. Reactions in which only a single stereoisomer can be detected are generally called stereospecific, but because of some controversy around this term we are not using it in this book.

18

1.4 Seleciiviiy in Chemistry

In the absence of intermolecular interactions between solute molecules the numerical value of enantiomeric excess is equal to optical purity (op) i t . the ratio of observed specific rotation and the specific rotation of the pure enantiomer ( [ aImax):

For all practical purposes the two terms are equivalent. Nowadays, when high resolution NMR techniques and various chromatographic techniques find ever increasing application to the determination of enantiomeric purity, it is more appropriate to use the term enantiomeric excess. It is not only for its simple relationship to optical rotation that the ee value rather than the percentage distribution should be given. Ee is in fact the upper limit of the amount of the major enantiomer which can be obtained by crystallization from a mixture of enantiomers forming a separate racemic solid phase (racemic compound), because for each mol of the minor component one mol of the major one is included in the crystals of the racemic phase. According to Jacques, unfortunately about 90% of the chiral compounds examined in this respect form such solid phases, while the rest form conglomerates or solid solutions 1261. On crystallization conglomerate forming-enantiomers behave as different compounds, and therefore the upper limit for the isolable major enantiomer is its percentage and not its excess. Product selectivity also comprises those reactions in which products are not stereoisomers, but these aspects are outside our interest. Preferential formation of one constitutional isomer over other possible isomers is called regiosefectivity. Regioselectivity arises from competition between groups (molecular sites) of different connectivity. An example is the chlorination of toluene which could lead, in principle, to four isomeric monochloro derivatives, of which practically only two (0 and p ) are formed.

I .4.2.1 Formation of Stereoisomers What we are really interested in is product stereoselectivity, a prerequisite for which is the fact that the formation of at least two stereoisomers from the same substrate should be possible. Therefore we should first examine the symmetry requirements for the formation of stereoisomers. 1 . Transformations of homotopic groups or addition to homotopic faces, even those in chiral molecules, gives rise to the same product. Thus bromination involving or hydride addition any of the homotopic methyl groups marked with 0 and 0, to any of the homotopic faces in acetone gives a single product (Fig. 1-10). 2. Transformation of enantiotopic groups or addition of a ligand different from the existing ones to enantiotopic faces usually produces stereoisomers. If the reaction

1.4.2 Product Setectivity

19

Fig. 1-10. Transformation of a compound with homotopic groups and faces.

does not involve the permanent attachment of a chiral element to the substrate, the products are enantiomers, otherwise they are diastereomers. For example, mono-esterification of glycerol with achiral propionic acid produces the enantiomeric esters @)-I3 and (S)-13, whereas with (R)-2-~hloropropionicacid the diastereomers (R,R)-14and (S, R)-14 result (Fig. 1-1 1). Addition to enantiotopic faces is more complex than transformation of enantiotopic groups and need not involve the formation of stereoisomers. Stereoisomers are only formed when the entering ligand is different from those already attached to the prohciral center. Also, changes at both ends of a double bond have to be considered

%OH

c1

H0,C n M e CQOH

CYOH (S,R)-(14)

HO+HCY0,C

y'

(~~~1414)

Fig. 1-11. Transformation of a compound with enantiotopic groups.

20

1.4 Selectivity in Chemistry

and situations in which only one or both bridgehead atoms are prochiral should be distinguished. For example, in styrene only one of the bridgehead atoms is prochiral and addition of a new achiral ligand produces a pair of enantiomers.

Fig. 1-12. Addition to olefins in which both sp2 carbons are enantiotopic.

1.4.2 Product Selectivity

21

If the incoming ligand contains a chiral element the products are diastereomeric. In non-terminal olefins usually both sp2 carbons are prochiral centers and the stereochemical outcome of additions to such double bonds depends on the symmetry of both substrate and reagent. The complexity of this situation can be illustrated by the addition of oxygen, chlorocarbene and water to (Z)- and (E)-2-butene (Fig. 1-12). (Z)-2-Butene has a C, axis coplanar with the double bond and a mirror plane bisecting it. Therefore attachment of a point-like ligand, such as oxygen, from either side gives the same stereoisomer, namely, the achiral(2R, 3S)-2,3-epoxybutane. Addition of chlorocarbene, which is of lower symmetry (CJ, gives rise to a pair of diastereomeric but still achiral chlorocyclopropanes (15). Finally, water is a reagent which forms different bonds (C-H and C - 0 ) at the two prochiral centers and therefore these have to be considered separately. C-H bond formation does not create a chiral center, while attachment of OH from opposite sides yields (R)-and (S)-2-butanols. Hydration of (E)-2-butene proceeds similarly but now there is a C, axis perpendicular to the plane of the double bond which renders the two centers, but not the two faces of the molecular plane, equivalent. Therefore both epoxidation and chlorocarbene addition lead to enantiomers. Note that C-I in the chiral cyclopropanes (2R,3R)and (2S,3S)-15 is not a chiral center. An important case is when in a reaction two planar trigonal atoms having enantiotopic faces are joined to form a pair of adjacent asymmetric centers. If the partners are different two pairs of diastereomers may be formed: OLi

1

R e i R e Ik

.Sf t S I

.1

ix

RriSi

1

it/

In order to characterize unambiguously both the mode of approach of the partners and the relative configuration of the products, Prelog and Seebach suggested a notation system based on the sequence rules [8] illustrated in the foregoing reaction. If the approaching faces of the atoms to be linked carry the same descriptors, i.e. both are Re or both are Si, bond formation is labelled as ‘‘like’’ (Ik), conversely if the corresponding descriptors are different (Reand S i ) the label is “unlike” (ul).The relative configuration of the centers in the product, following each other in the order

22

1.4 Selectivity in Chemistry

of their IUPAC approved numbering is labelled as like (I) for R, R and S, S pairs, and unlike ( u ) for R, S and S, R pairs. The system has been devised as to be applicable to reactions and products involving more than two centers as well. More complex cases, such as addition to unsymmetrical, non-terminal olefins, attack by reagents which provide two new and different ligands (e.g. HOHal) or by reagents containing a chiral element can be, with some patience, analyzed along these lines. 3. Transformation of diastereotopic groups or the addition of a new ligand to a center with diastereotopic faces always gives rise to diastereomers. Depending on the symmetry of both substrate and reagent the products may be chiral or achiral. Examples for both situations are given in Fig. 1-13. The first two illustrate the transformation of diastereotopic groups, the second two those of diastereotopic faces. Me A

-co, Me

Me

Me 1,3 - cis

1,3 - cis

Fig. 1-13. Transformations of molecules with diastereotopic groups or faces.

1,3

- tmns

1,3 - Vans

1.4.2 Product Selectivity

23

The situation just described is not more complicated when the reagent is chiral, provided that it is a single pure enantiomer: but two stereoisomers are formed which are both chiral. The Prelog-Seebach notation can be conveniently applied to reactions in which a trigonal center with diastereotopic faces in a molecule containing a chiral center is transformed to a chiral tetrahedral center. Re+R, and Si+S combinations are labelled “like” (lk),while Si+ R and Re+S combinations “unlike” (ul). Formation of (IR, 3s)-and ( I S ,3R)-3-methylcyclopentanols are both ul (more precisely a 1,3-ul) addition of hydride anion giving rise to a u product. Note that ul processes need not yield u products. Since the relationship of a given group or molecular face can be enantiotopic to one partner and diastereotopic to another, transformation of a single kind of group or face may lead to as many as four stereoisomers. Thus geminal hydrogens in 16 (Fig. 1- 14) are diastereotopic, whereas those associated with different carbons form two enantiotopic pairs. Monobromination should give four stereoisomers, namely two racemic pairs of diastereomers [(2R,3R)- and (2S,3s)-further (2S,3R)- and (2R,3S)-2-bromo-3-methylglutaric acid]. In the reaction of norbornene (17) with ethene four stereoisomers could be expected, but in fact only approach from the less hindered ex0 face i.e. the formation of (1S,2S)- and (IR,2R)-18 was observed (cJ: Section 6.4.2).

Fig. 1-14. Transformations at groups and centers which are involved in both enantio- and diastereotopic relationships.

1.4.2.2 Conditions Necessary for Stereoselectivity While it is rather difficult to predict the degree of stereoselectivity of a given transformation, it is very simple to define conditions which must be fulfilled that stereoselectivity should be possible at all. The formation of diastereomers involves diastereomeric transition states, which, being geometrically different, are of different free energy. Therefore, whenever the formation of diastereomers is expected, it can also be anticipated that they are

24

1.4 Selectivity in Chemistry

formed in unequal amounts. This applies equally to reactions of diastereotopic groups or faces with any reagent and to such reactions of enantiotopic groups or faces in which a reagent containing a chiral element forms a permanent bond with the substrate (e.g. the formation of 14 in Fig. 1-11). Since the ground state free energy of diastereomers is different, the observed product ratio is either the result of unequal rates of formation (kinetically controlled selectivity) or subsequent equilibration (thermodynamically controlled selectivity) or of both. Formation of a pair of enantiomers from a single substrate always involves the transformation of enantiotopic groups or faces. When such groups or faces interact with an achiral agent two different transition states arise which are related as mirror images. Consequently the free energies of activation associated with the two transition states and thus the rates of formation of the two products are equal. In other words, no selectivity can be expected in such reactions. Interaction of enantiotopic groups of faces with a chiral agent (reagent, catalyst, solvent etc. ) can be envisaged as the formation of a pair of transient compounds, and we have just shown that such pairs are diastereomeric. Thus the same applies as to the formation of diastereomers, i.e. they are formed at unequal rates and their decomposition gives enantiomers in unequal amounts. All this can be nicely illustrated by two-dimensional models. In Fig. 1-15 the substrate with enantiotopic groups is represented by a triangle having two equal sides (Su), the achiral reagent (Rg)by a rectangle, and the chiral reagent by a scalene

(t)-rr

Ti

’

(-)-

7\

TY2

Fig. 1-15. Two-dimensional representation of the reaction of molecules with enantiotopic groups or faces with (a) achiral ( R g ) and (b) chiral ( R g * ) reagents.

25

1.4.2 Product Selectivity

triangle (Rg”).It can be seen that any analogous association of Rg with the “enantiotopic” sides of Su leads to a pair of mirror image (“enantiomeric”) figures (+)-TS and (-)-TS (“transition states”), while a similar operation with Rg* gives T S ‘ and TS2, i.e. figures of different shapes (diastereomeric). It follows from the above discussion that no reaction is genuinely enantioselective; there are only diastereoselective reactions which ultimately lead to a non-racemic mixture of enantiomers. The diastereomeric nature of the selectivity-determining interaction becomes increasingly apparent in the series of examples shown in Figs. 1- 16 and 1-17. * In example (a) photoenergy transfer from a chiral sensitizer is the selectivity-determining step [28], while in example (b) it is the preferred abstraction of a proton by a chiral base from one of the enantiotopic methylene groups [29]. Fig. 1-1 6 (c) illustrates enantioselective methanolysis catalyzed by quasi enantiomeric chiral alkaloid bases [30]. In example (a) (Fig. 1-17) acetophenone is reduced with the bulky complex hydride (R)-19 prepared from (R)-2,2‘-dihydroxyl-l, 1 ’-biphenyl. Here it is possible to formulate two diastereomeric transition states, one of which ( TS2)is un-

APh

hv

(a) Ph

-

@*

3 %ee

Ph

‘Ph

Me

~

H0,c

CO,M~

MeOH+ PhMe cinchonidine

Me

(2R,43 64% ee

MeOH + PhMe cinchonine

~

Me0,C

(2R,4R) 70% ee

Fig. 1-16. Reactions involving enantiotopic groups or faces in the presence of chiral catalysts.

* For mixtures of stereoisomers the major one is depicted.

C0,H

26

1.4 Selectivity in Chemistry

Me

H

I ' C '-"OH

A

Ph

- 25 O C

*

&f3IPC2

- 97.5%

+

- 2.5%

(1R,2R) - (21) Fig. 1-17. Reactions of compounds with enantiotopic groups and faces with chiral reagents.

favored. In fact at - 100°C the (R)-I-phenylethanol was obtained in high purity [3 I]. Finally, in example (b) addition of (-)-diisopinocampheylborane [( -)Ipc,-BH, 201 to the enantiotopic ethylene groups of the substrate gives a mixture of

1.4.2 Product Selectivity

27

two diastereomers” which were not separated but oxidized directly to an enantiomeric mixture of alcohols (21). Here the selectivity-determining step is clearly enantio- and diastereoselective hydroboration, since the product ratio is not altered by oxidation which serves to remove the remnants of the chiral carrier molecule.

1.4.2.3 Concept of Stereodifferentiation The above classification of stereoselective reactions according to the symmetry of products is useful inasmuch as it gives insight into the process but suffers from being not structured enough. Probably this prompted Izumi to suggest, in 1971, a new classification based on the symmetry of substrates [32, 331. Reactions giving unequal amounts of stereoisomers were called stereodifferentiating and prefixed according to the nature of the substrate as enantiomer- and diastereomer-, enantiotopos- and diastereotopos-, further enantioface- and diastereoface-differentiating reactions, according to whether stereoisomers, groups, or faces were differentiated. Note that the first two types cover substrate selective transformations, while the last four product selective ones. Izumi’s classification is rather appealing because the conditions of selectivity can be defined very simply: enantio-differentiation requires chiral means, whereas diastereo-differentiation does not. Also very enlightening is Izumi’s observation that the key for enantio-differentiation is provided by the environment (reagent, solvent, catalyst), while that for diastereo-differentiation is within the molecule (steric hindrance by, or electronic and other effects of groups around the reaction center).

1.4.2.4 Methods for Inducing Stereoselectivity In the foregoing sections the fundamentals of stereoselective reactions have been discussed, but little information has been provided about the methodology of such reactions. For our topic, however, the various ways and means by which stereoselectivity can be generated is of primary importance.

Enantioselectivity Selectivity in the transformation of enantiotopic groups and faces can only be achieved, as explained before, by chiral means. These may be characterized as follows: (i) Photochemical transformations induced by “chiral’: i.e. circularly polarized, light. This is an impractical method giving only a few percent ee or less.

* Note

~~

(i) that the process is totally diastereoselective giving only trans addition products and (ii) that oxidation proceeds with retention of configuration.

1.4 Selectivity in Chemistry

28

(ii) Reactions conducted in a chiral solvent [34]. Chiral solvents are, compared with their generally low efficiency, very expensive and thus do not qualify for practical stereoselective synthesis. (iii) Reactions carried out in thepresence of a chiral udditive, Additives in this context are substances which emerge unchanged from a reaction and range from cosolvents ( c j Fig. 1-18 (a)) [35] through photosensitizers to catalysts. Enantioselective reactions carried out in the presence of chiral additives are illustrated by the photoisomerization of (2)- to (E)-cyclooctene in the presence of a chiral sensitizer (Fig. 1-1 8 (b)) [36], the electroreduction of a dibromocyclopropane in the presence of a chiral base (c) [37], base-catalyzed intramolecular aldol condensation (d) [38], and, finally, homogeneous hydrogenation with a rhodium catalyst (e) [39] prepared from chiral phosphines.

(4

c&x

&+& Me0

Li

N'

-

b

0 85%ee

C0,R

(b)

C02R hv, pentane, - 87 O C

('' Ph A

R = (R)-2-0ctyl 53% ee

b

2 eemrtine + H ~ - -B ~ G

Mew B

Pi

(d 1

Me0 PhMe, - 45 O C 2) He

r

Ph A ,, Pi

Br

B

r

44%ee

H

(S)-proline

DMF, -20°C

*

93% ee

0

".KO Metf Me

2 h P Ph2 ' "

[ ( B P P M ) ( C U HHzU ) C I ) R ~93%

0

*

86%ee

0

N

I

COztBu

BPPM Fig. 1-18. Induction of enantioselectivity with the aid of chiral solvents or additives.

1.4.2 Product Selectivity

29

(iv) Reaction of enantiotopic groups or faces with chiral reagents involving the transfer of an achiral species. This approach has been mainly exploited for redox processes. The chiral complex hydride 19 (Fig. 1-17) is just one of the many chiral hydride ion donating reagents developed in the past two decades. Rather often the hydride ion is transferred to the substrate from the chiral center itself, whereby the latter is destroyed. Such processes are called “self-immolative” [27], and one of them is shown in example (a). Oxygen transfer from chiral peracids is generally less efficient; however, selectivity is high in the oxidation of the enolate in Fig. 1-19 (b) with a chiral oxaziridine [40].

Me

(b)

Me0

OMe

Me

OMe

Me0

96% ee

Fig. 1-19. Transfer of hydrogen or oxygen to substrates with enantiotopic faces.

(v) The substrate with enantiotopic groups or faces is linked with a chiral auxiliary compound to form a derivative in which the former become diastereotopic. Thus the necessary condition for stereoselectivity with any reagent (chiral or achiral) is met. After the required transformations have been carried out, the chiral auxiliary is removed to give rise to a non-racemic mixture of enantiomers. An ideal chiral auxiliary substance is one which (i) provides high asymmetry in the selectivity determining transition state, (ii) is recoverable, and (iii) is readily available in both enantiomeric forms in high optical purity and at low cost. Examples for recoverable (a) [41] and non-recoverable chiral auxiliaries (b) [42] are shown in Fig. 1-20. The above examples require the following comments: (i) It has to be noted that condensation of substrates with enantiotopic faces with a chiral auxiliary usually leads to the formation of E-Z isomers, although one on them may be highly preferred. This was avoided in example (a) because the original substrate had homotopic faces and in (b) because the chiral auxiliary had homotopic groups. It has to be borne in mind that no serious rationalization of stereoselectivity observed in a certain reaction is possible without knowing the configuration (E or Z ) of the intermediate. (ii) In the stereoselective step the primary product is formed as a mixture of diastereomers which is then converted to a mixture of enantiomers by removing the chiral auxiliary.

30

1.4 Selectivity in Chemistry

I

66% ce

S? Fig. 1-20, Enantioselective synthesis with the aid of chiral auxiliary compounds.

(iii) The enantiomeric excess in the end product does not necessarily reflect the ratio at which the diastereomers were originally formed for the following reasons: (a) equilibration of the diastereomers, (b) enrichment of one of the diastereomers during work up (crystallization, chromatography etc. ) and (c) racemization during the removal of the chiral auxiliary. An example for situation (a) is the addition of cyanide ion onto the Schiff-base 23. The product ( 2 4 ) crystallizes out in 100% diastereomeric purity and can be converted to enantiomerically pure amino acid. * This result, however, neither reflects the ratio of formation rates nor implies a much higher thermodynamic stability of 24 in solution. In fact, in chloroform a 2 : 1 equilibrium ratio of epimers was established [43].

Q

,&

'Me Me Me

NaCN MeOH + AcOH

*

Me Me

CN

100% ee

*

Such epimerizations associated with the crystallization of one of the epirners are called second order asymmetric transformations in the older literature.

1.4.2 Product Selectivity

31

(iv) It is assumed that as chiral auxiliaries enantiomerically pure substances are used. In practice optically impure reagents are often employed, resulting in correspondingly lower ee values. * In this case, namely for reasons of symmetry, each of the diastereomeric intermediates is contaminated by exactly as much of its enantiomer, as was the chiral auxiliary substance used. In the extreme case, with a racemic chiral auxiliary, the end product will also be racemic, regardless of whether diastereomers are separated or not. On the other hand, the ratio of diastereomers obtained before the removal of the chiral auxiliary is equal to the enantiomeric purity of the end product which would have been obtained with an enantiomerically pure auxiliary. Thus, if one wishes to examine the performance of a chiral auxiliary, it is sufficient to employ it in the racemic form and determine in some way the ratio of the diastereomeric intermediates [44]. This also has the advantage that results are not influenced by possible racemization in the decomposition step. A rare but intriguing phenomenon is asymmetric amplification, i.e. when the enantiomeric purity of the product is higher than that of the chiral aid [45-471. E.g. in the following Lewis acid catalyzed ene reaction an almost threefold amplification of ee was experienced. It was postulated that the catalytically active species is a dimer of (BINOL)TiBr2 of which the homochiral ( R , R )one is much more active than the racemic (R, S ) one. In fact with the optically pure catalyst the initial rate of the reaction was by a factor of 10 higher than with the racemic one [47].

CY

Ph

A+

0

( R ) - BINOL (33% ee) + Ti(OiPr),Br,

HACO,~

(1 mol%), 4.4 MS, CyCl,, - 30 O C

Cq

OH

)ph 91% ee

Diastereoselectivity It has been demonstrated in the preceding section that enantioselectivity can always be traced back to diastereoselectivity. The selective preparation of diastereomers for their own sake is an even more important task and has been the subject of innumerable studies, mainly in the natural product field. Therefore, it would be very useful if it were possible to formulate some general rules controlling diastereoselectivity. Owing to the diversity of organic reactions, chances for an allembracing correlation are poor, but several rules of more or less limited scope are known and will be discussed in due course. The factors controlling diastereoselection can be identified as being of stereoelectronic and purely steric nature. For example, the well-known fact that addition of bromine to maleic acid gives exclusively (2R, 3S)-2,3-dibromosuccinic acid [42] can be well explained by the mecha-

* In this book, in order that data should be comparable, whenever possible ee values were corrected for enantiomerically pure reagents: eecorr= eeobs/eereagent

32

1.4 Selectivity in Chemistry

nism of the reaction (stereoelectronic control), while attack at the Si face in example (b) of Fig. 1-20 is rationalized (less convincingly) by shielding of the Re face by the phenyl group at C-6 (steric hindrance). In the following reaction the preferred formation of the thermodynamically less favored axial alcohol [43] can be ascribed both to steric hindrance by the axial 3-methyl group and to the bulkiness of the reagent. However, this explanation also implies the stereoelectronic requirement that hydride transfer should occur in a plane perpendicular to the ring and not laterally.

Me

hie 96%

Me 4%

Diastereoselectivity was first recognized by Emil Fischer [50], who pointed out that the ratio of diastereomers arising by the formation of a new asymmetric center in a molecule was biased by those already present. No successful attempt was made to predict even qualitatively the direction of this bias until the seminal papers by Prelog [51] and by Cram and Elhafez [52] were published in the early fifties. The rules of Cram and Prelog based on these papers and their further developments are closely connected with the addition reactions of carbonyl compounds (cJ Section 3.2.1). Among the very few quantitative approaches to the prediction of diastereoselectivity the semi-empirical method developed by Ruch and Ugi on the basis of group theory should be mentioned [53]. Stereoselectivity in nucleophilic addition to ketones was predicted by computational methods by Wipke and Gund [54] (cf. Section 3.2.1) and by Hirota et al. [55], who also modelled the oxidation of chiral sulfides. A thorough study of the temperature dependence of the photoaddition of electron rich olefins to a-ketoesters by Scharf et al. led to the recognition of the S.C. isoinversion principle [56]. It was found that, contrary to common belief, lowering of temperature did not always result in higher selectivity (see e.g. the reaction in Fig. 1-21 (a)). Moreover with some systems (e.g. Fig. 1-21 (b)) an inversion of diastereoselecindicating tivity was observed at a certain temperature (inversion temperature, Ti,”) a change of dominance in the influence of enthalpy vs. entropy on the reaction. In such processes two sets of activation parameters exist, one above and one below Ti,, and for a set of related reactions their differences (&AHf and SAS’) show a linear relationship, called isoinversion correlation. The slope of this straight line (having a temperature dimension) defines the isoinversion temperature (Ti = d @AHf)/ (6’6AS’) a constant characteristic for the selectivity of the whole set of analogous reactions. For systems of high selectivity the values of Ti,,and Tj are close. Inversion of selectivity is a feature of reactions with two discernable transition states both involved in stereoselection. The principle of isoinversion was successfully applied to in-

1.4.2 Product Selectivity

(4

33

Ph

Fig. 1-21. The isoinversion principle.

terpret chiral amine catalyzed addition of alcohols to ketenes, homogeneous phase enantioselective catalytic hydrogenations and several other reactions. The traditional term for diastereoselectivity experienced in chiral molecules is asymmetric induction. This term is useful in a way that it points to the source of selectivity, but also somewhat narrow because it excludes diastereotopicity in achiral molecules. Clearly the structural features which may induce stereoselectivity in the chiral and achiral substrates shown in Fig. 1-13 are essentially the same. It is much more controversial to take a stand on another traditional term, i.e. asymmetric synthesis. In Marckwald’s formulation [57] “asymmetric syntheses are those reactions which produce optically active substances from symmetrically constituted compounds with the intermediate use of optically active material but with the exclusion of all analytical processes”. The examples in Figs. 1-16 to 1-18 all correspond to this definition, although what Marckwald probably had in mind was the use of chiral auxiliary compounds as exemplified in Fig. 1-20. In their book “Asymmetric Organic Reactions” [58] Morrison and Mosher extended the definition to transformations of any kind of substrate containing prochiral groups or faces. This was a logical development, since, from the point of view of the selectivity-determining step, it was irrelevant whether the presence of the inducing group in the molecule was transient or permanent. The scope of this definition is, however, rather wide and points towards a third usage of the term, i.e. for synthetic sequences in general starting from and arriving at a chiral compound. In view of the above mentioned ambiguities, whenever possible we are going to replace “asymmetric synthesis” by the more specific terms enantioselective and diastereoselective synthesis, respectively. In order to assess asymmetric induction in reactions involving an achiral reagent we may as well use a racemic substrate since modern analytical methods (most conveniently GLC, HPLC or NMR) permit the determination of the ratio of diastereomeric racemates ( [ R R ]+ [ S S ] / [ R S+] [SR]).In the absence of intermolecular interac-

34

1.4 Selectivity in Chemistry

tions between substrate molecules (a resonable assumption in dilute solution), rates of formation for enantiomers must be equal and therefore

Transfer of hydride anion to racemic 2-methylcyclohexanone (Fig. 1-22), a typical ex-

+

(IS,ZR)

(IR.2R)

cis (u )

tram ( I )

TSRS

(I R .2 S )

t

h“

+ HD*

TSd

TS:

Fig. 1-22. Simple and double asymmetric induction.

\+

b TSZ

1.4.2 Product Selectivity

35

ample for this situation, is illustrated also in two dimensions (b). Diastereotopic faces of the chiral substrate are represented by different sides of a scalene triangle, the achiral hydride donor (HD),e.g. sodium borohydride, by a quadrangle. It is apparent that the transition states TSRR and TSss, as well as TSsR and T S R S , are mirror images, which results in equal rates of formation of the enantiomeric products the ratio of cis and trans alcohols being the same ( 2 : 1 ) [59] with both the racemic substrate and any of the pure enantiomers. However, with a chiral reagent four different transition states can be envisaged e.g. when a chiral hydride donor (HD*) is used. In the lower part of the figure such a reaction is depicted, the reactive side of the reagent being marked with a heavy line. Transition state representations are all of different shape, i.e. the relationship of any two is diastereomeric. Consequently, even the rates of formation of enantiomeric products are different, in other words the product is optically active. Thus reduction with (- )-diisopinocampheylborane produced 92% of the cisalcohol containing the 1R,2S enantiomer in 1.8% ee and 8% of the trans-alcohol with 13.3% ee for the 1S,2S enantiomer [60]. This phenomenon is called double induction and has considerable synthetic utility. * In practice, usually a certain stereoisomer is needed and very often it is difficult to predict which would be formed in excess with a given reagent. For reasons of symmetry, when one enantiomer of the product is in excess with a certain chiral reagent, the same excess of its antipode must be obtained using the antipode of the same reagent. GuettC and Horeau [59] have shown that in the reaction of a racemic substrate with a chiral reagent the optical purity of a given diastereomer is inversely proportional to its yield:

Another important feature of reactions with double asymmetric induction is that, when the substrate is non-racemic, reaction with one or the other enantiomer of a reagent gives different diastereomeric ratios. This can be easily recognized by exchanging the roles of substrate and reagent in Fig. 1-22 (b). Taking HD* as substrate, it is apparent that kr/k,*# k$/k$. In this respect it is irrelevant whether the transition states relax by dissociation of the chiral reagent (to give enantiomers) or the reagent remains attached to the substrate to give diastereomers. Examples for the latter are given in Section 5.2.1. Finally, the technique of chirality transfer ** or more generally, transfer of configuration should be mentioned, in which the stereoselective formation of a new stereogenic element is coupled with the elimination of another.

* For a review see ref. 62. ** Self-immolative asymmetric

synthesis in Mislow’s terminology 1621.

36

1.4 Selectivity in Chemistry

Intermolecular versions of this approach are usually of poor selectivity [ c j example (b) in Fig. 1-1 91, while intramolecular chirality transfer reactions are mainly sigmatropic rearrangements which usually proceed with high stereoselectivity. In all the examples in Fig. 1-23 transposition of a double bond is involved. Due to the concertedness of the reactions, formation of the new double bond is totally stereoselective. Example (a) [63] and those in Figs. 1-24 and 1-25 involve a six-membered cyclic transition state, while (b) a five-membered one 1641. Enantioselectivity depends on preference for one of two competing diastereomeric transition states. In example (a), the more favored one of them, i.e. attack at the Re face, is shown. Attack at the Si face would force the phenyl group into an axial position, which is less favored. Example (b) is out of line, since here not a chiral center is sacrificed for the creation of another one, but a double bond of a given diastereomeric configuration is stereoselectively converted to a pair of chiral centers. The products are of course racemic.

Fig. 1-23. Stereoselective synthesis by intramolecular transfer of chirality.

1.4.3 Stereoselective Synthetic Strategies 1.4.3.1 Enantioconvergent Synthesis The synthesis of optically active compounds containing several chiral centers from an achiral starting material poses serious problems of economy, even at a laboratory scale. Apart from less than quantitative chemical yields, at least half of the material is lost when generating the first chiral center, while losses in the creation of the other chiral centers are rather unpredictable owing to the diastereoselective character of the

1.4.3 Stereoselective Synthetic Strategies

a

t

b

a f b

-C

*

a

c*

C*

y:02Me Ho D

O

b

X

D2

R*=

t

H

I R*COCI

FDW 2) OH'

(1R,2R)-(25) Fig. 1-24. Enantioconvergent synthesis according to the Fischli scheme.

37

38

1.4 Selectivity in Chemistry

transformations. At least one highly enantioselective step or resolution of enantiomers at some stage is unavoidable, and, since the more material involved the higher the costs due to labor, chemicals and energy, this operation has to be carried out at an as early stage as possible. Enantioselective synthesis, even if it produces some of the unwanted enantiomer, is certainly the more efficient approach, since the

QH !

aSOn

Li/liq.NH,

OH

Me

?H

iBu

i I

iBu M -e I

MeC(OEt)i,

A

?Et

Me

iBu

1

I I

A

JI

Fig. 1-25. Enantioconvergent synthesis according to the scheme of Cohen and Saucy.

1.4.3 Stereoselective Synthetic Strategies

39

maximum yield of resolution cannot be higher than 50%. If the chiral compound obtained either by resolution or enantioselection contains a single element of chiralit y it may be possible to racemize the unwanted isomer, whereas the racemization of diastereomers is only very rarely possible. In 1975 Fischli et al. [65] proposed an ingeniously simple scheme by which a starting material with enantiotopic groups can be converted completely to a predetermined enantiomer. The original scheme is shown in Fig. 1-24 (a). Its essence is the reversal in the order of transformations. First, a bifunctional prochiral starting material is reacted with one equivalent of a chiral auxiliary ( C * ) to give a pair of diastereomers (0,and D2). These must be separated, and then the unreacted group (X) is transformed to group B in diastereomer D,and to A in D2. Removal of the chiral auxiliary group and transformation of the liberated functional groups to groups A and B, respectively, gives, in both cases, the same enantiomer. Interchanging the order of transformations at both branches provides an antipodal end product. In practice the scheme is fraught with considerable difficulties, and in the original paper no perfect example for its realization was presented. Apart from the problem that synthetic operations cannot be interchanged at will, it is not always easy to establish the absolute configuration of DI and D2 which is essential for the choice of the correct order of transformations. A relatively simple enantioconvergent scheme, somewhat different from the original proposition, was reported by Terashima and Yamada [66] and is shown in Fig. 1-24 (b). Note that any of the intermediates can be readily converted to any enantiomer of the end product (25). The synthetic sequence of Cohen et al. (Fig. 1-25) [67], which leads from a pair of enantiomeric alcohols, obtained by resolution, to a single enantiomer corresponding to the side chain fragment of tocopherol, features two important steps: (i) Diastereoselective reduction of a triple bond to either a (2)-or an (E)-olefin. It is the configuration of the olefinic bond which is traded-in in the next step for a chiral center. (ii) Chirality transfer in a concerted sigmatropic process, i.e. the Claisen-rearrangement. An interchange of the reduction methods would give the (R)-(23)isomer from both starting materials.

1.4.3.2 Selective Preparation of Both Enantiomers from a Single Substrate Often there is a need for both enantiomers of a compound, or only one is required but it cannot be foreseen which will emerge as the major product from a stereoselective reaction. Since resolution is not always practical, methods have been devised by which a chiral substrate can be converted to both enantiomers of a product or which can be conducted in a way that the configuration of the product could be controlled. Both approaches can best be illustrated by examples (Fig. 1-26). Example (a) [68]

40

1.4 Selectivity in Chemistry

OPh

OPh

O//p\--OEt

O//p,--OMe

I

I

0 Me

OEt

(S)

( R)

Ph 2I ) ) Mel, LDA, -98'C -78°C

Me-CHr--(o~ .\,/Me

Fig. 1-26. Selective preparation of both enantiomers from a single chiral substrate.

resembles the Fischli scheme inasmuch as the configuration of the major product can be controlled by the order in which the different alcohols are applied. Diastereoselectivity in this reaction is moderate, (+)- and (-)-esters are formed in a ratio of 2 : 1. Meyers and his coworkers have described several methods by which both enantiomers of an end-product can be prepared. In example (b) [69] a chiral oxazoline is first alkylated (for details c$ Fig. 6-16) with Me1 and then with BuI. Alkylation with BuI first and then with Me2S0, gave the enantiomeric @)-acid in 70% ee.

1.5 Kinetics and Thermodynamics of Stereoselective Reactions

41

1.5 Kinetics and Thermodynamics of Stereoselective Reactions Before concluding this introductory chapter it is appropriate to restate some well known facts and concepts of reaction kinetics and thermodynamics as applied to stereoselective reactions. In reactions giving rise to more than one product, the product ratio may be controlled either by the relative rates of formation (kinetic control) or by the equilibrium constant of the products (thermodynamic control) or both. The contribution of the two effects depends on the relative magnitudes of the activation free energies for the formation and equilibration of the products. When equilibrium is reached, the ratio of products, in our case of stereoisomers, is determined by the equilibrium constant K, which in turn depends on the ground state free energy difference of the products:

A

+

X

*

Y ; K = [Y]/[X]-RTInK= G",G'$=

AG&

At the other extreme, when the equilibration between products is infinitely slow as compared with their rate of formation, i.e. under pure kinetic control, the product ratio is determined by the relative concentrations of the activated complexes, which are related to their free energies by an analogous equation: Y'

Y-A-X

X*

Since both X and Yare derived from the same substrate, free energies can be replaced by free energies of activation (AG * = G * -Go). Thus the correlation of the free energy with both thermodynamically and kinetically controlled selectivity can be expressed by a common graph (Fig. 1-27). This graph should be kept in mind whenever constructing stereochemical models for the explanation of preference for a certain product. Free energy differences at low selectivities are very small and fall into the range characteristic for solvation, conformational changes and other unaccountables. Consequently the predictive power of stereochemical models constructed for reactions of low selectivity is very limited, and since they usually lack any physicochemical or spectroscopic evidence except the observed product distribution itself, they may not point to the real motives of selectivity. Therefore in this book stereochemical models will be employed only in a few justified cases, mostly in conjunction with reactions of high selectivity.

42

1.5 Kinetics and Thermodynamics of Stereoselective Reactions Product ratio (%Y/%X)

52.0 I .oo 11.52.0 , I

5

20

10

lo3

lo2

50

AGO o r AAGx Y

4

- 15

-m

- 10

y 2 - 5

I

excess of major product (%Y-%X) i 1

1

1

I

I

I

I

o l n o m o m o m

I 0

I IA

C

w

2 m

m I

I

o

I

.

2 o.

% of major product (Y)

Fig. 1-27. Relationship between free energies (AGO for equilibria and AAG * for kinetically controlled reactions) and product distribution at 25 " C .

The above relationships are based on the assumption that one single substrate is transformed to two or more products. This holds for conformationally rigid substrates but it is an oversimplification for those which can exist in several conformations. Such systems behave as a mixture of substrates and in situations when, for stereoelectronic or other reasons, one of the products can only be derived from a certain conformer and the other from another conformer, the question about the influence of the conformational equilibrium on product distribution inevitably arises. By simple argumentation Curtin and Hammett demonstrated [70] that for such systems product distribution only depends on the difference of the free energies in the transition state, i.e. : In ([ Y ] / [ X ]=) -€@(G;

-

-

G Z ) = - W A G ;x

Rr

RT

The energy diagram for this situation shown in Fig. 1-28 and the Curtin-Hammett principle itself invites several comments: (i) Product distribution and ground state stability are, in principle, unrelated. A less stable conformer may lead to a transition state of smaller energy. This is not an unlikely case since free energies of activation for chemical transformations are generally much larger than free energy differences between conformers (if A G g y > 11 kJ.mol-' the minor conformer is practically undetectable). A typical example is the rhodium-chiral phosphine-catalyzed enantioselective hydrogenation

1.5 Kinetics and Thermodynamics of Stereoselective Reactions

G:

43

-

A d A d

G'>

-G ;

Fig. 1-28. The Curtin-Hammett principle.

of 2-acylaminoacrylic acids (cJ Section 2.1.1.5) which proceeds via an undetectable minor intermediate. (ii) If one has good reason to assume that the free energies of activation for the competing products (AG; vs. AG;) are very close to each other, then the product distribution can be predicted on the basis of conformer distribution. Thus, in the hydrogenation of some macrocyclic exo-methylene ketones the ratio of cis and trans products was predicted by Still and Galynker [71] on the basis of the calculated conformational distribution of the substrate (Fig. 1-29).

-

+A'-- -

- \

u

93: 7 (calc)

k

0

Me

4e

/

91 : 9

Me

tmns

K4e CIS

Fig. 1-29. Example for the control of product ratio by substrate conformation.

(iii) The same as in (ii) applies when AG;, is of the order of AG *. In this case it is unlikely that the lower stability of the minor conformer could be compensated for by a smaller free energy of activation.

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

2 Stereoselective Catalytic Reductions 2.1 Stereoselective Homogeneous Hydrogenations with Rhodium-, Ruthenium- and Iridium-Phosphine Catalysts * The development of stereoselective homogeneous phase catalytic hydrogenation methods is one of the glorious chapters of synthetic organic chemistry. The role played by rhodium complexes was overwhelming, and therefore most of this chapter (Sections 2.2.1.1 to 2.1.1.5) will be devoted to them. The “rhodium-phosphine story” is briefly as follows: In 1966 Wilkinson and his coworkers discovered that chlorotris(tripheny1phosphine)rhodium, a complex soluble in apolar solvents (e.g. benzene), can be used as an efficient hydrogenation catalyst [2]. Not much earlier, methods for the preparation of optically active phosphines with phosphorus as the chiral center were developed by Horner and Mislow [3 - 51. The idea of replacing triphenylphosphine in the Wilkinson catalyst by a chiral phosphine, notably by methylpropylphenylphosphine ( I , Fig. 2-l), was reported independently by Knowles and Sabacky at the Monsanto laboratories in the USA [6, 71 and by Horner and his coworkers in Mainz, Germany [8]. Enantioselectivities achieved with 2-phenylacrylic acid and 2-phenyl1-propene as substrates were low (15 and 8% ee resp.) but promising. In 1969 Abley and McQuillin improved the record to more than 50% ee with a rhodium catalyst prepared by reduction of a Rh(II1) complex in a chiral amide solvent [9], but this method could not be developed further [lo]. Morrison et al. [ I l l were the first to recognize that phosphines achiral at phosphorus but carrying a chiral substituent may also serve as ligands for enantioselective hydrogenation. Meanwhile, much progress was made with P-chiral phosphines as well. Replacement of propyl by o-anisyl(2, PAMP) enhanced enantioselectivity to 50 - 60%, while an exchange of phenyl against cyclohexyl led to a ligand (3, CAMP) which was al-

* For

a monograph on homogeneous catalysis with rhodium and iridium catalysts see ref. [ l ] .

46

2.1 Stereoselective Homogeneous Hydrogenations

ready eligible for practical application, since it gave 80- 88% ee 1121. The search for better catalysts became more efficient when it was recognized that the best substrates for rhodium-catalyzed enantioselective hydrogenations were 2-acyl-aminocinnamic acids. The best among Fchiral ligands proved to be a diphosphine (5, DIPAMP, Fig. 2-1) [13]. The discovery that biphosphines are more selective catalysts than monophosphines can be credited to Kagan and his group [14, 151, who developed one of the most effi-

heR

P ---__

Ar'

.Ar P ' L

I

Me (S) - (1) (S) - (2) (PAMP) (S) - (3) (CAMP)

(9Ar = Ph ((-)-DIOP) (7)Ar

Ph

( S , R ) - ( 4 ) (PPFA)

Z-MeOW Z-MeO-

= 3,5-Me,-4-MeOC6H6

cHex

+IR Ph,

(DIOP-M)

Me

[19,201

R P(cHex), Me

P(cHex&

(10) ~ 4 1

(12) R = iPr [17]

( R , R ) - ( 5 ) (DIPAMP) Ar = 4-MeOCaH,

2.1 Stereoselective Homogeneous Hydrogenations

MeY PPh2

47

I

(CH2)2

MeAPPhl

PPh, (16) (CHIRAPHOS) n = 0 (17) n = 1

PPhz

PPh,

5

(18) ((R)-BINAP)

PPhz 0 9 ) (DIOXOP)

(qp-.:,

‘PPh,

R

(2/)R=Et[17] (22) R = iPr [ 161

NHPPh, NHPPh,

PPh,

(24) (PROPHOS)

Fig. 2-1. Chiral phosphine ligands.

cient ligands (-)-DIOP (6) which could be easily prepared from natural tartaric acid in five steps. By this time the period of fundamental discoveries was more or less over, and researchers turned to working out the details. The literature became flooded with papers on new ligands, only very few of which proved to be superior to existing ones. Also, intensive research began on the mechanism of the reaction, which shed light on many important details of the process, but still failed to explain the very high selectivities achieved with certain rhodium-chiral phosphine catalyst systems. One major deficiency of the method, which may be inherent to highly specialized systems in general, is that the spectrum of substrates which can be hydrogenated with high selectivity, not only with a given catalyst but also with the whole class of chiral rhodium-phosphine catalysts, is very limited. The monopolistic position of rhodium(1) as complexing metal is also surprising, in view of the enormous amount of research into organic transition metal complexes in recent decades. In the past few years ruthenium has become a strong competitor to rhodium, especially in enantioselective ketone reduction, while the use of iridium is still very limited.

48

2.1 Stereoselective Homogeneous Hydrogenations

It is not easy to separate essential from inessential when faced with the bewildering number of papers published on this topic. First we briefly describe the ligands which proved to be useful in the long run. Among the innumerable ligands described in the first edition of this book and reported since, only a handful are in general use. This is followed by a description of the most important experimental aspects and the range of substrates amenable to enantioselective hydrogenation. Finally, we summarize the present state of knowledge concerning the mechanism of the reaction.

2.1 1 Hydrogenation of Olefinic Bonds 2.1. .I Chiral Phosphorus Containing Ligands Ligands containing one or two phosphorus atoms of the phosphine, phosphinite and aminophosphine type, with chirality at phosphorus or carbon have been synthesised. Chiral ligands may also be incorporated into a polymer. Altogether, the preparation and testing of well over a hundred different ligands has been reported [25]. Most of them, however, have been applied only with a few standard substrates, most often with (2)-2-acetamido- and (2)-2-benzamidocinnarnic acid, their methyl esters, 2-acetaminoacrylic acid, and itaconic acid. Comparison of the selectivity of different chiral ligands is not easy, since both the set of test substrates and the experimental conditions (solvent, preparation of catalyst, temperature and pressure) are widely divergent. Also, a decline of ee with increasing conversion has occasionally been observed. Selected examples of enantioselective hydrogenation of typical substrates are compiled in Table 2-1, where prevailing configurations obtained with selected substrates and rhodium(1) catalysts prepared using the ligands shown in Figs. 2-1 -2-5 are listed. ee Values under optimum conditions are quoted, not only because data obtained under identical conditions were rarely available, but also because the main purpose of this table is to give information about the practical usefulness of the individual ligands. Namely, being highly specialized systems, the performance of rhodium-chiral phosphine catalysts is very sensitive to reaction conditions. Very few of the ligands tested gave satisfactory selectivities. It is generally accepted that ee values in excess of 80% are necessary to secure by recrystallization acceptable yields of an optically pure end-product (c$ Section 1.4.2). In addition, several practical requirements have also to be considered, such as ease of preparation, availability and optical purity of the starting material, chemical stability etc. At present CAMP ( 3 ) , PPFA ( 4 ) , DIPAMP ( 5 ) , DIOP ( 6 ) , CHIRAPHOS (16), DIOXOP (19),some members of the PPM family, such as BPPM (13) and BINAP (18) can be seriously considered as ligands for enantioselective hydrogenations.

49

2.1.1 Hydrogenation of Olefinic Bonds

Table 2-1. Enantioselectivities in the Hydrogenation of Standard Substrates with Chiral RhodiumPhosphine, -Phosphinite, and -Aminophosphine Catalysts under Optimized Conditions. (AcNH-CA = (Z)-2-acetamidocinnamic acid, BzNH-CA = (2)-2-benzamidocinnamic acid, AcNH-CE = methyl-(Z)-2-acetamidocinnamate,AcNH-AA = 2-acetamidoacrylic acid, ITA = itaconic acid) ~

Ligand Substrates AcNH-CA Vo ee (Conf.)

BzNH-CA '70 ee (Conf.)

AcNH-CE Vo ee (Conf.)

AcNH-AA Vo ee (Conf.)

ITA Vo ee (Conf.)

Monophosphines 3 89 ( S ) Diphosphines 4 67 (S) 5 96 (S) 6 7 10 12 13 16 17

18 19 24

81 ( R ) 98 ( R )

91 89 98 84 90 91

~~

References

70 ( R )

(R) (R) (R) (S) (S) (S)

69 ( R )

Prepara- Applition cation

12

12

44 26

45, 46 6, 26, 27 39 20 24 16 48-51 28, 29" 30 31, 33 41 -43 30, 35, 36

37, 38 20

73 (R) 96 ( R ) 93 ( R ) b 93.6 ( S )

16 47 28, 29 a 30 31 -- 34 40 30

" Refers to (R,R)-(16). For the methyl ester.

One of the reasons rhodium-phosphines are extremely active is that they are soluble in the solvents used for hydrogenation. Since they are also very sensitive to air, their solubility makes it practically impossible to recover them unharmed. Therefore, many attempts have been made to combine the activity and selectivity of soluble catalysts with the ease of recovery of heterogeneous phase catalysts [52-611 (Fig. 2-5). Although much interesting information was gained from the study of such polymer-supported catalysts, from a practical point of view they have not yet fulfilled expectations. A unique and selective non-phosphine ligand was prepared by linking N-3-[triethoxysilyl-propyll- proli line amide to zeolite [62, 631. Enantioselective hydrogenations with ruthenium catalysts were of marginal interest [64] until Noyori and his coworkers did not introduce ( R ) - and (S)-BINAP as the most suitable ligands [65]. Ruthenium catalysts for homogeneous hydrogenation are almost exclusively prepared with BINAP and usually applied as the BINAPRu-dicarboxylate complexes such as e.g. the diacetate 25 [66, 671.

50

2.1 Stereoselective Homogeneous Hydrogenations

The drawback of ruthenium-based catalysts is that they require high pressure and sometimes elevated temperature. Therefore their best field of application is where rhodjum catalysts d o not excell, i.e. in carbonyl reductions (see Section 2.1 .I .3). In the hydrogenation of dehydro-acylamino acids ruthenium catalysts are inferior to rhodium complexes [68]. An exception is the ionic complex 26 and its congeners, which are highly enantioselective catalysts [69, 701. Iridium diphosphine complexes are structurally completely analogous to the rhodium complexes but more stable and are useful in diastereoselective hydrogenations (see Section 2.1 .I .4). A mixed rhenium-rhodium chiral complex gave unexceptional enantioselectivities with standard substrates [71]. Rhodium-phosphine catalyst systems The classical Wilkinson catalyst, i.e. [Rh(Ph,P),Cl], has a square planar structure. By analogy, rhodium monophosphine catalysts were formulated as 27 (Fig. 2-2), where P* represented the chiral monophosphine. 27 may be in equilibrium with its dimer in which the P*-Rh ratio is only 2: 1. On addition of hydrogen 26 is supposed to be transformed to a coordinatively saturated (1 8 electron) octahedral species such as 28. This is in equilibrium with a solvated species (29) and the substrate-catalyst complex (30). The dihydride complex 28 could be directly observed by 31P-NMR [72]. CI H

C‘hkP* I H‘

P*’

P*

P*

h : P*

P*’

-

p*>/cl>/p* P* C ‘I

H clA/p* ‘ P*’ I H Sol

+

H

+ P*

\

2p*

‘P*

CIA/

P*

p*P + Sol

(29) (30) Fig. 2-2. Interaction of rhodium-phosphine catalysts with solvent, substrate and hydrogen.

2.1.1 Hydrogenation of Olefink Bonds

51

From a practical point of view rhodium-phosphine complex catalysts are applied in two forms. The catalyst is either prepared in situ in the hydrogenation solvent, or in an isolable ionic form. Neutral “in situ” catalysts are prepared by mixing, with the rigorous exclusion of air, a rhodium-olefin complex, such as [RhCl(cycl~octene)~]~, [RhCl( 1,5-cycloo ~ t a d i e n e ) ] ~[RhCI(norbornadiene)1,, , with an excess of the diphosphine. There is an optimum for the Rh: P ratio which varies with the ligand and other parameters but is generally around 1 : 3. On hydrogenation the olefinic ligand becomes saturated, loses its affinity to the metal and is displaced by the solvent and/or the diphosphine. The isolable cationic catalysts, e.g. rhodium( 1,5-~yclooctadiene)(-)-DIOP tetrafluoroborate (31) [73], are made by mixing one of the above mentioned neutral diene complexes with a slight excess of the phosphine and adding an equivalent of NaBF,. This catalyst is air sensitive, too. In such complexes both ligands must be cis coordinated. It should be noted that not every ligand is suitable for the preparation of ionic complexes.

Differences in rate and enantioselectivity between ionic and neutral complexes of the same ligand have often been recorded [e.g. 74, 751. Catalyst preparation is only one of the many experimental variables which have to be optimized for maximum selectivity. As will be seen later, the search for useful general correlations is a frustrating undertaking. Correlating ligand structure and enantioselectivity is an arduous task too. So far, two points have become firmly established: (i) chirality at phosphorus is not necessary for enantioselectivity, and (ii) diphosphines are more efficient ligands than monophosphines. Since, except for CAMP ( 3 ) and PPFA ( 4 ) , no monophosphine ligand can qualify for practical application, we shall restrict the present discussion to diphosphines. Diphosphines efficient as ligands form five- or seven-membered chelate rings with rhodium, and very efficient catalysts can be found in both groups. The conformation of the chelate ring is of crucial importance, since this is the lever by which the chirality of the ligand is transmitted to the substrate. A five-membered chelate complex of rhodium, even with the achiral bis-diphenylphosphinoethane, is intrinsically chiral, although mirror image conformations of C, symmetry undergo fast interconversion (Fig.2-3 (a)) [76]. In the case of a chiral ligand, one of the conformations may be highly preferred, and thus a stable chiral environment is provided for complexation with the prochiral ligand. Very few efficient chiral phosphines forming six-membered chelates have been described. Comparison of analogues forming five- and six-membered chelates is some-

52

2.1 Stereoselective Homogeneous Hydrogenations

Ph

(a )

CH 2

\

\

CH2

Ph

Ph

Ph

(b)

c":!, P

Me M @ F Ph.p Ph

(32)

\\

Ph

Ph

Rh

(331

Fig. 2-3. Conformation of rhodium-phosphine complexes forming five- and six-membered chelate rings.

what bewildering. CHIRAPHOS (16) and its homologue (17) both give a highly selective catalyst (ee>89%), while the homologue of PROPHOS ( 2 4 ) is a poor ligand. For an explanation it was assumed [76] that for high enantioselectivity the chelate ring must take up a twist-chair conformation (Fig. 2-3, 32), which provides an antisymmetrical orientation of the phenyl rings as is the case with the complex of diphosphine 17, while the compound with one methyl group less should be in a chair conformation with mirror image symmetry for the phenyl rings (33). Diphosphines forming seven-membered chelates constitute by far the most populous group and the majority of highly efficient catalysts belongs here. Selective ligands are generally cyclic compounds, a notable exception being the binaphthyls e.g. 10 and 18 and ferrocenes such as 8 and 20. According to an analysis by Brown and his coworkers [76], two chiral conformers should be considered for the seven-membered chelate ring, the chair (C) and the twist-boat form ( T B ) (Fig. 2-4 (a)). (One or both of the atoms marked with a dot are substituted.) The most important difference between the two conformations is that the quasi-axial and quasi-equatorial dispositions for the pro-R and p r o 4 phenyl groups are interchanged, and the helicity of the CH, - P- Rh - P- CH2 sequence is inverted (Pin C and M in TB). Since the latter is the very feature of the complex which controls enantioselectivity, prediction of the configuration of the product in excess may become feasible. This requires that both the prevailing conformer of the ring and the mode of substrate binding should be known. According to the empirical rule suggested by Brown, conformer ( P ) - C should give @)-amino acids from (2)-enamides, while conformer (M)-TBprovides the R enantiomers. Note that the conformational equilibrium C @ TB is controlled by the constitution and relative configuration of the ligand, and helicity by its absolute configuration. Thus, for the antipode of the ligand the enantiomers of the conformers shown [ie. (M)-Cand (P)TB] should be considered.

2.1.1 Hydrogenation of Olefinic Bonds

53

Inspection of models suggested that (-)-DIOP and its carbocyclic analogues preferred the (M)-TBconformer. In fact, all give @)-amino acids in excess, and for the homologues selectivity decreases with ring size. Eight-membered and larger rings seem to be too flexible for effective transfer of their chirality to the substrate. No direct evidence concerning the conformation of diphosphine-rhodium complexes in the absence of substrate has surfaced so far.

(P)-C

( M ) -7'6

Fig. 2-4. Conformation of rhodium-phosphine complexes forming seven- and eight-membered chelate rings.

Influence of experimental conditions In the field of rhodium-phosphine-catalyzed hydrogenations any attempt to establish correlations between experimental parameters and enantioselectivity is a controversial undertaking, and the final conclusion from all what follows here is that each catalyst-substrate system has to be optimized individually. There is, however, one feature common to all of the catalysts, namely that, though not pyrophoric, all are sensitive to oxygen, since phosphines are readily oxidized to phosphine oxides.

Solvent effects Alcohols (mainly methanol), benzene, benzene-alcohol mixtures, tetrahydro furan, and even aqueous alcohols can be used as solvents. A catalyst prepared from a cholesteryl-phosphine (14) is active even in cyclohexane [77]. Solvent effects are rather unpredictable. For example, no or insignificant changes were observed in the hydrogenation of pulegone with Rh-DIPAMP (DMF or MeOH) [78] and of the standard substrates with Rh-PROPHOS (24) (THF and EtOH) [35]. Moderate ( < 15%) and substrate-dependent effects were recorded, e.g. for RhCHIRAPHOS (16) (THF and EtOH) [28, 291, but for many systems the effect was dramatic. Since it has been shown that solvent molecules are integral parts of the catalysts, it is the absence rather than the presence of a solvent effect that is surprising. No

54

2.1 Stereoselective Homogeneous Hydrogenations

general conclusions can be drawn from available data and even with the same catalyst, the optimum solvent may change with the substrate. Pressure effects Since one of the important steps in the hydrogenation process is the uptake of dihydrogen by the square planar Rh(1)-phosphine complexes, whereby they are transformed to octahedral dihydrido Rh(rI1) complexes, it can be anticipated that hydrogen pressure should be a crucial parameter. In heterogeneous catalytic hydrogenation of substrates with a single reducible group, the only effect of enhanced hydrogen pressure is a higher rate of reduction. Acceleration of hydrogen absorption under higher pressure is of course a welcome effect in rhodium-phosphine catalysis, but it is, unfortunately, often accompanied by loss of selectivity. However, it was found that the latter can be partly compensated for, although at the price of lower

%ec 90

50

-IV+B

0

.Il+A 50 I atm @

’\.I S A IV+C

-8 -50

IVfB

-90

Fig. 2-5. The effect of pressure on enantioselectivity in rhodium-phosphine-catalyzed hydrogenations. 0 Without base. 0 Et,N added. 0 I-Phenylethylamine added. Catalysts: I = BPPM, I I = (-)-DIOP, III = DIPAMP, I Y = DIOXOP. Substrates: A = (Z)-2-benzamidocinnamic acid, B = 2-acetamidoacrylic acid, C = methyl ester of B.

2.1.1 Hydrogenation of Olefinic Bonds

55

rates [79], by adding a base, usually triethylamine, to the system. Since higher pressure may often result even in an inversion of the preferred configuration, the pressure effect is clearly a manifestation of competition between two different mechanisms. Some characteristic data are given in Fig. 2-5. Contrary to expectation, added base has a stabilizing effect even when the substrate is an ester. For an interpretation of the pressure effect, c j Section 2.1.1.2.5.

Effect of temperature Very few data are available on this topic, and these show that selectivity often increases on lowering the temperature [80, 81, 821. Since the latter is usually accompanied by a sharp decrease in rate, this measure is seldom practicable. A study by Sinou [83] revealed that higher temperature may result in an increased selectivity too. Increase of selectivity on elevation of the temperature was also reported by Ojima et al. [ 5 2 ] . An interpretation of this phenomenon was proposed by Halpern [82] ( c j Section 2.1.1 S),but not for the opposite effect. Certainly both originate from a competition of at least two pathways.

Effect of added base Attenuation of the pressure effect by added base has already been mentioned, and from Fig. 2-5 it is also apparent that this is accompanied by a net increase in enantioselectivity. Triethylamine and other bases obviously act by converting the carboxylic substrates to the anions. How many moles of amine are optimal has again to be determined experimentally.

Substrate dependence of enantioselectivity Although rhodium-phosphine-catalyzed hydrogenation is one of the most elegant among methods to prepare chiral compounds in high optical purity, the range of substrates to which it can be successfully applied is quite narrow and it seems that its significant widening cannot be expected. This becomes apparent from Tables 2-2 and 2-3 and Fig. 2-6, in which some typical results obtained with substrates, other than those shown already in Table 2-1 are compiled. This survey contains a few simple prochiral olefins, some additional enamines and a series of 2,3-unsaturated carboxylic acids. Ketones will be discussed in Section 2.1.2. In the presence of the ruthenium catalyst 25 geraniol (n = 1) and homogeraniol (n = 2), but not nerol (n = 3) could be regio- and enantioselectively hydrogenated [971.

n n

= 1, = 2,

geraniol homogeraniol

n = 1, ee 96% n = 2. ee 92%

56

2.1 Stereoselective Homogeneous Hydrogenations

Table 2-2. Enantioselective Hydrogenation of a,P-Unsaturated Carboxylic Acids and Esters with Chiral Rhodium or Ruthenium Phosphine Catalysts. (Ru Complexes are Marked with *). (For Atropic Acid see Fig. 2-17)

H.p((2R4

H,

R'

RR M : 3 ° R2 b L *R * 4 R

R'

R2

R3

R4

L*

H H H

H H H

CH2C02H CH,CO,Me CH,CO,Me CH,COP CH2C02Me

H Me Me

DIPAMP DIOP-M 10

CO2H H

Me H Me H Me H

C02H Ph

H H H H H H Ph H H

Ph H Ph Pr Ph Me Me Me iPr

Me H AcO

Me Ph

F Me H Ph NHBz

Me H H H H H H H H

4-Cl-C,H, OAc H

H Et H

CH,CO,Me CH,C02H CH,C02H

F

Yo ee

BPPM BPPM DIPAMP DIOXOP DIOP-M (S)-BINAP * (S)-BINAP * (S)-BINAP * (S)-BINAP * (S)-BINAP * (S)-BINAP * (S)-BINAP * (-)-DIOP CHIRAPHOS PROPHOS PPFA DIPAMP DIPAMP

(Conf.)

ref.

38 91 99 78 88 55 88 90 10 3 33 65 1 58 (?)

88 90 90 56 91 85 93 65 83 87 85 90 < 10

Table 2-3. Enantioselective Hydrogenation of N-Acylenamines with Chiral Rhodium-Phosphine Catalysts.

R' CN Ph

Ph H Me

H

R3

L*

% ee

(Conf.)

Ref.

Ph Me

DIPAMP (-)-DIOP ( - )-DIOP ( -)-DIOP ( - )-DIOP ( - )-DIOP

89 45 83 >73 85 14

(S)

26 100 1 00 100 100 100

Ph iPr OEt

(R) (R1 ( R1 ( R1 (S)

2.1.1 Hydrogenation of Olefinic Bonds

Ph

57

NHCOR R H

~

~

Mc

CF3

iPr

~

(-)-DIOP

tBu

~

69 ( R )

58 ( R )

22 ( S )

R’

R2

L*

Ph

Et

Bu Ph Ph C0,Et

Et OMe CO,H OAc

C0,Et

CF3

(-)-DIOP CPPM CPPM (-)-DIOP (-)-DIOP (-)-DIOP DIPAMP BPPM (-)-DIOP DIPAMP

% ee (conf .)

~

0

15 ( R )

Ph ~

35 ( R )

Ref.

R’

R*

R3

R4

L*

% ee (conf.)

Ref.

Me NHAc NHAc

C0,Et Ph Me

NHAc H H

Et Me H

BPPM BPPM BPPM

57 (R,R) 37 (R) 55 ( S )

[@I I861 [861

Fig. 2-6. Enantioselective hydrogenation of olefins with chiral rhodium-phosphine catalysts. Data refer to To ee values, configuration of the product in excess in parentheses.

As a group of eminent practical importance not included in either place, substituted 2-acylaminocinnamic acids have to be mentioned. At Monsanto the hydrogenation of the 3-acetoxy-Cmethoxyphenyl derivative using the DIPAMPrhodium complex as catalyst has been developed to an industrial process for the manufacture of (S)-3,4-dihydroxyphenylalanine[@)-DOPA] [6, 991. Enantioselectivity is slightly sensitive to substitution at the aromatic ring of the substrate. This is not due to steric effects but rather to subtle changes in the coordinating ability of the double bond.

58

2.1 Stereoselective Homogeneous Hydrogenations

Application of enantioselective hydrogenation to a wider range of acylenamines is barred by low yields in the preparation of precursors with an aliphatic C-3 substituent. Correlation of substrate structure and enantioselectivity has been thoroughly investigated in the laboratories of Glaser, Brown, Halpern, and others. These studies have permitted the proposal of some cautions generalizations about the correlation of substrate structure and enantioselectivity. (i) A polar group attached to the double bond promotes selectivity. The effect can be attributed primarily to electron attraction, but sometimes also to coordination with the metal center [79]. Accordingly, hydrocarbons have given poor enantioselectivities with all the rhodium-phosphine catalysts tested so far. (ii) Carboxylic acids are generally preferred to esters, amides or nitriles, although occasionally esters may be equivalent to (e.g. with DIPAMP) or much better than the acids (e.g. with BPPFA [46]). On one occasion inversion of the configuration on changing to the ester was observerd [101] (Fig. 2-6 (a)). (iii) Usually N-acetyl- and N-benzoylenamines are employed as substrates, and generally ee differences between the two are marginal. There are, however, cases in which one of them (more often the benzoyl compound) performs much better. Results on a series of methyl 2-acyl-aminocinnamates (Fig. 2-20 (a)) obtained with the DIOP-rhodium complex [I021 prompted Glaser and Geresh to come to the conclusion that selectivity drops with increasing bulk of the N-acyl group. Once again the validity of the correlation appears to be of rather limited scope, as shown by data obtained with other catalysts. Sensitivity to the nature of the N-acyl group is not surprising, since it has been shown (cJ Fig. 2-8) that the amide carbonyl is coordinated to rhodium in the substrate-catalyst complex. (iv) Concerning esterifying alcohol some catalysts (e.g. DIOP 11031 are rather indifferent in this respect, while with others a bulky alcohol enhances selectivity [loll. Although it is believed that the ester group is not bound to the metal in the substrate-catalyst complex, significant double induction was observed in the hydrogenation of (2)-(-)-menthyl-2-acetamidocinnamate: ( + )-DIOP afforded 76.8% ee, while (-)-DIOP only 52.5% [104]. (v) Scott et a/. subjected a series of methyl (2)-3-substituted-2-acetarnidoacrylates to hydrogenation with four efficient catalysts. No logical pattern emerged from their results [ 2051, except for a vague indication that ligands forming five-membered chelates are less susceptible to substrate variation than those giving seven-membered chelates. (vi) Although of the stereoisomeric 2-acylamino-3-arylacrylic acids only the 2 form is readily accessible and with the 3-alkyl analogues both diastereomers are difficult to prepare pure, in order to elucidate substrate-catalyst interactions it was of interest to compare the behavior of stereoisomeric (E and 2 ) substrates. It was observed with DIPAMP that in alcohols (Z)-2-acylaminocinnamic acids and esters were hydrogenated both much faster (by a factor of 16- 100) and more selectively than the corresponding @)-acids [26] (e.g. 94 vs. 47% ee for benzoylamino acid). One of the reasons why the E isomers are inferior to 2 ones may be their different mode

2.1.1 Hydrogenation of Olefinic Bonds

59

of complexation. Thus (E)-2-benzamidocinnamic acid was shown to bind with the carboxy CO to rhodium, while with the 2 isomers the amide CO was coordinated [ 1061. Since it was recognized that rhodium-phosphine catalysts also promote E - 2 isomerization of the substrate, the real significance of these data remained obscure until the isomerization process could be monitored. This was accomplished in an elegant way by Koenig and Knowles [I071 who reduced E and Z substrates with dideuterium, thereby creating a second chiral center at C-3. Thus different diastereomers arose from the E and 2 isomers, and these could be identified by NMR (Fig. 2-7). D

*

D2

Rh-P

'"fl ,

,C02H

H'

(El

NHBz (2R,3S)

p -

D NHBz

Fig. 2-7. Identification o f substrate configuration ( E or 2 ) by the addition of dideuterium.

These studies revealed the following interesting facts: (a) Stereoselectivity for (E)-2-benzamidocinnamic acid, when corrected for isomerization, was even less (30%). (b) No isomerization took place in benzene in the presence of any of the four catalysts examined. (c) In benzene, enantioselectivities for the E isomer were only marginally lower than for the Z isomer. The existing chiral center in compounds of type 34 has been found to play a secondary role; it modifies diastereoselectivity but does not control the prevailing configuration at the new chiral center. For the hydrogenation of dipeptide precursors, DIOXOP is the ligand of choice since it gives products with the natural configuration. Unfortunately, the most efficient ligand (BPPM) gives the unnatural configuration. The synthesis of its antipode is known [47] but extremely laborious. CONH -?H-C02R3

/

6

Ph

N HCOR'

II

R'

Diastereoselective hydrogenation, of course, does not require that the ligand should be chiral. Thus N-acetyldehydropeptides can be hydrogenated with the rhodium complex of (Ph2PCH2)2CH(CH2)2NMe2 giving de values of 78 -98% [98].

60

2.1 Stereoselective Homogeneous Hydrogenations

In summary, the effects of substrate and catalyst structure, as well as of experimental parameters are rather complex and, as will be seen in the next section, only part of them can be rationalized by the mechanism of the reaction.

Mechanistic considerations

It has often been stated that, as regards efficiency, rhodium-chiral phosphine catalysts emulate enzymes. Their capricious response to variations of substrate structure and reaction conditions is also reminiscent of enzymes. No wonder that the elucidation of their mechanism is a formidable task, and after the publication of well over 40 research papers devoted partly or completely to this subject the problem has not yet been solved completely (for a review see ref. [76]). Mechanistic studies have been carried out with the rhodium complexes of almost all of the more important ligands such as DIPAMP [26, 84, 89, 1081, CHIRAPHOS [109-Ill], DIOP [84, 111, 1121, DIOXOP [83, 1131 and BPPM 1114, 1151. Before the origin of stereoselection could be unraveled, some fundamental knowledge about hydrogenation with rhodium-diphosphine ligands had to be accumulated. For this purpose Halpern and his coworkers studied the hydrogenation of methyl-2-acetamidocinnamate (MAC) catalyzed by the rhodium complex of an achiral ligand, 2,2-bis(diphenylphosphino)ethane (DIPHOS) [I 10, 116- 1181. Their findings are summarized in Fig. 2-8. The ionic diene-adduct catalyst precursor (CP) enters the catalytic cycle by taking up two moles of H,. On saturation the diene loses its affinity to the metal and is replaced by solvent molecules. The solvated catalyst is an unstable species whose structure was derived mainly from 31P-NMR data. When isolated, it transforms to a binuclear complex containing no solvent [17]. On addition of the substrate a complex ( C - S u ) is formed at high rate in which the substrate is coordinated by its double bond and amide group. The structure of C-Su has been established both by NMR and X-ray crystallography [I 191, and this coordination pattern proved to be quite general for the 2-acylaminoacrylic acid and -cinnamic acid and ester complexes of other catalysts, too [76, 108, 110, 111, 120- 1231, but is not necessarily valid for the corresponding anions [51, 1241. By the way, coordination of the carboxyl group would involve a four-membered chelate ring, and this is unfavorable. At room temperature, oxidative addition of H2 to C-Su giving the species C-Su-H2 was the rate-determining step, and therefore there was no chance of detecting any of the downstream intermediates at this temperature. Due to its higher activation enthalpy, the product-forming step becomes rate-limiting at low temperature and, in fact, at - 75 'C the hydridoalkyl complex C- Su H - H accumulated and could be characterized by NMR [I 181. Brown et al. studied the more stable iridium complexes by 3'P-NMR and were able to observe at -70 'C an analogous complex [125], while Halpern et al. were able to detect diastereomeric hydrido complexes of DIPAMP at -35 "C [126]. With iridium-DIPAMP complexes of methyl 2-N-benzoylaminocinnamate the ratio of

2.1.1 Hydrogenation of Olefinic Bonds

61

k-1

S

Me

c --su

[H2]

J

ki

CHzPh

<

k' S

Me C--SuH-H

C---Su -H2

Rate constant

k (25°C)

AH' (kJ . rnol-I)

AS' (J . rno1-I K-I)

k' (mol-' . sec-I) k-l (sec-') k2 (rnol-' . sec-I) kl (sec-') k4 {sec-lj

14000 0.52 100 >1 23

26

17

+8.4 -117

71

+25

Fig. 2-8. Kinetics of hydrogenation with a rhodium-phosphine catalyst containing an achiral diphosphine.

diastereomeric complexes was 1 : 1, but one of them reacted faster with hydrogen [125]. It should be borne in mind that the scheme shown in Fig. 2-9 leads exclusively to an R product, but, since this is only one of the two possible schemes related as mirror images, the actual product is racemic. Formation of the hydrido-alkyl complex C- Su H - H is irreversible, as demonstrated by deuteration experiments with BPPM as catalyst [5 I].

62

2.1 Stereoselective Homogeneous Hydrogenalions

+

ACozR

Ph

NHCOMe

N HCOMe

su

c*- sus w

Me

I

Fig. 2-9. Hydrogenation of an alkyl (2)-2-acetarnidocinnamate with a catalyst containing a chiral diphosphine of C , symmetry (PAP) (S = solvent).

2.1.1 Hydrogenation of Olefinic Bonds

63

After the above results the stage was set for studying the hydrogenation of the same substrate with a chiral rhodium-diphosphine catalyst. The situation is less complex for chiral ligands with C, symmetry, such as CHIRAPHOS or DIPAMP, and is shown in Fig. 2-9. Now, because they are no longer mirror images, both possible modes of coordination have to be drawn up, giving two unequally populated sequences (R and S ) . Experiments with rhodium-@, S)-CHIRAPHOS, which hydrogenates ethyl (2)-2acetamidocinnamate (EAC) with over 95% ee to the @)-amino acid, furnished an astonishing result, namely, that the predominant catalyst-substrate complex ( > 95 Yo) was the one which was supposed to lead to the S product, i.e. C*-Su, [108, 1101. Analogous results were obtained with the Rh-(DIPAMP)-MAC system, where the minor diastereomer could be detected also [ 1 1 1, 1 16, 1261. Two conclusions may be drawn from these findings, namely, that either the proposed mechanism is wrong, or that the thermodynamically favored substrate-catalyst complex (C *- Su,) is kinetically disfavored and the main pathway proceeds via a hidden intermediate i.e. C*-Su,. Halpern opted for the second alternative, which could be supported by the following arguments. It has been mentioned that selectivity decreased with increasing hydrogen pressure. Since, being a bimolecular process, oxidative addition of hydrogen depends on H, concentration and on increasing the pressure, a point may be reached where the rate at which the olefin complex is formed becomes comparable to its rate of oxidation. Then k i / k k would gain control of the process, and this places the S pathway at an advantage. Interconversion of diastereomeric catalyst-substrate complexes has a much higher activation enthalpy than hydrogen uptake (see above). This means that on lowering the temperature the rate of interconversion declines much more rapidly than that of hydrogen uptake. This fact again favors the minor pathway ( S ) . An increase in selectivity on elevation of the temperature has been observed [51, 891; on one occasion it increased from 0% at 0 ° C to 60% ee at 100°C [83]. The foregoing rationalization is a striking demonstration of the Curtin-Hammett principle (cf. Section I S ) , but unfortunately no explanation based on structural features can be offered as yet for the discrepancy between stability and reactivity of the catalyst-substrate complexes. Since direct observation of both diastereomeric dihydride species had been unsuccessful, Brown et a/. modelled the situation by molecular mechanics calculations [127]. The substrate-olefin-Rh(1) complex leading to the minor product was found indeed more stable, while the opposite was found for the dihydride species. Equilibration of diastereomeric diphosphine-enamide-Rh complexes by full dissociation of the substrate was contrary to dynamic NMR studies carried out on Ph2P(CH2),PPh2 and DIPAMP complexes [ 1281. Halpern’s scheme is in accordance with many of the phenomena encountered in rhodium-phosphine-catalyzed hydrogenations, but does not apply to any substrate or any catalyst, and his views about the background of the pressure effect are not shared by certain authors. Based on a careful study of this effect involving several catalysts, substrates, and experiments when triethylamine was added to the system,

64

2.1 Stereoselective Homogeneous Hydrogenations

Ojima and his coworkers concluded that two competing pathways shown below are operative [ 5 11.

Rh

2

(6)

product

Another reason for a change in selectivity could be the acceleration of the C*-Su-tC*-Su-H2 transformation. Then it is the next step which becomes ratelimiting, and, if this is reversible, equilibration with a diastereomer of C*-Su- H2 leading to an enantiomeric product may take place. This possibility was excluded by showing that using D2 only a 2,3-dideuterio product was formed from 2-acetamidoacrylic acid, the substrate which showed the highest pressure-effect. It should be noted that the increase in selectivity on elevating the temperature may also be used as an argument in support of the dual mechanism [83]. As evidence for the emergence of pathway B at high pressure, it was mentioned that this was the established mechanism of olefin hydrogenation catalyzed by (PhP),RhCl. The crucial intermediate in the high pressure pathway (C*- H2- S u ) could be captured in the case of DIOP [I 131. Further support for Ojima’s hypothesis was invoked from the base-effect ( c j Section 2.1.1.3), a phenomenon which Halpern’s theory failed to explain. It was argued that the anion generated by the addition of triethylamine has a much greater affinity to rhodium than the corresponding acid, and thus diversion to route B is discouraged. No experimental proof for stronger binding of the anions to rhodium has, however, been presented. In order to determine in which step chiral recognition took place, hydrogenation of the dicarboxylic acids 35-37, all giving 2-methylsuccinic acid with an ionic rhodium-BPPM catalyst, was examined. When, according to the mechanism in Figs. 2-8 and 2-9, hydrogen is first transferred to C-3, and stereoselectivity is controlled by the structure of C*-SuH-H or in subsequent steps, then the outcome of the reaction should be indifferent to the configuration of the substrate. In other words 35,36 and 37 should all give the same product composition. Since this is not the case, chiral recognition must take place in the step of olefin complexation.

HozcYHozcYMe \

HOzC’

(351

(36)

COzH

(37 )

Few studies are available on the mechanism of ruthenium-phosphine-catalyzed hydrogenations. By deuteration experiments syn addition of hydrogen was established and with carboxylic acid substrates exchange with the original carboxylate ligand of the catalyst and a two-step transfer of hydrogen was postulated [129].

2.I.1 Hydrogenation of Olejinic Bonds

65

Hydrogenation of olefinic bonds with soluble catalysts containing other metals It should be pointed out that useful results with chiral complexes of metals other than Rh, Ru, and Ir are scarce.

Cobalt Cobalamin (Co *), the Co(r)-containing porphyrin complex contained in vitamin B,2, has been shown to be capable of mediating enantioselective hydrogenations (ee up to 79%) [130, 1311. Addition of a chiral amine to a bis(dimethylglyoximato)Co(rI) complex gives rise to a chiral complex capable of enantioselective catalysis 1132, 1331. Up to 34% ee could be obtained with such systems [134].

Titanium Cyclopentadienyl complexes of titanium catalyze the hydrogenation of olefins. Bis[$-( -)-menthylcyclopentadienyl]-titaniumdichloride is a unique catalyst in two ways: (i) It contains as a chiral ligand a hydrocarbon, and (ii) in contrast to other catalysts it only transfers hydrogen to hydrocarbons [15% ee ( S ) with 2-ethylstyrene] 113.51. Treatment of the sandwitch complex 38 first with BuLi and then with PhSiH, provides a catalyst with which a-substituted cyclic imines can be reduced with high selectivity [1361. (38), H,, 133 atm,

Ph R*O-+-OR*

I

Me ( 3 8 ) (R;= (R,R)-BINOL)

2)n

THF, 65 O C H n = 1, 2 e e > 9 7 %

66

2.1 Stereoselective Homogeneous Hydrogenations

2.1.2 Enantioselective Catalytic Hydrogenation of Ketones and Imines In this section reductions with hydrogen are described. Catalytic hydrosilylation, which leads ultimately to secondary alcohols as well, will be described in Section 2.2. Rhodium-diphosphine-catalyzed hydrogenations of ketones are much less satisfactory than those of amino acid precursors, and the method is not yet competitive with enantioselective reductions by chiral hydride reagents. The most important catalysts are again rhodium-chiral phosphine complexes, and an overview of the results obtained with a series of simple ketones is given in Table 2-4. The ligands have all been described before (cf. Fig. 2-1).

Table 2-4. Reduction of Prochiral Ketones (R'COR2) to Chiral Secondary Alcohols by Catalytic Hydrogenation with Chiral Rhodium-Phosphine Catalysts. R'

R2

070

Me

Et iPr t Bu Hex 1-Naph 2-Naph C0,Pr Bn CHPh2 Me

12 60 39 48 84 69 76 20 32 80

Ph

ee

(Conf.)

Ligand CAMP/iPrCO,H 39/C,H6 39/C6H6 39/C6H, (+)-DIOP + Et3N ( +)-DIOP + EtjN BPPM/TMF CAMP/iPrCO,H CAMP/iPrCO,H ( + )-DIOP + Et3N

Ref. 137 138 138 138 139 137 137 126 126 126 -

Enantioselectivities with ketones bearing no other functional group are generally poor [137, 138, 140, 1451. Results with ketoesters and aminoketones [22, 1461 are satisfactory. Thus with the N-methylcarbarnate analogue of 15 methyl pyruvate was reduced giving @)-lactate of 87% ee [21] and the precursor of propranolol, an important antiarrhythmic drug with 96% ee [146]. As with the hydrogenation of olefins, the best results were obtained with substrates containing two groups capable of coordination with the catalyst. Thus, the method was most efficient when applied to the preparation of pantolactone (40). With a neutral complex of BPPM in benzene the @)-lactone was obtained in 87% ee [141, 1421, and in a patent [I431 98.9% ee was claimed when the ligand was the N-phenyl analogue. 40 is a key intermediate in the industrial synthesis of pantothenic acid. Selectivities in the hydrogenation of waminoacetophenones to the (R)-alcohols with the ferrocenyl alcohol 20 as ligand were also high 60-95% [144]. The products are important intermediates in the synthesis of adrenergic drugs.

2.1.2 Enantioselecfive Catalyiic Hydrogenation of Ketones and [mines

PPh,

ph2p&

Met c

I

HO1 " I

o+co2Bn Bn

67

,b 0

( R ) - (44

(39

Several other ligands were tested for ketone hydrogenations, but they all proved to be less selective than the ones already mentioned. The most rewarding field of application of Ru-BINAP catalysts is the reduction of ketones. An attractive feature is that the prevailing configuration is predictable, the catalyst with (S)-BINAP giving usually (S)-alcohols. Thus ketones of the type R'COCH2R2(R' = Me, iPr, Ph, R2 = OH, CH20H, CO,Et, CONMe2, COSEt) give the corresponding alcohols of >92% ee [147, 1481 except for PhCOCH2C02Me (85% ee) [149]. In the case of non-enolizable a-substituted P-carbonyl compounds dia- and enantioselectivity is difficult to discern. Most probably there is a kinetic preference for one of the interconverting enantiomeric substrates, although epimerization at C-2 in the product cannot be excluded. Be as it is, enantioselectivity is generally outstanding, while diastereoselectivity is dependent on the a-substituent [70, 150- 1521.

R2 R' R* %de ofsyn

VQee of syn

I I

I

3,4-OCH,O-C6H,

Me

NHAc

NHAc

CH2NHBz

98

98

84

94

98

98

With a-substituted P-ketoesters enantioselectivity is high, but diastereoselectivity is poor [149], while with a-ketocycloalkane carboxylic esters (n = 5-7) enantio- and anti-diastereoselectivity are both high (> 86%) [153]. A wide range of 1,3-diketones give anti-diols, generally with excellent selectivity [147, 152, 1541.

I

Me

Me

Me

Me

Et

CH,CO,Me

Me

Et

iPr

Et

VQ de of anti

60

98

88

94

96

% ee of anti

78

z 99

94

98

96

R' R*

I

68

2.1 Stereoselective Homogeneous Hydrogenations

Very few examples for homogeneous phase hydrogenation of imines have been reported. Benzoylhydrazones were reduced with the rhodium complex prepared with ligand I1 [ 161.

~1

R2

%ee

1

1

1

Ph

C0,Et

C0,Et

cHex

Me

Me

Ph

Me

92

89

91

72

2.1.3 Diastereoselective Hydrogenations in Homogeneous Phase High enantioselectivities achieved with simple amino acid precursors stimulated the adaptation of rhodium-chiral-phosphine-catalyzed hydrogenation to dehydro diand tripeptide precursors. Most of these are in fact diastereoselective transformations. Here the first step is an enantioselective reaction, while the second one is subject to asymmetric induction. The influence of the existing chiral center has been thoroughly investigated on compounds of type 34. After saturation ee values were consistently lower when configuration of the existing center was R , but depending on substitution and catalyst, the difference varied in the range of 8 - 66% [155 - 1601. Hydrogenation of a,Pdisubstituted acrylic acids using ferrocene-based ligands, e.g. 9, was highly enantioselective but of varying diastereoselectivity [ 171.

R'

Ph

CD,

Et

Me*

R2

Me

Me

Me

Et

92

98

91

z 94

Voee

I

*At C-2

E.g. when R' = Me and R2 = Et, interchanging the substituents completely destroyed selectivity (2S, 3R/2S, 3s = I : 1) at C-3, while concerning C-2 de was still > 94%.

2.1.3 Diusteroselective Hydrogenufions in Homogeneous Phase

69

This finding supports that hydrogenation catalyzed by rhodium complexes is a twostep process (cf. Section 2.1.1.2.5) and not a concerted syn transfer as generally accepted for heterogeneous catalysis. Diastereoselective homogeneous phase catalytic hydrogenation using rhodium and iridium complexes, mainly of achiral phosphines has been studied in several laboratories and reviewed by Brown [161]. The classical Wilkinson catalyst has been soon superseded by the biphosphine rhodium complex 41 [I621 and the iridium complex 42 [163].

High selectivity requires an oxygen containing or other polar group, the best substrates being allylic and homoallylic alcohols, a , p and /I punsaturated , esters. Prediction of the configuration of the major product is usually straightforward, assuming binding of the polar functional group to the catalyst and approach of hydrogen from the same side. This is not unexpected with the homoallyl alcohols 43 (R' = H, R2 = Me or R' = Me, R2 = H) [162- 1641, but holds also with substrate 44, where hydrogen adds from the more hindered re face (de 97%; see Fig. 2-10). R

R

R = H, Me, iPr, de 94-99% Me

Me

(45) R = H, Me

(46) R = Me, Ph

Me

Fig. 2-10. Homogeneous phase diastereoselective hydrogenations with rhodium- and iridiumdiphosphine catalysts.

70

2.2 Catalytic Hydrosilylation

Selectivity is less when the functional group is not directly attached to a ring, as in 45 (de 70-80%) [163], i.e. if the link between the anchoring group and the olefinic bond is mobile. In the hydroxyesters 46 the directing effect of the hydroxy group dominates (de 94-98%) [165]. A hydroxyl or protected amino group in a-position can secure, however, even in acyclic substrates high diastereoselectivity [ 166- 1681. Use of chiral catalysts may result in kinetic resolution. E.g. on hydrogenation of the a-methylene esters 47 with the Rh-DIPAMP only one of the enantiomers are saturated to give the syn product (48) in > 96% de and > 83% ee, while the unchanged olefin can be isolated in 98% ee [169].

(47)

(48) R = Me, Et, Ph

Finally, some examples for diastereoselective hydrogenations with the achiral catalyst [(Rh(NBD)(Ph2P(CH2),PPh2)]+BF;should be mentioned: 3-methylenecyclohexanol [1701, 3-methylcyclohex-2-en01 and 4-methylcyclohex-3-en01 [ 17I] all gave more than 98% of the trans product.

2.2 Catalytic Hydrosilylation * Hydrosilylation is the addition of a silicon hydride across a C = C , C = O or C = N bond (Fig. 2-1 1). The 0- and N-silylated products derived from ketones and imines respectively can be readily hydrolyzed, and therefore the final result is equivalent to hydrogenation. The reaction is catalyzed by noble metal complexes, for olefin hydrosilylation chloroplatinic acid is most commonly used [172, 1731. It was discovered by Ojima in 1972 that Rh(Ph3P),C1 was an efficient catalyst for ketone hydrosilylation [174, 1751. The use of chiral phosphines as ligands, resulting in enantioselective hydrosilylation was first reported in 1971 [I 761.

* For reviews, see refs. [178- 1801.

2.2 Catylytic Hydrosilylation

R'CH=CHR'

R'RZC=O

+

+

71

R3SiH % R'CH2CHR2 + R'CHCH2R2

ISiR3

H+

R3SiH % R'R'CHOSiR3

R'R2C==NR3+ R3SiH

**R1R2CH(

I

SiR3 --+

R'R'CHOH

R3

R'R2CHNHR3

,

SIR,

Fig. 241. Hydrosilylation of olefins, ketones and azomethines.

Initial efforts for the hydrosilylation of olefins gave disappointingly poor enantioselectivities and were fraught with low rates and regioselectivity problems [ 178, 1 8 I]. Recently, however, Hayashi et al. achieved practically useful enantioselectivities using the monophosphine ligand 49 (MOP) (Fig. 2-1 31) [182, 1831. In an intramolecular Rh-BINAP-catalyzed version of the reaction Bosnich et al. [ 1841 achieved exellent enantioselectivities with 0-hydrosilylated allylic alcohols, when the dialkyl group at silicon was -(CHz)4-or -(CH&-. Interestingly configura-

OH

A

R2

SiCI,

I

1) EtOH + Et,N

2) H P ,

9 : 1

OH b

R

R = Bu, Hex, Cl,H,,, BnCH,, cHex, ee >94%

Fig. 2-12. Enantioselective hydrosilylation of olefins.

72

2.2 Catalytic Hydrosilylation

tion of R' ( E or 2 ) seems to have no influence on ee values and configuration of the product. In this system hydrosilylation proceeds with silyl-olefin insertion which is the rate limiting and enantioselective step. Contrary to rhodium-catalyzed hydrogenation ( c j p. 63) it is the major diastereomer which leads to the enantiomer in excess [185]. 1,4-Addition of HSiCl, to 1-arylbutadienes catalyzed by the palladium complex of PPFA ( 4 ) were quite regioselective, giving 1 -silylated 2-butenes but enantioselectivity remained poor (up to 64%) [I%, 1871. The most thoroughly studied aspect of stereoselective hydrosilylation is that of ketones catalyzed by rhodium-chiral phosphine complexes. Not counting the solvent, three components, i.e. the substrate, the silane and the catalyst participate in the reaction, and the structures of all three are important. Despite considerable efforts to find optimal combinations of parameters and novel ligands [I88 - 1931, for ketones lacking an additional functional group, enantioselectivities remained generally poor (ee < 60%). Therefore this approach became no viable alternative to reductive methods. Intramolecularity of hydrosilylation effectively contributes to selectivity with the silylated a-hydroxyketone 50 [ 1941.

Ligands specially designed for hydrosilylation, some of them free of phosphorus, are shown below. Me

Me

I

( 8 ) R = Me (MPFA) R = Ph (PPFA)

(54 R = H (52) R = Me (53) R = P h

Prevailing configuration may even be inverted by taking a different silane (CJ Table 2-5). Predictions concerning the configuration of the major product based on models, with no or very little experimental background, have been proposed for hydrosilylations catalyzed by [(R)-BPMP],RhCi and [( +)-DIOPIRhCl by Ojima, Yamamoto and Kumada [ 1781 and Glaser [ 1951, respectively.

2.2 Catalytic Hydrosilylation

73

Table 2-5. Enantioselective Hydrosilylation of Ketones (R'COR*) in the Presence of Chiral Rhodium Catalysts.

R'

R2

Silane

Ligand

Me

Et Bu tBu Bn CH,OBz Me

1 -NaphPhSiH, Et,SiH, Ph,SiH, Ph,SiH, 1 -NaphPhSiH, Ph,SiH, PhMe,SiH Ph,SiH, Ph,SiH2 PhMe,SiH PhMe,SiH Et,SiH, EtMe,SiH PhMe,SiH 1-NaphPhSiH, 1 -NaphPhSiH,

(-)-DIOP ( R)-BnPhMeP

tBu Ph

Et iPr iBu tBu CH2Cl CH,Br a

070

ee

(Conf.)"

Ref. 204 205 206 206 201 208 205 191 205 205 205 178 205 209 207 207

51

53 (-)-DIOP (R)-BnPhMeP (R)-BnPhMeP 54

(R)-BnPhMeP (R)-BnPhMeP (R )-BnPhMeP ( - )-DIOP ( R)-BnPhMeP (R)-BnPhMePb ( - )-DIOP (-)-DIOP

Configurations obtained with ( + )-DIOP and (S)-BnPhMeP were inverted. Cationic catalyst.

Aralkyl ketones are generally more selectively reduced than dialkyl ketones, but with a proper choice of ligand the difference is not as conspicuous as with other enantioselective carbonyl reductions (cJ: Sections 2.1.2 and 3.1). Ketoesters are better substrates for hydrosilylation than simple ketones, 2- and 4-ketoesters can be reduced more selectively than 3-ketoesters [196, 1971. Enantioselectivity is dependent on the structure of the ketoacid, the alcohol moiety, the silane and the catalyst. Some of the best results are shown, more data are listed in Table 2-5. It should be noted that double induction [R = (-)-ment] effectively improves selectivity with the phenylglyoxylate but not with the pyruvate and levulinate. 0

I1

1-NaphF'hSiH,

Me Pr n 0 %ee 185 RI

R2

Ph Ph

Et 0 39

Ph Me Me cHex (-)-Ment Et iBu 0 0 1 2 47 77 26 84

As mentioned above, Schiff-bases of ketones undergo rhodium-catalyzed hydrosilylation to yield, after methanolysis, secondary amines. Conditions are usually mild (0- 100°C, depending on the silane) and yields are good (>85%) [198]. Using a chiral catalyst enantioselectivity was moderate with acetophenone-derived Schiff-

74

2.2 Catalytic Hydrosilylation

bases (ee up to 65% with (DI0P)RhCI and PhSiH,), and poor when both substituents of the ketone were aliphatic [loo, 299, 2001. Rhodium- or platinum-catalyzed hydrosilylation of cyclic homoallyl alcohols by tetramethyldisilazan is highly enantioselective, that of the acyclic analogues is less so, but still acceptable [201].

phv/phmMe 1) (Me,HSi),NH

2) H,PtCI,

de70%

OH

OH

OH

Syn vs. anti selectivity can be controlled by the catalyst in the following reaction [202]:

Ph&

Me 'ONE$

PhMe,SiH CF,CO,H

PhMe,SiH, 0 OC

'ONEt2

[Me,N(CH,)]3S+.SiF2Me3-m

Me de > 98%

Ph

7

CONEt,

Me de > 98%

Intramolecular hydrosilylation can also be exploited in the reduction of P-hydroxyketones where I ,3-induction overrules 1,2-induction and disregarding configuration at C-2 the 1,3-antiproducts are formed in 2 9 8 % de (except for the 2,3-syn-2-ethyl substrate when de was but 66%) [203].

4J iPrzHSiO

iPr,SiClH

R'

R'

Et3N

R2 (2,3-syn-or anti)

R2

Si iPr2

R'

qs R2

,R'

I,

T @HFx z r

SnC&

R1 CH2CI2, - 80 OC

R'

R' G

R2 R

1

*

2.2 Catalytic Hydrosilylation

75

Enantioselectivities in hydrosilylation for a series of substrates obtained with the best combination of silane and catalyst, further, a few additional data of interest are compiled in Table 2-5. The catalytic cycle involved in the hydrosilylation of ketones catalyzed by rhodium-phosphine complexes can be envisaged as shown in Fig. 2-13 [ISO].

+

\

FSIH

w

I

\

.

LnRh ( --SI)H /

B

\

/

L,,Rh"'( TC-OSi \)H

D

111

\ .

\

L,W ( ~ S I (C ,)

=O)H

C

Fig. 2-13. The catalytic cycle in hydrosilylation of ketones with rhodium-phosphine complexes.

The rhodium(1) catalyst is applied as a phosphine-olefin complex, or, when prepared in situ, it contains coordinated solvent molecules. In the first step the olefin or the solvent is displaced by the silane, which becomes coordinated to rhodium in an oxidative addition establishing simultaneously a metal-hydrogen and metalsilicon bond. Thereafter, the ketone becomes coordinated and inserted into the metal-silicon bond. Finally, internal hydrogen transfer from metal to carbon and dissociation to a silyl ether concludes the cycle. This is a plausible scheme but, unfortunately, very little experimental work has been undertaken to explore its details. Kolb and HetflejS studied the kinetics of the hydrosilylation of the non-enolizable ketone tBuCOPh with Ph2SiH2 in benzene catalyzed by [Rh(COD)( -)DIOPI'C10; [244] and [Rh[( -)-DIOPI2]+C10; [210]. The authors observed an induction period in the hydrosilylation which disappeared if the ketone was added 30 min after the catalyst and the silane had been mixed. The induction period was associated with the oxidative addition of silane to rhodium and could be followed by UV spectroscopy. Initial rates showed a linear dependence on catalyst and silane concentration, but correlation with substrate concentration was non-linear. The kinetic behavior of the system suggested that the rate determining step was either the B-tC or the C-tD transformation, the latter being more likely. The absence of a significant deuterium isotope effect when Ph2SiD2 was used excluded process A -tB as the rate determining step. Temperature dependence of rate was also in agreement with the proposed scheme. No information can be gleaned from the above experiments concerning the origin of stereoselection. Catalytic hydrosilylation of cyclic ketones can be remarkably diastereoselective, and the predominant mode of approach is influenced by the bulk of the silane. In

76

2.3 Heterogeneous Stereoselective Catalytic Hydrogenations

(a )

endo

silane

endolexo

PhSiH, Ph2SiH2 Et&H

9O:lO 73:17 30:7O

eq

silane

axieq

iprl f 0

PhSiH, EtMelSiH PhMezSiH

9o:lO 50:50 0:loO

ay

(b)

trans

ris

Fig. 2-14. Diastereoselective hydrosilylation of ketones.

Fig. 2-14 (a) the ratio of axial and equatorial (or endo and exo) attack is shown for hydrosilylations of camphor and menthone catalyzed by (Ph,P),RhCl [211]. Reduction of (R)-carvone with I-NaphPhSiH, in the presence of a DIOP-rhodium catalyst (Fig. 2-1 4 (b)) revealed that in this double-induction experiment it was the chirality of the catalyst and not that of the substrate which controlled stereoselectivity: with (+)-DIOP the transkis ratio was 77:23, while with (-)-DIOP it was 21 :79 [212]. Double induction also enhanced stereoselectivity in the hydrosilylation of N-2-ketoacyl-2-aminoesters [213].

2.3 Heterogeneous Stereoselective Catalytic Hydrogenations * In this section hydrogenations over classical solid phase catalysts will be discussed; metal-complex catalysts immobilized by linking to a polymer matrix have already been mentioned in Section 2.1.1.1.

* For a book o n this subject see ref. [214].

2.3.1 Enantioselective Heterogeneous Catalytic Hydrogenations

77

For efficient chiral catalysis a highly ordered and very specific structure of the catalyst is necessary, and it is very unlikely that traditional non-complexed metal catalysts could ever meet this requirement. The most such a catalyst can contribute to stereoselectivity is a syn transfer of dihydrogen to the substrate, and in fact this is performed with very high selectivity. All the rest has to come from the substrate side, and we shall see that while diastereoselective hydrogenations are common, examples for genuine enantioselective heterogeneous catalysis in hydrogenation can hardly be found.

2.3.1 Enantioselective Heterogeneous Catalytic Hydrogenations The development of enantioselective heterogeneous catalysts was pioneered by Izumi and his school, who prepared a silk fibroin-supported palladium catalyst [215]. Up to 70% ee could be realized with cyclic amino acid precursors, but the method could not be developed to a practical procedure. An enormous amount of work has been invested in studies with chirally modified catalysts by Izumi, Harada and their coworkers in Japan (38 papers up to 1984) and

OH

0

____)

O

2

M

H

69y0 ee

e

4-

xMe Me 57% ee

Me

t 6M u e'

Meq

____)

tB u

88% ee

ee R - I C I I H ~ 85'): ,

MeL

C02Me

R=Me

M C O ze, - ,RJ,

RL c 0 2 M e

78: 22

MeL C 0 2 M e

Me 64% ee

740/0ec

Fig. 2-15. Stereoselective hydrogenations over Raney nickel modified with ( 2 R ,3R)-tartaric acid and NaBr.

78

2.3 Heterogeneous Stereoseleclive Cutulytic Hydrogenations

by Klabunovski and his group in the Soviet Union (more than 100 papers up to 1983) [216] and publications on this subject continue to trickle ever since [e.g. 217-2191. Both groups were primarily concerned with the hydrogenation of b-dicarbonyl compounds (almost exclusively of methyl or ethyl acetoacetate) on metal catalysts modified with (-)-(2R, 3R)-tartaric acid. Harada et al. found that the enantioselectivity of tartaric acid-modified Raney nickel was greatly enhanced by the addition of NaBr [220]. In Fig. 2-15 some results obtained with this system are shown. Enantioselectivities with linear P-ketoesters are in the useful range [220, 2211, those with a P-ketoalcohol [222, 2231, a ketone [224] and P-ketosulfonic acids [225] are less good. With both 1,3-dicarbonyl compounds [226, 2271 and 2-alkyl-3-ketoesters [228] diastereomers are formed. In the latter substrate the center of chirality is unstable, and the stoichiometry of the product reveals kinetic preference for the hydrogenation of the 2 s substrate.

2.3.2 Diastereoselective Heterogeneous Catalytic Hydrogenations Hydrogenation over classical metal catalysts of substrates having diastereotopic faces is a well established stereoselective procedure which involves the syn addition of dihydrogen preferentially to the less hindered face of the substrate. In this section we concentrate on studies which were aimed at the synthesis of optically active amino acids. In some of the examples (e.g. the hydrogenation of dehydropeptides) the inducing group became an integral part of the product, while in others it was removed after it had fulfilled its task (chiral auxiliaries). The basic requirement for a chiral auxiliary is that its inducing power should be high. It is also desirable that its chirality should not be destroyed in the course of the synthesis, i.e. it should be either recoverable or recyclable after simple transformations. In the following examples all three situations will be illustrated.

Hydrogenation of enamines Cyclic enamines of type 55 (Fig. 2-16 (a)) can be readily prepared from azlactones. If R' is an aromatic group and the chiral auxiliary proline, hydrogenation proceeds with high enantioselectivity, while with R' being alkyl the optical purity of the product is low [229]. This is somewhat contradictory to later reports, which claimed high diastereoselectivity for 2-alkylidene diketopiperazines unsubstituted at N( 1) [230, 23 11. Technically important is the stereoselective hydrogenation of the biotin precursor 56 (Fig. 2-16 (b)) [232].

2.3.2 Diasrereoselective Helerogeneous Catalytic Hydrogenations

R"

H '

79

R~NH-CO,H

(55) R'

R?

R'

R4

Q J ~de

Ref.

Fig. 2-16. Diastereoselective hydrogenation of enamines.

Hydrogenation of ketones Surprisingly few studies deal with the catalytic reduction of prochiral ketones. A series of chiral pyruvamides were investigated, of which the (S)-I-(I-naphthy1)-ethylamide could be hydrogenated with the highest selectivity giving, at -3O"C, the (S,S)-lactamide in 96Yo de [234]. Hydrogenation of C = N bonds Transfer of chirality from a chiral amine to a ketone to form a new chiral amine has been thoroughly investigated. * The general scheme of this procedure is shown in Fig. 2-17 (a). The chiral amine is most often I-phenylethylamine, in which the nitrogen atom is linked to a benzylic position and therefore the inducing group can be removed by catalytic hydrogenation. This requires more stringent conditions than the saturation of the C = N bond and destroys the original chiral center. ~

* The method is often called asymmetric transamination, but cess is only formal.

~

~

the analogy to the biochemical pro-

80

2.3 Heterogeneous Sfereoselective Catalytic Hydrogenations R* ' 7 1

u

/ , . 4

xR NU

l

e

T

u

1

LA"-,.-

/mt

2

n=2 R / M e Et Ph Bn %ee 96 92 94 98

n

=

1, Me, Et, iPr, tBu, Ph, Bn, ee >98%

Fig. 2-17. Transfer of chirality between amines and ketones.

When applied to a-arylketones the method is suitable for the preparation of psychotomimetic amines. Diastereoselectivities are usually moderate when simple chiral amines are used, but can be upgraded by recrystallization of the diastereomeric intermediates [235 - 2371. Much improvement can be brought about by using as chiral amines 2-amino acid esters with bulky groups [238]. In this case the inducing group was removed via Nchlorination, elimination and hydrolysis. With acetophenone results were similar. Transformation of racemic 2-alkylcyclohexa- and pentanones to the less stable cisamines with high enantioselectivity has been reported (Fig. 2-17 (b)) [239-2411. This result can only be explained if one assumes equilibration of the enantiomeric ketones or of the diastereomeric azomethines. Harada and his school investigated the transamination of 2-ketoacids involving chiral sec-amines very thoroughly [e.g. 242 - 2451 but satisfactory selectivities could not be realized and the chiral center of the amine component was destroyed by the procedure.

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

3 Stereoselective Non-Catalytic Reductions

The subject of this chapter is almost exclusively the non-catalytic reduction of carbony1 compounds to alcohols. Conversion of ketones to secondary alcohols, as well as of azomethines to amines by catalytic hydrogenation, and the hydrosilylation of ketones have already been discussed in Chapter 2.

3.1 Enantioselective Reductions Two main approaches to the enantioselective reduction of C = X double bonds (mainly of C = O bonds) are known: (i) the use of aluminum or boron hydrides having chiral ligands, and (ii) transfer of chirality by exchange of a hydride ion attached to a chiral center. The latter requires activation of the C - H bond usually by a metal atom (e.g. A1 or Mg) in P-position or by an alkoxide group (e.g. OA1,,3) in a-position. The first method is much more efficient and also more generally applicable. Catalytic methods for the transfer of H- are still in their infancy (see Section 3.1.3).

3.1.1 Chiral Lithium Aluminum Hydrides * 3.1.I. 1 Reduction of Carbonyl Groups The idea that lithium aluminum hydride (LAH) can be converted to a chiral reagent by exchanging hydrogens for chiral alcohols can be credited to Bothner-By

* For

reviews see ref. [ l ] .

82

3.1 Enantioselective Reductions

(1951) [2]. He reacted LAH with (+)-camphor and reduced with the resulting reagent prochiral aliphatic ketones. Although it was shown later that the optical activity of the product was due to contamination by the reagent [3, 41, the idea was born, and chiral lithium aluminum alkoxides have been developed to one of the most efficient enantioselective reducing agents. The factors controlling selectivity are still not well understood; the main reason for this is the complexity of the LAH-alcohol system. Reagents are usually prepared by mixing LAH with the appropriate alcohol in a solvent and are used without isolation. The equation for the formation of the reagent i. e. :

is only formal since several species with different numbers of alkoxide groups may be formed consecutively, and the situation is further complicated by disproportionation producing achiral LAH, e. g. : 2 LiAlH,OR*

* LiA1H2(OR*)2+ LiAlH,

Also, the alkoxides often precipitate from solution and may undergo ageing [e.g., 51. Thus, different results can be obtained not only with different molar ratios of LAH and alcohol but also with a freshly prepared reagent and with one left standing. Therefore, careful optimization of conditions is necessary in order to achieve high selectivities. The experimental protocol followed by the majority of authors is (i) testing a series of reagents with a single substrate, (ii) optimization of the conditions with the best reagent(s), (iii) testing the best reagent(s) under optimum conditions on a range of substrates. Bothner-By's choice of a chiral auxiliary, a monohydric alcohol with no other functionality, was unfortunate, and ever since attempts to develop, efficient reagents with such alcohols have been unsuccessful [4, 6 - 121. An exception seems to be the reduction of some /?-aminoketones with (-)-menthol-LAH (3 : 1 ) * [13].

&)-menthol

Ph

+ L A H ( 3 : 1)

Et*O. 0

=c

0" Ph- w e * 77% ee

The application of bifunctional compounds as chiral ligands was pioneered by Landor and his coworkers in the sixties. Not only was a highly enantioselective reagent discovered, but the ideas on which it was based could later be successfully adapted to the preparation of other reagents. In order to block the reappearence of LAH by disproportionation, instead of a monohydric alcohol, diols and triols derived from carbohydrates were used as chiral

*

Expressed here and later as molar ratios.

3.1.1 Chiral Lithium Aluniiniuin Hydrides

83

ligands [4, 141. Among them, 3-0-benzyl-1,2-O-cyclohexylidene-a-~-glucofuranose provided the highest enantioselectivity: propiophenone was reduced to ( S ) -1 -phenylpropan-1-01 in 57% ee. The presence of the 3-0-benzyl group seems to be essential for acceptable selectivity. The 1 : 1 complex (2) may take up a conformation in which one of the transferable hydrogens (H’) is shielded by the phenyl group and therefore the other one is preferentially transferred to the substrate, for which a preferred orientation should be assumed as well. An alternative explanation is that interaction of the substrate with the benzyl group assists the approach of the former in a specific orientation. Although it was not rigorously proved which of the hydrogens was more active, the hypothesis was fruitful and suggested that replacement of the more exposed hydrogen by an alkoxy group would give a reagent which was less active but showed increased and inverted selectivity. In fact, when ethanol (I .3 - 2.5 mol) was added to I , a monohydride with the tentative structure 2 was formed, and a dramatic increase in selectivity with inverted configurational preference was experienced [ 151. A further increase in the bulkiness of the complex proved to have an adverse effect [16]. Enantioselectivities of up to 71 Yo were realized, and therefore the method is only of historic interest.

Some monohydroxy sugar derivatives gave only very low enantioselectivities [ 171. The selectivity of 3 could not be reached with complexes prepared from other carbohydrate-based [ 17, 181 or terpenic diols [ 19- 221. The front runner among LAH-diol complexes, and one of the most selective chiral reagents at all, is based on 2,2’-dihydroxy-I ,Ir-binaphthyl, BINOL, and was discovered by Noyori et al. [17, 181. Addition of an alcohol to provide a third ligand, preferably methoxy or ethoxy, was indispensable; the reagent with BINOL as the only ligand was non-selective. The reagent prepared from (S)-BINOL can be formulated as 3. Results were optimized using acetophenone as substrate. High selectivity required low temperatures. Replacement of ethanol by 2,6-di-tert-butylphenol not only reduced enantioselectivity to 44% but inverted the configuration of the major product to R. Even more mysterious was the effect of 3,3,3-trifluoroethanol, which also gave the (R)-alcohol in 42% ee [19]. BINAL-H reduced with high selectivity not only aralkyl ketones [17] but also alkynyl alkyl ketones [20, 211, deuterated aldehydes [I 7, 221 and vinyl alkyl-ketones [18], some of which were useful as prostaglandin synthons [23] and prostaglandins

84

3.1 Enantioselective Reductions

( 3 ) (S)-BINAL-H (R=Me, Et)

proper [24].Note that the microbiological reduction of bromovinyl pentyl ketone gave the corresponding (S)-alcohol in but 80% ee and 10% yield [23]. Representative examples are shown in Table 3-1. Simple dialkyl ketones gave low enantiomeric excesses [17]. It is convenient that both enantiomers of BINOL are available. The absence of an ageing effect and the linearity of the In R / S vs. l/Tplot suggested that a single reducing species was involved. This and the C, symmetry of the diol may be the clues to the exceptional selectivity of BINAL-H. Similar reagents with 2,2’-dihydroxy-6,6’-dimethyl-biphenyl [24],a phenanthrene analogue [25] and the spiro compound 4 [26] were prepared, but proved to be only marginally better than Table 3-1. Reduction of Carbonyl Compounds (R’COR’) with (S)-BINAL-H (3) to (S)-AlcoholsaXb. R‘

R*

% ee

Ref.

Ph (E)-Me,C = C(Me) - C H = C H (2)-Me,C = C(Me) - C H = C H Ph

D

82 84 72 95 98 100 71 100 44 84 84 90 51 84 90 87

17 21 21 19 19 19 19 17 19 20 21 21 20 21 21 20 18 18 18 23

BuC-C MeO,CC=C (E)-BuCH = CH

Me Et Pr iPr Bu tBu (CH,),CO,Me Pent Oct iPr Me Pent Oct Me Pent

19 91 92 96

(E)-cPentCH = C H (E)-CH = CH ~~~~~~

a

~

~~~~~~~~~~~

Experiments were performed in THF at - 100°C. As additive, EtOH was used in ref. [19] and MeOH in ref. [21]. In some experiments (R)-BINAL-H was used and thus the (R)-alcohols were obtained.

3.1.1 Clirral L~thiurnAlutnitirurtr Ilyclrrdrs

85

BINAL-H. Incorporation of BINOL into a crown ether improved selectivity with non-aromatic ketones (e. g. 85% with iBuCOMe) [27]. Replacement of hydroxy groups in BINOL by NHR groups, however, gave poor selectivity [28]. Fundamental work in the field of enantioselective reductions was carried out by cervinka and his school, who first employed aminoalcohols, notably alkaloids, as chiral ligands [29]. Although enantioselectivities were modest (up to 39% ee with PhCOtBu and 48% with PhCOMe), interesting correlations between the configurations of the aminoalcohol and product were discovered (for details see ref. [30]). Efficient aminoalcohol-type chiral auxiliaries are all tertiary amines and replace one hydrogen of LAH per mole. The role of the nitrogen atom in enhancing selectivity is not yet clear, it may lend a more rigid structure to the complex or to the transition state, or may inhibit disproportionation of the complex to regenerate LAH. Landor's idea of adding a second but achiral ligand also proved to be fruitful in the hands of Vigneron and coworkers [31, 321. While reaction of LAH with 1 , 2 and 3 moles of (-)-N-methylephedrine ( 5 ) gave reagents with increasing but low selectivities (40% ee of the R alcohol at best with acetophenone), the addition of 2 moles of a hydroxy compound, optimally 3,5-dimethylphenol highly increased enantiomeric purity. The reaction has an optimum temperature at around - I5 "C, ee values drop sharply at both higher and lower temperatures. This points to the existence of several reducing species of different selectivity at equilibrium, the concentration of the most selective being at its maximum at the optimum temperature. Characteristic examples for reductions with LAH-( -)-N-methylephedrine-3,5-dimethylphenol 1 : 1 : 2 are displayed in Table 3-2. High selectivity was not only achieved with aralkyl ketones [31, 321 but also with a,pacetylenic alkyl ketones [34], and moderate selectivity with 2-alkyl-1,3,4-cyclopentatriones[35, 361, intermediates in the synthesis of allethrolones. As usual, the reagent failed with dialkyl ketones (ee 14- 19% with unbranched methylketones, somewhat higher with others, see Table 3-2) [32]. Enantioselectivity of the LAH-( -)-N-methylephedrine complex was also enhanced by the addition of N-alkylanilines, most effectively by N-ethylaniline (up to 90% ee ( S ) with aralkyl ketones) [37]. A variation of the method, in which the modifier is 2-ethylaminopyridine, is of greater interest, since this enables the reduction of cyclic a,punsaturated ketones to the (R)-alcohol with 73 - 98% enantioselectivity (98% for cyclohex-2-enone) [38]. The first aminoalcohol which was found to form a highly selective reagent with LAH without the aid of a third component was Darvon-alcohol (6) [39] and was used for the reduction of aralkyl ketones [ 5 , 401. Its real potential, the enantioselective reduction of a,pacetylenic ketones [41] was exploited for the synthesis of tocopherol [33]. The triple bond seems to play the same role as aryl groups, since the configuration of the major enantiomer is the same in both cases and selectivity drops sharply when the triple bond is replaced by a double bond or an aliphatic group. The reagent performs well, even with highly complex polyenic substrates [33, 411. Ten analogues of Darvon alcohol were prepared and tested later but none of them came even close to 6 [33].

86

3.1 Enantioselective Reductions

Table 3-2. Reduction of Ketones (R'COR2) with Reagents Prepared from LAH and Chiral Aminoalcohols. (A: LAH + (-)-N-methylephedrine ( 5 ) + 3,5-dimethylphenol ( 1 : I : 2), Et20, - 15°C; B: LAH+(-)-N-methylephedrine (5)+N-ethylaniline (1: 1 :2), Et,O, -78°C; C: LAH + Darvon-alcohol (6) ( I :2 : 3), - 78 "C) R'

R2

A Vo ee

Ph 4-MePh Ph

1 -Tetralone iPr tBu Bn cHex HC=C

MeC=C Me3SiC= C Pent - C=C

Me CF3 Et Pr iPr Bu iBu tBu cHex

B (Conf.)

Vo ee

C (Conf.)

Vo ee

(Conf.)

84 63 30 85 89 17 78 84 31 11

57d Me

Et iPr Bu iBu tBu Pent iBu Pent

41 20C 45 79 87 86 85 88 90 84 82

At - t00"C. At -65OC with fresh reagent, with aged reagent at r.t. 75% ee but S ! At 0 ° C with fresh reagent. 63% ee at 0°C. At 0°C. At -78°C. g From ref. [33].

a

LAH-Darvon alcohol displays all the caprices characteristic for modified LAH reagents. On addition of the aminoalcohol to a solution of LAH in ether, a precipitate was formed and when this was used immediately at -65 "C (R)-alcohols were obtained in excess. On standing, the precipitate dissolved and at room temperature this reagent gave a product with opposite rotation with some of the substrates (e.g. PhCOPr, PhCOiPr and tBuCOMe) but not with others (PhCOCF3 and PhCOtBu) [ 5 ] . Reaction of LAH with chiral secondary amines gives chirally modified reagents similar to those obtained with alcohols. In a series of reagents prepared from LAH and 1-3 moles of the amines 7 [42], as may be anticipated, amines with no second binding sites (X = H, Me) provided very poor reagents, while those with polar X

3.1.1 Chiral Lithium Alurniniun? HydrideJ

B

Ph NMez (5) (-)-(1R32S)-N-methylephedrine

n

/

HO

87

w Mez

Ph

(6) Darvon alcohol

(S)-(7)

X=H,Me, OMe, NMe2, SMe

NMe

OH

( I ] ) X=Me2N, piperidino

Fig. 3-1. Chiral amines and aminoalcohols used for the preparation of chiral lithium aluminum hydride complexes.

groups (OMe, SMe and NMe2) and, surprisingly, the mesitylmethyl derivatives were much better but still unsatisfactory. A series of 12 chiral auxiliaries with two secondary amine functions ( 8 ) were prepared from L-proline by Mukaiyama ef ai.,of which the ones with Ar = Ph [43, 441 and Ar = 2,6-dimethylphenyl [45] were the best giving high ee values for aralkyl ketones, but still inefficient with dialkyl ketones. Only one hydrogen is available for hydride transfer in the 1 : 1 complex, the other, presumably the one in endo position of the rigid bicyclic structure (9),is efficiently shielded both by the aryl group and the pyrrolidine ring. Li' may also be an essential part of the complex, since addition of agents complexing with Li+ (1,Zdimethoxyethane or TMEDA) lowered selectivity. A tridentate ligand was prepared from aspartic acid [46]. Its remarkable feature was that it reduced the black sheep substrate PhCOtBu with not only the highest selectivity reported (86% ee) but more selectively than it did other ketones. The reagent is supposed to have the rigid structure 10. Cyclohex-2-enone was reduced by 10 with complete enantioselectivity ( S ) [47]. A further increase in the number of functional groups seems to do more harm than good. Thus Seebach and his coworkers prepared a series of ten 1,4-bisdialkylamino-2,3-butanediols from (R,R)-tartaric acid and reacted them with LAH [48, 491. Among them, only two (11, X = Me2N and piperidino) were useful.

88

3.1 Enantioselective Reductions

3.1 .I .2 Reduction of C =N and C-C Bonds Although the first successful experiment for enantioselective reduction with a chiral LAH complex involved the reduction of immonium salts with LAH-(-)menthol reagent by cervinka in 1961 [50], not very much attention was paid to reducing C = N bonds with chiral LAH complexes. In fact, early results had been rather disappointing, reduction of immonium salts with LAH-( -)-menthol complex gave <6Vo ee [50], that of imines > 10%. Even 2 failed to produce a useful degree of enantioselectivity in the reduction of azomethines [5I]. The same reagent and a close analogue [52] were, however, moderately selective with ketoximes, which all gave (S)-amines in 9.5-5670 ee and >6O% yield [53]. Essentially the same results were obtained with O-methyl- and 0-tetrahydropyranyl oximes. It was remarkable that with aliphatic ketoximes selectivity was higher than with the aralkyl analogues. Reduction with the ethanol-modified reagent gave the ( R)-amines.

3.1.2 Chirally Modified Hydridoborates * Much less attention and success is associated with sodium borohydride-based enantioselective reductions than with those involving LAH. Only a few chiral ligands have been reported as yet which lend acceptable enantioselectivity to NaBH,. 1,2 : 5,6-Diisopropylidene-a-~-glucofuranose was selected out of six monosaccharide derivatives by Hirao et ai. Even this was inefficient, giving only 40% ee, at least with aralkyl ketones [55]. Addition of a Lewis acid, especially of ZnC12, both improved selectivity and inverted the prevailing configuration from R to S [56, 571. Soai et al. prepared fairly selective reducing agents by reacting (R)-N-benzoylcysteine and (R)-N,N'-dibenzoyl-cystinewith LiBH, and tert-butanol [58 - 601 or ethanol [62], which gave up to 90% ee of (R)-aralkyl carbinols. Sodium triacyloxyborohydrides prepared from N-benzoyl- or N-benzyloxycarbonylproline reduced dihydroisoquinolines with up to 86% ee [ 1151. The solubility of NaBH, in water and its stability to alkali make it an obvious candidate for phase transfer catalysis experiments, and several attempts at enantioselective reductions with NaBH, in the presence of chiral phase transfer catalysts have been reported. Unfortunately, so far no viable process has emerged from these experiments [62, 631. This is not quite unexpected, since the involvement of the chiral cation in the hydride transfer step may not be very intimate.

* For a review see ref.

[54].

3.1.2 Chirully Modified Hydridoborates

89

The water solubility of NaBH, permits the use of proteins as chiral aids in reductions. While results were modest with lecithin (no ee values reported), selectivities achieved in the presence of bovine serum albumin were remarkable [64] (up to 78% ee with aralkyl ketones) [65, 661. The ee values increased with protein concentration and dropped sharply on denaturation. Addition of a chiral amine (12) prepared from pyroglutamic acid to the CoC12NaBH, system provided remarkable selectivities in the reduction of p, pdisubstituted acrylates. Reduction of the 2 isomers gave antipodal products with similar enantioselectivity [67]. Me

NaBH, + COCI,

Dh4F - EtOH

R

Me

+ (12) R

A reagent prepared by mixing 2-amino-I , I -diphenyl-3-methylbutan- 1-01, NaBH4 and ZrC1, reduced acetophenone 0-methyloxime to (S)- 1-phenylethylamine of 95 Yo ee. Reductions of other oximes with (S)-valinol as chiral aid were less selective (42-66'5'0) [68]. Reducing agents with tetracordinate boron combining 9-borabicyclo[3.3. I ] nonane (BBN-H) * with a chiral moiety were prepared by Brown et al. While the reagent containing the isopinocampheyl group (13) was of poor selectivity [69], the one incorporating the inexpensive diacetoneglucose (14) proved to be highly selective with ketoesters (86- 98% ee), aralkyl ketones (78 - 100% ee), and 2-alkylcyclohexanones ( > 9 4 % ee), but failed (as usual) with dialkyl ketones (3-70'70 ee) [70-721.

(-)-IPC-BBN

tBdi

THF,-18 O C

*

*

a

B

-

O

&

The parachute-like symbol in the formula represents the carbon skeleton of 9-borabicyclononane (9-BBN).

90

3.1 Enantioselective Reductions

A similar reagent prepared from the borane (24) (see later) showed unexceptional selectivity in the reduction of aralkyl ketones but was relatively efficient with dialkyl ketones (70- 79% ee with unbranched ketones). The prevailing configuration ( S ) was opposite to that obtained with the parent borane [73]. Dialkoxyl-alkyl-hydridoboratesinvolving chiral diols gave moderately to low enantioselectivities in the reduction of ketones [74].

3.1.3 Chiral Boranes and Boronates Reaction of borane with a-pinene gives diisopinocampheylborane (IPC2BH), a highly enantioselective hydroborating agent ( c j Section 8.3). Reduction of ketones with IPC2BH gave, however, carbinols with relatively low ee [75,76]. (-)Monoisopinocampheylborane [( -)-IPCBH,] gives comparable ee values [78,79]. Boranes, as Lewis acids, readily form complexes with amines and, when the amine is chiral, the complex becomes a chiral reducing agent. Despite several attempts no highly enantioselective amine-borane reagent could be developed [80 - 851. With a-amino rert-alcohols borane forms compounds which were initially formulated by Itsuno et al. as of the type R*OBH2 [86-891, but later found by Corey et al. [90] to be oxazaborolidines. These have in themselves no reducing power but are acting as efficient catalysts when complexing with excess borane (15).

(Id) R = H ( I ? R = Me (18) R = Bu

B

R’

R2

%ee

R’

R2

% ee

Me

iPr Bu

60

Me Et

2-Naph iPr Ph

52

iBu

tBu Ph

55 61 78 94

Bu

94

60 > 99

3.1.3 Chiral Boranes and Boronates

91

15 derived from (S)-valine reduced acetophenone (E)-0-benzyloxime to ( S ) -1 -phenylethylamine of 91 -95% ee [89, 911 and azomethines of phenyl alkyl ketones to the corresponding (R)-sec-amines of 71 -87% ee [92, 931. A borane complex of norephedrine reduced 0-alkyl ketoximes with high ( 5 92%) enantioselectivity [94]. New knowledge about the complexes stimulated the preparation of the (S)-proline derived oxazaborolidine donor ligands 16-18 [95 - 981, bearing no transferable hydrogen and which could be applied in catalytic amounts (0.1 mol eq.) in reductions with borane [95] and catechol borane [99, 1001 as the hydride source. With the latter system alkyl and aryl trichloro- and trifluoromethyl ketones could be reduced to (R)-alcohols of 95 -98% ee. A polymer-linked a-aminoalcohol proved to be an efficient catalyst for enantioselective borane reduction of aralkyl ketones [ l o t , 1021. The 2-naphthyl analogues of 18 and 19 catalyze enantioselective BH, - T H F or catecholborane reductions of both aralkyl and dialkyl ketones (84- 97% ee) [107]. A series of ab initio calculations on reductions involving 19 have been carried out [103-1061. 2,5-Dimethylborolane (19) generated in situ is remarkable for its high selectivity in the reduction of dialkyl ketones [108, 1091.

The aminoalcohol 20 (both R and S ) [110] and the phosphine 21 [ l l l ] are both efficient chiral ligands for borane reduction of aralkyl ketones.

Ph

Me

(22)

%ee (conf.) 56

Ph

Ph

Ph iPr

iP r Me

Ph Me

(R) 73 (365 ( S ) 18 (-) 71 (-) 72 ( R )

92

3.1 Enantioselective Reductions

3.1.4 Enantioselective Reductions with Hydride Transfer from Carbon Carbon atoms may transfer a hydride ion to a carbonyl group when activated by a metal atom two bonds away which also provides a pair of electrons to form a double bond. This general pattern is illustrated by one of the first processes of this kind discovered, the Meerwein-Ponndorf-Verley reduction. The six-membered cyclic transition state for the mechanism of this and similar reactions is a very obvious, and in most cases adequate, representation, though exceptions are known.

The activating atom can also be attached to carbon, and enantioselective reductions by transfer of hydrogen from a P-carbon in trialkyl boranes as well as in alkyl beryllium, aluminum and magnesium compounds are known.

3.1.4.1 Chiral Trialkylboranes * Usually trialkylboranes show reducing properties only under vigorous conditions [96]. Midland and his coworkers have developed a reagent, B-(3a-pinanyl)-9-borabicyclo[3.3.l]nonane (23, Alpine-borane@**), readily availably from 1,5-cyclooctadiene, borane and a-pinene, which reduces carbonyl groups to alcohols under very mild conditions and generally with high stereoselectivity (Fig. 3-2) [I 18- 120, 1241. For example, at room temperature 23 reduced deuteriobenzaldehyde to (S)-benzyla-D-alcohol of 100% ee [120]. Reduction of the substrate is coupled with the elimination of a-pinene, which may then be recycled. As terpene partners, (-)-P-pinene [121], (-)-camphene and (+)-3-carene were also tested, but all produced less selective reagents [120]. Only the borane derived from 0-benzylnopol (24, NB-enantrane) was of comparable selectivity, but it was less reactive [122, 1231. Alpine-borane@(23) is outstandingly selective in the reduction of aliphatic, allylic and aromatic aldehydes [I201 and of a,pacetylenic ketones [119, 1241. With other _________________

~

* For a review see refs. [78, 1151 and [116]. ** Trade name of Aldrich Chemical Co.

93

3.1.4 Enantioselective Reductions with Hydride Transfer f r o m Carbon

Me

. -

Me

Me

Alpine-borane@ (23) (-)-IPC-BBN

IA..

fi DB0 Fen

B

(24) NB-enantrane

Me

Me

Me

b

l

c:'

Me

(25)

Fig. 3-2. Enantioselective reductions with chiral trialkylboranes.

ketones reduction is too sluggish and accompanied by the dissociation of the reagent to a-pinene and borabicyclo[3.3.1]nonane. Since the latter is an achiral reducing agent, enantioselectivity suffers [125]. This can be remedied by performing the reaction at high pressure [ 1261 or by applying the reagent neat which permits the reduction of any kind of ketone, albeit with varying enantioselectivity [127]. The neat Table 3-3. Reduction of Carbonyl Compounds (R'COR2) with ( +)-B-(3a-Pinanyl)-9-borabicyclo[3.3.l]borane (23) at Room Temperature in THF or Neat (*). R'

R2

To eea (Conf.)b R'

Pr Pent tBuCH2 (E)-PhCH = CH Ph 4-NO2 - C,H, 4-Me0 - C,H, Ph Me iPr tBu Me Pent Ph

D

(S) (S) (S) (S) (S) (S) (S) (S) (R) (R) (S) (R)

100 89 98 84 98 100 82 C-CBu 89 C=CPh 100* C=CH 99* 99 C=CC02Et 77 92 100

(R) (S)

R2

TOeea (Conf.)b

(E)-PhCH CH Me 1-Cyclohexenyl Ph Et iPr tBu Me Et iPr Ph Ph Hex

97* 64* 87* 92' 43* 62* 83 I* C02Me 86* C0,tBu 100* loo* loo* loo*

CH2CN CH,CN

98* 84*

(S) (S) (S) (S) (S) (S) (S) (S) (S) (S) (S)

(I?) (R)d (R)d

a Corrected for enantiomerically pure reagent. All products derived from a,B-acetylenic ketones are homochiral. At 600 atm. The product is the corresponding aminoalcohol.

94

3.1 Enantioselective Reducfions

reagent reduces a-ketones to a-hydroxyesters [ 1281 and P-ketonitriles [ 1291 with good to excellent enantioselectivity. The most important results obtained with Alpineborane@ (23) are summarized in Table 3-3. Note that the reduction products of a,pacetylenic ketones are all homochiral. Reagents prepared with the ethyl and propyl homologues of a-pinene offer no extra advantages [130]. Treatment of [( -)-IPC],BH with HCI gas generates a new versatile reducing agent [( -)-IPC],BCl giving, except for unhindered dialkylketones [I 3 I], good to excellent enantioselectivities [ 132- 1351. Selectivities with the ethyl analogue of [(-)-IPCI2BCl were even better (e.g. >95% for MeCOiPr and MeCOcHex) but the reagent is less accessible. A similar chloroborane (25) prepared from nopol benzyl ether gave with dialkyl ketones the antipodal alcohols with similar enantioselectivities [ 1371. [( -)-IPC],BCl can be converted to reagents of the type [( -)-IPCIRBCI of which the t-butyl compound reduced aryl methyl ketones with opposite enantioselectivity as compared to the parent borane [I 171. Table 3-4. Reduction of Ketones with ( - )-Diisopinocampheyl-chloroborane ( 2 5 ) in THF at - 25 "C. R'

R2

Yo ee

R'

R'

Yo ee

R'

R,

Yo ee

Ph

Me CH2C1 CH,Br CHJ Me

98 96 86 61 4

iPr tBu CH,CO,Et (CH,), (CH,),

Me Me

32 95 84 98 91

SiMe,

Me iPr Bu

96 98 96

Et

3.1.4.2 Chiral Metal Alkyls The standard reaction of metal alkyls with carbonyl compounds is the addition of the alkyl group to form carbinols. It was, however, recognized very early that Grignard reagents with a hydrogen available at the /?-carbon, especially those with a bulky alkyl group, may transfer this hydrogen to the substrate with the concomitant formation of an olefin. Trialkyl aluminum, dialkyl zinc and dialkyl beryllium compounds are also amenable to reduction of carbonyl compounds. When the alkyl groups are chiral, the reduction may be enantioselective. Selectivity is generally poor and therefore, although of much theoretical interest, such reagents hardly qualify for practical application. For the interesting history of these studies we refer the reader to the first edition of this book. Enantioselective reactions with Grignard reagents have been extensively reviewed by Morrison and Mosher who covered the literature up to 1968 [30].

3.1.4 Enantioseleclive Reductions with Hydride Transjer f r o m Carbon

95

3.1.4.3 Chiral Metal Alkoxides Metal alkoxides with a hydrogen atom available in a-position, i.e. two bonds away from the metal atom, are capable of transferring a hydride ion to a carbonyl group. For practical purposes secondary alcoholates of aluminum and magnesium have proved to be of value. Reduction with achiral aluminum alcoholates (the Meerwein-Ponndorf-Verley reduction) has long been exploited as a diastereoselective method. Aluminum alkoxides formed with chiral secondary alcohols are very interesting for mechanistic studies but not for practical purposes, since their enantioselectivity is generally very low ( < 2 2 % ) [138]. Chiral alkoxyaluminum dichlorides, obtained by mixing 1 mole of LiAlH, with 3 moles of A1Cl3 and thereafter adding 4 moles of chiral alcohol [I 391, are much more efficient reagents, but are still not competitive with LAH-complexes except for special cases [ 1401. Thus (-)-menthyloxyaluminum dichloride reduces phenyl trifluoromethyl ketone to the (S)-carbinol in 77% ee [141], and (-)-bornyl-aluminum dichloride was used for the enantioselective reduction of aryl aminoalkyl ketones (60-9270 ee) [142]. Chiral halomagnesium alkoxides are rather similar to the alkoxyaluminum dichlorides, both in their scope and efficiency [139, 143, 1441.

3.1.4.4 Chiral 1,4-Dihydropyridines * Dihydropyridines, primarily nicotinamide-adenine-dinucleotide phosphate (NADPH), play a central role in biochemical redox processes. The reaction, which is shown schematically below, has been the subject of highly interesting model studies, and although the results are not yet of immediate practical interest, they deserve to be surveyed briefly. Since dihydropyridines themselves are not potent hydride donors, activation of the acceptor, usually by a divalent cation such as Mg2+ or Zn2+, is necessary. Enantioselectivity in the hydride transfer step requires, of course, that some chiral functionality should be attached to the dihydropyridine nucleus. With NADPH this is a dinucleotide moiety linked to the nitrogen. The standard experiment for the testing of NADPH models is the reduction of ethyl benzoylformate to ethyl phenyllactate in the presence of Mg(C104)2 in acetonitrile. Some model compounds are shown in Fig. 3-3. Not unexpectedly, highest selectivity was observed with 26, in which the hydridedonating carbon was also a chiral center. The configuration at the side chain was irrelevant to both configuration and ee. Thus (R,R ) - and (S, R)-26 both gave the

* For a review see ref.

[145].

96

3.1 Enantioselective Reductions

R

R

Fig. 3-3. Stereoselective reductions with chiral dihydropyridines.

(S)-alcohol in nearly the same ee (97.6 and 96% resp.) [146], while the analogue (27) with no chiral center in the dihydropyridine ring was quite unselective (ee 16%). Without added Mg2+ selectivity dropped to 5270, which suggested that Mg2+ not only activated the acceptor group but contributed to the rigidity of the activated complex. Chiral dihydropyridines were also used for the reduction of several other substrates. Among them conversion of tBuCOC0,Me with total selectivity to the @)alcohol and of PhCOCF3 with 70% ee to the (R)-alcohol by the 2-methyl analogue of 26 [I461 should be mentioned.

3.1.5 Correlation of Substrate Constitution and Enantioselectivity One of the aims of studies concerned with enantioselective reductions has always been the elucidation of the mechanism by which enantiotopic faces of a prochiral carbonyl group are distinguished. An obvious way to gather some information about the nature of the transition state is to determine selectivities with a range of substrates under identical conditions. In our case this approach exposed certain tendencies but failed to provide a real insight into the process. In Table 3-5 enantioselectivities obtained with the most important chiral reducing agents and a selected set of carbonyl compounds was compiled.

(Darvon alcohol) fresh reagent [5,40,411 aged reagent [5] LAH+(IR,2S)-N-methylephedrine + 3,5-dimethylphenol [31, 32, 341 LAH + (lR,2S)-N-methylephedrine +N-ethylaniline [37] LAH + (S)-4-anilino-3-methylamino-I -butanol [46] LAH + (S)-N-methyl-I-phenylethylamine [149] LAH + (S)-N-(2-dimethylaminobenzy1)-1-phenylethylamine [42] LAH + (S)-2-anilinomethylpyrrolidine [43]

1,2-diphenyl-3-methyl-2-butanol

LAH+ quinine [147, 1481 LAH + (2S,3R)-4-dimethylamino-

14 ( S )

0

41 (S)

6 (S)

21 ( S ) 21 ( S )

1 1 (S) 28 ( R )

35 ( R )

68 (S)

51 (S)

52 ( R ) 85 (S)

43 ( R ) 92 ( S )

50 (S)

77 (S)

90 (S)

84 ( S )

85 ( S )

78 (S)

85 ( R )

30 (R)

27 (S)

71 ( R )

20 (S) 17 ( R )

44 (S)

46 ( R )

98 ( R )

75 ( S ) 84 (R)

48 ( R ) 75 ( R )

42 (S)

12 (S)

8 (S)

71 ( R )

95 ( R )

18 (R)

82 ( R )

47 ( R )

86 (S)

28 ( R ) 63 (S)

2 (S) 36 ( R )

21 (S)

30 ( R )

MeCOEt MeCOiPr MeCOtBu PhCDOa PhCOMe PhCOEt PhCOiPr PhCOtBu PhCOCF,

LAH + (R)-2,2-dihydroxy-l ,l'-binaphthol+EtOH [17] LAH + 3-O-benzyl-l,2-~yclohexylidene-a-D-glucopyranose +EtOH [15] 24 ( S ) LAH + (2S,3S)-1,4-bis(dimethylamino)butane-2,3-diol [48, 491

Reducing agent

66-91 ( R ) 79-90 ( R )

57 - 90 (S)

R1C=CCORZc

Table 3-5. Enantiomeric Excess (070 ee) and Configuration of the Major Enantiomer in Reduction of Carbonyl Compounds with Chiral HydideDonating Reagents.

4

W

4

9

LiBH, + N-benzoyl-(R )-cysteine4 +tBuOH [151] LAH + (S)-2-(2,6-dimethylphenylamino)methyl-pyrrolidine [45] ( -)-Monoisopinocampheylborane [781 ( - )-Diisopinocampheylborane [771 (- )-Diisopinocampheylchloroborane [134] Potassium 9-0-(1,2,5,6-di-O-isopropylidene-a-D-glucofuranosy1)9-boratabicyclo[3.3. llnonane [73, 741 Lithium-B-(3a-pinanyl)-9-borabicyclo[3.3.1Inonylhydride [71] (S)-5,5-Diphenyl-4-isopropyl1,2,3-0xazaborolidin + BH, [89] NaBH, + 1,2,5,6-diisopropylidene-a-D-furanose (a) +ZnCI2 [56, 571 (b) +iPrCO,H [152] (c) +NaBH, serum albumin [65] (+)-B-(3a-Pinanyl)-9-borabicyclo-[3.3.llborane (Alpineborane) [120, 153, 1271 68 (S) 85 ( R ) 78 ( R )

SO (S)

98 (S)

94 ( R )

94 ( R )

43 (S)

13 ( R )

17 ( R )

29 (S)

78 ( R ) 45 ( R ) 85 (S)

92 ( R )

78 ( R )

3 (R)

4 (S)

9 (R)

13 (S)

30 ( R )

15 ( S )

96 S)

95 (S)

22 ( S )

88 ( R )

62 ( R )

87 ( R )

65 ( R )

28 ( S ) 57 ( R ) 66 ( R )

89 (S)

57 ( R )

43 ( R )

22 ( R )

MeCOEt MeCOiPr MeCOtBu PhCDOa PhCOMe PhCOEt PhCOiPr PhCOtBu PhCOCF,

LAH + (4S,5S)-2-ethyl-4-hydroxymethyl-5-phenyl-2-oxazoline[l 501

Reducing agent

Table 3-5 (continued)

77 - 100 ( R )

R'C=CCORZC

17 (S)

a

56 ( R ) 63 ( R )

20 ( R )

23 (R)

82.6 ( R )

13 (S)

12 (S)

47 (S)

22 (S)

30 (S)

22 (S)

50 ( R ) 16 (S)

Results obtained with deuterated reagents were transformed as if obtained with PhCDO and the protio reagent. Note the change in priority: CF,>Ph. R‘ and R2 alkyl groups, configuration given for priority of C-C over R’.

10.5 ( R )

44 (S)

38 (S) 59 (S)

38 (S) 15 (S)

6 (S) 24 ( S )

4 (S)

32 ( R )

13 (S)

70 ( S )

( - )-a-Pinanyl-aluminum dichloride [158]

2 (S)

8 (S)

76 ( S )

MeCOEt MeCOiPr MeCOtBu PhCDOa PhCOMe PhCOEt PhCOiPr PhCOtBu PhCOCF,

Tris[(S)-2-methylbutyl]aluminurn [156,1571 Tris[(S)-2,3,3-trimethylbutyl]aluminum [157]

(S)-2-Methylbutylmagnesium chloride [ 1541 (S)-2-Phenylbutylmagnesium chloride [1131

NB-Enantride [67]

Reducing agent

Table 3-5 (continued)

64 - 85 ( R )

10 (S)- 30 ( R)

RICE CCOR”

a

a

2

5a

@+ 0,

z2s.

?

0

&

L

+J

100

3.2 Diastereoselective Reductions of Carbonyl Groups

The substrates given in Table 3-5 have, as the only group capable of coordination, the carbonyl function, and the substituents attached to it are either apolar or of medium polarity. Therefore it is not unreasonable to anticipate that the structure of the transition states should be controlled by steric effects. Thus one of the enantiotopic faces of the carbonyl compound should be the more preferred the greater the difference between the effective size of the substituents. Thus stereoselectivities should increase in the order Et < iPr < tBu in the methyl ketone series, and decrease in the same order in the phenyl ketone series. This expectation was not borne out by experiment. In the dialkyl ketone series the general trend tends to be Et < iPr > tBu and in the aromatic series Me s Et > iPr, but in both exceptions are common. Since tBu is certainly more bulky than phenyl, with PhCOtBu there should be a change in the prevailing configuration relative to PhCOMe, but this occurs only rarely. Perhaps the irregular behavior of tert-butyl ketones can be traced back to the fact that this group cannot turn out in a way that a C - H bond should point towards the inside of the transition state complex (as is possible, e.g., with the isopropyl group). The behavior of trifluoromethyl phenyl ketone is also capricious. The product of the reduction is sometimes homochiral, in other cases, however, heterochiral to that obtained from acetophenone. The rod-like ethynyl group is one of the sterically least demanding groups, nevertheless enantioselectivities with a,,&acetylenic ketones are generally high and the products are (with one exception) all homochiral when the same reagent is used. A common trait of the phenyl and ethynyl groups is high n-electron density, which may be the clue to the strong stereodirecting effect characteristic for both. In order to be able to resolve the anomalies associated with enantioselective hydride transfer, a model based on some experimental evidence other than product distribution is called for. The retrospective construction of transition state models “SO as to fit results” is especially futile in this field, where very often even the exact structure of the reagent is uncertain.

3.2 Diastereoselective Reductions of Carbonyl Groups Diastereoselective reduction processes are highly interesting both for theory and praxis. The practically important cases involve almost exclusively the reduction of C = N and C = O bonds. The diastereoselective reduction of C = N bonds is most often carried out by catalytic hydrogenation and was discussed in Section 2.3.2. Thus, here we only deal with the reduction of the carbonyl group and discuss even

3.2.1 Stereochemistry of Diastereoselective Ketone Reducrions

101

this subject much more briefly than its importance would justify. Not only is our space limited, but also the most important aspects of this subject have already been assimilated by the chemical community and relatively few important novel contributions have been made recently to this field. Chemical reduction of a carbonyl group can be discussed in terms of a nucieophilic addition of hydride ion to an electrophilic carbon center and is therefore closely related to carbon-carbon bond-forming reactions involving nucleophilic addition. In fact the theoretical background for the two processes is very similar and the basic mechanistic aspects of both will be discussed in the present chapter.

3.2.1 Stereochemistry of Diastereoselective Ketone Reductions Although the necessary requirement for a diastereoselective reduction of a carbony1 group is that its faces should be diastereotopic (CJ Section 1.3), a significant degree of selectivity is generally only achieved when the carbonyl group is a member of a ring or when there is a chiral center close, preferably adjacent to the carbonyl group (1,2-asymmetric induction), or some interaction, e.g. by metal complexation can be established between the reaction center and the inducing chiral center. There is a sharp contrast in the behavior of the two systems, and much effort has been invested to offer an interpretation which would apply to both. A series of typical examples for asymmetric 1,2-induction in acyclic carbonyl compounds is shown in Fig. 3-4 [159, 1601. * Example (a) [I611 illustrates the effect of the constitution of the inducing group on selectivity, in example (b) [ 1621 the achiral ligand of the carbonyl group is varied, and finally in example (c) [163, 1641 the same substrate is reacted with a series of different nucleophiles. In Fig. 3-5 examples for the addition of nucleophiles to cyclic ketones are shown [165, 1661. It can be seen that even with quite bulky hydride reagents the more stable equatorial alcohols are formed from t-butylcyclohexanone by axial attack while with metal alkyls equatorial attack is preferred. With 2,5,5-trimethylcyclohexanone,however, equatorial attack prevails with any reagent. Inspection of Figs. 3-4 and 3-5 should convince anybody that the interpretation of stereoselection, even in such simple systems, is difficult, and the situation becomes

*

Here and throughout this book, when discussing diastereoselection involving a chiral compound, only one enantiomer is shown, but arguments apply equally both to its antipode and the racemic mixture.

102

3.2 Diustereoselective Reductions of Carbonyl Groups

R

Me

Et

iPr

tBu

%anti

74

76

85

98

(c 1 P

h

G

RLI, -50°C Et20

P

h

H

h

+

R

PhL

OH

OH ant/

\)'/I

R

* In THF at -78°C [166].

R

%syn

1

I

Me*

Bu*

iPr

tBu

Ph

34

90

87

86

84

Fig. 3-4. Asymmetric 1,2-induction in nucleophilic additions to acyclic carbonyl compounds.

even more complex when the whole range of reported examples, the effects of solvent, temperature, molar ratio etc. are also drawn into the discussion. From the fifties on, some of the best organic chemists have been challenged to give rationalizations both for the direction and the degree of stereoselections observed. The first successful attempt to predict the direction of stereoselection in acyclic systems was Cram's rule published in a paper by Cram and Elhafez in 1952 [ 1671. A similar interpretation of 1,4-asymmetric induction in nucleophilic additions to the a-ketoesters of chiral alcohols was independently provided by Prelog in 1953 [I 681. Cram's rule is summarized in Fig. 3-6. Substituents attached to the chiral center were classified as being small (S), medium (M) and large (L), and it was assumed that, due to complexation with the reagent, the carbonyl group became a bulky substituent and therefore in the most stable conformation the C = 0 bond bisected the S-C-M angle. Preferred attack from the side of the small (S) substituent follows from this model and was actually observed in a large number of additions [159]. Cram's rule is an example for a not uncommon situation in which a correct conclusion is reached on incorrect premises. First, as was pointed out later by Cornforth [169] and Karabatsos 11701, the conformation in which R is eclipsed with the

3.2.1 Stereochernisiry of Diastereoxlective Keione Reductions

(a) tBu

do ___) Nu

Nu NaB& LiA1& LiAI(OMe)3H LiAl(OtBu),H

t Bu &OH+

t Bu &NU

trans

cis

(axial attack)

(equatorial attack)

% tram

Nu

% trans

Nu

% tram

86 90 56 89

MeLi PhLi MeMgBr iPrMgBr

35

tBuMgBr PhMgBr

0 51

42 40 18

trans Nu %tram

I

I03

cis

NaBH,

LiAIH,

LiAI(0Me)SH

LiAl(OtBu),H

MeLi

MeMgBr

43

20

8

12

0

0

Fig. 3-5. Diastereoselective addition of nucleophiles to cyclic ketones.

L substituent is not the one preferred either in the ground state or the transition state. Second, the free energy difference between ground states is small compared with the free energy of activation. Therefore, the Curtin-Hammett principle (cJ Section 1.5) is applicable, and ground state conformations have little relevance to selectivity. Finally, substituents influence the direction of attack not only by their size, but also by electronic interactions with the incoming nucleophile. Actually the preferred conformation for rotation around an sp3-sp2 bond (e.g. in propene or acetaldehyde) is the one in which one of the ligands at the sp3 center is eclipsed with the double bond [171]. Corollaries of Cram’s concept are that: (i) stereoselectivity should increase when the difference in size between M and S substituents becomes larger and (ii) when more bulky reagents are applied; further, (iii) selectivity should diminish with increasing size of R since this would force ligand L out of its eclipsed conformation. There are not enough data available in the literature either to confirm or refute the first prediction. A confirmation of prediction (ii) is, e.g., the addition of methyl-, ethyl- and phenylmagnesium bromide to 2-phenylacetaldehyde giving the expected diastereomers in 33, 50, and 60% de resp. [167]. Prediction (iii) is clearly incorrect, as shown by example (b) in Fig. 3-4. Cram’s rule, as first formulated [167], is only valid when there is no chelating group in the substrate other than the carbonyl group, and none of the groups attached to the chiral center is highly polarizable. For cases not fulfilling these conditions, the “cyclic model” [I721 and the “dipolar model”, shown in Fig. 3-6 (b) and (c), respectively, were suggested, the latter by Cornforth in 1959 [169]. Some reac-

104

3.2 Diastereoselective Reductions of Carbonyl Croups

L

Fig. 3-6. Cram's rule. (a) The basic model. (b) The cyclic model (chelation control). (c) The dipolar model (Cornforth's rule).

tions which obey these models are shown in Fig. 3-7 (a) and (b). The validity of the cyclic and dipolar models (b) and (c) is even more subject to reaction conditions than that of the acyclic model (a). This is illustrated by example (c) in Fig. 3-7. Not only is the degree of induction dependent on the nature of the nucleophile but the predominant configuration may be inverted too. While the limitations of Cram's rule are obvious, its outstanding importance for the development of the theoretical background of stereoselectivity cannot be questioned. Existence of chelate intermediates has been detected by NMR for the addition of MeTiC1, [173] and for TiC14-catalyzed addition [174]. A kinetic study of Me,Mg addition to a series of a-trialkylsilyloxy ketones demonstrated that reaction rates and diastereoselectivities run parallel. With the bulky SiPr3 group where chelation by the reagent is precluded, the reaction is slow and the prevailing configuration inverted [ 1751. Chelation-controlled additions have been modelled by ab initio calculations 1176, 1771. An extension of Cram's rule deals with the kinetically controlled addition of carbanions with three different substituents to prochiral ketones [178].

105

3.2.1 Stereorhemistry of Diastereoselective Ketone Reductions

S

1) Be H (S')

2) 0

Me*

H

4 Me (L')

OH

H

Ph

90% de

1) Be 2) tBuCHO *

M~

54% de

tBu

I

Me

syn

anti

S

R

NuZ

% syn

S

R

NuZ

'70syn

H

pTol

LiAlH, PhMgBr PhMgBr MeMgBr

83 98 96

Me

Ph

MeMgI MeLi PhLi

66 89

Me Ph

88

96

R ' K o t i CI R'

R2

NuZ

Me

Ph

Et Ph

Bu Me Ph

LAH4 AI(OiPr)3 NaBH, LiAIH, LiAIH,

+

R

' CI

R2

w

u

R2

% svn ~~

PhM %syn

I I

75

66 80 43 SO

anti

syn atm ::)

PhLi (Et20)

PhlMg (Et20)

PhMgBr (EtZO)

PhMgBr (THF)

83

87

62

36

58

Fig. 3-7. Dependence of diastereoselectivity on reaction conditions and the nature of the reagent.

106

3.2 Diastereoselective Reductions of Carbonyl Groups

One-electron transfer-mediated reductions take an anti-Cram course, e.g. reduction of methyl 1-phenylethyl ketone with lithium in liquid ammonia gives the syn alcohol in 32% de [179]. In 1967 Karabatsos suggested an alternative approach to the problem [170], still based on ground state conformations, but taking into account newly acquired information about the conformations of carbonyl compounds [ 1711.

1

X

D

E

Fig. 3-8. The Felkin-Anh stereochemical model for the addition of nucleophiles to acyclic carbonyl compounds.

Karabatsos calculated that with R = Me rotamer B was about 3.3 k J - m o l - ' less stable than A and therefore only a minor component of the equilibrium. Thus product ratios can be correlated with the relative stabilities of A and C. Approximate relative stabilities of ground state conformations were calculated and were found to correctly reflect the trends of stereoselectivities [163, 1701, but because they were based on a groundstate model, they were unable to account for the effect of the reagent. Felkin and his coworkers were the first who not only directed attention to the transition state but also attempted to rationalize both acyclic and cyclic stereoselection by a common concept [162, 1801. They, too, adopted the hypothesis that, owing to the pronounced exothermic nature and high rate of the reaction, the transition state was essentially substrate-like rather than product-like, but also considered torsional strain generated by interaction between existing full bonds and the partial bond N u . * . C - . -0.The requirement that the approach of the nucleophile should be both

107

3.2.1 Stereochemistry of Diastereoselective Ketone Reductions

perpendicular to the R-CO-C, plane and staggered to the directing group, can only be fulfilled by conformations of the type shown in Fig. 3-8 (a). Here, out of the six possible rotamers those three are depicted in which the developing bond is close to the small group. The smallest steric interaction occurs in rotamer A , which leads to the product predicted by Cram’s rule, while B and C give rise to its epimer. To these considerations two more assumptions were added, namely (i) that unlike with the Cram and Karabatsos models, the dominant interaction was that between the entering group and the achiral group R attached to the carbonyl, and (ii) that due to polar effects those transition states are stabilized in which separation between NuZ and the electronegative group at the directing center was the largest (ie. the L group or the one of the highest electronegativity). The latter is a necessary amendment to the dipolar model, since it was shown that in the ground state the carbonyl group and the halogen atom are eclipsed rather than antiperiplanar [ 1811. The validity of this statement, however, depends on the relative size of the polar group and the M group, as was demonstrated by Canceill and Jacques [I611 for a series of ketones shown below. With R=Me and X = C 0 2 M e or CN the major product arose from an attack antiperiplanar to X, while with R = i P r or tBu, X took up the role of the M group.

.&”

MeOH KBH4+

=;

R

Jph

i

X

1

13 tBu

CN

COzMe :34;

5

43 5 0

55

20 16

Felkin’s model correctly predicts the trend observed when increasing the size of R and the lack of selectivity experienced with methyl 2-butyl ketone (L=Et, M=Me, S = H) for which rotarners A and C are almost of the same stability. A b initio calculations by Anh and Eisenstein for the approach of HO to 2-chloro- and 2-methylpropanal [182, 1831 established that transition state A according to Felkin represented an energy minimum, while those proposed by Cram, Cornforth and Karabatsos were all of higher energy (at least by 11 kJ-mol-’), i.e. they account for less than 1% of the total yield. Anh’s calculations showed that in the favored transition state the incoming nucleophile is antiperiplanar to one of the groups attached to the vicinal atom. The magnitude of this “antiperiplanar effect” follows the order CI > CH3> H. Later ab initio calculations by Houk on NaH addition on propionaldehyde contested some of the conclusions of Anh [184]. A weak point in Felkin’s argumentation was that the significant preference for rotamer A over B lacked adequate experimental support. An explanation was provided later by the hypothesis of “non-perpendicular attack” based on X-ray evidence [I851 and ab initio calculations [186], from which it was concluded that the nucleophile approached the carbonyl at an angle of about 107” rather than at 90”. This gives a clear preference for rotamer A over C (cJ: Fig. 3-8 (b)).

1 08

3.2 Diastereoselective Reductions of Carbonyl Groups

A b initio calculations showed that the difference in the free energy of activation for an attack by H- against the two faces of Ph(Me)COR differed by 0.61 and 0.95 kcal.mol-‘ for R = Me and iPr resp. [187]. The behavior of cyclic ketones in stereoselective nucleophilic additions cannot be rationalized by any of the models proposed by Cram, Cornforth or Karabatsos [188]. Although the general observation that reduction with NaBH, or LiAIH4 gives predominantly the equatorial alcohol with unhindered, and the axial alcohol with hindered cyclohexanones had been disclosed by Barton in 1953 [189], a global rationalization of this phenomenon has not been forwarded and it is rather unlikely that it will ever be possible. An excellent analysis of the situation was given by Wigfield [I901 who reviewed not less than seven different treatments of the problem, to which he added that of his own. Dauben, Tonken and Noyce argued that with unhindered ketones the transition state was product-like and the direction of addition was controlled by product stability (“product development control”), while with hindered ketones axial approach of the reagent was hampered (“steric approach control”) [191] (for cyclohexanone cJ: Fig. 3-9 A and B ) . This rationalization was, however, not adequately supported by facts. The most pertinent criticism out of many was that kinetic preference often surpassed thermodynamic preference for an equatorial alcohol in the product [192]. On the other hand, steric approach control was adequately documented by experiments and therefore became widely accepted. Thus cyclohexanones with an axial substituent at (2-3 are preferentially attacked equatorially, and this selectivity increases with bulky reagents (cJ: Fig. 3-5 (b)).

A

C

B

NuZ ---%

.O”+

D

E

Hi6 -

F

Fig. 3-9. Models for the interpretation of stereoselectivity in nucleophilic additions to cyclic ketones (demonstrated for cyclohexanone). A : Interactions in steric approach control. B : Interactions in product development control. C: Steric repulsion by vicinal axial hydrogens. D :The torsional strain effect. E : The antiperiplanar effect. F : Mechanism of borohydride reductions.

3.2. I Stereochemistry of Diastereoselective Ketone Reductions

109

Preference for axial attack in unhindered ketones was attributed by Richer [I931 to repulsion by axial hydrogens at C-2 and C-6 (Fig. 3-9, C). If we take into account that the attack is non-perpendicular, this interaction is even more significant than was assumed by Richer. A similar conclusion can be reached by applying the Felkin model to cyclohexanones [180]. The reagent approaching the CO group at 90" becomes almost eclipsed with axial C - H bonds at C-2 and C-6 (shown for C-Ha, in Fig. 3-9, D). These arguments were inverted by Anh et al. [I671 who argued that due to the above mentioned axial C - H bonds equatorial attack was not really inhibited, but, due to the antiperiplanarity effect, rather the axial attack was promoted. Flattening of the ring enhances this effect, and in fact axial attack is slower in some rigid polycyclic ketones in which ring flattening is difficult. It should be noted that in a nonperpendicular mode of attack the antiperiplanarity effect is much diminished. In an important theoretical treatment of the problem Wipke and Gund developed empirical functions to quantify steric congestion (i.e. ground state steric hindrance) and torsional effects at the reaction center. Results strongly supported that the effect of axial a-hydrogens was torsional in nature. Calculations for 52 cyclohexanones, cyclopentanones, steroidal and polycyclic ketones were carried out, and correlation with experimental values was good [195]. A calculation based on the linear combination of steric strain and product stability was recently published by Rei [196]. It was pointed out by Wigfield in his critical review [I971 that none of the above rationalizations took the mechanism of the reaction seriously into account. In a series of studies concentrated mainly on sodium borohydride reductions Wigfield established that it was futile to conceive any single rationalization even for hydride reductions, since the mechanism of the reaction was changing with the reagent. More specifically, borohydride reductions in alcohols proceeded by an acyclic mechanism, probably also involving the solvent (Fig. 3-9, F ) via a product-like transition state. The mechanism of reduction by LiAlH4 was not discussed but it could be established that a substrate-like transition state was involved and, finally, the transition state with LiAl(OtBu)3H was midway between the two types. Lack of product development control in reductions by LiAlH4 was demonstrated for bicyclic ketones by Ashby and Noding [198]. Kinetic isotope effects with LiAl(OtBu),H showed no variation with ketone structure and indicated that the character of the transition state did not change with the substrate. The ratio of rates for equatorial attack in 3,3,5-trimethyl- and 3,3,5,5-tetramethylcyclohexanone should reflect the contribution of product development control to the transition state, since in the latter the effect should be enhanced by increased 1,3-diaxial interaction. In fact no such effect was observed in LiAlH, reduction, while the ratio with LiAl(OtBu),H and NaBH, was 1.8 and 5.8 respectively. It must be emphasized that interactions in a product-like transition state cannot be equated with those in the product, since in the former the reagent is still present. Thus in borohydride reductions of unhindered cyclohexanones interaction with hydrogens at C-2 and C-6, while in hindered cyclohexanones interaction with axial group@) at C-3 and C-5 are dominant (c$ Fig. 3-10).

1 10

3.2 Diaslereoselective Reduclions of Carbonyl Groups

H

e BHI

-

a

H I

H

Ffg.3-10. Borohydride reduction of unhindered and hindered cyclohexanones

In LiAlH, reductions there is an intrinsic preference for axial attack due to the antiperiplanarity effect. This may be compensated for or overruled by steric interaction in the case of hindered ketones. Calculations by Houk gave further support to the hypothesis that equatorial attack was disfavoured by torsional strain in the transition state [184]. In a thorough study on a series of nine 3-substituted cyclohexanones Cieplak et al. [ 1991 found that in the addition of nucleophiles (mainly RLi) electronegativity of the 3-substituent increased the proportion of axial attack. Also sensitivity of product ratio to C-3 substitution decreased with increasing electron deficiency in the transition state. In the Cieplak model stereoelectronic control is associated with electron donation into the o**,i.e. the vacant orbital associated with the incipient bond.

3.2.2 Practical Aspects of Diastereoselective Ketone Reductions 1,2-Induction in acyclic systems Although studies on diastereoselective nucleophilic addition to ketones were of outstanding importance for the development of theories about asymmetric induction, their contribution to the solution of practical problems is less impressive. While methods of excellent stereoselectivity have been developed in the past decade for the nucleophilic addition of carbon to carbonyl in acyclic systems (cJ:

3.2.2 Pracfical Aspects of Diastereoselective Ketone Kerlucfions

111

Chapter 5 ) , stereoselectivity in the reduction of acyclic carbonyl compounds is seldom at a high level. In Table 3-6 the most important data available in this field have been compiled. Interpretation of the results in terms of the concepts discussed in the preceding section is not always straightforward. The role of reagent size is rather elusive. Thus K(sBu)~BHis more selective than LiAlH,, but Na(Me0)3BH and the highly hindered 9-BBNH are less so. The outcome of the reduction of 3-ketoesters is highly dependent on the counter ion: KBH, gave a moderate excess of the anti product, while this was the only product with Zn(BH,),. Since Zn2+ is a much better complexant than alkali metals, it is not curprising that Zn(BH,)2 is superior to other complex hydride donors [200] in the reduction of a-hydroxy- [201], phydroxy- [202], a-carboxamido- [203], and a-methylthio- and phenylthioketones [204]. Unexpectedly, with the latter Zn(BH,)* gave the antialcohol, while Li(sBu),BH afforded the syn product. The stereoselectivity of Zn(BH4), is only paralleled by L-Selectride@,i.e. Li(sBu)3BH, which is highly selective not only in the presence of a polar neighboring group, but also with vinylketones [205]. This property has been exploited in the synthesis of prostaglandins. Recently, efficient stereodirection in the borohydride reduction by an a-sulfonium group was demonstrated [206]. Combination of enolate acylation in the presence of a chiral auxiliary group and subsequent diastereoselective reduction permitted the synthesis of 2-alkyl-3-hydroxycarboxylic acids in high enantiomeric and diastereomeric purity [207]. CH20CH20Me

C.HzOCH2OMe syn 98% 'Yo ee > 94%

'Yo

I) BuLi, THF, -711OC 2) R'COCI, 3) Z n ( B H &

0

CHzOCH20Me

OH

0

CH20CH20Me

R'=Me, Et, Bn, R2=Et, IPr, tBu, Ph

With a-aminoketones selectivity is much influenced by substitution at the nitrogen. Owing to the importance of arylethanolamines to the pharmaceutical industry, research on this subject has been intensive [216, 2331. Lack of an adequate rationalization did not prevent utilization of the effect of N-substitution. Generally, product B (cJ:Table 3-6) (being the anti isomer with amines) is the prevailing product with primary and secondary amines, while with tertiary amines A is in excess. Thus reduction of the aminoketone 28 gives the syn isomers (debenzylated to 29) in excess, while the parent amine gives mainly the anti product (30) (Fig. 3-1 1). As an explanation it was proposed that hydrogen bonding stabilizes conformation 31 in primary and secondary amines, while the transition state for tertiary amines has to be envisaged as 32. Interestingly, even tertiary amines gave the anti alcohols as the major product when the hydrochlorides were reduced in methanol at less than 25 "C [204]. Recently several examples of reductions were described in which change of reagent and/or conditions inverted the prevailing configuration. Thus with the nucleophilic

1 12

3.2 Diastereoselective Reductions of Carbonyl Groups

Me

I

Me

OH

(30)

I

Ar

R’

Fig. 3-11. Reduction of a-aminoketones.

K(sBu)~BHketoamide 33 gave only the anti alcohol, while with the electrophilic (iBu)2A1H mainly the syn alcohol was formed [222].

%h C02Et OH de 92%

--

K(FF

C0,Et I 6H &-looO/o

%h C0,Et 0 (33)

A similar reagent-dependent selectivity was observed in the reduction of amide 34 [236]. Me

R*

LR

Na(sBu),BH *THF, - 78 OC

OH

5zkEFE-

Qs.& 0

(34)

0

ZnBH, Et20, - 20 OCm

Me

R*

bH

Reduction of the 1,3-diketone PhCOCH(R)COPh (R = Me, Bn) with Li(tBuO),AlH is solvent dependent, in THF at -30°C the anti,syn-diol (de 96 and 92% resp.) while in CH2C12the syn,syn-diol is formed in excess (de 94%) [237].

3.2.2 Practical Aspects of Diastereoselective Ketone Reducfions

1 13

Table 3-6. Diastereoselective Reduction of Acyclic Ketones with a Chiral Center Adjacent to the Carbonyl Groupa.

(A R'

R2

Apolar neighboring groupsb Ph Me

OH sy"

R3

R4

Reagent

H

Me

AIHC12/Et20 lIAIH4/THF LiBH4/THF Li(sBu)3BH/THF K(sBu)~BH/THF Al(OiPr)3/iPrOH Na(MeO)3BH/Et20 9-BBNH/THF NBH4/Et20 + HO /aq Na(MeOCH2CH20)2AlH /C6H6 NaB(CN)H3/MeOH + H' LiAlH,/Et,O, 35°C LiA1H4/Et20, 35 " C LiAIH4/Et20, 35°C LiA1H4/Et20, - 70°C LiAIH, LiAIH, LiAIH, LiAIH, LiAIH, Al(OiPr)3 LiAIH,, - 70°C LiAIH,, 35°C NaBH, LiAIH, Al(OiPr)3 LiAIH4 LiAIH, LiAIH, LiAIH4/THF LiBH4/THF Li(sBu)3BH/THF Li(sBu)3BH Li(sBu)3BH L~(sBu)~BH L ~ ( s B u ) ~ B H-,78 "C Zn(BH4)2, - 30°C

Et iPr tBu

Ph

iPr Et

Et

Me

Me

Ph iPr Ph Me Et Me

cHex cHex

Me

H

Et iPr tBu

cHex

Me

H2C = C H (Z)-BuCH = C H (E)-BuCH = CH Me Me

Ph(CH2)2 SiMe2(iPr)2 OBn

Polar neighboring groupls) CI Et Ph Me OH OMe OH

Ph Me CH20H Me

(B) anti

H H

Me PentC-C

H

Bu Me Ph

H

pTol Ph H Pent iPr

NaBH4 LiAIH, LiAlH4 Al(OiPr)j LiAIH, NaBH4 NaBH4 LiAID,, - 42 "C LiAIH, ZnBH4 (iBu)3AI

% A

86 82 80 94 85.4 61.5 67.1 53.7 67.3 69.5 81.1 76 85 98

Ref.

80 61.5 65 69 78 96 99 99 97.5 85'

[212] [212] I2121 [212] (2131 [213] [213] (2131 [213] [213] [213] [162] 1162) [I621 [I621 [I671 [209] [209] (2101 12101 [208] [208] [208] [211] [162] [211] [I621 [I621 11621 (2121 [212] [212] I2051 [205] [205] (2211 [232]

80 43 75 66 84 73 58 60d 64 77 61

[I691 [224] [224] [224] [225] 12141 [214] [226] [227] (2011 (2281

100 80 96 69 75 75 56.5 51 48.5 62.5 61.5 34.5 66.5

1 14

3.2 Diustereoselective Reductions of Curbonyl Groups

Table 3-6 (continued) R'

U2

R3

R4

tBu Ph Et Pr Bu Pent

Bu Pr Et Me

iPr tBu Ph Me iPr tBu C02Me

Me

H

Ph

C02Me

Me

H

CH2C02H CONHPh SMe

Me

H

Ph (E)-CH = CHMe Ph(CH2)2 Ph Me

Et

Ph

iPr CH20H CH20Ac SPh

Me

Bu iPr

S @ Me2 S@MePh

Et Bu iPr Nitrogen as neighboring group,c CH2NH2 NH2 CH2OH NHBn Me NMe2 NMeBn NHAc NHBz NHAc N-Phthaloyl NHAc CH2OH Me NH2 CH2NH2 CH20H CH20Et Me NHMe NHBn NMetBu NBn2 Ph NiPr2 NMeBn Me Piperidino

H

Ph

4-N02 - C6H4 Ph H

Reagent

Ref.

A

ZNBH& LiAIH4 LiAIH4 Zn(BH4)2 ZnBH4 ZnBH, ZnBH, LiAIH, Zn(BH4)2 (iBu)3A! Zn(BH& (iBu)3Al LiA1H4 LiAIH, (iBu)3Al Zn(BHd2 A!(iBu)3 Al(iBu), LiAIH, KBH, Zn(BH4)2 Zn(BH4)2 Zn(BH4)2 DIBAL/ZnC12 Zn(BH& Zn(BH4)2 Li(sBu)3BH Zn(BH& Li(sBu),BH Zn(BH4)2 Zn(BH& Li(sBu)jBH Zn(BH& Li(sBu)3BH Zn(BH& Li(sBu),BH NaBH4 NaBH4 NaBH4

85 87 87 98 89 > 99 87 70 87 83 85 7s 75 80 82 90 85 91 90 27 97 96 56 85 98 6 > 99 4 96 93 27 96 82 98 61 99

NaBH, NaBH, NaBH4 NaBH, NaBH4 NaBH4 NaBH4 Al(OiPr)3 Al(OiPr), Al(OiPr), NaBH4/EtOH NaBH4/EtOH NaBH4/EtOH NaBH4/EtOH NaBH4/EtOH NaBH4/EtOH NaB H4/E tOH LiAIH,/Et20 LiAIH4/Et20 NaBH4/EtOH NaBH4/MeOH

36 43 12 54 7s 23 23 28 67 47 5 50 19 8 7

1

20 1

0

97 100 100 88 94

I 15

3.2.2 Practicul Aspects of Diustereoselective Ketone Reductions Table 3-6 (continued) R'

R2

C02Me

NBn2

Et02CN(CH,), -

OBn

I

EtO&N(CH,),

-

R3

R4

Reagent

H

Me Ph Hex Ph

NaBH4/EtOH NaBH4/EtOH NaBH4/EtOH K(sBu)~BH

H

Ph

DIBAL

H H H

Ph Ph. Et'

PhMe2SiH PhMe2SiH K(sBu)~BH

% A

Ref.

93 98 96 100

[221] (2211 [2211 [222]

92

[222]

>99 >91 >98

[71] [71] I2321

I

Et2NOC Et2NOC

--OMOM

a

'

Me Me Me

For reviews, see refs. 1217) and 12331. If R' = L, R2 = M and R3 = H A is the anti and B the syn isomer, A is preferred according to Cram's rule. If R' is a polar group (OH, C02R, SR, NR2 efc.)and R3 = H, A is the syn and B the anri isomer. Under chelation control A is formed preferentially. Replace H by D in formulas A and B . Under kinetic conditions. For R' = Hex, iPr and tBu % A 94, 92 and 50% resp. 12321. For R' = Me, Ph, and CH2C02Et Vo A 70, 94 and 92% resp. 12331. In the presence of Me2N(CH2)2SCSiFZMe;. I n the presence of CF3C0 H at 0°C. . tBu, cHex, Ph. Same selectivity also for R4 = IPr,

1,3-Induction in acyclic systems

1,3-Induction, i.e. the stereodirecting effect of a chiral center separated by two bonds from the carbonyl group, plays an important role in C - C bond-forming reactions (cJ: Section 5.1.2), but in hydride transfer reactions in the absence of a polar group the effect is generally too weak for practical application. For substrates without a polar directing group slight to moderate excess (up to 56% de) of the 1,3-anti product was found in reductions with LiAlH4 [238]. A chiral sulfur atom is a moderately effective directing group, which can then be removed by Raney nickel desulfurization to give an optically active alcohol. For this purpose both the sulfoxides and the sulfoximides are useful, but sulfoxides are more accessible and with the proper choice of reducing agent better selectivities and control of relative configuration can be achieved. In reduction of sulfoximides with NaBH4 at room temperature [239] or with BH3.THF at -78°C [240] up to 69% de could be realized. Removal of the chiral auxiliary group by treatment with Raney-nickel is simple in both cases. Results with a sulfoxide as the directing group and LiAIH, as the reducing agent are rather similar [241]. SolladiC et al. discovered that reduction with DIBAL not only secured high diastereoselectivity, but configuration at the new chiral center could be inverted by prior complexation with ZnC12 [242].

1 16

3.2 Diastereoselective Reductions of Carbonyl Groups

OH

Lowering of the temperature to - 100 "C improves selectivity to 94-98% ee with the ZnCl,/DIBAL method [243]. The method has been extended to enones [244, 2451, aryl-P-ketosulfoxides [246] and to chloromethylketones [247]. In a repeated relay-like 1,3-induction process anti-l,2-diols can be prepared [248].

RU

\

p

T

o

LR* QH

0

DIBAL - 78 O(2C eq.L

R

,

Me4#8(OAc),H AcOH, R = Me*

xR* R

D -Raney

=

OH

- Ni

Me

Me, Et02CC%, Ph, Mede > 95% QH

88% ee

Me

Me

High diastereoselectivity in the reduction of the boronates 35 is attributed to the formation of a six-membered chelate rather than to the effect of the /3-substituent [249]. Me 1) BH,.THF,

- 78OC

2) H202, OHe

R'

QH

R'

RZ

&OH

(35)

In /3-hydroxyketones the directing effect of the hydroxy group can be efficiently amplified by transient esterification with an alkoxyborane leading after reduction with NaBH4 to syn-1,3-diols [250,251].

R'

R'

1-

R'

R2

R', R2 = Bu, Ph, de > 96%

3.2.2 Practical Aspects of Diastereoselective Ketone Reductions

1 17

Zn(BH,)2 in ether at -20°C [251] or DIBAL in T H F at -78°C [252] are also syn selective with similar substrates, but give lower selectivities (de 64 - 84% and 84 - 90% resp.). In the reduction of the a-substituted-/I-ketol 36 1,3-induction dominates over 1,2-induction [253] giving the 1,3-anti products (37) in 96% de, while in the 0silylated derivatives 38 the configuration of this center has little influence on the outcome of the reduction, diastereoselectivity at the new center being >88% [254].

(36) 2,3-~ynor anti

q

Me,tBuSiO

Rl

R3

R2

(38) (2,3-syn or anti)

.TR, (37)

1) LiAlH,, EGO, - 7 8 i C 2 ) H@

R1 = iPr, vinyl, Ph R2 = Me, Bu; R3= Et, iPr, Ph

R2 1,2-anti

Ketalization by (R,R)-2,4-pentanediol and reductive ring cleavage was employed with success by Yamamoto et al. [255,256]. Configuration can be controlled by proper choice of reagent. A drawback is that removal of the chiral auxiliary is cumbersome.

OR*OH

Me Me Bu Bu Ph Ph iBu cHexMe Et Me Et

* R, and R,

not specified in ref. [256].

Fashionable but expensive Sm12 catalyses the Tishchenko type reduction of /Iketols with acetaldehyde to anti-diols (de > 98%) [257].

1 18

3.2 Diustereoselective Reductions of Curbonyl Gsoups

Asymmetric induction by remote centers in acyclic compounds 1,4-Induction played a very important role in the development of our understanding of stereoselective reactions. At the turn of the century it was discovered by McKenzie [258] that Grignard addition to esters of 2-keto acids with chiral alcohols gave, after hydrolysis, optically active 2-hydroxy-2-alkyl acids. In 1953 Prelog reviewed the accumulated data and formulated a prediction known as Prelog's rule, by which the prevailing configuration of the product could be deduced from the constitution and configuration of the chiral alcohol [168]. As in Cram's rule, that of Prelog also operates by drawing the molecule in a fixed conformation and classifying substituents at the inducing center as small, medium and large. The rule is illustrated below:

phkyph'L) Hoyph OH

2) I ) NaBH, ne

0

Ph

Me H ( S ) (M)

+

Me H

75% ee

Poor selectivity of the original method (generally < 20%) could be dramatically improved by using the proper combination of reducing agent and chiral auxiliary [259, 2601. Even control of configuration may be possible [260]: OH

I

COPe

1)DIBAL. LiBr 2) Ha

Ph*CO,H

Ph

OH 1) LiBH,, LiBr TW, - 78 OC; 2) H S m P h A m , H

66 YOee

87% ee

An example for combined 1'5- and 1,3-, or with some imagination even 1.7-induction is the following reduction [261]:

NaBH,, E%BOMe THF, - 78 OC,+ r.t. Md

OH

OH@ ____)

$)H

Ho p 2c-h

100% de > 95 9'0 ee

P-Ketoesters of the above chiral auxiliary can be reduced to both ( R ) - and ( S ) alcohols with high selectivity using Zn(BHJ2 +ZnCI2 or DIBAL and 2'6-ditert-butyl-4-methylphenol resp. [262].

3.2.2 Practical Aspects of Diastereoselective Ketone Reduclions

119

A systematic study of the LiAlH, reduction of non-enolisable diketones revealed that 1,2- and 1,4-diketones give the meso-diols, while that of 1,3- and 1S-diketones the racemates in excess [263, 2641. A rare case of 1,6-induction is shown next: Tos

1) K(sBu),BH

2)H'

*

Ar

40% ee

The following reaction is an example both for 1,7-induction and metal-catalyzed stereoselective pinacol cross coupling [266]:

92% ee

Reduction of cyclic ketones With cyclic ketones it is impractical to discuss 1,2-, 1,3- and 1,4-induction separately, so these cases will be discussed together. With cyclohexanones, including their complex derivatives such as steroids and terpenes, there are useful practical methods available to produce both the equatorial and the axial alcohols. It can be seen from the foregoing discussion that a different approach has to be applied with unhindered and hindered cyclohexanones. When the objective is to prepare an equatorial alcohol from an unhindered ketone LiAlH, or NaBH, is the reagent of choice (cJ Fig. 3-5 (a)), while very high selectivity for the axial alcohols can be achieved with very bulky reducing agents, e.g. with Li(sBu),BH (L-Selectride) 12671 or K(tBuBBNH) [ 1821. Hindered ketones give axial alcohols preferentially with any hydride donor, and this effect can be enhanced by applying a bulky hydride donor (cJ Fig. 3-5 (b)). Finally, equatorial alcohols cannot be obtained from hindered cyclohexanones by a kinetically controlled process, but by equilibration, e.g. over Raney nickel [268], it is possible to convert an axial alcohol to the more stable equatorial alcohol. An obvious limitation to stereoselectivity is when the substrate is conformationally not homogeneous. Lowering of temperature may then be an effective measure. It was shown [218] that at -78 "C a methyl group is almost as good a conformation biasing group as tert-butyl. Selectivities achieved by bulky hydrides at low temperature are impressive, as demonstrated by data in Table 3-7. Surprisingly, Li(iBu)2tBuH, which should not be less bulky than Li(Siam),H, is a less selective reagent [274, 27-51.

120

3.2 Diastereoselective Reductions of Carbonyt Groups

Table 3-7. Reduction of Alkylcyclohexanones with Bulky Hydride Donors. Substituent(s) (Temp. "C)

4-tBu 4-Me 3-Me

(0) (-78) (0) (-78) (0)

(-78) 2-Me (0) 3,3,5-Me3 (0) a

9'0 Isomer from equatorial attack

Li(sBu),BH ~701

LiMes,BHZa Li(Mecp),BH 12721 12711

93 96.5 80.5 90 85 94.5 99.3 >99

94

Li(Siam)BHC K(tBuBBNH)d 1273, 2741 12821

> 99

> 99

98

98

99 99

> 99 > 99 > 99

> 99

94

94

99 99

Mes = 2,4,6-trimethylphenyI. Mecp = trans-2-methylcyclopentyl. BBN = 9-borabicyclo[3.3.1]borane.

98

>99 > 99

Siam = 3-methyl-2-butyl.

Reagent

Vo trans

Ref.

LiAlH, Li(OtBu),AIH Li(sBu),BH Li(Mes),BH;

84 72 98 98

I2801 I2811 ~701 I2711

em

"Mes = 2,4,6-trimethylphenyI

endo

Reagent

% ex0

Ref.

LiAlH, Li(OtBu),AlH Li(sBu),BH Li(Mes),BH;

91

I2801 P811 I2741 I2681

93 98 > 99

Fig. 3-12. Stereoselective reduction of 2-methylcyclopentanone and camphor.

A series of reducing agents obtained by mixing LiH with magnesium alkoxides, prepared from hindered phenols or alcohols such as 2,6-di-tert-butylphenol and 2,2,6,6-tetramethylcyclohexanol,also exhibited almost total selectivity for axial at-

3.2.2 Practical Aspects of Diastereoselective Ketone Reductions

121

tack with 2-methyl- and 3,3,5-trimethylcyclohexanoneand over 80% preference with 4-tert-butylcyclohexanone [276, 2771. LiAI(OMe),H and even LiAl(OtBu),H are in turn only moderately stereoselective hydride donors, presumably because of the intervening oxygen atom [ 15 1]. Potassium 9-(2,3-dimethyl-2-butoxy)-9-boratobicyclo[3.3.l]nonane,prepared by Brown et al. [278], was a highly selective reagent, but less so than e.g. Li(Siam),BH. Cyclopentanones are conformationally mobile molecules and thus it is more difficult to reduce them stereoselectively than cyclohexanones. Usually only one of the epimeric alcohols can be obtained in large excess. Some results are shown in Fig. 3- 12 (a). Product distributions for the reduction with LiAlH, of a series of 2-alkylcyclopentanones was successfully predicted using the Ruch-Ugi approach [279] and in the case of 2-methylcyclopentanone by Wipke’s calculations [ 1951. Camphor can be regarded as a typical sterically hindered bicyclic ketone and has been used as a test substrate for the evaluation of hydride donors. Endo attack is preferred with all of them. Typical results are shown in Fig. 3-12 (b).

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

4 Stereoselective Oxidations

For structural and mechanistic reasons the range of stereoselective oxidations is rather narrow. Oxidation of a hydroxy group to a carbonyl group is a stereodestructive transformation, while oxidation of hydrocarbons is a radical process of high activation energy with little chance to be conducted stereoselectively. Oxidation of sulfides to sulfoxides and of amines to amine oxides may create a chiral center, but this aspect has been studied little. The stereoselective oxidation scene is dominated by studies on epoxidation and syn-hydroxylation by osmium tetroxide. Recently oxidation of enolates leading to a-hydroxy carbonyl compounds gained some importance. Epoxides are very important intermediates which can be easily transformed to many other compounds.

4.1 Enantioselective Oxidations 4.1.1 Epoxidation with Chiral Oxidants Henbest was the first to use a chiral peracid, percamphoric acid, for enantioselective epoxidation [I]. Enantioselectivity was disappointingly low even with purified peroxycamphoric acid ( I ) (< 15%) [2] and useful levels of enantioselection in oxidations with chiral peracids could not been attained ever since. Chiral N-sulphonyl oxaziridines are enantioselective oxidizing agents, as shown first by Boschelli et al. [3]. Among them the camphor-derived sultams 2 [4, 51, 3 [6] and 4 [7] were used successfully for the oxidation of enolates [4, 6, 81 and sulfides [7, 91 as exemplified next:

124

4.1 Enantioselective Oxidations

Ph M ,(ef,

1) NaHDMS,THF,-78 "C, 2 ) (3)

Ph

kMe .

95 % ee

OH

4.1.2 Oxidations in the Presence of Chiral Catalysts Epoxidation in the presence of chiral phase transfer catalysts was actively investigated by Wynberg et al. and revealed interesting correlations between catalyst and product configuration [lo], but failed to turn out a practically useful method. Surprisingly high enantioselectivities (up to 96% ee) were observed in the epoxidation of chalcones with 3-chloroperbenzoic acid in the presence of poly-(S)-alanine N-butylamide using a triphasic system. Optical purity increased with the molecular weight of the peptide [Ill. Adapting the method vanadyl-acetylacetonate-catalyzed epoxidation of allylic alcohols with tert-butyl hydroperoxide Yamada et al. [I21 and Sharpless et al. [13, 141 concurrently disclosed that similar complexes of molybdenum and vanadium respectively containing chiral ligands (5 and 6 ) catalyzed enantioselective epoxidation of allylic alcohols (Fig. 4-1).

XO"

R2

RZ

tBuOlH, Ti(OiPr),, CHzCIz, - 2 O T (+)- d ie t h yl tartrate, 45 - 87 '%,

R3 R' H

CioHzi H

R2

R'

Et Me Me H H CIOHZI Ph Ph

125

LO"

4.1.2 Oxidations in the Presence of Chiral Catalysts

"'.o

R3

%ee

R1

R2

R3

YOee

r95 >95* > 98 91 99

H

H Me02C(CH& Me,C=CH(CH,), Me

cHex H

> 95'

Me

Me& = CH(CH,),

95

95 94

* With diisopropyl tartrate.

Fig. 4-1. Enantioselective epoxidation of allylic alcohols with the Sharpless method.

Although selectivities were unsatisfactory (up to 50% ee), these processes deserve mention because they were among the first examples of efficient asymmetric catalysis. All these results were soon overshadowed by the discovery (Katsuki and Sharpless 1980) that in the presence of titanium (Iv) isopropylate and (+)- or (-)-diethy1 tartrate allylic alcohols could be epoxidized with very high enantioselectivity [15, 161. A selection of epoxidations carried out with this simple and relatively inexpensive method is shown in Fig. 4-1. It is apparent that the method not only gives high yields and excellent enantioselectivities, but is also very flexible as regards the structure of the allylic alcohol, except that if R' or R4 are tBu enantioselectivity drops sharply [17]. The procedure proved to be extremely useful for the synthesis of insect pheromones and other natural products and by now the number of papers describing its application may amount to more than a hundred. Since tartrates are available as both enantiomers, there is no restriction as to the configuration of the product. Anhydrous conditions are essentiell [ 181, and sometimes diisopropyl tartrate gives better results than the ethyl ester. Addition of 4 A molecular sieve permits the use of only catalytic amounts (5 mol%) of Ti(OiPr)4 [I91 and the generally slow rate of oxidation can be accelerated 4 - 5 times by the addition of CaHz and silica gel [20]. Since participation of the hydroxy group of the substrate is evidently essential for selectivity, it is not surprising that homoallylic alcohols are oxidized much less selectively than allylic alcohols and give the opposite prevailing configuration (ee 27-55%) [21]. Zr(OiPr), performs in this case somewhat better than Ti(OiPr)4 (ee 10-77%) [22]. A search for ligands other than tartrate esters revealed that at least one ester group and the chiral vicinal diol function is indispensable for high enantioselectivity [23]. Sharpless and his coworkers studied the mechanism of the oxidation and established the following characteristics of the reaction [24- 281: (i) Enantioselectivity is at its optimum at a 1 : 1 molar ratio of Ti(OiPr), and tartrate ester. (ii) Mixing of the two reagents results in a rapid release of 2 moles of alcohol whereby a dimeric species is formed.

126

4.1 Enantioselective Oxidations

sRTCO,R (7)

(8)

Fig. 4-2. The stereochemical model of the Sharpless epoxidation.

The dimeric species was initially envisaged as a 10-membered ring [24], but as evidenced by IR, and NMR spectroscopy as well as X-ray studies on the complex of tartaric acid benzylamide [26] structure 7 was suggested to represent the reagent in the absence of substrate and oxidant. 7 is a fluxional structure with rapid exchange of the carbonyl groups coordinated to titanium. C, symmetry about titanium is apparent from the figure. (iii) As shown by kinetic experiments *, on addition of the substrate and the oxidant (both are alcohols) the remaining two isopropyloxy ligands are displaced by allyloxy and tert-butylhydroperoxy groups. The reactants are thus assembled on the metal in a chiral environment prior to reaction. Epoxidation takes place, as the rate * determining step on the chiral titanium template and the reaction cycle is completed by displacement of the product and the spent oxidant by new substrate and hydroperoxide molecules. The importance of the 1 : 1 (actually 2: 2) complex 7 was underlined by experiments with tartaric acid benzylamide and 2,3-diphenylallyl alcohol. On changing the Ti/5 ratio from 1 : 1-2 to 2 : 1 the configuration of the major product became inverted [29]. This was explained by disruption of the dimer by excess of tartrate ester. Added substrate and oxygen displace two moles of alcohol from the complex and when this occurs at the same metal core, a reactive intermediate is formed. Regarding the structure of this species opinions diverge. In Corey’s opinion [30]the dimer disproportionates to an ion pair, the anion of which is binding both substrate and oxidant and is the site of oxygen transfer. This concept explains opposite enantioselection with homoallylic alcohols, but is conflicting with the 2nd order kinetics of inhibition by excess alcohol [25]. The loaded titanium complex and transfer of oxygen is subject to the following constraints: (i) The substrate must occupy one of the two open quadrants available around each titanium atom. (ii) The hydroperoxide is not only fixed to titanium by the O,-Ti bond but also by coordination to one of the enantiotopic lone pairs at

*

rate

=

k [tBuO,H] [Ti (OiPr), (tartrate)] [allylic alcohol] [inhibitor alcoholJ2

127

4.1.2 Oxidations in the Presence of Chiral Catalysts

0,. In this way chirality of the catalyst is extended onto the oxidant. (iii) Transfer of oxygen requires perpendicular attack of 0, at the n bond, while the TC*orbital must overlap with one of the lone pairs at OD. A model constructed along these lines (8) is in accordance with the observed enantioselectivity. In contrast to epoxidation of allylic alcohols that of olefins without a hydroxyl group has been only partially successful. Oxidation with hypochlorite catalyzed by a chiral manganese complex is promising [3 1,321: Ph

Ph

R

Ph

v /I

\

tBu

/ tBU

cat*

A similar complex catalyzes the epoxidation of dihydronaphthalenes with oxygen in the presence of N-methylimidazole [33]. To the enantioselective oxidation of unsymmetrical sulfides to sulfoxides the Sharpless system has been adapted by Kagan et al. who used cumene hydroperoxide as oxidant [34]. In contrast to epoxidation requiring strictly anhydrous conditions, water is an essential component of the system. Diethyltartrate may be replaced by 2,2’-binaphthol [35]. Me.

Me

0

Several chiral media have also been tested for the same purpose, among them bovine serum albumin (BSA) with H 2 0 2 [36-381, dioxiranes [38] or periodate [39, 401 as oxidant. Osmium tetroxide is well-known as a reagent for the syn hydroxylation of olefins. * Hentges and Sharpless developed an enantioselective version of the method by adding, as base, dihydroquinine acetate instead of pyridine. From modest beginnings [42] (poor selectivity except for stilbene, stoichiometric amount of Os04) the method has been developed, mainly by Sharpless and his group to an efficient catalytic method. Most work has been done with 9-0-acyl derivatives of the quasi enantiomeric pair of quinuclidine alkaloids, dihydroquinine (DHQ) and dihydroquinidine (DHQN) giving enantiomeric diols as shown for styrene and the 4-chlorobenzoates of the alkaloids [43] (Fig. 4-3 (a)).

* For

a review see ref. [41]

128

4.1 Enantioselective Oxidations

Ph

OsO, (0 2 - 0.4%), Ac - DHQ Me,CO, YO, NMO Ph-

&Ph f

OH

, 2 - 0.4%), Ac-DHQN Ph 0 ~ 0(0 Me,CO, q0,NMO

AC = 4-Cl-C6&C0

(SJJ ce 78%

P h q OH P h

(H,R)ee88%

OAC -

Ac - DHQN (9R)

Ac - DHQ (94

(b)

’PR2

R’

(C)

OsO, + INDCO - DHQN K3[Fe(CN),I, W O , , t h 0 H - H20 (1 : 1)

kR2

R’

%ee 72

OH

R2

OH

Bu(CH,), Eh C0,Et Ph

N-N

R O q

F O R (9) R = DHQN (for p - AD - mix) (10) R = DHQ (for a - AD mix)

-

Fig. 4-3. Enantioselective dihydroxylation of olefins with osmium tetroxide,

H

H

H

129

4.1.2 Oxidations in the Presence of Chiral Catalysts

The process has been improved by optimizing experimental conditions [44], replacing NMO by inexpensive K3 [Fe (CN),] [45,46] and the 4-chlorobenzoyl group by others [47]. Thus with the 9-phenanthryl or 4-methylquinolyl [48] derivatives highly selective oxidation of 1-alkyl- [48] and 1,2-dialkylethenes [47] as well as of trisubstituted alkenes [48] was realized. An acyl group specially developed for the oxidation of unsymmetrical cis-olefins (symmetrical ones give meso-diols!) is the l-(2,3-dihydroindolyl)carbonyl(IND-CO) group [49] (Fig. 4-3 (b)). Lately a very efficient system, called AD-mix, has been found by Sharpless et al. using the phthalazines 9 and 10 as catalysts, and K3 [Fe (CN),] as reoxidant. With trisubstituted olefins methane sulfonamide was added as rate enhancer [50] (Fig. 4-3 (c)). Enantioselectivities with the j?-AD-mix are by a few percent lower. With the same system enolethers can be oxidized to a-hydroxyketones ( R with /3and S with a-AD-mix) in 2 84% ee [51], (E)-enynes and 1,3-dienes to the corresponding diols in 73 -97% ee and 93 - 99% ee resp. [52, 531, as well as unsaturated esters to lactones [54] e.g.:

R = Pr, h, Pent, C,,E&, ee> 95%

0

R = l3.h u . Ph ec > 96

Several non-alkaloid chiral catalysts, mostly diamines, have been developed for dihydroxylations with OsO, (11 [55], 12 [56], 13 [59], 14 [57], 15 [58]) but their practical value is less than that of the latest Sharpless procedure, first of all because they require a full equivalent of the expensive OsO, .

R 2

(14) R = C & t h

(IS) R = Pent

Two different views collide concerning the mechanism of the reaction and since amines of rather different structure are involved as catalysts, perhaps there is no single answer to the problem. X-ray structures for a DHQN-OsO, complex (without

130

4.2 Diastereoselective Oxidations

olefin) [60] and for a product complex with diamine 14 [57] are available, but, of course, nothing in between. According to one hypothesis the reaction proceeds as a concerted [3 + 21 cycloaddition involving a 5-membered transition state [e.g. 611, while the other, advocated mainly by Sharpless [42] and Tomioka [62] is a two step mechanism. First, in a fast reversible [2 + 21 cycloaddition a four-membered intermediate with a C - 0 s bond is formed which rearranges in the rate limiting step to an osmate ester with a 5-membered ring. Molecular mechanics [63] and FMO calculations [64], both for the more tractable diamine complexes, support the concerted mechanism. Kinetic studies by Sharpless et af. on the catalytic cycle with NMO [44] as reoxidant indicated the existence of a second reaction cycle of low selectivity, becoming prominent when addition of the olefin was fast. This second cycle is absent when for reoxidation K, [Fe (CN)6] is used in a two-phase system [46].

4.2 Diastereoselective Oxidations 4.2.1 Diastereoselective Epoxidation Fundamental studies in this field were initiated in the late fifties by Henbest and his coworkers, who investigated both simple cycloolefins and cyclic allylic alcohols [65, 661. They established that with simple cyclic olefins or allylic ethers epoxidation by peracids was influenced primarily by steric hindrance and modified by strong solvent effects. With the free allylic alcohols, in turn, results could be interpreted by assuming stereodirection by hydrogen bonding between the peracid and the hydroxy group. Later studies were practically restricted to the epoxidation of allylic alcohols. Two techniques of diastereoselective epoxidation will be discussed here, viz. that using peracids and that using hydroperoxides in the presence of transition metal catalysts. With more advanced techniques of analysis at hand, Whitham et al. [67] and Itoh et a[. [68] established that cyclic allylic alcohols gave predominantly the cis-epoxides with the VO (acac), - HzOz (or - tBu0,H) system, while with peracids cyclohex2-en01 and cyclohept-2-en01 gave the cis-,the 8- and 9-membered analogues mainly the trans-epoxides (Fig. 4-4(a) [67]). When the hydroxy group is blocked by acetylation stereodirection by means of hydrogen bonding with the peracid, or ester formation with the metal does not operate anymore and the trans-epoxide is formed in slight excess. If a bulky group interferes with cis epoxidation, as in cis-5-tert-butylcyclohex-2-enol, using the VO (acac),-

13 1

4.2.1 Diastereoselective Epoxidation

MCPBA (A) or VO(acac)2+ t B u 0 2 H

(B)

(CHZ)~

(CHZ)~

(CH$n

cis

R2

R3

H Me H

H

H

Me

Me H Me H Me

Me

syn:anti

63:37 38:62 >98:2

1

2

A

68

90

22

B

98

>99

299

% cis

R'

trans 0

n=

JYn

3 <1

94

4


anti

5:95

1o:m

4455 90: 10

Fig. 4-4. Diastereoselective epoxidation of allylic alcohols.

tBu02H system, instead of a trans-epoxide 5-tert-butylcyclohex-2-enonewas formed [42]. Potassium peroxomonosulfate in the presence of a crown ether gave, from cycloalken-2-01s in a two-phase system, mainly trans-epoxides [69]. Epoxidation of 2- and 3-cyclohexenols with CF3C03Hin CHzC12 at - 40°C is highly syn selective [70]. Pierre, Chautemps and others studied the epoxidation of acyclic allylic alcohols with 4-nitro-perbenzoic acid and found that stereoselectivity depended on the substitution pattern of the double bond (Fig. 4-4(b)) [71,72]. Results were interpreted by considering the relative stability of those ground state conformations in which one of the bonds attached to the a-carbon was eclipsed with the double bond. Epoxidation of (Z)-4-acylamino- [73] and (Z)-4-trimethylsilylallylic alcohols [74] with MCPBA and t B u 0 2 H + V 0 (acac), resp. was also highly syn selective. Sharpless, Nozaki and others reported on the VO (acac),- and Mo (CO)6-catalyzed epoxidation of the same compounds by tBu02H [75,76]. Catalysis by Mo6+ proved to be less selective than that by V5+.With certain substrates stereochemical preference is opposite and selectivity higher than with peracids, whereas with others metal catalysis also gives the syn-epoxide in excess but in lower selectivity (Fig. 4-4(b)). Similar results were recorded by Mihelich [77]. Takai et al. described the epoxidation of allylic alcohols in the ~BLIO,H-(~BUO)~A~ system, but, except for

132

4.2 Diastereoselective Oxidations

special cases, selectivity was lower than in V5+ catalysis [78]. An interpretation of the stereochemistry of allylic epoxidation by transition state models based merely on product analysis was proposed by Narula [78, SO]. The Sharpless method was also applied to substrates with diastereotopic faces. As expected, the reaction is highly sensitive to the constitution of the substrate. While high syn selectivity was observed by Kishi et af. with P-substituted allylic alcohols [81] anti-epoxides were obtained from a-methyl allylic alcohols [82]. Acetals of a-methyl-P,yunsaturated aldehydes were oxidized in the tBu02H/ TiC13(OiPr) system at - 78 "C to the trans-epoxides with total diastereoselectivity 1831. Sharpless epoxidation is perhaps the only non-enzymatic method which can be used for practical kinetic resolutions [84, 851 (see also Fig. 1-13 (e)). Exceptionally effective kinetic resolution (k,/k, = 16- 104) of racemic allylic alcohols of type 16 were realized by Sharpless and his coworkers, permitting the convenient preparation of allylic alcohols in high optical purity (>96% ee at 50% conversion). As in other double-induction experiments, diastereoselectivity is highly dependent on the configuration of the substrate (or vice uersa of the reagent) (Fig. 4-5) [86]. In the absence of diisopropyltartrate diastereoselectivity dropped sharply with P-unsubstituted or (E)-allylic alcohols but was high (> 95% syn) with (Z)-allylic alcohols.

t B u a H , Ti(0iPr)r (+)-diisopropyl t a r t r a t e , SO'%,

OH

(+ 1- (16)

M

*

converSlon

~

c

t

0

bH

l

e

x+ M e , q . f H e x

'

(IS) de 9804

OH

>20 : I

( I R ) de 24%

98% ee + MeTc OH >

(R )

Fig. 4-5. Kinetic resolution by allylic epoxidation.

The method has been extended to 1-alkyl-3-silylated allylalcohols [87] and furfury1 alcohols too [88]. Epoxidation of homoallylic alcohols acquired great importance in the synthesis of macrocyclic antibiotics, first of all in the hands of Kishi and coworkers. As a prelude, Sharpless and Michealson found almost complete cis selectivity in Mo6+-catalyzed epoxidation [89]. Later Kishi applied both the tBu0,H-VO (acac), system and rn-chloroperbenzoic acid (MCPBA) to various chiral homoallylic alcohols and obtained the anti-epoxides (or their transformation products) in excess [90,91]. Some examples are shown in Fig. 4-6.

13 3

4.2.2 Various Diastereoselective Oxidations

anti % anti

CHzOH CHZOBn

z 96

83

Fig. 4-6. Diastereoselective epoxidation of homoallylic alcohols.

4.2.2 Various Diastereoselective Oxidations Dialkyl sulfides having a chiral center in the a-position are preferentially transformed to syn-sulfoxides by peracids. As an explanation the synclinal disposition of the large substituent was assumed, for which some NMR and X-ray evidence could be furnished (Fig. 4-7(a)) [92]. Periodate oxidation of 1,3-dithianes proceeds with strong equatorial preference [93]. Effective 1,2-induction was experienced in the oxidation of 2-alkylthianes to sulfylimines. Product configuration could be controlled by reaction conditions [94] (Fig. 4-7 (b)).

TosNCINa MeOQ 20 OC

c

%syn

76

78

98

66

71

\I

I) tEhOC1, MeOQ - 78 o C * p A L R 2) T o s ~EbN, , -78 OC

R = Me, Et, IPr, tBu

Fig. 4-7. Diastereoselective oxidation of sulfides.

75

N-Tos

R

de 80 - 90 %

I

de 80 - 88 %

134

4.2 Diastereoselective Oxidations

Kishi and his coworkers explored the osmium tetroxide oxidation of chiral allylic alcohols. Besides confirming the well-known syn stereochemistry of hydroxylation, they established anti preference, as related to the inducing center and the adjacent hydroxyl. 2,3-Anti selectivity was higher with polar directing groups and (2)-allylic alcohols (Fig. 4-8 (a)) [95]

OH tWIMe,SiO

OSMe,tBu

&O,C+ q Eto,cA-

co2Et 0SlMc2t&

oso,

c

t&Me,SiO

0

*

&

OH

de > 99 % Fig. 4-8. Stereochemistry of the osmium tetroxide oxidation of allylic alcohols.

The anti rule is quite general and applies to both free [96, 971 and protected [97 - 991 allylic alcohols, e.g. [IOO] (Fig. 4-8(b)). A variety of chiral auxiliaries, such as alcohols [IOI, 1021, amines [ l o 3- 1051 and hydrazines [ 1061 can be used to enable diastereoselective a-hydroxylation of enolates, e.g. [103]: 0

0

Yde 98 90 90

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

5 Stereoselective Carbon-Carbon Bond Forming Reactions by Nucleophilic Addition to Carbonyl Groups While diastereoselective C - C bond formation in cyclizations and in the transformation of cyclic systems is a traditional area of stereoselective synthesis, some twenty years ago no efficient methods for the enantioselective formation of such bonds or procedures for diastereoselective C - C bond formation in acyclic systems were available. Stimulated by the rapid advancement of stereoselective synthesis in general, further by an endeavor to elaborate methods for the synthesis of macrocyclic antibiotics and similar natural products containing a multitude of chiral centers in a noncyclic environment, a whole arsenal of highly selective enantio- and diastereoselective methods has been developed. The abundance of information available poses serious problems of systematization. Our own approach to classification was based on types of reactions rather than on target molecules. Stereoselective pericyclic reactions are dealt with in Chapter 7, while the rest of the material is distributed between Chapters 5 and 6. Chapter 5 is essentially devoted to nucleophilic additions to C = 0 and C = N groups, whereas Chapter 6 is a collection of diverse methods ranging from Michael addition to Grignard cross-coupling and other catalytic enantioselective reactions.

136

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds 5.1.1 Enantioselective Additions One of the fundamental methods to establish a C-C linkage is the addition of simple carbon nucleophiles, such as Grignard reagents, lithium alkyls, etc. or a cyanide ion to a carbonyl group. While diastereoselective addition involving first of all chiral carbonyl compounds but often also chiral nucleophiles has been the subject of extensive studies and led to the discovery of the rules of Cram and Prelog (cJ:Sections 3.2.1 and 3.2.2), enantioselective addition to prochiral carbonyl compounds remained, for long, a rather neglected field. Enantioselective additions can be realized in two ways: (i) an achiral nucleophile is used in the presence of a chiral additive, such as a chiral solvent, complexant or catalyst, or (ii) by transfer of an achiral alkyl group from a chiral reagent. Additions in the presence of a chiral complexant or catalyst

It is well known that Grignard reagents readily form complexes with ethers, while alkyl lithiums coordinate to tertiary amines. Therefore, in the presence of chiral complexants such alkyl metals may add enantioselectively to prochiral carbonyl compounds. Such an effect, albeit very small, was first demonstrated with (+)-2,3-dimethoxybutane by Cohen and Wright in 1953 [I]. While endeavours to carry out the addition of Grignard reagents with acceptable enantioselectivity remained fruitless for long, addition of lithium alkyls in the presence of chiral ditertiary diamines could be developed, at least for certain models, into a reasonably enantioselective procedure. Some of the ligands used are shown in Fig. 5-1. Compounds 1-3 were selected by Seebach et al. from 21 tartaric acid derivatives [2- 51 but enantioselectivies were < 53%. Mukaiyama and his coworkers first discovered that 4 and chiral diamines in general were rather efficient and versatile aids [6] and thereafter prepared 8 less successful analogues [7]. High ee values (up to 95% with 4 ) were realized in the practically irrelevant addition of BuLi to benzaldehyde at - 123 "C [8]. Addition of a carbanion prepared from ally1 p-tolylsulfone to acetone in the presence of 4 proceeded with 80% ee ( S ) [9], and the scope of the reaction was extended to lithium acetylides; the products were converted into optically active butenolides (Fig. 5-1) [lo, 1I]. An early example for remarkable enantioselectivity in additions of Grignard reagents is the reaction of cHexCOPh with MeMgBr in the presence of diacetoneglucose [70% ee (R)] [12].

1 37

5.1.1 Enantioselective Additions

Y-p

( I ) R1= M e , , R2 = Me (2) R' = W e , , R2 = (CH,),NMe, (3) R' = OMe, R2 = Me

Rl&R'

OR2

OH

(3) R = Me

RCHO, ( 4 ) Me,O, -123 O C

iMe3Si -C=C--Li

OH O c t y C 0 2 H

R = O L I , I ) OH" 2) B u L i , 3 ) CO1

I) H r 2 ) He

3-

OH Fig. 5-1. Chiral additives used in the enantioselective addition of alkylmetals to carbonyl compounds.

A tartaric acid-derived diol (TADDOL) (5) with many other applications also provided high selectivity in Grignard additions [I31, although the process was wasteful in respect of the reagent. Ph

3R3Mgk

+

(5)

R'COR2

- loooc

____)

&

RR?

OHR2

(5)

for R1 = Ph, R2 = Me, R3 = Et, Pr, Bu ally\ ee > 98%

Meg$ii ""fMe

Ph

Ph

The past decade witnessed a burst of activity directed towards enantioselective addition of zinc alkyls onto aldehydes and well over 50 papers were published on this topic. It became almost a matter of prestige to invent a new chiral additive but most of the authors were content to try those merely on the addition of Et2Zn to benzaldehyde. Here only methods of wider scope and/or providing outstanding enantioselectivity will be quoted. Seebach et al. prepared 5 and analogues of C, and C, symmetry and tested them in the addition of Et2Zn to benzaldehyde. Also an X-ray analysis on some of their spirotitanate esters was carried out. ee Values invariably exceeded 90070, but ketals with non-symmetrical ketones performed better (ee196070) [14, 151. Results with the most readily accessible diol ( 5 ) were as follows [16, 171:

AR2 OH

R'CHO + R2,Zn

(5) + Ti(0iPr)q PhMe, -78 - -30 O c ) R'

R1 R2 %e

Me IPrc€kx Ph Ph Ph Ph Ph Hex Me E?u Me Et Pr Bu Oct 70 90 90 94 98 94 98 98

138

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

Dialkylzinc compounds unbranched in a-position can be generated in situ from Grignard reagents and ZnClz [ 171. Vicinal tert-aminoalcohols, among them prominently ephedrine derivatives are the most popular chiral catalysts for dialkylzinc additions. Out of 12 N,N-dialkylnorephedrines Soai et al. selected the N,N-dibutyl compound (6, DBNE) as the most effective additive [18 -201.

R1 R2

&I Hex Hex Fkx cHex Ph Ph Oct

Et

%ee 87

Me Et Pr 90 88 90

Et

93

‘ E t IPr Et 90 75 87

Efficient aminoalcohol catalysts were prepared from L-proline (e.g. 7 [21,22]), camphor (e.g. 8 [23, 241) and ferrocenes (e.g. 9 [25]).

In the addition of Me,Zn to benzaldehyde catalyzed by 8 Noyori et al. discovered an intriguing amplification effect [24]. While with optically pure catalysts a product of 97% ee was obtained, with (-)-DAIB of only 14% ee optical purity of the product increased with catalyst concentration and levelled off at 8 mol%. Me2Zn and DAIB form 2 :2 dimeric catalyst complexes of which the racemic one is the more stable, but it is catalytically less active than the homochiral dimers. Chiral diamines, such as trans- 1,2-diarninocyclohexane sulfonamide [26] and 2,5dialkylpiperazines [27], as well as chiral 2,2’-bipyridyls and 2-pyrrolidinyl-pyridines 128, 291 offered no extra advantages as catalysts. An enantioselective Friedel-Crafts type reaction was discovered by the Casnati group [30]. OH

OH

OH

5.1.2 Diastereoselective Additions

139

C h i d reagents donating alkyl groups

By reacting a metal tetraalkyl aluminate or alkyl titanate with a chiral alcohol, a chiral reagent capable of enantioselective alkyl transfer to carbonyl groups is formed but selectivities achieved so far are not of practical value [31 -331. Enantioselective addition of hydrogen cyanide to benzaldehyde in the presence of quinine, discovered in 1912 by Breding and Fiske [34], was probably one of the first enantioselective carbon-carbon-bond forming reactions ever reported. Despite extensive investigations, among others by Prelog and Wilhelm [35], enantioselectivities did not reach practical levels for long. Addition of cyanide, mostly in the form of Me3SiCN and catalyzed by chiral esters of Ti(1V) or Sn(I1) is a relatively recent development. The ligands, e.g. diisopropyltartrate [36, 371 or cinchonine [38] are well known from other enantioselective procedures. As an illustration the use of a TADDOL analogue is shown. On mixing the diol with TiCl,(OiPr), a mixed titanate (10) is formed to which the components are added one after the other: Ph

TiCl,(OiPr),

RCHOR

x,

-

lOct ckkx Ph Eh WICH, %eel 76 68 98 77 91

R

A reagent prepared by mixing N-tosyl-norephedrine with TiC14 and isopropanol transferred a methyl group onto aldehydes with up to 76% enantioselectivity [39].

5.1.2 Diastereoselective Additions In this section addition of simple nucleophiles, mainly of organometallic compounds and carbanions, to carbonyl groups with diastereoselective faces, i t . to chiral carbonyl compounds and substituted cyclic ketones, will be discussed. First, asymmetric induction by an adjacent chiral center in acyclic substrates (1,2-induction) will be described, followed by induction by more remote centers in acyclic systems (first of all 1,3-induction) and additions to cyclic ketones. Finally, the addition of chiral nucleophiles to prochiral carbonyl compounds will be covered.

140

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

Table 5-1. Diastereoselective Addition of Simple Organometallic Reagents to Acyclic Carbonyl Compounds with a Chiral Center Adjacent to the Carbonyl Group.

( A1

(B

R'

R2

R3

R4

R~M

Ph

Me

H

H

MeMgBr EtMgBr PhMgBr MeTi(OiPr), Et4Pb+ TiCI, MeMgI EtMgI iPrMgBr PhMgBr EtLi MeLi MeMgBr iBuMgBr MeMgI EtMgBr BuMgBr MeLi MeLi PhLi PhMgCl PhLi PhMgBr MeLi PhMgBr PhMgBr MeMgI MeLi MeMgI PhLi BuMgBr BuMgBr MeMgCl MeTiC1, iPrMgBr -2 HMPTe

Et iPr Et

Me Me Et

Et

Me

H

H Me

cHex CI Ph

Bu Et OH OMe OH

H Me

Ph Me

OMe cHex Ph

OH

Me H

cHex To1 Me Ph

Me

PhC Hept Me

OBn OBn OMEM NBn,

Me Ph OBng

OBn OSitBuMe,

BnOCH,OCH,

Me

H

Me H Me Bu H

P h C Z CZnBr Et,Pb + TiC1, Me,Zn + TICI, MeTi(OiPr), allylSiMe, Me,CuLi MeLi Bu,CuLi CH, = CHCuBu,P

% Aa

70 75 80 88 93 71 75 69 66 71 89 60 64 65 70 80 89 66 91 76 90 95 89 98 96 96 89 66 91 0 100 0 6 0

95 2 96 I 81 97 41 94 89

Ref.

5.1.2 DiastereoselectiveAdditions

141

Table 5-1 (continued). R'

R2

BnOCH,O Ph

Me Me

w11 I

I

R3

H

, o w

R4

R ~ M

(70 A "

Ref.

SiMe,

MeMgBr BuLi MeLi

91 >99h 87

[65] [66] [67]

MeLi

>98

[67]

88

[68]

yw3

s2 s9 (5W4

SiMe,

Me

H

MeMgBrj

Me% O

A

Me NH2 NBn,

H

Ph 4-MeOC,H4 H

MeTi(OiPr), TolMgBr PhMgBr MeTi(OiPr),

<1

'

EtMgBr Ndo

NEt, NiPr, NMe,

Ph

NMe, NMePh

Me

Me Me Bn,N

CONMe, CONMe," Me ~~~

~

PhMgBr Me

EtMgBr

Ph Phm H

PhMgBr PhMgBr + MgBr, MeMgBr MeMnCl MeAlCI, Me,SiCN + ZnBrzo

[69]

99 50 3

WI

100

[53]

66

[53]

48 0 100 55 100

[53] [531 [53] 1531 [54J [551 [56] 1561 [56] 1571

0

93 10 88
1701

[70]

[SS] [59]

-~ ~

Normalized to 070 A + 070 B = 100. If R4 = H, A is a syn and B an anti product according to the usual notation! For R5 = Bu and Ph analogues and with allylTi(NEt,), 83, 75 and 93% syn resp. [42]. Yo syn >96% also for R' = Me and iPr, and R2 = Me and Ph. In the presence of ZnBR2. 87 and 90% syn with EtMgBr and BuMgBr resp. 99 and 86% with R' = iPr and BnOCH, resp. g ~ 9 7 % syn for RSM = allylSiMe, and methallylSiMe,. %A>94% for R' = cHex and cyclohexenyl, and R'M = MeLi and allylSiMe,. R = Me, Bu, Ph, allyl. 96 and 85% syn with EtMgBr and allylMgBr resp. In the lactol form. %A<6% also for MeMgBr, PhMgBr, iPrMgBr, tBuMgBr and allyITi(NEt,), when R' = Me and for MeTi(OiPr), when R' = iPr, iBu and Bn. %Syn>99% also for R4 = Me, Et, tBu. " For R' = Et, iPr and R S = Me, Et, '%syn>93% [SS]. O With TiCl, as Lewis acid 72% anti. a

J

'

142

5.1 Addition of Simple Nucleophiles to Curbonyl Compounds

1,2-Asymmetric induction

Along with reductions by hydride donating reagents, a study of the stereochemistry of the addition of carbon nucleophiles to carbonyl groups with an adjacent chiral center played a central role in the development of theories concerning asymmetric induction. * The essence of these theories has already been discussed in Section 3.2 and need not be restated here. The most important results obtained under optimized conditions have been compiled in Table 5-1. To this the following comments may be added: (i) For compounds with carbonyl as the only polar group, Cram’s rule applies. Stereoselectivities are, by present standards, moderate. Asymmetric induction in 2-chloroaldehydes can be interpreted by the antiperiplanar effect (cf: Section 3.2.1). (ii) A practically useful degree of stereoselectivity can be realized under chelation control (cf: Section 3.2.1). An oxygen atom (in alcohol or ether function) in a-, possibly also in P-position, or an amino group in a-position can serve as an anchor for the metal atom of the reagent. Since the solvent may compete with the reagent for chelation and the coordinating ability of various organometallic compounds is also different, such reactions have to be carefully optimized for solvent, reagent, temperature, etc. [50, 511. Thus, addition of butyllithium to an a-alkoxyketone gave 41 -75% of the syn-diol (type A ) (lowest de in THF), while that of butylmagnesium bromide afforded 90- 100% (highest de in THF!) [61]. As was shown, among others by Still et al., the nature of the protecting group was also important, and the (2-methoxyethoxy)methyl (MEM) group was particularly useful owing to its “crown-ether” effect. The authors realized very high stereoselectivities with models having a chiral center in a- and an ether function in a- or in P-position [61, 651. Comparison of models 11-13 revealed that the more remote chiral center played a secondary role in stereodirection (Fig. 5-2 (a)). Interestingly, no induction was exerted by the p-center alone, but its effect on induction by the a-center was significant 1611. Addition of Grignard reagents to a-aminoketones is much influenced by the substitution of the amino group 1721. Primary amines and dialkylamino compounds in general give an excess of the diastereometer expected under chelation control. An exception was 2-diisopropylaminopropanal, which gave exclusively product B. The behavior of phenylmethylaminoketones was rather unpredictable, and stereoselectivity could be inverted by the addition of MgBr,. In the addition to a-N,N-dibenzylaminophenyl-acetaldehydeMeTi(OiPr)3 and Grignard reagents give the anti, while MeTiC13 the syn-aminoalcohol in excess (see Table 5-1) [62]. The coordinating ability and bulkiness of Ti(1V) compounds can be modified by varying the ligands. This was exploited by Reetz and his coworkers who studied the addition of various metal alkyls to benzyl (Fig. 5-2(b)) [73]. Clearly no chelation

* For a detailed discussion of the literature up to ref. [71].

2968

CJ

ref. [40], pp. 84- 132. For a review see

5.1.2 Diastereaselerfive Additions

3 43

Me+

BnO-0

OH

BnO-0

OH

95% syn,anti

Me

M d C H O

M

e

F

H

MeZCuLl

O

70% a n t i , a n t i

MenM

MerCu Li

50% syn

BnO..,-,D

BnO-0

OH

(13)

ph)(.l*

HO

(b)

Ph

MeM

Li

MgBr

Zr(OiPr),

Me

+

ph")h

HO

0 M

Me

Ti(OiPr)3

r

he Ti(OEt),

H

ph$Ph

HO TiMe,

Bn

OH

Me

TiMe(OiPr)*

1

Fig. 5-2. Asymmetric 1,2-induction in the addition of metal alkyls to aldehydes.

control was operative in the addition of methyl-alkoxytitanium reagents. Note, however, that with phenylacetaldehyde as substrate MeTi(OiPr), only gave 88% of the syn product [74]. It is also remarkable that the zirconium analogue led to the arztidiol in moderate excess [731. Amouroux el al. found that anti-Cram addition becomes predominant when Grignard reagents complexed with 2 moles of HMPT are added to aminoaldehydes [63] (Table 5-1). On the other hand, with MeTiCl,, in which coordination is promoted by the electrophilic chlorine atoms, the chelation-controlled product was formed in high excess

[W.

144

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

Finally, examples for 1,2-asymmetric induction by removable chiral auxiliary groups should be discussed (Fig. 5-3 (a) and (b)). The ketone 14 was prepared from relatively inexpensive mannitol; Grignard addition proceeded under chelation control [76]. Eliel and his coworkers utilized the chiral oxathianes 15 [77], 16 [78] and 17-19 [79, 801 as auxiliaries (cJ:Section 6.2.2). 16 was prepared from (+)-camphor, 17 and 18 from (+)-pulegone. It was supposed that Mg, as a hard acid, would complex preferentially with a hard base (oxygen) rather than with a soft base like sulfur. The starting ketones were prepared by addition of the lithiated oxathianes to the aldehyde followed by oxidation of the mixture of epimeric alcohols.

Ph

A* 0

Ph

MeMgl

THF. -78'c

Me

*

OMe

97% ee

Me

Me

(17) x = s (18) x = N B n

Fig. 5-3. Addition of alkylmetals to ketones attached to a removable chiral auxiliary group.

The a-hydroxy acids obtained (cJ:Table 5-2) may be converted to glycols [81] [79, 821 (cJ:Table 5-2). The high selectivity obtained with the combination R' = Me, R2 = Et suggests that it is not the difference in the steric requirements of groups R' and R2 which is essential for stereoselectivity.

145

5.1.2 Diastereoselective Additions

Table 5-2. Addition of Metalorganic Reagents to Chiral Dithianes.

R'

R~M

R*

Ph

MeMgI

1s 17 15 15 17 19

EtMgI EtMgBr iPrMgI Me

a

EtMgI iPrMgI

% eea

98 96 96

98 100 98

IS

86

19 15

98 94 34

7 tBu

Bu2Mg allylMgBr HCZ-MgBr PhMgBr allylMeMgI MeMgI PhMgBr MeMgI

R*

Vo eea

19

14 82 91 80

18

1s 15

90 52 98 92

Prevailing configuration as shown above.

A rather exotic way of asymmetric induction by shielding of one of the faces of an aromatic compound by complexation is shown next: 1) Cr(CO), 2) resolution

1) RMgI, 2) THF, o,, hv-78 O C

q

R

CHO

OH R = Me, Fit ee > 99 %

There is hardly a field of enantioselective synthesis where proline derivatives do not play an important role. Indeed, Mukaiyama and his coworkers found that addition of Grignard reagents to aminals prepared from 20 [derived from (S)-proline] and 2-ketoaldehydes was highly diastereoselective. After hydrolysis, the chiral auxiliary could be recovered (Fig. 5-4(a)). In the case of R' = Ph the aminal could be prepared by direct condensation [84], otherwise it was obtained from the ester 21 by Grignard addition [85]. The aminals obtained by the two methods consist of a single diastereomer. Both enantiomers of the end product can be prepared by reversing the order of the reagents [86]. When the ester 21 was reduced to an aldehyde chiral 1,Zdiols could be prepared with equally high enantioselectivity (Fig. 5-4(b)) [87]. Analogous 2-acetyloxazolidines prepared from L-prolinol can be transformed to (S)-2-hydroxyaldehydes when reacted with RTi(OiPr)3 (R = Me, Bu, cHex, ee>92%) or to the enantiomers with lithium alkyles (de = 60%) [SS].

146

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

Alkylation of chiral2-phenyloxazolidineswith Grignard reagents give after oxidative removal of the inducing group primary amines [89]:

Efficient 1,3-induction leading from P-ketols to anti-1,3-diols can be realized (except for R' = cHex) under chelation control with alkyltitanates. With the silylated ketols giving the syn-diols in excess selectivity is poor with MeTi(OiPr), and varying with lithium alkyls [90]:

HO

R3

OH

R3Ti(OiPr),

R'

R3 OH OH

R'

R2 = Ph, f i x : R3 = Me, WI, de 6492%

R1 = Pr, Ph, BnctE,

R2 = Ph, R3 = Me, Eb de 98%

Cyanide addition can be directed very selectively both by a P-hydroxy- [91] or a /3-sulfoxy group [92] e.g.:

mR2 uR2 KCN, &I2

R'

Me3SiCN

OH

R'

R1 =Et, iPr, Ph, Bn R2 = Me, iPr, t h , de >95%

1,3-Induction Until recently it appeared that in acyclic compounds a chiral center in P-position relative to a carbonyl group could not effectively control the stereochemistry of addition to the latter. The addition of Grignard reagents and alkyl lithium compounds gave poor or medium (up to 76% de) selectivities. The trend of diastereoselection for simple ketones can be predicted by a modified acyclic Cram model, while that for hydroxyketones is in accordance with the chelate model. In 1983, however, Reetz and Jung found very good selectivities under chelate control employing MeTiCl, or a combination of BuzZn and T i c & [93]. The following addition is unique in a way that the metal atom is attached to a chiral center. Under BF3*Et20catalysis the anti-alcohol is formed in 22-78070 de {157].

Ipbh3 qR2 141

5.1.2 Diastereoselective Additions

Asnh3 OMc

R'

1) l3UL 2) h,PbBr, -78

R2CH0

OC'

R1

R'

OH

R' = B p t , c k x R2 = Ph, c b x , k p t . Pr de 94 - 99 YO

I) R'MgX

>

MgCI?

Ph

n

x,,

R2

OH

R'

R'

R2

%ee

R'

R2

%ee

Me

Ph Et

99 78

iPr Ph

Ph Me

>Y4

9s

R

Et

iPr

Bu

Pent

%ee

96

91

94

83

Asymmetric induction over three or more bonds

The first example of asymmetric 1,4-induction, the addition of a Grignard reagent to a chiral-2-ketoester, was described by McKenzie in 1904 [94]. These and later results by both McKenzie and other authors, as well as his own data on this reaction, were systematized in 1953 by Prelog and formulated in a rule, which has become known as Prelog's rule [95] * ( c j Section 3.2.1). For a long time, due to the low level of stereoselectivity, the method was used for the determination of the absolute configuration of the chiral alcohol rather than for synthetic purposes. Excellent selectivities were observed recently by Whitesell et al. using the more readily available chiral

*

For developments in this field up to 1968 see ref. [40], pp. 50-83.

148

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

auxiliary (-)-8-phenylmenthol [96] (Fig. 5-5 (a)). Additions to the corresponding pyruvate and phenylglyoxylate also proceeded with 90% de [97]. Meyers and Slade investigated I74-induction in a series of chiral 2-oxazolidinyl ketones, which were readily available by oxidizing the anions of the corresponding alkyl compounds 1981. Selectivities with oxazolines prepared from ( R)-phenylalanine, ( R )-phenylglycine and (S)-mandelic acid were inferior to that for the example shown in Fig. 5-5 (b). Amides of 0-methylprolinol show strong 1,binduction (de L 95 Yo) when reacted with metalorganic compounds and the prevailing configuration can be controlled by the choice of the metal [99]: 0

OH Me R'

'

R2MgX

R' R2 = Me, Ph

OMe

One of the diastereotopic faces of the carbonyl group in the ketals below are very effectively shielded by one of the methoxymethyl groups in the following Grignard additions [1001: MeOb.o-

OMe

, H@,

R2MgX THF, -78 O C R'x(Me

0

Rl+e

-

R2 R' = Et, Ph R2 = Et, vinyl, Ph, Me,$

HO

ee 194%

An example for 1,6-induction with reagent control of configuration is the following reaction [IOI]:

99% ee

100% ee

Complexation by titanium proved to be useful in 1,4-induction, too, as shown in Fig. 5-5 (c) [75]. Addition of Grignard reagents to chiral acyclic ketoesters with the inducing center more than three bonds away proceeds with very poor stereoselectivity.

5.1.2 Diastereoselective Additions

149

R=Me, Hex, cHex, Ph ee 3 99%

M e T P h Me

I ) LDA, 2) 0 2 -7XOC

BuCO--(

'.IOMe

OMe Me\

MeMgBr.TMEDA Eli0

*

rph

>*uB

OH

N

Ph

He

Mx

Bu

COzH

15% ee

Me2Z n -950c

iMe

H

Me L

M

H

e

85% urrti

OH Fig. 5-5. Asymmetric induction over three or more bonds in the addition of alkylmetals to carbonyl compounds.

Finally, an example for high selectivity without inducing groups should be quoted when two molecules, both with enantiotopic faces, are joined diastereoselectively [102].

150

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

Additions to cyclic ketones Over the decades following the pioneering work of Barton on the problem which is now called stereoselectivity in cyclic systems, an impressive amount of material has accumulated on the addition of organometallic compounds to cyclic ketones. * The theoretical background of organoalkyl addition is very much the same as that of hydride reduction ( c j Section 3.2.1), and here we only wish to point out some distinguishing features. (A) Whereas in the hydride reduction of unhindered cyclohexanones axial attack is generally preferred, in organometal addition equatorial attack is predominant both with hindered and unhindered ketones. The only exception is the addition of acetylide anion to unhindered ketones in ionizing solvents in which attack is predominantly axial. Axial attack was also highly preferred when LiC CH was added to a conformationally stable 2,3-cyclohexenone [ 1041. (B) Addition t o ketones with an axial substituent at C-3 is totally selective for equatorial attack, even with the acetylide anion, while with hydride reduction selectivity for the same approach is high but not total. Some typical results for organometal additions to cyclohexanones are summarized in Table 5-5. For aliphatic Grignard reagents and unhindered cyclohexanones preference for equatorial attack ( A )increased with the bulk of the alkyl group (Fig. 5-6(a)) [105], while PhMgBr only exhibited stereoselectivity with hindered ketones [ 1031. The factors governing selectivity are not always clear. Thus addition of MeLi to tert-butylcyclohexanone at -78 "C proceeded with 68% equatorial attack while a 3 :2 mixture of Me2CuLi and MeLi showed 88% equatorial preference according to one report [I061 and 94% according to another [107]. Under optimized conditions addition of the non-chelating titanium reagent MeTi(OiPr)3 proceeded with 94% equatorial attack, while ally1 Ti(NEt2)3 added preferentially from the axial side to 4-tert-butylcyclohexanone giving 80% of the cisalcohol [109]. Axial attack by metalorganic reagents can be enforced by prior complexation of the carbonyl group with a bulky Lewis acid [I lo]: R

RMI Meh EtMgEk hM&& AuylMgEk Yde( 98 100 100 82

The situation with the bicyclic ketones shown in Fig. 5-6(b) is straightforward: organometallic reagents add with 100% or near to 100% diastereoselectivity from the direction indicated by the arrows [103].

* Results on ketones with no other functional groups have been summarized [103], while cyclic aminoketones have been reviewed up to 1981 in ref. [72].

up to 2974 in ref.

151

5.1.2 Diastereoselective Additions

Generalizations are less easy to make in the cyclopentanone series. Steric approach control seems to be the rule for the addition of alkyllithiums and Grignard reagents, except acetylenic ones, when the behavior of 2-methylcyclopentanone is considered (Fig. 5-6(c)). One would expect the improvement of selectivity when the substrate is 2-tert-butylcyclopentanone, however, the percentage of the A-type product does not exceed 66%.

R

RM

I MeLi

%A

70

MeMgBr

EtMgBr

BuLi

PhLi

PhMgBr

60

75

86

95

99

Me

Et

iPr

tBu

Bii"

Ph

60

71

82

100

76

49

aMgC' =-Na 92

9

"For BnMgCl

Fig. 5-6. Diastereoselective addition of alkylmetals to cyclic ketones.

Torsional strain takes control over the approach of the nucleophile in the case of ethynyl reagents which add preferentially in the cis mode. These are linear molecules with but a single other atom bonded to the entering carbon atom. To 2-methoxycyclopentanone in liquid ammonia, however, HC = CLi (in fact HC C - ) adds predominantly in the trans mode due to repulsion of the negative charge and the oxygen atom. In T H F the reagent is not ionized and again cis attack predominates [ 1031.

152

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

Table 5-3. Addition of Organometallic Reagents to Substituted Cyclohexanones [104]. Vo Equatorial attack" / Stubstituent(s)

Reagent

4-tBu

4-Me

__

MeLi/Et,O PhLi HC-CLi/THF MeMgBr/Et,O EtMgBr/Et,O PhMgBr/Et,O HCECMgBr a

+ NH,

65 58 12b 53 71 49

Normalized to Vo eq+ Vo ax = 100.

3-Me

2-Me

3,3 ,5-Me3

84 88 45 84 95 91 45

100

~

53 53

44 18

54

68 59

For H C E C N a .

100 100

100 100

From ref. [I061

Addition of chiral nucleophiles to prochiral carbonyl compounds

It has been demonstrated before (p. 19) that the permanent association of a chiral reagent with a substrate with enantiotopic groups or faces results in the formation of a pair of diastereomers. The selective formation of one of the diastereomers may be an end in itself, but often the original chiral center can be removed later to give a pair of enantiomers with one of them in excess. This removal may result in the destruction of the original chiral center, but ideally it should be conserved and the chiral auxiliary recycled. For the exploration of selectivity experiments with a racemic chiral aid yield valuable information, since the distribution of the diastereomers, formed in the first step, puts a limit to the enantiomeric purity of the end product. Chiral sulfoxides should be well suited for the transfer of chirality because (i) they can be relatively easily prepared in an optically active form [109], (ii) hydrogens in the a-position are acidic and therefore a-metalated sulfoxides are readily formed, and (iii) the sulfur center can be smoothly removed by Raney nickel. Addition of lithiated methyl-p-tolylsulfoxide to simple ketones was, however, rather unselective (de 0-40%) [IIO]. The I-naphthyl analogue fared better (Fig. 5-7(a)), at least with aralkyl ketones [I 1I]. Recently diastereoselectivity was much improved by converting the lithio-compound to a zinc alkyl [I 121. Solladie et al. found that a-sulfinylesters metalated with tBuMgBr added with acceptable selectivity to both ketones and aldehydes (Fig.5-7(b)) [113, 1141. (S)-2-Anilinomethyl-pyrrolidine,a versatile chiral auxiliary, was used by Asami and Mukaiyama to prepare chiral phthalides in high optical purity (Fig. 5-7 (c)) [I 151. In the following examples chiral but racemic metalorganic compounds were added to aldehydes (Fig. 5-8 (a) and (b)). The first method is a typical kinetically controlled process, if the reaction mixture is not quenched immediately diastereoselectivity rapidly vanishes [116]. A method for the preparation of syn vicinal diols by McGarvey and Kimura was also based on organotin compounds but was less selective [117].

1 53

5.1.2 Diastereoselective Additions

//”

(a)

.-! S l - ~ a p h / ‘M~

1) EQNL, THF

HO

2) PhCOR, 4 0 OC, ) b e y - N i

*

Me

Ph xR

R

1Me Et

P r h IBU t h e x 50

Leo*.

%ec1100 100 44 100 52 50

It

1) tBuMgBr, THF, -78OC

T ~ \ - - -S- -COztBu -

2) R’COR’,

!.

R’----

>

3) AI/Hg

R’ R2

R2

IH I Ph

% e e X l

H

Me

Me

CF3

Hept

Ph

cHex

Ph

68

95

20

8

6

OH

‘Ph r ’.

\

R

1

%ee

I

88

W

87

20

OH Fig. 5-7.Addition of chiral nucleophiles to carbonyl compounds with enantiotopic faces.

Certain specially substituted allylstannanes, when activated with TiCl, or SnCl,, add to aldehydes without rearrangement (cf.Section 5.2.5) with 1,4- or even 1,5-induction [118, 1191, e.g.: OH

R = Pr, Lpr, CWX,Ph, crotyl, CO,Me, dc>96%

Hoffmann et al. found that lithium alkyls with a geminal heteroatom are sufficiently stable at very low temperature to conserve their configuration until quenched

154

5.I Addition of Simple Nucleophiles to Carbonyl Compounds

Me

Fig. 5-8. Diastereoselective addition of racemic metalorganic compounds to aldehydes with enantiotopic faces.

with a carbonyl compound [120]. This was exploited in the following reaction in which high diastereoselection in the exchange of one of the geminal bromine atoms was remarkable 1121, 1221:

5.1.3 Additions to C = N Bonds Asymmetric induction in the addition across a C = N bond can be quite effective and gives rise to primary or secondary amines. Attempts to utilize 2-phenylethylamine as chiral aid in the addition of cyanide to azomethines (Strecker synthesis) met with varying success. While Harada and Okawara [I231 reported up to 5370 ee with aliphatic aldehydes, Stout et al. [I241 reported 75-7870 ee for aliphatic, and 69-8570 ee for aromatic aldehydes. Weinges et al. described complete diastereoselectivity of cyanide addition to azomethines prepared from 3-aryl-2-propanones and 4-aryl-2-butanones as the carbonyl

155

5.1.3 Additions to C = N Bonds

component and 2-phenylethylamine [I251 or an amine 22 [126, 1271 as prepared from an intermediate of chloramphenicol synthesis. The key to high selectivity is that the reaction is reversible, and one of the diastereomeric products crystallizes out. The method, as applied to the synthesis of a-methyl-3,4-dihydroxyphenylalanin,an important antihypertensive agent, is shown in Fig. 5-9(a) [126]. Grignard reagents add to -HC = N - bonds to form secondary amines. Selectivity is usually not bad but removal of the chiral auxiliary is problematic. This can be overcome using a hydrazone prepared from ephedrine (Fig. 5-9(b)) [128, 1291. (a)

0 HCI

NaCN

+ R*-NH2

Ar&Me

x:E

HO

H2

Ar

R*-NHz=

Me

AcOH, McOH

Me

Raney N i

>

OH? [OJ

Me0

(221

%ee

a-Me-DOPA

&=

0

RM

___)

A r A C 0 2 H iOO% ee

Me)(05NH2 Me

NH2

+

I I

MeMgI

EtMgI

iPrMgBr

BnMgCl

PhLi

>96

>96

76

44

91

Fig.5-9. Stereoselective additions to C = N bonds.

a-Silyloxy aldehydes can be transformed with (Me3Si)*NLi to silylated azomethines which then react with metallorganics to provide, depending on the reagent, synor anti-aminoalcohols [ 130, 13I]:

Y

& Me'

H

I Eu Yde I >98

R

6 ally1 cPr c€kxC&

>98

90

>98

156

5.1 Addition of Simple Nucleophiles to Carbonyl Compounds

Complexation of the alkyl lithium inverts relative configuration in the following additions [132]: Ph

R1 k C H , BnCK Ph

7 Yde 98

94

>98

Addition of lithiated chiral sulfoxides to achiral azomethines is fairly selective (de>60%) provided that one of the substituents is aryl [133, 1341. Alternative syn or anti selectivity can be achieved by choice of metal in the following additions [ 1351: W

1Pr

r

NFhPr

de >98%

de 86 YO

With a-N,N-dibenzylamino azomethines diastereoselection can be inverted by reducing the donor character of the N atom. Removal of the N-substituents leads to 1,2-diamines [ 1361:

m*

R'

LR3 * R2 = Tos R3MgX

H

m o s

&O, -78 O C

m

R' = h4e, d'r, 1Bu, W R3 = Me, Et, Eh,Ph, vmyL ally\ de 70-90%

Addition of Grignard reagents to N-acylazomethines of glyoxylate esters of 8-phenylmenthol give protected (S)-a-aminoacids (de > 82%) [ 1371. Achiral esters of the same can be highly selectively alkylated with chiral enamines [138], e.g.

I) BzN=CHCO,Me 2) H@

de>96 % ee of anti > 96 YO

5.2.1 General Aspects of Allylrnetal Addition

157

5.2 Additions Involving Allylmetal and Allylboron Compounds * 5.2.1 General Aspects of Allylmetal Addition The majority of stereoselective methods developed in the last two decades for application to acyclic systems was based on scheme A.

One realization of scheme (A) is the aldol reaction (B) (Section 5.3) and another, to be discussed here, is the nucleophilic addition of an ally1 group to a carbonyl group with concomitant allylic rearrangement (C), which is the carbo analogue of scheme (B). A typical reaction of this kind is the following [139]: OR*

ann anh

* For reviews see refs.

[140-1421.

158

5.2 Additions Involving Allylmetcrl and Allylboron Compounds

The reaction proceeds with complete 3,4-anti diastereoselectivity as a consequence of its cyclic mechanism (see below) and with 92% 4,5-syn diastereoselectivity due to cooperative asymmetric induction, on the one hand by the chiral center of the aldehyde (1,2-induction) and on the other by the chiral boronate ester moiety of the allylic partner (1,6-induction). By ozonolysis, epoxidation and a score of other reactions the vinyl group can be transformed to a wide variety of useful derivatives. Before embarking on a detailed discussion of the allyl addition to carbonyl compounds, let us first discuss (i) the symmetry and stereochemistry of the reaction and (ii) the role of the leaving group. Regarding the symmetry and constitution of the partners several different situations can be envisaged in which stereoselectivity is possible. The following structural features of the reactants have to be considered: (i) The allyl partner can either be substituted at C-I or not. Substitution at C-2 does not contribute to the stereochemical complexity of the reaction. (ii) The allyl component can be achiral or chiral (either the leaving group or C-3 may be chiral). (iii) The faces of the carbonyl partner can be enantiotopic or diastereotopic.

96% de

OH H ~ O ~ I O H ~

Ph

-

=-250:1

Fig. 5-10. Stereochemistry of the addition of allylmetals to aldehydes (I).

5.2.1 General Aspects of Allylrnetal Addition

159

When a simple allyl compound is reacted with a carbonyl compound having enantiotopic faces, the reaction can only be enantioselective when the leaving group is chiral (e.g. Fig. 5-10(a)) [143]. If the carbonyl partner is a compound with diastereotopic faces, such as a chiral aldehyde, diastereomers are obtained as a result of asymmetric induction, regardless of whether the allyl compound is chiral or achiral (Fig. 5-10(b)) 11441. With allyl compounds substituted at both ends an additional chiral center is created at the point of attachment and even with achiral carbonyl partners a mixture of diastereomers is obtained. When none of the partners is chiral, the products are racemic and, provided that geometrical isomers of the allyl partner are stable under the conditions of the reaction, the relative configuration of the major diastereomer is determined by the configuration of the allyl partner (E or 2 ) (Fig. S-lO(c)) [145]. With chiral 1-substituted allyl compounds non-racemic diastereomers arise, and the enantiomeric purity of the individual diastereomers may differ. The chirality of the leaving group is transmitted less efficiently than that of C-3 ( c j Figs. 5-1 1 (a) and (b)) [146, 1471. Note that in example (b) chirality at the silylated center is completely transferred to the C-Me center, whereas diastereoselectivity is high but not complete.

98% s , y n 60-70% ee iPrCHO TiClr

-

Me +

OH anti,syn

72 1 2 8

Ph

Me .

OH anti,anti

Fig. 5-11. Stereochemistry of the addition of allylmetals to aldehydes (11).

160

5.2 Additions Involving Allylmetal and Allylboron Compounds

When a chiral carbonyl compound is attacked by a 1-substituted allyl reagent the relative configuration at the ends of the new bond (at C-3 and C-4) is primarily a consequence of the mechanism (mostly cyclic) of the addition, while the relative configuration at C-4 and the chiral center of the carbonyl partner is controlled by the asymmetric induction of the latter. The two effects interact, of course. In our example (Fig. 5-1 1 (c)) [I481 3,4-anti selectivity was complete, while 4,5-syn selectivity was modest. The most complex combination is the reaction of a chiral 1-substituted allyl compound and a chiral carbonyl compound. This has already been exemplified on p. 157 and obeys the rules of double induction, i.e. diastereoselectivities are different for the ul (R,S or S , R ) and Ik (R,R or S , S ) combinations of the partners. For an understanding of stereoselection in allyl additions the mechanism of the process has to be analyzed 11411. It has been established that under kinetic conditions the reaction proceeds with allylic rearrangement. When addition is uncatalyzed and the metal atom in the leaving group is able to coordinate with the non-bonding pair of the carbonyl group, a six-membered cyclic transition state is highly probable (Zimmerman-Traxler model [ 1491). In this case the major product can most often be deduced by evaluating steric interactions in such a transition state. Although a boat-type transition state cannot be dismissed as totally improbable, for our purposes it is sufficient to compare the chair-type transition states shown for a (2)-ally1 compound in Fig. 5-12(a). If R2 = H, transition state (C,) predominates in two forms. These are enantiomeric when the partners are both achiral while diastereomeric when either of them is chiral. In the latter case stereoselectivity can only be partially deduced from the present model. If R2 is other than hydrogen, gauche interaction between R' and R2 in (C,) should be weighed against 1,3-diaxial interaction between R' and L in (C2).Similar considerations apply to (E)-ally1 compounds. A consequence of this model is that high diastereoselectivity can only be expected with aldehydes but not with ketones for which a 1,3-diaxial interaction would be present in (CJ, too. * Also, as was recognized by Evans et a[. [ 1501, this interaction should be more severe when the metaloxygen distance in (C,) is shorter, and this distance can be inferred from ground state bond lengths. Thus a decrease of diastereoselectivity in the series boron, silicon, titanium, tin, and zirconium can be anticipated. These considerations may not be applicable to Lewis-acid-catalyzed additions in which the catalyst is competing with the metal in the leaving group for coordination with the carbonyl oxygen, a linear mechanism, as shown in Fig. 5-12(b), may be preferred in such cases, and then the syn diastereomer should be predominant regardless of the configuration of the allyl compound.

* Fortunately reactions do not read the literature and sometimes diastereoselectivity may be quite high with ketones too [ 15I].

5.2.1 General Aspects of Allylmetal Addition

161

Experiments by Denmark and Weber [152, 1531 on a model which could not take UP the arrangement required for the linear mechanism demonstrated that the latter was not exclusive for Lewis-acid-catalyzed additions (Fig. 5-1 2 (d)). Although a cyclic transition state is also improbable with ate complexes, reaction of such a compound (23) even with acetophenone gave anti products in high excess (Fig. 5.12(~))[154]. Allylmetal addition to carbonyl compounds has been studied recently by computational methods [155, 1561.

%!mfi

I

>99

TEI, : 82% syn CF3C02H: 99% ~ y n Jyn

Fig. 5-12. The mechanism of allylmetal addition to aldehydes.

91

162

5.2 Additions Involving Allylmetal and Allylboron Compounds

When working with I-substituted allyl compounds it is important that one should be able to prepare the geometrical isomers pure and that they should be configurationally stable. Allylmetals can be either rnonohapto ( q l )or trihapto ( q 3 )bonded, i.e. the metal-carbon bond is either localized at one end or delocalized over all three carbon atoms. Both types may be configurationally stable, stability being much dependent on the character of the metal and its substituents. Allyllithium, -magnesium, -zinc and -cadmium compounds undergo fast E / Z isomerization via a trihapto intermediate and there is not much one can d o about it. Dialkyl-ally1 boron compounds also undergo fast isomerization, but the introduction of one amine or two oxygen substituents at boron completely inhibits E / Z isomerization at room temperature. As is apparent from the foregoing discussion, allylmetal addition to carbonyl compounds is a highly versatile method which can be tuned to the actual task by a proper choice of the metal, its ligands and the reaction conditions. There is a bewildering amount of information available about this topic and our own way of ordering it is division according to the leaving group. This is not ideal since it cuts across classification according to the stereochemical outcome of the reaction.

5.2.2 Addition of Allylboron Compounds * Applications of dialkyl allyl boranes is dominated by works originating from the Brown laboratory and involve almost exclusively the use of diisopinocampheyl and other terpene-derived boranes. B-Allyl-( + )-diisopinocampheyl-(( + )-24) and (+)-diisocaranylboranes add to simple aldehydes with almost complete enantioselectivity when working at -100°C [159, 1601 and somewhat less at -70°C [143, 1441 (Fig. 5-13 (a)). In addition to chiral aldehydes strong double induction was observed. While addition of (+)-24 to (S)-2-methylbutanal produced the anti-homoallyl alcohol in 90% de, ( - ) - 2 4 led to the syn product in 92% de (e.g. Fig. 5-13(b)). With the corresponding (E)-crotylboranes complete 2,3-anti- and 2,3-syn selectivity resp. was realized while 1,2-selectivity was still high and in the sense observed with the allyl compounds [ 161 - 1641. The reaction was successfully extended to boranes in which the allyl system was part of a 6 - 8 membered ring [165, 1661. Syn selectivity with a (Z)-l-methoxy analogue of (-)-24 was complete and combined with high enantioselectivity (Fig. 5-1 3 (c)) [ 1671. Stereoselectivities with the 2-methallyl analogues were in the same range [168]. Addition of B-allyl-9-borabicyclo[3.3.llnonane to azomethines of 2-phenylacetaldehyde gave 2 9 2 % of the syn amines [169,170].

* For a review see ref.

[158].

5.2.2 Addition of Allylboron Compounds

163

OH

(4

R = Me, Et, Pr, IPr. t k , vm) I. Ph ee >96%

OH

I

Me de 92 Yo

de 90 %

OH

1

R = Me, Et, IPr, Ph, vinyl, ce >88%

R = Et, P r , t b , c e x , crotyl, ec >96%

I

SiMe, (25)

Fig. 5-13. Addition of allylboron compounds to carbonyl compounds (I).

When the reaction was extended to azomethines of achiral aldehydes with 2-phenylethylamine [ 170, 1711 asymmetric induction (de > 84%) was experienced. With B-crotyl-9-borabicyclo[3.3.llnonane and achiral azomethines as partners anilides gave predominantly anti products, while azomethines with aliphatic amines mainly syn amines [ 1721. Since simple dialkyl-allylboranes undergo fast E/Z isomerization, they are not very promising reagents for diastereoselective additions. Nevertheless the reaction can be selective when one of the geometrical isomers is highly preferred (e.g. Fig. 5-13(b)) [173]. When the ligands are bulky enough, (e.g. BBN), excellent diastereoselectivity can be attained even with ketones [ 1741. With the aid of N-methyl-pseudoephedrine the chiral borane 25 can be conveniently prepared and reacts very selectively with aldehydes (Fig. 5-13 (d)) [ 1751.

164

5.2 Additions Involving Allylrnetal and Allylboron Compounds

de~900/b

Me Me

; Cl

RCHO ------+

OH

o

YOee of Z W-93 Me

CI R=Me, Et, iPr, Ph

Fig. 5-14. Addition of allylboron compounds to carbonyl compounds (11).

Reaction of chiral monooxygenated boranes with aldehydes was reported by Midland and Preston to give >96% of the anti product with moderate to good enantioselectivity (50-85'70 ee) [176]. The addition of ally1 boronates to aldehydes has been very thoroughly studied by Hoffmann and his group [141, 1771 and in the last decade by Roush et al. While the first group concentrated mainly on boronates of pinacol, the second explored the potentials of tartrate derivatives. By the proper selection of the diol component high dia- and/or enantioselectivity can be achieved. Preparation of the boronates and typical examples for their use are illustrated in Figs. 5-14(a) and (b). Using pinacol as the diol component, addition to aldehydes was completely diastereoselective: the (2)-crotylboronate gave the syn-, and the E isomer the anti-homoallylic alcohol. Enantiomeric purity was limited by the diastereomeric purity of the boronate (93 -95%) [145, 1781. The method can be extended to y-alkylthio-allylboronates and (2)-and (E)-y-alkoxy (e.g. methoxymethyl) boronates. The latter give 3,4-anti- and 3,4-syn-diols resp. of 60-96'70 de [179]. y-Unsubstituted allylboronates of chiral diols such as 26 (from (+)-camphor) [180, 183, 1841, and 27 [I851 or the inexpensive 28 [181, 1821 add to achiral aldehydes with varying enantioselectivity (Fig. 5-15 (a)).

5.2.2 Addition of Allylboron Compounds

HO

(27) R

=

165

iPr

(28) R = COjPr

HO Me

(R,R)-(29)(R* = 28)

R Me Et tPr t&u Ph -CH=Cy CEX R*=26 Yiee 86 77 70 45 36 50 R*=28 %ee 82 71 87

?

Fig. 5-15. Addition of allylboron compounds to carbonyl compounds (111).

An example for high selectivity and transformation of the sterically very modest triple bond to a bulky group is as follows [188]: CHO

* \ ee92%

&pt’

6

166

5.2 Additions Involving Allylmetal and Allylboron Compounds

Reaction of 29 with a-chiral aldehydes involves double induction. With (R)-isopropylidene-glyceraldehyde (R,R)-28 gives the anti product in 96% de, while its enantiomer the syn-alcohol in 86% de [187]. In the addition of 1 -substituted allylboronates (crotyl analogues) to a-chiral aldehydes, high Cram-selectivity is epxerienced with the (E)-boronates while the 2 isomers react predominantly in the anti-Cram sense. Contrary to what should follow from the 6-membered chair-like transition state model, the size of the 1-substituent exerts not much influence on selectivity (Fig. 5-15(b)) [189]. a-Substituted allylboronates of pinacol of which predominantly the a-chloro compounds have been studied by Hoffmann et al. [190- 1921 yield the 2 isomer of a homoallylic alcohol. Combination of a chiral, but less bulky diol used for chirality transfer and pinacol for stereodirection in the addition step permitted the enantioand diastereoselective synthesis of such alcohols (Fig. 5-14(c)) [ 190, 1911. Cooccurrence of the indicated configurations is a proof for the cyclic mechanism of the reaction. Selectivities with the Br, OMe, and SEt analogues fall into the same range [192]. Reaction of the crotyl boronate involving diol 26 with benzaldehyde was highly diastereoselective (98% syn) but not sufficiently enantioselective (60% ee) [ 1461. In the addition of the same to (S)-2-methylpropanal complete 2,3-anti- and high 3,4-syn selectivity was observed (Fig. 5-15 (b)), while the R + E , S + Z , and R + Z combinations all gave poor 3,4-diastereoselectivities [ 1101. On addition of the (E)- and (Z)-crotylboronates and y-silylated allylboronates with 28 as the chiral group to achiral aldehydes, excellent diastereo- and moderate enantioselectivities (55 - 86%) could be realized [186, 1941. The silyl derivatives lead via oxidative desilylation ultimately to diols [ 195, 1961. n-Face selectivity with a-chiral aldehydes followed Cram's rule but was moderate to poor [197-1991. Reaction of (E)-and (Z)-a-chloro-, and a-methylcrotylboronates of pinacol studied foremost by Hoffmann et al. are highly selective concerning the relative configuration around the newly formed single bond and, except for the (Z)-a-methyl compound, totally selective at the new double bond [200 - 2041: Me

OH

R' CI OMe Me R2 Me Et P r Ph MeEt P r Ph Et iPr Ph %de 92 96 95 96 94 90 94 90 96 92 94 Me

Me

OH

Me

5.2.3 Addition of Allyltitanium Compounds

161

With a-chiral aldehydes homoallyl alcohols with three consecutive chiral centers can be generated with high selectivity from (Z)-boronates [200, 203 -2051, with the (E)-boronates diastereofacial (Cram) selectivity is very poor [ 177, 2061. (E)-crotylboronates add at high pressure to aldoximes to give anti-a-methyl-yoximinoolefins (de 50- 90'70),while the Z isomers provide the syn-oximines (de 75-94oio) both convertible to amines [207]. The addition of an allenic boronate to aldehydes is anti selective and probably involves a cyclic transition state, too (Fig. 5-1 5 (c)) [208]. The following reaction discovered by Corey et a/. excells not only by high selectivity, but also by the novelty of the products [209]: Snh,

OH

5.2.3 Addition of Allyltitanium Compounds Titanium can form monohapto tetracoordinate allyltitanium compounds, pentacoordinate ate complexes, further trihapto allyltitanium complexes and thus offers ample room for variation [210]. The monohapto amide 30 was found to add to a chiral aldehyde with high syn selectivity (Fig. 5-16(a)) [21 I]. The amide 33 of undefined E / Z configuration adds to aldehydes with complete 2 selectivity [212]

Several alkoxy and aryloxy titanium compounds were tested for their diastereoselectivity by Widler and Seebach, of which the triphenoxy compound was the best. Addition was anti selective (Fig. 5-16(b, c)) [151, 1371. Addition to ketones was also selective and could be interpreted, by considering the difference in the size of the

168

5.2 Additions Involving Allylmetal and Allylboron Compounds

R'

JJ R2

R1 = Me. Pr, b. R* = P r , &CH Ph. dc 192%

c%x cFkx Ph Ph Ph tBu tBu *x Me Ei iPr Ph Me Me Me Ei 86 60 %de 82-92 14 56 >96 >96 70

L S Me

substituents. The addition of crotyl-bis(cyclopentadieny1)-titanium halides (32) can be conducted either uncatalyzed or under boron trifluoride catalysis. In the uncatalyzed reaction stereoselectivity depends on the nature of the halogen and is supposed to proceed via a cyclic intermediate. With X = Br the (E)-crotyl compound gave an anti product exclusively [213]. In the catalyzed reaction syn preference was observed (Fig. 5-17(a)) independently of the E / Z ratio, an indication for a linear mechanism (cJ: Fig. 5-12(b)) [214]. Addition of the ate complex 23 has already been mentioned (p. 161). Trihapto titanium complexes can be prepared via ally1 anions, and the addition of 1,3-disubstituted representatives was studied by Sat0 et al. Anti-preference and an interesting dependence of double bond configuration on substitution at C-3 were observed (Fig. 5-17(b)) [215]. As found by Hoppe et al. diisopropyl carbamates of crotyl alcohols (e.g. 31) can be readily lithiated in a-position with retention of configuration. At low temperature,

169

5.2.3 Addition of Allyltitanium Compounds

Me anh X=CI

X=Br

91:9

1.99

91:9 84: 16

0:IOo

Ph

EtCHO

THF, -78OC

*

OH

Me anti

Cbm

= CON@r,

OH 1) Ti(OlPr),

2) PrCHO

(3R,45), 63 % ee

(3S,4R), 87 % cc

R’ = Me, Ph; R2 = OTHP, Me; R3 = Pent, cHex

anii Fig. 5-1 7. Addition of allyltitanium compounds to carbonyl compounds (11).

170

5.2 Additions Involving Allylmetal and Allylboron Compounds

due to complexation with the CO group the lithio compounds are configurationally stable and can be substituted with inversion by TiClR, (R = OiPr, NEt,). The resulting crotylitanates add syn selectively to aldehydes, absolute configuration of the product depending on the nature of the R group (Fig. 5-17(d)) [216, 2171. Lithiation of crotylcarbamate 31 was enantioselective when it was carried out in the presence of (-)-sparteine and after exchange for Ti (OiPr)3, reaction with aldehydes proceded with complete anti and excellent ( > 92%) enantioselectivity. The latter was due to second order transformation of the (R)-lithio compounds to a crystalline (-)-sparteine complex of the S enantiomer [218, 2191. When simple carbamates were lithiated and then treated with e.g. CITi(OiPr)3, the resulting titanates added with very high anti selectivity to pentanal and acetophenone ( d e r 9 4 ) [215]. Chiral allyltitanates reacting with high enantioselectivity with aldehydes can be prepared with diacetoneglucose as chiral alcohol [220]. A reaction related to allylmetal addition is depicted in Fig. 5-17(d) [221]. If azomethines were used as partners, the corresponding anti amines were obtained, in most cases with total selectivity [222].

5.2.4 Addition of Allylsilanes The addition of allylsilanes to carbonyl compounds requires activation by a Lewis acid, TiC1, being the most often used catalyst. Enantioselective addition of a chiral allylsilane to a chiral aldehyde was described by Hayashi et al. (Fig. 5-18(a))* [223]. Addition of the crotyl analogue has already been mentioned (Fig. 5-1 1 (b)). Kishi and his coworkers demonstrated that in an intramolecular ally1 addition chirality at the silylated carbon was completely transferred to the product (Fig. 5.18(b)) [224]. 1,3-Induction and configurational control by choice of the Lewis acid is shown below. Ally1 transfer is supposed to be strictly intramolecular with TiCI, and intermolecular with SnCl, [225].

OH

OH

OH

OH

TiC14 CYCh, -78 OC *R

I R = Me, de 84 %

* Ee values corrected for the optical purity of the silane (95%).

R = Me, Bu, de 80-84 %

5.2.4 Addition of Allylsilanes

..SiMe, OHC

/

2- 98

TICI., CHlCI2 -78T

ITUI1.S

2 9 8 % ee HO"

(c)

R ' w S i M e 3

'%I

171

&CHO, TEI. CHzCI2, -78°C

R' = M e , Ph RZ = M e , Et, iPr, tBu de = 86 - 98%

Fig. 5-18. Addition of allylsilanes to carbonyl compounds (I).

Addition of both (E)-and (Z)-crotylsilanes to achiral aldehydes proceeds with syn preference, but selectivity is higher with the E isomers (Fig. 5-18(c)) [226]. Crotylsilanes with pentacoordinate silicon add to achiral aldehydes with comparable syn selectivity [227]. Sometimes protic acid catalysis should be preferred to that by Lewis acids, as shown by Itoh et al. (Fig. 5-18(d)). The 2 isomer of 34 gave the trans-lactone in >93% [230]. Addition of allylsilanes to chiral carbonyl compounds, both under 1,2- and 1,341duction, was highly diastereoselective, while even this approach failed to produce useful results with chiral a-ketoesters [23 11. Examples for 1,2-induction are depicted in Fig. 5-19(a) [75, 232, 2331 and for 1,3-induction in Fig. 5-19(b) [93, 2341. Addition to the ephedrine-derived a-ketoamide 35 gives ultimately a diol (S)-(36). (R)-36 (ee 92%) can be obtained using the (S)-leucinol analogue of 36 [235].

172

5.2 Additions Involving Allylrnetal and Allylboron Compounds

w,

OH e e > 9 8 %

Ph

Intramolecular transfer of the ally1 group under chelation control was completely selective, while no selectivity was found with nonchelating BF3.Et,O [236]:

I , - 78 OC

~

W

C

;)2

q

,

R2= Me: €3:IPr: Ph,'H

A chiral alcohol is involved as a third partner in the procedure reported by Mukaiyama et al. [237] and with T i c & as Lewis acid by Seebach et al. [238]: OSMe,

RCHo

Ph

+

AMe

Ph,BOTf PhMe, - 7 8 0 C )

~

I

iPr c€kx W C 3 Ph de%I 99 85 93 56

R

Addition of allylsilanes to chiral acetals of protected a-aminoaldehydes produced 1,2-aminoalcohols [228]: Me

A m s M e 3 , TiCI9 CH$lz, - 78 OC

1 NHCbZ

R = Me, &h,Eh, de - 70 %

173

5.2.5 Addition of Allylstannanes

R

CIS

SNe,

TGI,

~

> 99 %

R1+

OW

O h

OH

R'

R2

%anti

Me

H Me

99

Bu

95

Fig. 5-19. Addition of allylsilanes to carbonyl compounds (11).

5.2.5 Addition of Allylstannanes Allylstannanes can be induced to react with aldehydes in the absence of a Lewis acid at high temperature, high pressure or by taking a very active aldehyde like chloral [229], but it is more convenient to activate the aldehyde. In both cases the syn products prevail [141]. Allyl-tributylstannane adds to a-hydroxyaldehyde derivatives with excellent stereoselectivity, which can be inverted by changing the blocking group and the Lewis acid (Fig. 5-20(a)) [144]. Similar phenomena were observed in the addition of allyl-tributylstannane to a alkoxymethyl aldehydes [471] and of crotylstannanes to a-alkoxyaldehydes [239]. Reacting the components (R = Bn) in 5 M etheral LiCIO, at RT is another way to obtain the syn product (dez9270) [240]. An analogue of the above reaction lacking, however, its alternativity is shown below. Existence of a chelate complex of the substrate with TiCI, of a conformation leading to the predominant product was demonstrated by NMR [241].

174

5.2 Additions Involving Allylmetal and Allylboron Compounds

Acetals of (R,R)-2,4-pentanediol can be cleaved by ally1 tributylstannane activated by a large excess of TiCl, and Ti(OiPr), to give homoallyl alcohol derivatives with almost complete selectivity [2421.

(a )

R=Bn

+

R= eSnBu MeltBuSi

.Hex+

.Hex+

OSitBuMez 95"4 anti

OBn

RCHO

R = Me, Et, iPr, Ph, de 292%

BF,. %O, - 78 oC Me

RI = CKOMe, Me; Rz = Et, Pr, Pr, B,Ph

Me

R*=8-phenylmenthyl

k$z?--E Fig. 5-20. Addition of allylstannanes to carbonyl compounds.

5.2.6 Addition of Other Allylinetals

175

Achiral crotylstannanes add to achiral aldehydes with high syn preference, irrespective of the configuration around the double bond (Fig. 5-20(b)) [229, 243-2461. Fig. 5-20 also shows the preparation of the stannane. Syn selectivity is also the rule for the BF3-Et,O-catalyzed addition of 1 -alkoxyallylstannanes to achiral [247] and to a- or /3-chiral aldehydes [248]. The alkoxy group may be methoxy [247], Me,tBuSiO [248] or even a chiral alkoxy [249]. Diastereofacial selectivity with chiral aldehydes is high (de 85 - 100%) [248]. Stereoconvergence of the reaction for ( E ) - and (Z)-allylstannanes is attributed to the linear mechanism presented in Fig. 5-12(b). In an uncatalyzed addition of an asubstituted stannane only the product anticipated by assuming a cyclic mechanism could be detected (Fig. 5-20(c)) [250]. At high pressure the role of boat type transition states becomes significant [252]. Yamamoto and his coworkers found that in the addition of a crotylstannane to glyoxylate esters diastereoselectivity increased with the size of the alcohol (90% syn with iPr), and when the alcohol was 8-phenylmenthol enantioselectivity was also acceptable (Fig. 5-20(d)) [253]. a-Alkoxycrotylstannanes and achiral aldehydes react under BF, Et,O catalysis with generally very high syn-(E)selectivity with aliphatic aldehydes, while with benzaldehyde, presumably due to interaction of the aromatic ring with the ether oxygen the predominant product is syn-(Z) irrespective of double bond configuration in the starting material [254].

5.2.6 Addition of Other Allylmetals It seems that, except in the presence of special stabilizing groups (e.g. SC = S(NMe,) [256]) no useful stereoselectivities can be realized with allylmetals in which the metal is lithium, halomagnesium or cadmium [140]. According to Mulzer et al. [257] diallylzinc adds to isopropylidene-glyceraldehyde with 90% anti selectivity. Zweifel and Hahn found high anti selectivity for the addition of diallenylzinc to achiral aldehydes (Fig. 5-21 (a)) [258]. Finally addition of an allenylstannane to aldeyhdes should be mentioned. High syn selectivity requires a-substitution of the aldehyde (Fig. 5-20(e)) [255]. Dialkyl-crotylaluminum compounds were studied by Hoppe and his coworkers [259, 2601. Both anti and E selectivities were high, especially with bulky aldehydes (Fig. 5-21 (b)). One of the rare cases of chiral catalysis involving an allylaluminum compound was reported by Mukaiyama et a/. [261]: -

1) Sn(OTQ2, 2) e A ( f i ) z C&CI,, -78 OC, 3) ArCHO + A r Me

ce 80 - 84 %

176

5.2 Additions Involving Allylmetal and Allylboron Compounds

(4 R~CH.-C-CH~

--

I ] tRul i . THF. - 00°C

2 ) LtiC11.

.>)

R-C'HO

R2

OH anti 92 - 99%

oh,

R'

R'=Pr, tBu, Pent, Hex, cHex Rz=H, Et, Pr, iPr, tBu

2I)) tRu1.i. ClAl(iF3u)2.

Me---YA'iBu2

*

- 78°C

OCONi Prz

bCON i Prz

OH 1 -

Me

91 86

%anti

%E

(Q-anh

R

Pr

iPr

tBu

97 93

94

95

Pent

83 89

Ph

Fig. 5-21. Addition of allylmetals (M = Zn, Al, Cr) to aldehydes.

Reaction of both ( E ) - and (2)-crotylbromide with chromium(I1) chloride gives a (not isolated) crotyl chromium compound which adds, as reported almost simultaneously by Buse and Heathcock [I481 and by Okuda et al. [262], with high to total anti selectivity (except for pivalaldehyde) (Fig. 5-21 (c)). Asymmetric induction in the addition of this reagent is modest with simple a-chiral aldehydes [148,474], but with aldehydes containing several chiral centers 4,5-anti selectivity can amount to 95 Yo [263]. 1,I-Disubstituted chromium ally1 intermediates generated from allylphosphates

add with high selectivity to aldehydes proving the validity of the Zimmermann-Traxler model even in this case [264]: OH R2CH0, 2 CrC12,

Bu

Ry Ykie 80

>98 86

5.3 Stereoselective Aldol Reactions

177

A chromium reagent obtained from a dichloroallyl compound adds with high anti-(2) selectivity to aldehydes while with the allyl analogue both regio- and stereoselectivity is poor owing to 1,3-allylic rearrangement [265]: Me

I

OH

i2sbzEz c€kx Ph WC& Oct

It appears as if few metals have escaped from being tested as a ligand for allyl addition. Thus for example the addition of crotyl-biscyclopentadienylzirconium chloride [93, 2661, of trihapto molybdenum complexes [267, 2681 and of tin and lead allyl compounds to aldehydes have been reported but the results seldom justify the often cumbersome preparation of such reagents.

5.3 Stereoselective Aldol Reactions * An aldol reaction, also called aldol condensation, is, in the broadest sense, characterized by scheme (D) and involves the establishment of a C-C bond between two carbonyl compounds to form a P-hydroxycarbonyl compound, i.e. an aldol. The similarity to the addition of allylmetal compounds has already been pointed out and is not merely formal, but concerns the mechanism, stereochemistry and also the nature of M.

The aldol reaction, discovered in 1838, is almost as old as organic chemistry itself, but its potential for stereoselective synthesis was only recognized in 1968 by the Dubois'. The idea was picked up by Heathcock, Masamune, Evans and many others and

* For

reviews see refs. [269-271, 281, 2841.

178

5.3 Stereoselective Aldol Reactions

developed into a powerful method for the stereoselective construction of open chain compounds containing several adjacent chiral centers. The fact that this reaction remained dormant for so long should not be blamed on lack of imagination among chemists of yesterday but rather on the fact that the techniques (high field NMR, HPLC, flash chromatography, etc. ) by which the complex mixtures of the heat-sensitive products could be analyzed and separated were simply not available earlier. Also, earlier chemists thought it unrealistic to tackle such highly complex synthetic targets as, for example, macrocyclic antibiotics, which much stimulated the development of the aldol methodology. The number of papers published on this topic is staggering and after more than 15 years of intensive research the flow of papers on aldol reactions is still unabated. As with allyl addition, several ordering principles can be invoked by which some shape can be given to the mass of facts. Such criteria may be (i) the symmetry of the partners (both achiral, one of them or both chiral), (ii) type of end product, (iii) the nature of metal component, and (iv) type of compound providing the enolate (ketone, ester, acid, etc.). We decided that a mixed approach would be least confusing, i.e. first we discuss how the stereochemistry at the ends of the newly formed bond can be controlled and then, in less detail, asymmetric induction by chiral tenters in the carbonyl and the enolate partners.

5.3.1 Stereochemistry and Mechanism of the Aldol Reaction Basic stereochemistry The theoretical possibilities for stereoselection in the aldol reaction can be analyzed along the same lines as those of allyl addition and will be treated here only briefly. (A) Addition of enolates, unsubstituted at the P-carbon, to achiral carbonyl compounds with enantiotopic faces gives no stereoselection unless the enolate is chiral. When chirality is imparted to the enolate by a chiral auxiliary, which is cleaved off after addition, enantiomers are obtained (Fig. 5-22(a)) [272]. Otherwise the products are diastereomers (Fig. 5-22 (b)) [273]. (B) Addition of enolates substituted at the P-carbon to carbonyl compounds with prochiral faces leads necessarily to diastereomers with configurations syn or anti at the ends of the new bond (Fig. 5-22(c)) [274]. A chiral center in any of the partners, even when these are optically pure, doubles the number of possibilities, since this center is disposed either syn or anti to the nearest new center (Fig. 5-22(d)) [275]. Note that in our example addition is completely

5.3.1 Stereochemistry and Mechanism of the Aldol Reaction

OTiCp(OR*)2

MeC02tBu l)LDA, Et20, -74 OC ~)CPT~(OR*)~CI

*

A

OtBu

QH

RCHO -74 o c

-

* R&a2tBu

R = Pr, iPr, iBu, tBu,cHex, Ph, ee>92%

R* = diacetoneglucose

(b)

119

OSiMe, Me,SiOTf Et,N

iPrCHO . CHzCIz,-78 OC *

iPrM e*

+

i

p

r

q

M

,tBu

CO,tBu 964

Yn

ariri

(d )

P h y C H O +

e

anti

8 5 : I5

Jyn

OH

MeJ

Me

tBu

0

P h y k f k

> 98% z

t Bu

syn, syn

+ P

8 6 : 14

h Me

q t Bu

Me

s y u . onri

Fig. 5-22. The stereochemistry of the aldol reaction (I).

syn selective concerning the centers at the new bond (no 2,3-anti products are detected), whereas asymmetric induction by the a-center of the aldehyde is incomplete. (C) Reaction between achiral partners in the presence of a chiral additive may give rise to an optically active aldol. The Reformatsky reaction in the presence of (-)spartein is a classical example [276].So far chiral solvents have failed to induce appreciable enantioselectivity [ 2 ] ,while some diamines derived from proline studied by

180

5.3 Stereoselective Aldol Reactions

R

1

Me

iPr

iBu

Me02C(CH2)2 Ph

%ee

I

85

>95

>95

>95

~

CN

x/.H-

OSnOTf

I

2) HCHO -95'C

*

R

I syn

%syn % ee

> 95

1)

Me

iPr

tBu

cHex

Ph

>95

100

75

90

80 80

86 80

I-Naph

Utlli

Fig. 5-23. The stereochemistry of the aldol reaction (11).

Mukaiyama were highly effective with tin(I1) enolates (Fig. 5-23 (a) and (b)) [277]. Mulzer et af. found that a carboxylic acid dianion generated with a chiral lithium base added to benzaldehyde with appreciable enantioselectivity (Fig. 5-23 (c)) [278] *. (D) Because simple 1,2-asymmetric induction generally failed to bring about high diastereoselectivity in the relationship of newly formed and existing centers of chirality, double induction in the aldol reaction has been thoroughly studied. It was assumed that a combination of chiral partners which produced the desired relative configurations in their reactions with achiral partners individually would provide high diastereoselectivities by synergism. Examples in Fig. 5-24 illustrate the confirmation of this hypothesis by experiment [279] **.

* In want of more practically important examples this case will not be discussed any further. ** Multiplication of the individual stereoselectivities [281], whether the result agrees with experimental findings or not, is a gross simplification.

18 1

5.3.1 Stereochemistry and Mechanism of the Aldol Reaction

Me Mc

cHe.xGCH0

-k

>

cHexA

Me

Me Me Ph

T‘K

OH

0

anti.syn,onti

’0

+

I

SiMei 8 9 : I!

Ph

M e Me

0

C k X

OH

0

I

SiMe,

syn. syn. syn

OSiMe3 (S ) Fig. 5-24. Double induction in the aldol reaction.

Note that if the (S)-aldehyde was combined with the (R)-enolate diastereoselectivity was, as expected, poor and the anti, syn, syn product was formed in excess. Occasionally very high selectivities could be realized with complicated substrates [270, 279, 2801. Since the main field of application of double induction is the synthesis of complex natural products, very few simple examples can be quoted, and therefore no separate section will be devoted to this topic. *

Transition state models In contrast to ally1 addition which is irreversible and thus under kinetic control only, the aldol reaction is reversible and may be controlled, according to conditions

* For more details, see also refs. [282, 283, 360, 4301.

182

5.3 Stereoselective AIdol Reactions

and the nature of partners, either kinetically or thermodynamically. The stereochemistry of addition under kinetic control can be successfully rationalized by considering the geometry of the transition state. In the majority of cases a cyclic transition state, as first proposed by Zimmerman and Traxler [I491 is probable. In such a transition state the metal atom of the enolate is coordinated to the carbonyl oxygen of the partner. The associated product configuration is governed by three factors: (i) ring geometry (chair or boat), (ii) enolate configuration (2 or E ) , and (iii) mode of approach (lk or ul). A change in any of these factors results in a change of product configuration (syn or anti). On circumstantial evidence the boat form is usually disregarded. The enolate configuration is mostly stable enough to be analyzed by spectroscopic methods. In Fig. 5-25 (a) cyclic chair-like transition states are depicted for the (2)-enolates. For the (E)-enolates similar transition states can be drawn up and the result is complementary, C, leads to a syn- and C2 to an anti-aldol. When R2 is small R‘ R3 1,3-diaxial interactions overrule R2 R3 gauche interactions, C, is the favored transition state and therefore (2)-enolates give syn-aldols and (E)-enolates anti-aldols as the major product. Also, stereoselectivity should increase as the size of R3 increases. In transition state C2R’ and R3 get closer to each other when the M - 0 bond length is smaller and/or Me - 0 C coordination is stronger. Indeed, boron enolats are more selective than lithium enolates. Table 5-4 provides ample evidence for the correctness of these predictions, but the diagrams in Fig. 5-25(a) do not account for the fact that when R’ is not large, syn selectivity with (2)-enolates is usually much higher than anti selectivity with the (E)-enolates. This can be explained by assuming that the torsional angle of C = C and C = 0 is larger than the ideal 60°, which is a reasonable proposition for metals with a long M - 0 bond (Li, Mg, Zn). In such a “skewed” transition state (Fig. 5-25 (b)), while R’ R3 repulsion becomes more severe in arrangements for both (2)-and (E)-enolates, the advantage of C, (E,Ik) over C2(E,ul) is diminished due to increased R2 R3 gauche interaction in C, (E,lk). These effects are amplified by nonperpendicular attack (at 1 lo”, cJ: Section 3.2.1) and explain why selectivity deteriorates when R2 is large. According to Evans et al. [284], a contribution by boat-type transition states cannot be excluded. A b initio calculations on enol-borates by Hoffmann et al. [285] pointed out that the reason for the often aberrant behaviour of (E)-enolates may be that they exist as both an extended and an U-shaped conformation, while (2)-enolates are conformationally more homogeneous. Molecular mechanics calculations by Gennari et al. on the reaction of (Z)-enolborinates with a chiral a-substituent suggested that the large group at the chiral tenter is oriented opposite to the incoming aldehyde, leading to an a,2-syn-2,3-syn product [286]. A similar study involving achiral (2)-enolates and a-chiral aldehydes non-bonded interactions are dominant in determining selectivity. When a bulkier group is attached to the chiral center 3,4-syn (“Felkin”) selectivity turns to 3,4-anti selectivity [287].

-

t-t

-

-

-

5.3.1 Sfeseochemistry and Mechunisin of (he A f d o fReaction

RZ

I83

R2

Fig. 5-25. Cyclic chair-like transition states for the addition of (Z)-enolates to aldehydes.

A study on the reaction of trisubstituted enolates with aldehydes leading to a quaternary carbon center revealed that for substrates giving rise to syn products regardless of enolate geometry a boat type transition state should be postulated [288]. For enolates giving the same predominant product irrespective of their configuration, or for those requiring Lewis-acid catalysis, an open-chain transition state, similar to that for ally1 addition (cJ: Fig. 5-12(b)), has to be postulated. An interpretation of ester and acid enolate addition by orbital interactions was proposed by Mulzer et al. [289]. Molecular mechanics calculations on enolate additions to chiral aldehydes correctly predicted the trend of the reaction [290].

184

5.3 Stereoselective Aldol Reactions

5.3.2 Generation of Enolates and their Addition to Achiral Aldehydes and Ketones In the knowledge of the mechanistic framework of enolate addition the problems to be solved for the stereoselective creation of the new bond are (i) the stereoselective generation of (2)-and (E)-enolates and (ii) the conservation of enolate configuration in the course of addition.* Enolates can be generated from a variety of substrates, such as aldehydes, ketones, esters, carboxylic acids, thioesters, amides and thioamides using several metals, such as Li, Mg, Zn, Al, Ti, Zr, Si or B. The character of the oxygen-metal bond rnages from purely ionic (e.g. Li, K) to covalent (e.g. boronates). The most important facts about enolate addition to a selected set of achiral aldehydes are summarized in Table 5-4. The correlation of stereoselectivity with substrate structure, counter ion and experimental parameters constitutes a multidimensional matrix from which we have chosen a few vectors for more detailed discussion. If under kinetic control a (2)-enolate gives the syn-aldol in excess, then the (E)-enolate leads to an excess of the anti product. When this rule is not observed and/or product composition is independent of enolate configuration, either thermodynamic control or a non-cyclic mechanism can be suspected. Thermodynamic control is probable when an anti adduct is preferentially formed from a (2)-enolate or from both, while a non-cyclic mechanism is likely when an (E)-enolate or both give a syn adduct. The Reformatsky reaction is usually under thermodynamic control. For example, the sydanti ratio in the reaction of the (2)-enolate (M = ZnBr) of propiophenone with benzaldehyde is a function of time (at - 10 "C 70 :30 after 5 s and 25 :75 after 16 min) 12751. Lithium aldolates equilibrate even more rapidly. Kobayashi et al. studied the addition of the tributylstannyl enolates to benzaldehyde, which was anti selective at - 50 "C and rather non-selective at room temperature. This was shown not to be a result of equilibration, which was slow at room temperature, but due to a change of mechanism [319]. Titanium [303] and zirconium [302] enolates are typical of the case when the syn product predominates irrespective of enolate configuration. Stereoselectivity is generally higher with the (Z)-enolates than with the (E)-enolates (Fig. 5-26(a)), and there are fewer methods available for the preparation of anti-aldols than for syn-aldols. Stereoselectivity much depends on the nature of M. The preparation of lithium enolates is the easiest, but it is often necessary to resort to other metals. Although pure Z or E lithium enolates can be prepared from the corresponding

* This does not apply to reactions in which aldol configuration is independent of enolate configuration.

5.3.2 Generation of Enolates and their Addition to Achiral Aldehydes und Ketones

A A

A

A h

w e

v v

3

0 0 cm w .l

00 00

h

h

e

w

v

v

m

00 o\

m

A

A

6z cn

6z m

A

A

v

v

NNN

E

5

w‘

Bn

Ph Me

Et

Ph

Me Et

tBu Me

Me

ZrCp2C1

ZnCl

AIMe,

iPr tBu Me Et iPr tBu

- (CH2)4 -

M

R2

R’

Table 5-4 (continued)

(86)

E

z E z

E Z

z

E E E E E b

E (86)

z(92)

C

b E

E

z Z z z z z

ration

Enolate configu-

100 (s)

100 (s)

100 (s) 100 (s) 100 (s) 100 ( a ) 100 (s)

Me

Et

88 (s)

iPr

R3

tBu

>95 75 >95 80

Ph

(0)

(s)

(0)

(s)

Comment Ref.

2

m

00

+

tBu

Et Ph iPr Me Et iPr PhS

Me

Et iBu iPr tBu cHex Ph Me,Si Et Bn CHEt, cHex

Ph

R'

- (CH,), -

- (CH2)4 -

- (CH2)4-

Me

Me

Me

R2

Table 5-4 (continued) Enolate

z (95) z(99) z(98)

z(96)

C

Z

z

z

E (55)

z (>97) z(>99)

E

Z

(%)a

configuration

%Me3

4:") 0

Me BClPh

Me

T '

E E Z

Z

Z

E BcPent.Thexd E

BBu, 9-BBN BBu,

M

Me

iPr

95 (s)

100 ( S ) e

100 ( s ) b

>99 (s) >99 (s) >99 (s)

90 (s)

>95 (s) >95 (s)

Et

R3

>95 (s)

tBu

>99 >95 >94 >99 >99 99 95 75 50 >98

75 >97 >97 56 >97 >97 >97 81 95 95 >99 >99 87 >96 95

Ph

(a)f ( a ) under BnMe,N@F" catalysis

(a)f

(s)

(s) (s)

(s)

Ref.

[I531 [153] 11531 [3 101 [3101 [3 101 [3101 [311, 3201 [311, 3121 [313]

[3061 [3061 [274, 3071 [2741 W41 [274 I2741 12741 [2741 [3081 13081 [3081 [3081 r2.741 P741 in petroleum ether at [309] 25 OC in CH,Cl, at -28°C (s) in CH,Cl, at -28°C (s) in CH,Cl, at -28°C (s)

(s)

(a) (a)

(s) (s) (s) (s)

(s) (a)

(s)

(s)

(a)

(s)

(s)

(a)

Comment

-1

co

e

w

s

6

4

a

s

r2

a

188

5.3 Stereoselective Aldol Reactions

n

Y

v

m 00

h

n

n

v

m W

r-

n

-3.3 m o m m

c3

3

m

A

A

h

3 m N

h

Y

v

W

r-

N

N

lu

41

5

I:

a

I

I

I

I

5.3.2 Generation of Enolates and their Addition to Achiral Aldehydes and Ketones

-

m

m N

E *

h

N

5 u,

p!

-d

5

I

I

I

I * F

WPI

189

m r4 m

PhS (c) Carboxylic acid dianions' OLi Et iPr tBu 1-adamantyl Ph Me OBn

OSiMe, OMe SPh tBuS

OMe

Me2N Me

OtBu OEt

Ph Me Et Me Et iPr cHex tBu (Me,Si),N Me

R2

R'

Table 5-4 (continued)

Li

Li ZrCp,Cl 9-BBN BBu, BcPent 9-BBN

ZnBr

SnBr

OSiMe,

M

E

C

C

C

C

C

C

C

C

E Z E (85)

Enolate configuration

Et

64 ( a )

Me

64 ( a )

52 ( a )

iPr

97 ( a )

90 ( a )

R3 Ph

>98 ( a ) >98 ( a ) >98 ( a )

50

tBu

92 ( a ) 80 ( a ) 98 (0)

55 ( a ) 55 ( a ) 58 (a) 88 ( a )

Comment

Ref.

4

i,

I/,

W 0

5.3.2 Generation of Enolates and their Addition lo Achiral Aldehydes and Ketones

h h h h h

h

h

33333 w w m m m

3

v

W

m 0

m o w m w

A

m

h h

h

33

0, v

mrA

0

Y

m

h

h

h

3

3

3

W

00

0

N

N

vl

r-

m

19 1

R2

Ph

Me

MgBr

z

Z Z Z Z

Li

iPr Me iPr Ph

Me

Et

-

iPr

R3 tBu

Ph

Comment

Ref.

a

Only quoted with mixtures. Stereoselectivity independent of E / Z ratio. Unknown. 1 ,1,2-trimethyl-l-propyl. Under BF,.Et,O catalysis. Under TiCI, catalysis. g DMP = 2,6-dimethylphenyl. BHT = 2,6-ditert.-butyI-4-methylphenyI. No E / Z isomerism possible. j Note priority of S over 0.

MeNLi MeOCH2CH,NLi

PhNLi

E Z

>97 ( Z ) >97 ( Z ) >91 ( Z )

ration

configu-

Enolate

Z Z

Li SiMe,

Li MgBr

M

iPr Me

- N(Me) - (CH,), NMe, Me

NMe,

(e) Thioamides and thioesters

R'

Table 5-4 (continued)

8

?2

2

h,

\D

L

193

5.3.2 Generation of Enolates and their Addition to Achiral Aldehydes and Ketones

trimethylsilyl enolates, when prepared directly lithium enolates tend to be formed as a 2-E mixture. The selection of the best M is not easy; no general guidelines can be given in this respect, since yields, ease of preparation, separability of products and last but not least the way in which the aldol can be transformed to the target compound all have to be pondered. 4. Correlation of enolate constitution and stereoselectivity is also rather complex. In accordance with the cyclic model, selectivity improves, both for ( Z ) - and (E)-enolates, with increasing size of R' (Fig. 5-26 (b)). Available data on the role of R2 indicated that when the aldehyde is bulky, on increasing the size of R2 product composition shifts in favor of the anti-aldol (Fig. 5-26 (c)). Finally, no generally valid correlation between the constitution of an achiral aldehyde and stereoselectivity has emerged yet. After having surveyed the aldol reaction in general, we now turn to the generation of enolates. Lithium enolates are prepared by reacting a ketone with a lithium base, most often lithium diisopropylamide (LDA). Occasionally lithiated cyclohexylisoproplyamine, hexamethyldisilylamine (LHMDS) and 2,2,6,6-tetra-methylpiperidine (LTMP) are used. The enolates are formed as 2-E mixtures. LHDMS promotes the formation of (2)-enolates, while LTMP that of the (E)-enolates [275, 2941. Pure ( E ) -

(4 hwcH=<

OM

PhCHO,

R

Ph

V R

R = Me, WI, Ph

+

P

h

U

w1

Bu

w

anti

(from E 75-80 %)

(from Z > 95 %)

.syn R2 %syn

I

R1 = tBu, M = MgBr, R3 = iPr Me Et iPr tBu

I

100

100

29

0

R

anti R2 %syn

I

1

R1= OMe, M = ZnBr, R3= Ph Me

Et

iPr

tBu

cHex

63

54

47

31

31

Fig. 5-26. Control of stereoselectivity in the aldol addition by the configuration and metal component of the enolate.

194

5.3 Stereoselective Aldol Reactions

and (2)-enolates can be prepared by treating ( E ) - and (2)-trimethylsilyl enolates with methyl lithium [292]. This, however, is only practical when (E)-or (Z)-silylenolates themselves can be obtained in sufficient purity [340]. Although esters give the (E)-enolates with LDA, syn-anti selectivity is poor except when the alcohol is a bulky phenol. N,N-Dialkylamides and thioamides give mainly the (Z)-enolates, but syn preference is only at an acceptable level with thioamides. Dianions of carboxylic acids are generated by treating the acids with LDA. The major product is the anti-aldol, but selectivity is only high when both R2 and the aldehyde are bulky [289,333], or when the cation is potassium complexed with a crown ether. Alkylamides and thioamides form dianions on treatment with butyllithium [339, 3411. A 2 configuration has been assumed for the thioamide dianions, and modest to good anti preference is observed [341]. Fleming et al. converted acrylates into both ( E ) - and (2)-b-silylenolates giving with aldehydes syn- and anti-aldols resp. in acceptable selectivity (de 70- 88%) [342, 3431:

C0,Me

R-k

(PhMe&CuLi t

1) W C L H,o* 2 ) LDA

Bromomagnesium enolates are probably the intermediates when a-bromoketones are treated with magnesium and give excellent syn selectivity when the substrate is a tert-butyl ketone [298]. Bromomagnesium enolates are also accessible by treating the substrate with Et,NMgBr [344] or a Grignard reagent, e.g. tBuMgBr. Stereoselectivity may depend on the nature of the Grignard reagent [345] and is often better than with the lithium enolates [149, 303, 3371. Bromozinc enolates are believed to be the intermediates in the Reformatsky reaction and are prepared by reacting a-bromoesters with zinc. House et al. transformed lithium enolates by the addition of zinc chloride to zinc enolates, which were sometimes more selective than the parent compounds [300, 3011, while in other cases only the yields increased at the expense of stereoselectivity. The (2)-and (E)-dimethylaluminum enolates of methyl neopentyl ketone prepared by Jeffery et af. [299] gave the pure anti- and syn aldols respectively with acetaldehyde, but with benzaldehyde both stereoisomers led to the pure anti adduct. Boron enolates have been intensively studied by Masamune and his group. Such compounds can be prepared from diazoketones with trialkylboranes as the E isomer and isomerized by lithium phenolate to the (2)-enolate (Fig. 5-27 (a) [306]. More conveniently (b), according to Evans et al. [274, 3461 the ketones may be reacted with dialkylboron trifluorosulfonates (triflates) in the presence of a hindered tert.-amine to give, in high purity, the (Z)-enolates from ketones and (E)-enolates from thio-

5.3.2 Generation of Enolates and their Addition to Achiral Aldehydes and Ketones

195

esters." The triflates can be replaced, without loss of selectivity, by PhBCI, [347]. The transmission of enolate configuration to aldolate configuration depends somewhat on the constitution of the partner but is in general very efficient (070 syn > 95). Boronates formed from phenylthiopropionate with the bulky 9-borabicyclononyl (9-BBN) group have a Z configuration and give pure syn-aldols [331]. Hoffmann and Ditrich prepared pure ( E ) - and (2)-enolborates, while Gennari et al. obtained mixtures of them. In an uncatalyzed addition both gave 81 -99% of the syn-aldols. Enolsilanes can be prepared by trapping lithium enolates with trimethylsilyl chloride [e.g. 338, 340, 3481 (Fig. 5-27 (c)). Enolsilanes only react in the presence of a catalyst. In example (c) the pure (Z)-enolsilane shows poor anti selectivity under TiCl, catalysis, whereas the E isomer (85% pure) exhibits medium to high preference for the anti-aldol. In the presence of BnMe3N+Fpit is possible to obtain, if R ' = tBu, the syn-aldol under kinetic and the anti-aldol under thermodynamic control [275, 3131. When the catalyst was (Et3N),)S@SiMe3F9,up to 95% of the syn products were formed under kinetic control, regardless of enolate configuration [349].

RiBOTf

R'=Et, iPr, IBU,tBu, Ph, StBu, SPh R*=Bu, cPent, 9-BBN

iPr,ErN

(c)

OSiMe3

p\a,

MeISiO I) L D A 2) McSiCl

MeCHzCOzEt

I ) LDA+HMPA 2) Me.SiCI

Me 85% E

loo%,

Cp2ZrC12

+

z

OZrClCpz M e d M e

92%Z

Fig. 5-27. Generation and aldol reaction of enolates (M = B, Si, Zr).

* In the latter, since S>O, the disposition of oxygen and R2 is gives, as expected, the anti-aldols in excess.

trans in the classical sense and

196

5.3 Stereoselective Aldol Reactions

When mediated by trimethylsilyl triflate, enolsilanes react with aldehyde dimethylacetals and give, if under kinetic control, syn-aldols in excess both from (2)-and (E)-silanes [250, 25 I]: In recent years silylated ester- and thiolester enolates, also called silylketene acetals and usually prepared from lithium enolates have gained importance for the introduction of CH(R2)C02R, groups. OH

OSMe,

MeC'COstwl

1) LDq THF, -78 2) Me3SICI

OC

&

Me

RCHO BF, EQO

Me

R I P r twI Pent Ph %de182 78 83 91

Enolate geometry may be inverted from 2 to E by adding HMPA [352] or DMPU [353] but since addition requires Lewis acid catalysis, product configuration is independent of enolate geometry. With acetals prepared from esters use of TiCI, + Ph,P as catalyst improves anti selectivity [344]. In the following example addition is coupled with reduction of a lactone [355]: R

snl,

Addition of silyketene acetals to N-alkyl [356] or N-tosyl azomethines [357] provides anti-a-alkyl-p-aminoesters (de 50- 86%). Enolstannanes can be prepared by reacting the lithium enolates with a chlorostannane and may be used in situ [316] or after isolation [317]. The reacting species is uncertain since 0-and C-stannanes equilibrate in solution. Selectivity appears to depend on the ligands attached to tin. According to Yamamoto et al., at low temperatures with the cyclohexanone enolate the syn product was slightly preferred ( - 70%) in the case of the triphenylstannyl derivative [316]. On the other hand, Stille reported [318] that the trialkyl analogeus yielded up to 92% of the anti-aldol at low temperatures and about 70% of the syn-aldol at 45 "C. Tin(I1) enolates, which can be prepared in situ by treating an a-bromocarbonyl compound with tin [320] or more conveniently by reaction of a ketone with Sn(OTf), [321], were introduced by Mukaiyama and Stevens. Syn selectivities in excess of 90% were achieved with ketones [320, 321, 3261, while selectivity was modest with esters [321, 3261. As mentioned in Section 5.3.1, the method is amenable to chiral catalysis [358, 3591.

5.3.3 Addition of Enolates to Ketones (the Cross-Aldol Reaction)

197

Titanium enolates of ketones (of which the triisopropoxy derivatives are of practical interest) are prepared by reacting the lithium enolates with (iPrO),TiCI [303J. Titanate composition reflects that of the parent enolate, but this is irrelevant, since the syn-aldols are the preferred products from both (E)-and (2)-enolates. Addition is under kinetic control and probably follows a non-cyclic mechanism. Treatment of enolsilanes with TiC14 affords trichlorotitanium enolates which react with aldehydes with poor to modest syn preference (inclusive cyclohexanone enolates). A boat transition state was proposed by Nakamura and Kuwajima to account for the results [304]. The last metal to be discussed here, but certainly not the last to be tested for the aldol reaction, is zirconium. Zirconium enolates, which are accessible from lithium enolates [302, 3301 (e.g. Fig. 5-27 (d)) react, irrespective of enolate configuration, with syn preference. Selectivity is modest with ketones [302] and good to excellent with amides [330]. With some chiral amide enolates ( c j p. 201) syn selectivity is excellent [360].

5.3.3 Addition of Enolates to Ketones (the Cross-Aldol Reaction) It follows from the cyclic stereochemical model for the aldol reaction that stereoselectivity depends, among other factors, on the difference in size of the ligands attached to the carbonyl group. Therefore much better selectivity can be expected with aldehydes than with ketones. Data on ketones are scarce and appear to confirm this point. In the Reformatsky reaction selectivity increases with increasing size of R (Fig. 5-28(a)) [328, 358, 359, 3611 and of the esterifying alcohol [328]. Under the conditions quoted, the reaction is thermodynamically controlled, while in methylal at -70°C [328] it is under kinetic control and, as expected, selectivity improves with bulky esters. The trend was similar in the Reformatsky reaction of a series of esters (RCHBrC02Me, R = Me, Et, iPr) with a- and P-aminoketones; behavior of the tertbutyl derivative was anomalous [362]. Promising results were reported by Mukaiyama et al. with tin(I1) enolates (e.g. (Fig. 5-28 (b)) [263, 2641. In Fig. 5-28 (c) an example for stereoselective addition to a chiral ketone is shown [365].

198

5.3 Stereoselective Aldol Reactions Me OH

+ A

AMe

Ph

R

Zn. C&

C02Me

>

rcflux

Et NT-\ I) Sn(0Tf)z.

EtCHzCOR

-u, >

Me OH 0

OcC

2) PhCOMe

Phq

R R kt

(4

lMe

(-) - MentOZC

+

\

/

-hOGC, TiClr93./b

%anti

!

Et

Ph

I

13

100

HO Me

x

(--)-MentOK I-Naph (-), 92% ee

Fig. 5-28. The cross-aldol reaction.

5.3.4 Addition of Achiral Enolates to Achiral Aldehydes Under Chiral Catalysis As with ally1 addition, proline-derived diamines developed mainly by Mukaiyama et al. [366] proved to be efficient chiral catalysts first of all in aldol reactions involving silyl ketene acetals. Useful catalysts of the sulfonamide type, e.g. 42, 43, and 44 have been described by Masamune et al. [367] and Corey et al. [368, 3691. Syn and enantioselectivity is usually high (see Table 5-5). The reaction was successfully extended to additions to a-ketoesters [370]. Configuration at the new chiral center (C-3) is controlled by the configuration of the catalyst and not the substrate in the following reaction [383]:

5.3.4 Addition of Achirul Enolates to Archirul Aldehydes Under Chirul Catalysis

199

OH

OSiMc,t&

Table 5-5. Addition of silylketene acetals to aldehydesd. OSiMe3 R3CH0 + Sn(OTD2

R' +R2

cat*

R'

R'

R2

R3

cat. *

H Me

SEt

Ph cHex PhC-Ce Et cHexg Eth iPr Prk

37

OBn OSiMe,tBu H

OBn StBu

R3 &COR2

41 41 41 41

38 38 42j

070

de

-

> 98 80 92 82 84 -

'70

ee"

82 92 96 67 96 82 > 98 92

Ref. 13721 [3811 13781 [3791 D741 13821 13731 13671

Of syn when R'f H. With R3 = iPr, tBu, BnCH, ee 95, >95 and 90% resp. For R3 = Hept and Ph de 100 and 86%, ee for syn >98 and 91% resp. Selectivity improved by using BuSn(OAc), instead of Bu3SnF [377]. Reactions were usually conducted at -78°C in CH,CI, in the presence of Bu,SnF or Bu,Sn(OAc), as promoters. For MeC-C, BuC-C and MeSi,C=C de 72, 86 and 90%, ee for syn 88, 91, and 91% resp. With SnO and Me,SiOTf as additive. For R3 = iBu, Hex, cHex and Ph, de 96, 96, 82, and 88%, ee of syn 94, 85, 68 and 91% resp. For R3 = Et, Ph, and PhCH = CH- de and ee for syn > 96%. For crotyl, Ph, Ph-CH=CH- de 94, 76, and 80%, ee of syn 94, 90 and 94% resp. ' At -95 "C in CH,CI,-mesitylene. For R3 = crotyl, cHex, and Ph ee 91, 97, and 89% resp. SiMe,tBu instead of SiMe,. Used as the BH, complex in THF. For R3 = cHex, Ph, BnCH, ee 81, 87, and 91% resp. a

J

200

5.3 Stereoselective Aldol Reactions

371, 372 373, 374

NR2R3

382 370, 376 Me

Me

hmosphAphH0 Pr

CF,S02N,

C0,H

,NSO,CF,

\B/

Al

H

IBU

iPr

Fig. 5-29. Chiral catalysts used in aldol reactions.

Efforts by Koga et al. to carry out enantioselective aldol reactions with the aid of chiral lithium amide bases have met with partial success [384, 3851, e.g.: Ph

Ll

yaMe NcFkx

1) TMEDA PhCOMe 2) BuLi 0 O C - -70 O C *

RCHO,

u

Ph

R %ee

I

1

t h c€kx Ph 1-Naph 74 65 82 86

5.3.5 Addition of Chiral Enolates to Achiral Aldehydes The purpose of adding a chiral enolate to an achiral carbonyl compound is usually the transfer of chirality to the product, and at the end of the process the original chiral moiety is removed. Chirality can be imparted to an enolate in several ways: (i) the ester of a chiral alcohol or (ii) the amide of a chiral amine may be used, (iii) chirality is associated with the metal component of the enolate or (iv) there is a chiral center adjacent or near the carbonyl group. As early as in 1949 in Reid and Turner described the enantioselective Reformatsky reaction of (-)-menthy1 bromoacetate with acetophenone (ee 30% ( S ) ] [377]. Despite repeated efforts useful level of enantioselection could only be realized recently e.g. with titanium enolates [387, 3881:

5.3.5 Addition of Chirul Enolates to Achirul Aldehydes

cr

Si' V Me2

201

OH 1) LDA - 78 O C , 2) CpTi(OR*),CI

C0,E

C0,Ei

3 ) RCHO, 4) YO

R//\/

rmr,

R = Me, Pr, t k , vinyl, Ph; de and ee 296 % R* = diacetoneglucose

Magnesium [389, 3901 and zirconium enolates for monoacetates and propionates of 1,1,2-triphenylethanol (prepared from mandelic acid) undergo highly selective (with the propionate also anti-selective) aldol reactions.

M

ZrCpzCl ZrCp&3

BBu2 BBuz

R, % ee of syn

a (40)

96 99

96 98

c (42)

>99 >99

>99 199

b (41)

d (43)

R %ee

I

I

conf.

98

97 >9¶ >99.

Me

iPr

iBu

tBu

299

95

98

90

2R 2.5 2R 2S

Fig. 5-30. Aldol reaction of chiral enolates with achiral carbonyl compounds.

The use of chiral amides was pioneered by Evans and his coworkers who prepared 45, 46 [360], and 47 from amino acids, further 48 from (IR,2S)-norephedrine [382]. Later, several other amide forming chiral auxiliaries were introduced by him and oth-

202

5.3 Stereoselective Aldol Reactions

ers. Note that in all of them the NH group is part of a five-membered ring containing a C=O, C=Si, or SO2 function too. The zirconium enolates of 45 and 46 as well as the boron enolates of 47 and 48 added to achiral aldehydes not only with excellent syn selectivity ( 296%) but also with high enantioselectivity (Fig. 5-30(a)). 43 with Ti(OiPr)3 as M gave similar results [392]. When the enolate 42 was formed using a slight excess of Bu20Tf and iPr2NEt the 2S product was formed with 3-naphthylthio-acroleine (de > 96%) [393]. With the chloroacetyl analogues of 42 outcome of the addition to benzaldehyde depended on the choice of the metal. With Li, Sn(1V) and Zn the (2S)-anti isomer was obtained in low dia- and high enantioselectivity, with Sn(I1) and Boron the (2S)-syn diastereomer was obtained in high purity [394, 3951. The anti product was assumed to arise from an unchelated boat-shaped transition state. Juglery with catalysts permitted Heathcock et al. to prepare both the syn- and anti-aldols from the same substrate [396]. When TiC1, was added at -78°C to the boronate 47 first and then the aldehyde, the syn aldols were obtained in 74-88% de. Adding, in turn, first Et2AlCl to the aldehyde and later the boronate, the antialdols could be isolated in 72-90% de. Dithio analogues of 47 (M = SnOTf) are useful in enantioselective aldol reactions involving a-alkyl-P-hydroxyacids (e.g. [397 - 4031). Of special interest are, however, aldol reactions involving amino acid-derived B-a-haloacyl- and N-a-thiocyanato-oxazolidinones leading to (2S)-syn- [404] and (2S)-anti-a-amino-fl-hydroxyacids [405]. Amides of the versatile sultam introduced by Oppolzer offer not only high primary selectivity, but also the advantage that purity of diastereomeric intermediates can be upgraded by crystallization [406-4081.

1) Et,BOTf, EtNlprz (2 eq.)

R'

CHZCI, - 5 OC; 2) R'CHO

R1= Me, Et, Bu R*

~2

R'

Rz = Me, Et, @r, Ph de 94%

Note that using the tributyltin enolate the enantiomeric products were obtained in 58-70'70 de. The use of chiral enolates of the type C = C-OR* is a relatively recent development. Compounds 49 [409, 4101, 50 [411, 4121, 51 [413-4161 and (-)-IPC2BOTf [417] were used to form boron enolates of ketones, esters and thioesters, while esters of N-methylephedrine [418-4211 and esters of the camphor derivatives 54 and 55 [422] were converted to silylketene acetals. Aldol reaction of esters of 52 and 53 yield epimeric P-hydroxyesters.

5.3.5 Addition of Chiral Enolates to Achiral Aldehydes

203

2,6-Dimethylphenyl propionate was converted to an enol titanate with (R *O),TiCpCl (R* = diacetoneglucose) which gave with achiral aldehydes the ( 2 R ) syn product in >91% ee [423]. In a process invented by SolladiC et al. [424, 4251 and improved later by others [426] optically active /I-hydroxy acids were obtained with the aid of magnesium enolates derived from chiral a-sulfinyl N,N-dimethylacetamides (Fig. 5-30 (b)). The chiral inducing group could be readily removed by amalgamated sodium. The use of chiral ketone enolates is only expedient when either the chiral center is destined to become an integral part of the target molecule or when it can be easily removed at a later stage. After exploratory studies by Seebach et al. [427] and by Evans and Taber [477], a series of chiral a-alkoxy ketones 54 [428, 4291 available as both enantiomers, further 55, 56 [283], 57 [430, 4311 and 59 [350] were developed in the laboratories of Masamune and Heathcock respectively (Fig. 5-3 1 (a)). After desilylation the aldols can be converted to a-methyl-P-hydroxycarboxylic acids by periodate cleavage of the ketol function. 54 was used as the boron enolate; asymmetric induction (diastereoface selection in respect of the enolate) increased in the series 9-BBN < BuzB < cPent2B [328] the (iPr0)3Ti enolate gave also the syn,syn-ketol in >96% de [432]. 55-57 were reacted as the lithium enolates and showed less diastereoface selection than the boronates. Syn selectivity at the ends of the new bond was complete, while the degree of 1,3-induction is shown in Fig. 5-3 1 (a). The methylketone analogue of 57 was totally unselective in this respect [430]. For the products arising from the lithium enolates 1,3-relative configurations have not been determined, but are assumed to be syn by analogy with X-ray results for a product obtained with a chiral aldehyde [433]. Ketones with a chiral iron center, introduced by Davies and his coworkers [434], proved their versatility in aldol addition also [435]. The configuration of the product was inverted when, instead of the aluminium enolate, the tin(I1) enolate was used [436]. Removal of the inducing group was rather cumbersome. Lithium and boron enolates differ not only in the magnitude but also in the mode of selectivity. Since both yield predominantly (Z)-enolates [437], this result has been interpreted by Heathcock by assuming complexation of the silyloxy oxygen by lithium as opposed to boron, which only complexes with the carbonyl group (Fig. 5-3 1(b)).

204

5.3 Stereoselective Aldol Reactions

( e OM

Me+

/”’

Si

R‘

ox

xe

R’

+ R’ f

Me

OX

Me

OX

Si

( 5 4 ) - (59)

OH I ) HF. MeCN 2 ) NalO,

Enolate

l x

Me R’

R2

%ee,

Et

SitBuMe,

H

cHex

SiMe,

Me

SiMe,

H H

Ph cHex tBu tBu iPr

Bu,B cPent2B Li

96

> 98

iPr >98

50 Bu,B

80 69

conf.

R3= Ph 91 95 56 50 86 34

Bn 3R 3R 62 92 >74

3s *

3 s* 3s* 3s

3s *

* Probable but not proved.

Fig. 5-31. Aldol reaction involving chiral enolates (I).

When the ( C H ~ X ) ~ was B C ~used for preparing the enolates, bulkyness of the reagent enforced an (E)-enolate which gave necessarily anti configuration at the new bond [438]:

R = Pr, iPr, Me, Mew

,A

de193 %

205

5.3.6Addition of Achiral Enolates to Chiral Aldehydes (a1 0

OH I ) LDA, 2 ) MX 3 ) RC'HO, -78°C

Me

CO

*

0

R

PPhj

MX = Et,AICl

MX = SnCl,

( A1

(RIL)

= Fe*COMe

OH

( B1 MX=Et*AICI

I

Me

R

Me

8,

R

R& OH

CO$

ann

YOee of anti

1

Et

iPr

tBu

cHex

92 77

91

94 79

9s

8S

84

Fig. 5-32. Aldol reaction involving chiral enolates (11).

Enantiomerically pure (I? )-3-hydroxy-2-methylpropionicacid was prepared by reacting the dicyclopentylboronate of (S)-54 with formaldehyde gas followed by oxidative cleavage 14291. Chiral boron enolates [439, 4401 provide good syn or anti selectivity, but enantioface selection is unsatisfactory (I 85%) (Fig. 5-32 (b)).

5.3.6 Addition of Achiral Enolates to Chiral Aldehydes When an enolate is added to a carbonyl compound with diastereotopic faces, usually to a chiral aldehyde, cases in which one or two new stereogenic centers are created should be distinguished (e.g. Fig. 5-33 (a) and (b)). In the first case the enolate is unsubstituted at the a-position and therefore only two stereoisomers are conceivable, and their ratio is dominated by the rules governing nucleophilic addition to chiral

206

5.3 Stereoselective Aldol Reactions

91 : 9

J1.n

OH

(c)

‘,y”” Me

OSithMe,

+

A

R

BF3 -78OC

atiti

0

~

Me

Fig. 5-33. Aldol reaction of achiral enolates with chiral carbonyl compounds (I).

carbonyl compounds in general (c$ Section 3.2.1). In the second case the formation of four stereoisomers is possible, only two of which were observed in our example. Here the high stereodirective power of the aldol reaction was combined with the lower efficiency of asymmetric induction. When there is no polar group in the aldehyde, Cram’s rule is valid and consequently the syn product dominates. The reaction was optimized for 2-phenylpropanal by Flippin and Heathcock, and dimethyl-tert-butylsilyl enolates proved to be the most efficient (Fig. 5-33 (c)) [250]. Unfortunately, under identical conditions, addition to 2-benzylpropanal was hardly selective, but this is exactly what can be expected from stereochemical models. Addition of a-substituted enolates to chiral aldehydes with no polar group follows the pattern of example (b) in Fig. 5-33. Namely, at the new bond it is strictly syn from ( Z ) - and anti from (E)-enolates, and the configuration of C-3 and C-4 can usually be predicted by Cram’s rule [270, 275, 294, 3251 (c$ Fig. 5-33(b)), although some inexplicable exceptions are known [441]. Interest in polyoxygenated aliphatics prompted several studies in chiral alkoxyaldehydes [269]. Selectivity in additions to a-alkoxyaldehydes (1,2-induction) is rather unpredictable and depends on M and the structure of both substrates. Enolates derived from acetates show no selectivity [250, 4331, while ketone enolates may be

5.3.6 Addition of Achirul Enolufes to Chirul Aldehydes

syn,yyyn

unri.nnri

M=SiMe, (+TiCI,) M=SiMe, (+Bu4N+F-) M=Ti (OiPr),

OH

97% syn, syn 82% syn, anti

87% anti, and

0 Ph

R = tBu, Ph de>98%

an h

OSMe3

\

AF% TiCI,,

- 78 OC

P

h

r:: TEI, w

R JOMc

+

92%~3"1,syn

Ph

O h

(d)

G

- 78 O C

Me

RP

F

'

h de cis > 92%

R = 2-. 3-, 4-Me, 4 - t h Fig. 5-34. Aldol reaction of achiral enolates with chiral carbonyl compounds (11).

207

208

5.3 Stereoselective Aldol Reactions

highly selective, and the configuration in excess depends on M and the catalyst [442] (Fig. 5-34) [75]. As another example, addition of methyl tert-butyl ketone as the lithium enolate to 2,3-isopropylidene-glyceraldehydeis 97% anti selective [433], whereas that of the trimethylsyl enolate under TiC1, catalysis to 2-benzyloxypropanal gives > 95% of the syn product [234]. Addition of various lithium enolates to both of the above mentioned aldehydes has been thoroughly investigated by Heathcock and his coworkers [270, 4331. Efficient chelation control under Lewis acid catalysis leads to anti addition of a-unsubstituted enolsilanes to a-alkoxyaldehydes (Fig. 5-34(b)) [93, 224, 4431. Use of a silylenolate of an a-thiomethylketone in a catalyzed reaction leads, after removal of the thiomethyl group, to a 3,4-syn product (Fig. 5-34(b)) [446]. Phenolates play the role of enolates in the following reaction:

R = H, 2-tBu, 3-OMe, 4-0me

With Ti(NiPr2)3 as the metal component the anti product is obtained in 2 7 6 % de [444]. Reetz et al. studied by NMR the TiCl,-catalyzed version of the reaction (R = tBu) leading predominantly to a syn product and found no interaction between OSiMe3 and C = 0 groups [445]. Outcome of the addition of silylketene acetals to a-thioalkylaldehydes depends on whether the Lewis acid catalyst is chelating (TiC14) or not (BF,*Et,O). SMe

Me

OSiMe,

R+cost,.-

fLE%

Me

OH de 100 %

a SMe

BF,'EtZO,

R

Me

COStA

:

OH de 82 %

Ester enolate addition to a ketone attached to a chiral auxiliary group permitted the enantioselective synthesis of a-hydroxyaldehyde esters (Fig. 5-34 (c)) [447]. High equatorial selectivity was observed in the addition of a silyl enolate to cyclohexanone acetals (Fig. 5-34(d)) [448].

5.3.7 Stereoselective AIdol Condensations

209

5.3.7 Stereoselective Aldol Condensations * An aldol reaction can be carried out not only by transforming the nucleophilic partner to an isolable enolate but also in the presence of an amine salt, when the product is usually dehydrated to an a,punsaturated ketone. By taking a chiral amino acid as catalyst the triketones 60 could be enantioselectively transformed via the aldol 61 to the enones 62 (4491. The method was improved later by several groups [450-4541 and became an important step in the enantioselective synthesis of steroids (Fig. 5-3 1 (a)). Other amines were less effective, except (S)-phenylalanine, which was occasionally better than (S)-proline 1455, 4561. Interestingly, with (S)-homoproline (63) as catalyst (R)-57 (n = 1) was obtained in 65% ee [457]. With non-cyclic analogues both yields and enantioselectivities were poor [458, 4591. Two pathways can be envisaged for the above condensations. One involves the selective formation of an aminal (64) with one of the enantiotopic carbonyl groups.

n

0

(64) Fig. 5-35. Enantioselective aldol condensation.

* The aldol reaction, in fact an addition, is often misnamed aldol condensation (e.g. ref.

[ 1561).

210

5.4 Catalytic Addition

of Isonitriles to Aldehydes

The other pathway goes through an enamine in which the carbonyl groups become diastereotopic ( 6 5 ) (Fig. 5-35(b)). A careful study by Agami et al. gave support to the enamine pathway [460, 4611.

5.4 Catalytic Addition of Isonitriles to Aldehydes The reaction to be reviewed here and studied intensively by Hayashi et al. is usually called an aldol reaction, although its mechanism is basically different from the ones discussed so far. It typically involves an isocyanoacetate ester or amide and is catalyzed by a gold or occassionally silver complex of a chiral bis(phosphino)ferrocenyldiamine (66- 6 8 ) [462 -4641. With the known catalysts trans-(4R, 5R)disubstituted oxazolines are formed predominantly, while enantiomeric purity of the minor cis isomer is poor.

(67)

R2= -(&J5- [458]

(68) R2= -(CHJ20(CHJz-

[457]

Activation of the CH2 group by an additional electronegative group is required. Essential results are summarized in Table 5-6. The oxazolines are usually hydrolyzed to syn-P-hydroxy-D-a-aminoacids [e.g. 4641. Selectivity with a-ketoesters is low and capricious [465]. Reaction of a-substituted isocyanoacetates with formaldehyde gives, after hydrolysis, (S)-a-alkylserines [466]. Ito formulated the transition state complex as 69 based on 3'P-NMR studies on the silver complex [467]. Kumada et al. demonstrated that configuration at the side chain of the catalyst has a minor role in controlling stereoselectivity [468].

21 1

5.4 Catalytic Addition of Isonitriles to Aldehydes

Table 5-6. Diastereomeric and enantiomeric excess values (in parentheses) for the (4R,SR)-trans products in the catalyzed addition of isonitriles to aldehydes.

R'

R2

cat.*

R3 Me

C0,Me

H

iPr

66 68

C0,Et

66

C0,Me

66

C0,Me

Me

68

P(O)(OPh),

H

67

CONMe,

67

C0,tBu

66

S0,pTol

67

a

Ref.

78 (89) 96 (92)

82 (98)

100 (77)

> 94 (> 95)

tBu

=C(Me)-

cHex

Ph 92 (80) 90 (95) 78 (93) 80 (94) 86 (941 > 94 ( > 95) 88 (94) 84 (93) 100 (83)

With AgCIO, as metal source. With R3= Et de 8%, ee 87%; with R3 = iPr the cis isomer is in excess, de 52%, ee 51%. With AgOTf as metal source.

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

6 Stereoselective Carbon-Carbon Bond Forming Reactions 6.1 Carbon-Carbon Bond Formation Involving Olefins 6.1.1 Addition of Carbanions to Activated Olefins This section deals with reactions covered by the following scheme:

The carbanion may be generated from a C-acidic compound or provided by an organometallic compound, while Y is an electron attracting group such as C = 0, C=N, S-0, NO2 etc.

6.1.1.1 The Michael Addition In the Michael addition a C-acidic compound adds to an activated double bond in the presence of a base. By the introduction of strong lithium bases, first of all of LDA, to Michael-addition chemistry its scope has been much extended and involves now ketone, ester and thioester enolates, as well as some enamines and metallated hydrazones. For carbonyl compounds the reaction can be formulated in general as follows:

where X may be C, 0, N, or S, and Y as defined above. The enolates may be generated in situ (as is usually the case with M = Li), or prepared as isolable intermediates (e.g. M=SiR3). The choice of acceptors has been rather restricted, most often

2 14

6.1 Carbon-Carbon Bond Formation Involving Olefins

enones were applied. 1,4-Addition is often accompanied by simple 1,2-addition to the carbonyl group. Early examples of Michael additions under chiral catalysis are the addition of C-acidic compounds to a ,punsaturated ketones catalyzed by quinine type alkaloids, which gave up to 76% ee [1,2]. A method using K + complex of a chiral crown ether providing very high enantioselectivity was of limited scope [3]. Chiral catalysis of quite different kind is present in the following addition [4]:

In the addition of donor compounds unsubstituted or symmetrically substituted at the terminal carbon atom only one new chiral center is generated and one way to induce selectivity is the combination of the donor with a removable chiral auxiliary as in the examples in Fig. 6-1 (a),(b), and (c) [ 5 -71. The removable chiral auxiliary may be attached to the acceptor too as in Fig. 6-1 (d) [8]. When both compounds are achiral but substituted at their bond forming terminals, syn and anti diastereomers arise and this may be associated with enantioselectivity. In an early intramolecular example for this situation Stork et al. demonstrated high diastereoselectivity of addition (Fig. 6-2(a)) [9]. Ketone enolates, either as the lithium enolates proper [lo], or after trapping by silylation [ l l - 131, add to enones or to 2-vinyl-1,3-dithiolanium salts [14] with anti selectivity (e.g. Fig. 6-2(b)) [15]. The percentage of anti products corresponds to that of the (Z)-enolate. (E)-enolates add with syn selectivity but are difficult to prepare. The lithium enolate may be transformed with Ti(OiPr)4 to an ate complex giving with a wide variety of enones anti products in 82-9470 de [16]. Addition of the enolsilanes requires Lewis acid catalysis e.g. (TiCl,, Ph,C+ClO,). Addition of ketone and ester enolsilanes to a-chiral enones proceeds in an anti sense and configuration of the resulting 175-diketonecan be changed by interchanging R' and R2 (Fig. 6-2(c)) [12]. In the addition of lithium enolates prepared from esters [17], and silylenolates derived from thiolesters [ 181 and dithioesters [ 191 enolate configuration ( E or Z)* controls product configuration. With lithium enolates of esters the (E)-enolates give with enones mainly the syn adducts, while the (Z)-enolates the anti products (Fig. 6-2(d)), [17,20]. The same ap-

* In some papers the sequence rule is disregarded

by ranking OLi>OC.

6.1.1 Addition of Curbanions to Acrivared Olefins

R Me Me Me X CO&* CN C0,Me 96 98 %dc >98

21 5

iPr

(CHJ2C02Me CN >!I9 >99

CN

plies t o silylated thioenolates when the acceptor is a cyclic enone, but with acyclic ones anti products are obtained from (E)-enolates [18]. Addition of a glycine ester enolate to acrylates was completely syn selective [21]:

Bn2N -C02tEk

1) LDA, THF, - 78 OC 2) R

+mMe

R = Me, Bll, Fkpt, de - 100%

Addition of lithium enolates of chiral propionamides to ethyl crotonate seems to be controlled by the amine components [22]:

2 16

6.1 Carbon-Carbon Bond Forrnafion Involving Olejins

R1 IEt iPr Ph Ph t h Ph Rz IPh Ph Me Me Ph

(a

Etw*R*

Ph

R,yoLi R2

+

THF, HMPA

E t O z C , & X & t

-18QC

OEt

R'

> 95% anti R'=Me, Et, Oct, R2=Me, Bu, Ph, C02Et

Fig. 6-2. Stereoselective Michael additions (11).

21 7

6.1.1 Addition of Carbanions to Activated Olefins

N*OCmMe

I

1) LDq THF, - 78 o c

Me de 82% ee 74%

N*OC Y C 0 , E t

de 87% ee >go%

N*OC a C O $ . i

Heathcock et al. thoroughly studied the addition of amide and thioamide lithium enolates to enones [23,24]. The choice of substituents in both partners providing satisfactory regio- (1,4) and diastereoselectivity (syn) is rather limited. Amide enolate formation is coupled with allylic rearrangement in the addition in Fig. 6-3 (a) [25].

R1 = y Me, Rz = Me, WI, Fkpt, Ph, dc >90%

Nd N L O M e 1) LDA, THF + TMEDq 0 OC 2) R 3 / n / ‘qMe , - 100 OC, 3)

o3

*

RI x

C

0

2

M

c

R1 = y Et, Ph, R2= Me, Et, Bn, R3 = Me. Et, Ph

NU:

Fig. 6-3. Stereoselective Michael additions (111).

218

6.1 Carbon-Carbon Bond Fornzniion Involving Olefins

Versatile SAMP, introduced by Enders, proved its usefulness also in Michael addition [26]. The metallated aldehyde hydrazones (R2 = H) add with almost total enantioselectivity (ee > 96%) onto 3-substituted acrylates. The corresponding ketone derivatives give rise to anti-5ketoesters of high ( > 90%) dia- and enantioselectivity (Fig. 6-3 (b)). Bernardi et al. applied molecular mechanics and ab initio calculations to predict stereoselectivity in the Michael addition to y-chiral acceptors [27]. In a conformation with the small (S) group “outside” ( I ) on approach of the nucleophile according to the Felkin model, anti selectivity is predicted for (E)-enolates and a lower syn selectivity for (2)-enolates (Fig. 6-3 (c)). In practice syn selectivity is much dependent on the constitution for the nucleophile.

6.1 .I .2 Addition of Metalorganics to a,PUnsaturated Ketones and Esters The scope of nucleophilic addition to activated double bonds has been much extended by the use of organometallic reagents. Enantioselective addition in the presence of chiral complexants lacking an exchangeable hydrogen has not been successful (e.g. [28]). Reagents obtained by mixing dialkylzinc [29] or lithium alkyl cuprates with a chiral aminoalcohol [30, 311 or sec-amine [32- 341 add with good to excellent enantioselectivity to enones (Fig. 6-4(a)) [3I]. A lithiated chiral azomethine efficiently mediates /?-addition of dimethylzinc to cyclohexenone and heptenone (ee 88%) [34]. Addition of lithiumalkyls to chiral substrates concerns either compounds with removable chiral auxiliary groups or those in which the original chiral center remains in the end product. As chiral auxiliaries a very large number of ester, amide, acetal and oxazolidine-forming compounds, as well as the sulfoxide group has been tested. Among them, chiral esters and sulfoxides were studied most thoroughly. 8-Phenylmenthol and other terpenic alcohols developed for enantioselective DielsAlder synthesis by Oppolzer and his coworkers proved to be potent stereodirecting groups also in conjugate addition, as shown in Fig. 6-4(b) [35,36]. Amides of Oppolzer’s sultam offer not only high inducing power but also the advantage, that the crude diastereomeric mixture can be readily purified by crystallization. As nucleophiles Grignard reagents [37, 381 or lithium alkyl cuprates [39] may be used, as e.g. in Fig. 6-4(c) [40]. In the reaction in Fig. 6-5 (a) the chiral auxiliary is eliminated by quenching the reaction with acid [41]. a,PUnsaturated esters of the alcohols 4 [42], 3 and 6 [43] also undergo highly enantioselective additions with RCu * BF, complexes giving products of opposite configuration. Attack of a cyclohexyl radical generated photochemically is efficiently directed to the si face in the amide 7 [44] and its analogues (Fig. 6-5(b)) [45].

2 19

6.1.1 Addition of Curbunions to Activuted Olefins

(R)-musconc. cc 96%

(b)

C02R* BU-

I ) MeCu. BF,

+

2 ) -R*-OH

Bu~P

>

Bu

( s ) R*=(-)-8-phenylmenthyI,

86 5%ee

R'=2,94% ee

Me

,SO,Ph

&$C6&-3,5-Me SO,N(c&x), (3)

Me

Me

Me 1

RZzCuLi . PBu3 PhMe, - 80 - - 40 O C

L

1)H'

~

2) LOH

R2

R'

(Et Bu Ph Me

RZ %de

1100 87 92 97

Fig. 6-4. Stereoselective addition of metalorganics to activated double bonds (I).

220

6.1 Carbon-Carbon Bond Formation Involving Olefins

Ephedrine can also play the dual role of masking an aldehyde as an oxazolidine and directing the attack of an alkyl metal (Fig. 6-5(c)) [46,47].

1) R2CuL Li&,THF - 90 OC; 2) H Q

R

'5 +

Me0,C

0

'

Me

R = Me, Et, h, Ph, vinyl n = 1, ee 47-85% n = 2, 3 ee >95%

CFkX

-

cHexCOONT

C3C12, hv' h3snH - 78 O C

*

(Me02C),CHnCOR*

S

de >99 %

Fig. 6-5. Stereoselective addition of metalorganics to activated double bonds (11).

The strategy of chirality deposition (see p. 235) has been successfully applied to alkyl metal addition too, as in the following palladium-catalyzed arylations of aminal 8 obtained from (S)-asparagine [48]:

Mukaiyama and his coworkers found that Grignard reagents add very selectively to ester-amides of ephedrine with alkylidene malonic acids giving, after hydrolysis and decarboxylation, 3,3-disubstituted propionic acids in 58 -99% ee [49,50]. Product configuration can be controlled by substrate configuration (E or Z ) . Surprisingly, open-chain amides gave even higher stereoselectivities (Fig. 6-6 (a)) [5I].

221

6.1.1 Addition of Carbanions to Activcrted Olefins

6H

RI=RZ=Me, Et, Bu, Ph, %ee 79-99%

R R %ee

(c)

Me-

C0,B

SnBu,

+

co,B

1 I

Et

iPr

Bu

cPent

Bn

93

85

93

92

35

Me SnCI, C&CI2 -78 - -30 OC

*

/ A C W C 0 , W 2 Me

de 84 YO

RI = fl Me, R2= Me, €3, h, R3 = Me, h Fig. 6-6. Stereoselective addition of metalorganics to activated double bonds (111).

Asami and Mukaiyama utilized (S)-2-anilinomethyl-pyrrolidine aminals to prepare 3-alkylsuccinaldehyde acids of high optical purity (Fig. 6-6 (b)) [52,53]. Only one diastereomer of the aminal was obtained in the first step, and the chiral auxiliary could be recycled. Allylmetals (M = 9-BBN-, Ti(OiPr)3, SnBu3) may add to strongly activated double bonds with 1,3-rearrangement (Fig. 6-6(c)) [54]. Complete transfer of chirality in a 1,3-rearrangement accompanies the formation of a quaternary center in the following a-addition. Configuration at C-5 and of the double bond plays no role in the outcome of the reaction (Fig. 6-6(d)) [55,56]. Tartaric acid N,N-dimethylamide proved to be a potent inducer in the addition of trialkylaluminum to its acetals with a,punsaturated aldehydes [57] or cyclohex2-enone [58]. Regioselectivity was solvent dependent (Fig. 6-7 (a)). Diastereoselective additions where the inducing group (mostly in y-position, 1,2-induction) remained in the product were also intensively studied. Some results are compiled in Table 6-1.

222

6.1 Carbon-Carbon Bond Formation Involving Olefins

MroYoNM

(a )

I

4:YNMe2 CHCIi

"CONMe2

Ph-

Me,Al __

CoNMe2

(b) To1

0

0

-,!* I) RlM:,

DME, -78°C

2) A1IH.g

G)- ( 9 ) R

1) Znh,

2) @ M ~ B ~

%ee

1

R Me

Et

CH,tBu

97

88

91

*With CH,

= CH,CH,MgBr

Ph

loo*

>Y8

in the presense of ZnBr, [go].

Fig. 6-7. Stereoselective addition of metalorganics to activated double bonds (IV).

1,3 and more remote asymmetric induction is mainly operative in cyclic enones, an early example being provided by House et al. [59] (Fig. 6-8(a)). Trans addition seems to be a general feature for both cyclohexenones [60, 611 and cyclopentenones [62, 631 and is explained by Corey et al. by complexation of the cuprate to the enone system [61]. Nevertheless, no clearcut generalization can be drawn from the data presented, the more so that stereochemical preference is often inverted by change of reagent.

a

Me

tBuMezSiO NBn, OH OMOM CH20H

MeMgBr Me,CuLi + Me3SiCld Me,Li, - 7 8 "Ce MeLi, -78°C' MeLi

- 7 8 "C

CO,C(Me),O,C C0,Et SiMe, PhSO, PhSO, SiMe, SiMe, SOPh

+ CuI,

+ Me3SiC1, - 78 "C

Me,CuLi

Me,CuLi

C02Et

H

anti sYn anri SY n sYn

SY n

sYn

anti anti SY n anti SY n anti sYn anti

Conf.

80

76 74 54 - 82 42 90 7s 94 83

Vode

For R3M = BuCu.BF, de 90%. With R' = Ph and tBu, Y = CO,Et de 84- 100%. Similar selectivities were found for R3 = vinyl. Syn selectivity was poor with the Z isomer, and 88% for X = Y = C0,Et. De>98% also for MeMgBr, PrMgBr and Bu2CuLi. For R' = Me, Bu, iBu and R2 = Me, Bu, vinyl, Ph, de 84 - 90%. Followed by desilylation. De>96% for R' = Me, R2 = tBu and R' = iPr, R2 = Me, Bu, tBu, de 68Vo for R ' = Ph, R' = Bu.

Pr Me BnCH, Me Bu

Me.

-a2'

-myM 0 e

Me C0,Me

CO2Et

OMOM

tBuPh,SiO

CO2Et COMe

OBn

-

CN

CN

Me - OC(Me),O

BuCu. BF, Me,CuLi3. BF, R:CuLi, - 7 8 "C allylSnBu + TiCl, MeCu(CN)Li.BF3a'b iPrCu, Et,O, -78°C iPrLi, Et,O, - 78 "C Me,CuLi + Me,SiCI, - 78 "C

H

CO,Et

Me

R3M

Y

X

Ph

R'

Table 6-1. Diastereoselective P-Addition of Organometallics t o Activated Double Bonds.

Ref.

w

h,

p3

224

6.1 Carbon-Carbon Bond Formation Involving Olefins

Fig. 6-8. Stereoselective addition of metalorganics to activated double bonds (V).

Addition of alkylmetals to chiral aryl a,pethylenic sulfoxides was exploited by Posner et al. [77]. A second activating group, preferably a carbonyl, is necessary, and there should be no hydrogen in a-position. Diastereoselectivity was modest with an acyclic substrate (up to 65 Yo) [78, 791 but high with 2-arylsulfinyl cyclopentanones. Selectivity depends both on the alkyl group and ring size (less with the corresponding cyclohexenones) (Fig. 6-4 (b)) [SO]. Addition of alkyl lithium to the cyclohexenone analogue of 9 proceeded with 65 -96% ee in 2,5-dimethyl-tetrahydrofuran[Sl]. Starting from 3-substituted analogues of 9,construction of a quaternary carbon center became possible, the configuration of which could be controlled by the order in which the groups were introduced (Fig. 6-7(c)) [82]. Still and Galynker correlated diastereoselectivity in the addition of lithium dimethylcuprate to 8-, 9- and 1O-membered a,punsaturated ketones and lactones with the calculated conformational composition of the substrate [83]. Some examples are quoted in Fig. 6-8 (b).

6.1.1.3 Addition of Metalorganics to a,PUnsaturated Azomethines and 2-Vinyloxazolines Koga and his coworkers established that among several amino acid esters tested the tert.-butyl ester of tert.-butylglycine was the most efficient inducer in additions of acyclic [84- 861 and cyclic [87- 891 a,punsaturated aldehydes to azomethines derived from these amines (e.g. Fig. 6-9(a)). One of the several areas in which the chiral oxazolines pioneered by Meyers can be used successfully is conjugate addition of metalorganics. Provided that pure (E)-olefins are used (which were prepared by a special method [90]), enantiomeric purity of the end product (except for R' = Et, R2 = Bu, ee 61%) exceeds 91% [91]. By interchanging R' and R2, both enantiomers of the acids are accessible (Fig. 6-9(b)).

6.1.1 Addition of Carbanions to Activated Olefins (a)

R'

R' V

I) R*MgBr, THF, - 5 S T

2) H? 11-56%

tOztBu

C

H

225

O

R'

% ee

903H

R1 R2-

LI

0

Me

R Li

TMEDA ( R ' = Me)

OH

OH

89% u n t i

Fig. 6-9. Stereoselective addition of metalorganics to vinylazornethines and vinyloxazolidines.

The outcome of the reaction was rationalized by postulating double coordination with the reagent (10). This was supported by the fact that in the case of oxazoline I1 a product of opposite prevailing configuration was obtained with BuLi [90]. To the P-position of chiral a-oxazolinylnaphthalenes lithium alkyls add enantioselectively and the intermediate anion can be trapped by electrophiles [92]: tFh

1) Q THF, -78

2) Me1

- -40

R OC

226

6.1 Carbon-Carbon Bond Formation Involving Olefins

The same chiral masking group permits the preparation of chiral 3,3-dialkyl propionaldehydes in 2 94% ee [93]. Norephedrine [94] and ephedrine [95] are less efficient as chiral auxiliaries in analogous additions. Addition of metalorganics to the a-position of allylic alcohols, described by Felkin et al. [96], belongs only formally to Section 6.1.1. While allyl and benzylmagnesium bromides add with anti selectivity (alkyl Grignards do not react) [97], in the presence of a cation complexant alkyl lithiums show medium to high syn preference (Fig. 6-9(c)) [96].

6.1.2 Allylic Alkylation Allylic alkylation, introduced into the synthetic arsenal by Trost [98], can be summarized by the following scheme: *

Either stoichiometric palladium complexes or catalytic amounts of the complex can be used, X is usually OAc or OPh, and Nu an amine, an enamine or a soft carbon nucleophile. Since it was established that the nucleophile attacks from the side opposite to palladium, then, provided that complexation takes place selectively at one face, diastereoselective alkylation can be expected, and, if the phosphine is chiral, one enantiomer of the preferred diastereomer may prevail. This is exemplified in Fig. 6-10(a) [102]. With unsymmetrically substituted allyl systems the situation can be complicated by lack of regioselectivity. Bosnich and McKenzie studied allylic alkylation in the presence of (S,S)CHIRAPHOS (12) [103] and demonstrated by 31PNMR that there is an equilibrium between the diastereomeric allyl complexes. Enantioselectivities obtained with sodium dimethylmalonate as nucleophile were roughly in accordance with the observed equilibrium constants (Fig. 6-10(b), K = 6). A series of chiral diphosphine ligands prepared specially for allylic alkylation were tested by Hayashi et al. [104]. Reasonable levels of enantioselectivity could be realized with the ferrocene derivative 13 (Fig. 6-10(c)) [105], but not with DIOP [106]. Ring size and reaction conditions influence both direction and magnitude of diastereoselectivity in the following reaction, applicable also to acyclic analogues [1071:

* For reviews, see refs.

[99- 1011.

61.2 Allylic Alkylation

227

A variant of the reaction using Ni(I1) complexes of chiral diphosphines and Grignard reagents as nucleophiles gave enantioselectivities up to 60% [ 1081. Expen-

,

trans

PhwPh

(b)

\

Ph

AcO

or

Ph\/\yPh (Me 02C)zCH

,

Ph

84% ee

Ph-YPh Ph OAc

0

P h z P q P h 2

Me

Me

(12) (CHI RAPHOS)

(xb:-."^""H Me Me

I

PPhz

(131

Fig. 6-10.Stereoselectivity in allylic alkylations (I).

0

81% ee

228

6.1 Carbon-Carbon Bond Formation Involving Olefins

sive PdC12 may be replaced by Bu,N+.[Fe(CO),NO]- [I091 but the scope of this reaction is still unexplored. Stereoselective alkylation of an ally1 ether anion with a recoverable chiral auxiliary prepared from atrolactic acid giving 3-phenylalkanals was developed by Mukaiyama et al. (Fig. 6-1 1 (a)). (a 1

Me OMe

ph-

Ph

I ) KNiPrz, €1~0 -IOOGC , 2) RX

>

Me OMe H’B

Ph

Ph

Ph R %ee

I Me I

85

Et

iPr

Bn

84

87

75

& 55

R

(C)

II 0

RX = Mel, EtI, WW, de >99%

Fig. 6-11. Stereoselectivity in allylic alkylations (11).

Takahashi et al. observed 100% transfer of chirality in intramolecular allylic alkylation (Fig. 6-1 1(b)). The Z isomer gave a 3s product [110]. Finally an example for alkylation of a chiral (2)-crotonoyl enolate should be mentioned (Fig. 6-1 1 (c)) [I 1 I]. Note that the E isomer undergoes mainly P-alkylation. Alkylation of the (E)-crotonoyl ester of (I?)-BINOL also give 2s products (de 8O-86%) [112].

6.1.3 Stereoselective Hydrocarbonylation

229

6.1.3 Stereoselective Hydrocarbonylation * Hydrocarbonylation, a reaction used on a very large scale in the petrochemical industry, can be briefly described by the following schemes:

>-( ><

+ CO

+

H,

+

+

ROH

CO

-*cat.

cat.

>L?< H

CHO

(hydroformylation)

(hydrocarbalkoxylation)

The first enantioselective hydroformylation, albeit in very low optical yield ( < 3 070 ee), was reported in 1972 by Botteghi, Consiglio and Pino [I 151. Continued efforts by several groups resulted in considerable improvement of the methods but there is still much to be done to elevate selectivities to practically useful levels. The method is handicapped by problems of regioselectivity and easy racemization of aldehydes with an a-hydrogen. As catalysts Pd, Pt, and Rh complexes of DIOP and its diphosphol analogue (14) [116, 1171 are used (Fig. 6-12(a)). With the Pt(I1) complex of polymer-bound 14 [I 181 and of ligand 15 [I 191 up to 65 and 85% ee could be realized in the hydroformylation of styrene. Regioselective cis hydrocarbonylation at C-2 of 3-alkylcyclohexene was catalyzed by a Rh ( O A C ) ~ + P ~ catalyst ~ P system [120]. Perhaps because there is one more parameter, i.e. the structure of the alcohol component, which can be varied, stereoselectivity is generally higher in hydrocarbalkoxylation than in hydroformylation. In addition, esters are less prone to racemization than aldehydes. For enantioselective hydrocarbalkoxylation, mainly the complexes of chiral phosphines with palladium have been used [ 11 31 (Fig. 6- 12(a)). Up to 50% ee could be achieved with other olefins and DIOP.PdC1, as catalyst [I 131. Enantioselectivity in the hydrocarbalkoxylation of a-methylstyrene in neat alcohols is increasing with the bulkyness of the alcohol, only sBuOH is out of line (MeOH < sBuOH <EtOH
* For reviews,

c$ refs. [101, 113, 1141.

230

6.2 Stereoselective Alkylations

Me

Ph

Me

PdCI, + (14) Ph

L C 0 2 t h

cc69 %

Fig. 6-12. Enantioselective hydrocarbonylations (a) and hydrocarbalkoxylations (b).

6.2 Stereoselective Alkylations This section covers a large number of methods, all based on the attack of an electrophile at a nucleophilic center. Whereas the electrophilic partner is usually very simple, almost exclusively an alkyl halide, a nucleophilic center can be created in several ways, and the proper method to do this must be combined with special structural features or other means which lend stereoselectivity to the reaction. It appeared to be practical to subdivide this section according to the constitution of the nucleophile.

6.2.1 Alkylation of Enolates Enolates prepared from esters, amides or oxazolines can be alkylated in a-position.

6.2.1 Alkylution of Enolates

23 1

Enantioselective enolate alkylation using a chiral alkyl-donating agent is an almost untouched area. Duhamel et a/. described the methylation of a glycine ester enolate with a chiral methanesulfonate (up to 62% ee) [122, 1231, further Yamashita et al. reported methylation of a similar enolate generated with lithium salts of chiral amines (max. 31% ee) [124]. Following pioneering experiments by Yamashita et al. [124], Koga et al. achieved useful selectivities when alkylating an enolate generated by a chiral lithium base (Be) [125]:

0

0

Diastereoselective alkylation of chiral enolates can, from a practical point of view, be divided into two groups: (i) substrates with a removable (and preferably recoverable) chiral auxiliary group and (ii) substrates in which the inducing chiral center remains in the end product.

6.2.1.1 Alkylation of Enolates with a Removable Chiral Auxiliary Group A wide range of chiral auxiliary groups has been recommended for direct alkylation of ester and amide enolates. The most effective ones are shown in Fig. 6-13. Selected results achieved with these groups have been compiled in Table 6-2. In 1980 highly stereoselective alkylation of enolates derived from (S)-prolinol (R* = 16) was simultaneously reported by Sonnet and Heath [ 1261 and by Evans and Takacs [127]. The latter authors also found that the enolate was configurationally homogeneous, probably 2. It was suggested that allylic strain prevented the formation of the (E)-enolate in the corresponding transition state (Fig. 6-14(a)) [128]. It could be anticipated that any effect which has an influence on the rotation around the C-N bond would also be reflected in the stereoselectivity of alkylation. In fact, substitution of the hydroxyl in 16 inverted the configuration of the major product [ 126, 1271. After trapping by silylation, enolates of amides derived from 20 were also found to be of 2 configuration [129]. Although Ireland et al. found that the prevailing configuration of enolates was often inverted on adding HMPA to the solvent [130], this effect was not observed with

Me Bn Bu Me

Me allyl Me Me

23

25

30 30 31 32

>98 ( S )

>90 ( R )

78 (R) 82 (R) 86 ( R ) a

74 ( S ) 86 (S)a 78 ( S ) 88 (Wa

Me1

>94 (R)'

76 (R) 88 (R)a

>90 (S)

EtI

93 ( S )

BuBr

>90 ( S ) >80 ( S )

BuI

96 ( R )

Br

iBuI

>98 (S)

BnBr

Ref

a

With the sodium enolate; Enolate formed in the presence of HMPA; No reaction; Enolate formation with 1 eq. of BuLi+0.1 eq. of iPrcHexNH. For R' = Me and R2 = CH,CO,tBu, MeOCH, and Pent, de 97, 48, and 95%. For RZ = Me and R' = allyl, Pent, and Bn, de 90,92. and 88%.

Me Et Oct

Oct

Et

R'

22

20

19

I8

17

I6

R*

Table 6-2. Diastereoselective Alkylation of Chiral Ester and Amide Lithium Enolates. (Values shown are % de (conf. at C-2) for various RZX.)

233

6.2.1 Alkylation of Enolates

(16) R=H (17) R=Me

(16) R=CH20(CHZ)20Me (17) R=tRuMeZSi

Me

I (22)

Me (23) R=X (24) R=Y ( 2 5 ) R=Z

X = PhN(Me)CO, -

(29) R=Y

(26) R=X (27) R=Y (28) R=Z

Z = 2,4,6-Me3 -C,H, -NSO,Ph

Y = 3,5-Me2 -CH,H,-NS0,Ph

le

(30)

(31)

Me

(32)

(33)

Fig. 6-13. Chiral auxiliary groups (R*) used in ester and amide enolate alkylations.

16 [127]. In contrast, addition of HMPA changed the predominant configuration when the chiral auxiliary was 23, 25 or 28. A versatile series of chiral auxiliaries was prepared from camphor by Helmchen and his coworkers [131- 1351. These compounds enable the control of configuration in several ways. Interchange of endo and ex0 configuration (e.g. 23 us. 26), or of substitution from 2,3 to 3,2 (e.g. 27 us. 29), a change of solvent from T H F to THFHMPA, and, finally, changing the order of introducing R' and R2 all invert the predominant configuration. It should be added that in the presence of HMPA selectivity often decreases. The best results were achieved with 24 and long chain alkyl iodides (Fig. 6-14(b)) [131]. Recently, the method was adapted to the 2-benzyloxy-

234

6.2 Stereoselective Alkylations H

(a)

0 H

Me

OLi

(Z )

Fig. 6-14. Alkylation of chiral enolates (I).

acetates of 27 [I 3.51. The reaction proceeded with remarkable S selectivity ( > 91 070) but was unaffected by the addition of HMPA. By acylation of the enolate of propionyl-20 and -22 followed by immediate quenching, it first became possible to prepare optically active P-dicarbonyl compounds (Fig. 6-14(c)) [136]. Later several other amines (30 [137], 31 [138], 32 [139], 21 [140], 33 [141]) have been added to the armory of chiral auxiliaries (see Fig. 6-13). Alkylation of amide enolates involving 34 permit the highly enantioselective preparation of both a-mono- and disubstituted carboxylic acids [ 142, 143]:

6.2.1 Alkylation of Enolates /

(34)

235

OMOM

H R2 Me Me Me Et l3n R3 Et ally1 Bu Me Me Me WlBn P r * O/oee 90 90 85 80 84 97 96 98 97

Tin(I1)-enolates prepared from N-acyl derivatives of 30 are transformed to pyrrolidones [I 371.

R* = (32)

n = 1, 2,R = Q SPh, (CW,CI, O h , de 91-9776

Enolates of N-protected a-aminoacyl derivatives of (&R)-2,5-bis-methoxy methoxy-methylpyrrolidine [ 1441, and 5 [ 1451 can be transformed to a-aminoacids (ee 89 - 99%). A special removable chiral group containing a chiral iron center was devised by Davies et al. and used as the enol methyl ether (Fig. 6-15(a)) [I 52, 1531 or the enolate ion generated directly by treating the acyl complex with BuLi. The latter can be alkylated stereoselectively with alkyl halides [ 1541 or epoxides [ 1551. In the reaction of the enolate with racemic halides almost complete kinetic resolution was observed (Fig. 6-15(b)) [156, 1571. The technique of “chirality deposition”, i.e. transferring the chirality of an enolizable chiral to another chiral center not affected by enolization, was first realized for mandelic acid by Frater et a/. [158]. The method was later extended to other a- and P-hydroxyacids [159] as well as a-aminoacids [160- 1641 mainly by Seebach et al. (Fig. 6-15(c) and (d)). Optical purity can be most often readily improved by recrystallization of intermediates. Stereochemistry of acetal and aminal formation has been thoroughly studied by Seebach et al. who established that allylic strain is decisive for predominant cis selectivity of the reaction [165]. The application of s. c. azaenolates, obtained by lithiation of chiral oxazolines, has been very successfully exploited for the stereoselective synthesis of a wide range of end products by Meyers and his coworkers [166, 1671 and many others. A typical example is shown in Fig. 6-16(a) [168]. Meyer’s procedure is an excellent example of “second generation” stereoselective methods which are not hinged upon steric hindrance but on directing the reagent to a specific face of the substrate by coordination

236

6.2 Stereoselective Alkylations

1) Me,O"BFp,

Ph,P

2) KOtBu

R'

-

-78°C

R'YC02H

lo1

R2X

R', R2 = H, Me, Et, ee >95%

R2

R'

(b)

tBu

tBu

R' cis

/o: Ii iMeMe 1:; I

R'

%ck

I ) R'X. -78OC 2) H*

R'__-R

I

Me

Ph

Bn

CHzCOaH

80

95

83

98

C02H

Bn

Et

%ee

94

%

>9o

R' = Me, Pr, MeSCH$3$, Bn, de 60-98% R2X = MeI, BnBr,de 80-98% Fig. 6-15. Alkylation of chiral enolates (11).

6.2.1 Alkylation of Enolates

237

to a metal atom. This approach is then combined with the specific formation of one of two possible geometrical isomers of an enolate. The chiral aminoalcohol (35)can be prepared from commercially available material and is recoverable. As demonstrated in Fig. 1-26(b) (p. 40), both enantiomers of a given a-alkyl-alkanoic acid are accessible by interchanging the acid and alkyl components. It is apparent from Fig.6-16 that the methoxymethyl group is an essential part of the system, and, indeed, selectivity dropped sharply with the methyl analogue of 35 [I691 and with (S)-4-benzyloxazolines [170]. Though without a clear rationale, the presence of the phenyl group is also essential [169]. The formation of the (2)-enolate is kinetically favored, and no equilibration took place up to -30°C [171]. Unlike with other enolates (cJ: Sections 6.2.1.1 and 7.2.1), the Z/E ratio was only somewhat diminished but not inverted by the addition of HMPT [172]. Selected results of the application of the above sequence are compiled in Table 6-3. The 2-alkyloxazolines (R' # H) themselves can be conveniently prepared by alkylation of the enolate of the 2-methyl compound [I 68, 1741. The rather uniform results for the dialkyl products also indicate that steric effects play a subordinate role in shaping stereoselectivity.

I

OMe

: h g Ph

r

-I

Et I

OMe

HMj{OPh Et

L Me !

He ____f

"1OMe

+

Et /\CO*H (S). 78% ee

I) LDA 2) B u l . 3 ) H "

*

Me

! (S). 75% ee

Fig. 6-16. Alkylation of chiral oxazolines.

(35)

238

6.2 Stereoselective Alkylations

Table 6-3. Enantioselective Synthesis of a-Alkylalkanoic Acids via Chiral Oxazolines.

I ) L D A , THF, -7X"c

N

R'

RZX

Me

EtI PrI BuI BnCl Me,SO, BnCl Me,SO, BnCl Me,SiO(CH,), BuI Me,SiO(CH,),

Et Bu

Bn

2) R'X. -7M0C--Y8"C,

.~ Yo ee

Ref.

3) H *

*

R'

R2X

Ph

Me,SO, EtI BuI iPrI Me,SO, EtI ally11 BuI Me,SO, EtI BnCl ally11

Me,SiO(CH,),

Me,SiO(CH,),

-

R2

Vo ee

Ref.

Meyers et al. prepared bicyclic amides, such as 37 from (S)-valinol [I75 - 2771 and the commercially available (S,S)-2-amino- 1-phenyl-I ,3-propanediol [ 1781 which gave on repeated enolate alkylation products with a quaternary chiral center (monoalkylation products racemized on hydrolysis (e.g. [ 1771): 0 1) LDA R'X, -78 O C 2) LDA, RzX, -78 O C

*

&OH

R2 R1

Ph

(37)

For the transfer of chirality from one amino acid to another, Schollkopf and his coworkers elaborated an efficient scheme based on mixed diketopiperazines composed of a chiral amino acid as the stereodirecting moiety and a simple amino acid, like glycine or alanine, which is alkylated [I 79- 1811. As stereodirecting group, even a methyl group is quite efficient (de > 90?lo) [ 1821, while isopropyl and tert-butyl provide selectivities, generally beyond the detection level of the minor stereoisomer. The sequence is concluded by acidic methanolysis and separation of the two amino acid esters (a weak point in practical applications). The method is illustrated for (S)-valine in Fig. 6-17 (a) [I 83 - 1851. Selectivities for the tert-butyl analogue are even better [ 1861, but unfortunately tert-leucine, though available commercially, is not a natural product.

6.2.1 Alkylation of Enolates

iPr

Rx ___)

Me0

'LlG

R -

MeOH

YN ACO,Me

R

%ee

(b)

I

75-80

91-95

iy?JP

Mto

2) I ) RCHO H? MeOH

__

lPr H,N AC02Me

(9

(R)

> 95

>95

+

60-65

*,;h;

TI(NMe2)?

,

OH anti

CO2Me

239

93

R=Me, iPr, Ph % ee at C-2 >95% % ee at C-3 >97%

Fig. 6-17. Transfer of chirality between amino acids.

Results were explained by coordination of Li@ with the delocalized anion at the face opposite to the substituent. Attack at this face by the electrophile is promoted both by complexation with Li@ and shielding of the other face by the bulky group at C-2. Proton abstraction seems to affect only C-5, even when this position is substituted by a methyl group [184]. 3,5-Dialkyl-bis-lactimethers [184, 187, 1881 are also amenable to alkylation, and de values in excess of 95% were obtained with the same electrophiles as were used with 38 [184, 1871. Note that the configuration at C-5 is lost on deprotonation. Addition of bis-lactimether anions to carbonyl compounds is highly selective at the center which is part of the ring but much less so (de <52%) at the side chain [189-1911. Reaction of lithiated 38 with C1Ti(NMe2)3gave a titanium compound, the addition of which was highly stereoselective with respect to both new chiral centers (Fig. 6-17(b)) [192]. Innumerable applications of the method have been published, almost exclusively by the Schollkopf team [203].

240

6.2 Stereoselective Alkylations

6.2.1.2 Diastereoselective Alkylation of Enolates Diastereoselective alkylation of the lithium enolates of achiral amides with chiral epoxides resembles the aldol reaction (Fig. 6-18 (a)) [193, 1941. Otherwise, the scene is dominated by studies on P-oxygenated compounds. Stereoselectivities are often high, and this has been ascribed to chelate formation involving the alcoholate function [195]. Some of the more remarkable results are as follows. Alkylation of the dianion of P-hydroxyesters was found by Frater to yield predominantly the anti product, both when the a-position was unsubstituted (Fig. 6-18(b)) [196, 1971 and substituted [198]. Frater's results were later confirmed on different P-hydroxyesters by several other authors [195, 199-2021. Anti selectivity was high with most of the alkylating agents, and this was rationalized by the formation of a chelated intermediate [195], which was attacked by the electrophile from the less hindered side. More than 97% of the anti product was obtained in bad to modest yields on alkylation of the dianion of 3-hydroxy-y-butyrolactonein the presence of HMPA at -78 "C [203].

1) LDA THF, 0 O C 2) M e 7

Me (b)

Me

Hx;Li

YYCoN(pr)2 OH

Me

L~o~E~

Me

>9 5 % v

OH

2) I ) 2RBr, L D A .HMPA THF

L C 0 2 € t ,

Me

R=Me, Bu, Bn, 2 9 5 % anti

Rk

OEt

Li

Fig. 6-18. Diastereoselective alkylation of enolates.

Diastereoselectivity in the alkylation of macrocyclic ketone enolates was successfully correlated with the conformational composition of the substrate by Still and Galynker in a similar way as has been discussed for other types of reactions elsewhere ( c j Sections 1.5 and 6.1.1.2) [83]. Ketone enolate alkylation is, however, not an attractive reaction due to problems of regioselectivity and dialkylation. A rule, similar to Cream's rule (in the interpretation by Houk), but of opposite outcome has been formulated by Fleming for electrophilic attack on enolates. The rule and a test for its validity [204] is shown next:

24 1

6.2.2 Alkylation of Stabilized Carbanions

R1 R'

R3

Ph Me Me 72

Ph

Pr

Pr

Me Me Ph R3 OMe Me Me 58 60 46 %deanh (H!J Y d e y (Me ) 20 10 50 74 R2

Me

Problems to ranking substituents by size are similar to those encountered with rules on nucleophilic addition to carbonyl compounds. Fleming et ai. studied in detail the stereochemistry of the attack of electrophiles on lithium p-silyl enolates generated in situ by the addition of a silylcuprate to a , /3 unsaturated ketones or esters [205]. The rule holds in most cases as e.g. [206]: PhMe,Si I ) (PhMe,Si),CuLi 2) Me1

G

R'

R

:

3

Me

By oxidation with MCPBA the silyl group can be replaced by a hydroxy group with retention of configuration. Molecular orbital calculations by McGarvey and Williams also predicted anti preference in enolate alkylation regardless of enolate configuration [207]. Enolates of tert-butyl-5-hydroxycarboxylatesare a-alkylated with > 84% syn- [208], the tert-butyl-4-hydroxy-3-alkylcarboxylates,in turn, with anti selectivity (de 60- 82%) [209].

6.2.2 Alkylation of Stabilized Carbanions It has to be noted in advance that the distinction of enolates and stabilized carbanions is often arbitrary. Here we discuss reactions of carbanions stabilized by S, SO, SO2, phosphonate and urethane groups. Eliel and his coworkers demonstrated that under thermodynamic control 1,3-dithianes form the equatorial lithio derivatives exclusively; these can be then alkylated with retention of configuration [210-2131. Examples are shown in Fig. 6-19. Prefer-

242

6.2 Stereoselective Alkylations

H

H

MeKTH \

THF. -2S'C BuL1

S

>

Me

H \

Me

\

M e F f i C O Me

H

Me /ratis

Fig. 6-19. Diastereoselective alkylation of 1,3-dithianes.

%syn

I

I Me

L

R2 = Pr, ally], En, C H , W - , ee >97%, for R2 = Me, ce 81%

94

84

13

0

R1 = Cl, Me R2 = Et, Ph, Bn, allyl, ee >98%, for R2= Me ee 80%

MeJN3 2% (c)

m,

HMPA 1) - 80 OTHF C ; 2) RX

* Me

Fig. 6-20. Stereoselective alkylation of carbanions.

I ) 2) h eHey N

T

9

6

6.2.2 Alkylation of Stabilized Carbanions

243

ence for the equatorial metallation is so strong that even a 2-terf.-butyl-l,3-dithiane was lithiated equatorially [211]. Methylation of the p-aminosulfones 39 was syn selective with tertiary amines and anti selective with primary and secondary amines (Fig. 6-20(a)) [214]. It is remarkable that methylation of syn- and anti-P-hydroxysulfoxides is controlled by the hydroxy group giving in both cases the 2,3-anti products in 83% de [215]. Trans-1,2-bis-methylaminocyclohexane is an efficient chiral auxiliary permitting the preparation of a-chiral phosphonic acids (Fig. 6-20(b)) [216, 2171. Chiral isoxazolines may serve as P-hydroxyketone equivalents and can be alkylated, via the anion selectively (Fig. 6-20(c)) [218]. When forming a carbamate from an alcohol the carbamate group stabilizes the corresponding a-lithiated compound and, as found by Hoppe et al., when lithiation is carried out in the presence of (-)-sparteine, the si-proton is abstracted and subsequent attack by an electrophile proceeds with retention and excellent selectivity [219- 2231. With bis-carbamates the procedure can be repeated and replacing (-)-sparteine by TMEDA in the second step inverts anti to syn selectivity (Fig. 6-21) [222].

1-

1) s w l b (-)-sparteine

Cbrn = (411, n = 1

PhMe, -78 OC 2) Me1 3) H e

H’ -OH Me

*

EbO, - 78 OC

2)

CbmO ,,(cHJ~yoam

R

n = 1,2, RX = Me,SiCl, Me,SnC1, MeI, cc 3 5 %

Cbrn = (40)

1) sl3uLi TMEDA PhMe, -78 OC 2) Me1 3) He

Me

Me

-CON Me

Fig. 6-21. Stereoselective alkylation of lithiated carbamates.

Y Mc O H

244

6.2 Sfereoselecfive Alkylafions

6.2.3 Alkylation of Enamines and Metallated Azomethines Enamine alkylation is a standard method for the preparation of 2-substituted carbony1 compounds. Since chiral amines are relatively readily available, the procedure can be easily adapted to enantioselective synthesis. An advantage is that the chiral amine is, at least in principle, recoverable. As amine components, the esters and amides of (S)-proline [224-2281, chiral 2-alkyl- and 2-pyrrolidinylmethyl-pyrrolidines [229 - 2321 and isobornyl enamines [233] were tested but only mediocre selectivities could be attained. The first efficient chiral amine component, (2S, 5s)-dimethylpyrrolidine, was introduced by Whitesell and Felman [3 151 and used to prepare 2-alkylcyclohexanones (ee 82-93'70) [314]. The same group proved to be very effective also in the alkylation of enaminolactones (Fig. 6-22 (a)) [235]. High diastereoselectivities were reported by Blarer and Seebach with methylprolinol (Fig. 6-22(b)) [234]. A promising chiral auxiliary is the amine 42 prepared from a by-product of drug manufacturing [236].

&..

Rigo 0

Me

1) w l 4 THF, - 78 O C 2) R2X

R*

-

-

H

R2

R1=H,Me R2 = Me, ally\ Bn, de 294% (b)

(c)

:3

(42)

CH(C02Et)z

(0..

ACO*,,

~

+

Ph

@ph

C0zEt

Fig. 6-22. Stereoselective alkylation of enamines (I).

de 7 9 5 %

245

6.2.3 Alkylation of Enamines and Metalluted Azomethines

Morpholinyl enamines add to 1-nitroolefins giving syn products [237]. Configuration of both reactants is irrelevant (Fig. 6-22(c)) [238]. While enamines involving secondary amines are structurally well defined, reaction of primary amines with carbonyl compounds gives a product to which either an enamine or an azomethine structure can be ascribed. Metallation shifts the equilibrium to the enamine side and permits alkylation at the a-position. In most cases, metallation was carried out with LDA and, for efficient stereodirection, an ether function in the amine, such as in 43 prepared from phenylalanine [239], appeared to be essential. 43 was an efficient auxiliary in the alkylation of cyclohexanone (Fig. 6-23 (a)) [240, 2411. The Whitesells, in turn, methylated, at - 100 “C, the bromomagnesium azaenolate of cyclohexanone formed with the 0butyl analogue of 43 and obtained the S product in 85% ee [242]. Enantioselectivity with 43 was slightly less with cycloheptanone and poor with cyclooctanone. Koga et al. found that tert-butylesters of isopropyl- and tert-butylglycine were powerful chiral auxiliaries for lithium azaenolate alkylations (up to 98% ee) [243, 2441 and observed that in the presence of an ester group in a-position the direction of alkylation could be influenced by additives (Fig. 6-23 (b)) [245, 2461.

(a)

Li NR*

N /R*

RX %ee

I I

Me1

EtI

PrI

-Br

BnBr

87

94

99

99

88

0

NH2

(43)=R*--NH z

K:

iPr

1 ) LDA, cosolvent

+

ci:”

&JY-+

PhMe, -78OC, 2) RX, 3) Ha

( A1 Rx Me1

e BnBr

B

r

cosolvent

% ee (conf.)

HMF’T

2998 92A 76 B 56 A >99B 71 A

THF HMF’T THF HMPT dioxolan

Fig. 6-23. Stereoselective alkylation of enamines (11).

----COB

(B)

246

6.2 Stereoselective Alkylations

Enantioselective a-alkylation of acyclic ketones via azomethines proved to be a much more demanding task than that of cyclic ketones. Meyers et a/. reported high stereoselectivities with symmetric dialkyl ketones using 43 as chiral auxiliary, provided that the enolates were converted, prior to alkylation, to the E form by refluxing (Fig. 6-24(a)) [247, 2481. Note that no inversion of configuration, only enhancement of enantioselectivity, was brought about by 2 E transformation. Azomethines of 43 and aldehydes were alkylated as the lithio azaenolates with poor selectivity ( <50%) [239]. Azomethines of terpenic ketones with esters of a-amino acids can be alkylated with moderate to high diastereoselectivity (Fig. 6-24 (b)) [249 - 2521. -+

R'

(b)

= R2 = Me,

Et, Pr ee 79 - 94%

Me OH

R'

Me

I) LDA 2) R*X. 3 ) H @

L

Me R'

RZ

I

H

H

H

Me

Me

iBu

Bn

Pr

Me

PI

Me

Fig. 6-24. Stereoselective alkylation of azomethines.

6.2.4 Alkylation of Metallated Hydrazones and Amidines * Stereoselective a-alkylation of lithiated chiral hydrazones was developed by Enders and his coworkers [256]. The key to success was the synthesis of a chiral hydrazine 44, available as both enantiomers (SAMP and RAMP). SAMP was prepared from @)-proline in 5 steps in 50% overall yield [256, 2571, while RAMP was

*

For reviews, see refs. [253-2551.

6.2.4 Alkylufion of Metalluted Hydrazones and Amidines

(a' w

EtKO

e

'f

I

Prl. THF -9ST

LMe LDA, THF OT,4h

NH2 ( S ) - (44)(SAMP)-R*NH2

*

241

*

Me

-780C

e

Et

N

Me

Pr 99.5%) ee

RZ

I Me

iPr

Ph

Ph

Me

iPr

Ph

%syn

I

86

82

88

92

96

96

90

I

NO

Fig. 6-25. Stereoselective alkylation of hydrazones.

obtained from the relatively expensive, but commercially available, (R)-glutamic acid in five steps in 30% overall yield [258, 2591. The general scheme of metallated hydrazone alkylation is shown in Fig. 6-25 (a). Repeated alkylation provides the anti-1,3-dialkylketones in outstanding selectivity (ee > 98%) [266]. Removal of the chiral auxiliary is also feasible by quaternization with methyl iodide, followed by two-phase acidic hydrolysis, but in this case only a part of 44 can be recovered [253]. 2 configuration of the metallated hydrazone was proved [260, 2611. The conformation around the N - N bond is a subject of speculation, but it is reasonable to assume that Li0 is coordinated to both the allylic type anion and the ether oxygen, and this feature is the clue to the exceptional stereoselectivity of the alkylations. So far the method has undoubtedly been the best approach to chiral a-alkyl carbonyl compounds. The most important results are compiled in Table 6-4. Addition of lithium alkyls to aldehyde SAMP hydrazones followed by hydrogenolysis of the N-N bond gives rise to primary amines of 81 -93% ee [262]. Addition of the same to a-alkoxyaldehyde N,N-dimethylhydrazones produces after hydrogenolysis syn-a-alkoxyamines in > 94% de [263]. The hydrazone method is flexible in two ways: The configuration of the end prod. uct can be controlled, both by that of the auxiliary and by interchanging the roles of R' and R2. Reduction of the alkylated hydrazones with catecholborane followed by hydrogenolysis of the N - N bond over Raney-nickel gives /3-chiral primary amines in >90% ee [260].

248

6.2 Stereoselective Alkylations

Table 6-4. Stereoselective a-Alkylation of Lithiated Hydrazones of Carbonyl Compounds (44, SAMP). (R’COR’) formed with (S)-1-Amino-2-methoxymethylpyrrolidine R’

R2

R’X

% ee

Me

Et Pr Et BnCH, BnCH, Et Hept i Bu En Et

EtI Me1 BnBr Me1 Me’BuSiOTf Hex1 Me1 Me1 Me2S04 EtI PrI Me,BuSiOTf MeO(CH,),Br Me1 Me1 EtI Me1 Me1 MeO(CH,),Br Me1 Me,BuSiOTf tBuO,CCH,Br Me,CO, Me,SO, EtI PrI allylBr Me1 Me1 allyl(CH,),Br

71 62 82 - 95 > 90 > 96 > 95 95 51 31 94 99.5 > 96 > 96 >98 2 98 2 98 20 30 > 96 10 > 96 2 95 86 99 > 98 > 86 73 94 61 -75 89

Me

Et

Pr Bu Bn

Et

(Conf.)

Ref.

While the addition of aldehydes to 0-enolates obtained from aldehydes is unselective, high syn selectivity, presumably via a non-cyclic transition state, was reported by Reetz et al. for titanium enolates of dimethylhydrazones (Fig. 6-25(b)) [268]. Whereas alkylation of lithiated chiral amidines is generally not enantioselective enough [269, 2701, in a special case it proved to be successful [271-2731:

A R = Me, Bu, allyl, BuCH, 2 93% ee

6.3.1 Grignard Cross-Coupling

249

Molecular mechanics calculations by Meyers et al. [274] indicated preferred stabilization by complexation with the heteroatom of that particular lithiated species which leads, by inversion, to the predominant enantiomer.

6.3 Enantioselective Grignard CrossCoupling and Other Stereoselective Catalytic Carbon-Carbon Bond Forming Reactions 6.3.1 Grignard Cross-Coupling * Grignard cross-coupling is the reaction of a Grignard reagent with a halogen compound in the presence of a transition metal catalyst, whereby a carbon-carbon bond is formed with the formal elimination of magnesium halide. In practice, only vinyl (occasionally aromatic) halides and nickel(I1)- or palladium(I1)-phosphine complexes are eligible as substrates and catalysts, respectively. If the Grignard reagent is prepared from a chiral but racemic secondary alkyl halide and the catalyst contains a chiral phosphine (P*), kinetic resolution may take place, and an optically active product is obtained. Since chiral Grignard reagents undergo fast racemization, in principle there is no limitation to the yield and optical purity of the product. Isomerization of the secondary Grignard reagent to a primary one may, however, diminish the yield of the chiral product. In 1973, the first asymmetric Grignard cross-coupling was reported by Consiglio and Botteghi, who used [( -)-DIOP].NiCl, as catalyst (ee< 17%) [277]. The reaction has been thoroughly investigated by Hayashi, Kumada and their coworkers, who prepared for this very purpose a large number of chiral ferrocenyl phosphines mainly of type 45. In a model reaction shown in Fig. 6-26(a) up to 65% ee was realized with the ferrocenyl monophosphines [278], 93% ee ( R )with the diphosphine 46 [279], and 94% ee ( S ) with 47 [280, 2811. Addition of ZnBr, inverted the prevailing configuration [282].

* For reviews, c$

refs. [IOO, 2751.

250

6.3 Enantioselective Grignard Cross-Coupling

h2

PPh,

&Ime2

*,

PPh,

Me

Me

tEu Y P P h , NMe,

RMe,

tBuCI, TiCI, CHzClr, O°C

(47)

*

1

J

(c)

+PhOTf

[P-{O c), + (R)-BINAP]

wJJ% c,H,

-k

(-)"."Ph 0

Me

-15

Fig. 6-26. Enantioselective Grignard cross-coupling.

F

'

h

89 : 1

93% ee

&;,,:

Q

69% ee

Me b OC

Me

95% ee

6.3.2 Stereoselective Codirnerization of Olefins

25 1

The model reaction has direct practical relevance, since on oxidation 2-arylpropionic acids are obtained which are important antiinflammatory agents. Thus the 4-isobutylphenyl analogue, the precursor for the well known drug Ibuprofen could be prepared in 80% ee [281]. Hayashi et a/. extended the scope of Grignard cross-coupling to a-silylated Grignard reagents [283]. No E-Z isomerization takes place during coupling, and the products, optically active silanes, can be transformed stereoselectively to useful intermediates by Lewis acid-catalyzed SE1-type electrophilic substitution (Fig. 6-26 (b)). Enantioselectivty with (Z)-vinyl bromides was poor ( < 24%). Somewhat related to cross-coupling is the enantioselective, but not quite regioselective Heck arylation discovered by Hayashi et al. (Fig. 6-26(c)) [284]. A useful method to prepare chiral binaphthyls was elaborated by the same group (Fig. 6-26(d)) [285]. Alkylzinc halides are also amenable to the asymmetric cross-coupling reaction, but no special advantage is gained by this modification [286].

6.3.2 Stereoselective Codimerization of Olefins * It has been recognized by Wilke and his coworkers in 1963 that phosphines were excellent cocatalysts in the codimerization of olefins catalyzed by n-allylnickel halides and aluminum halides. It was a logical development that this reaction should be tried with chiral phosphines. Indeed, codimerization of 2-butene and propene in the presence of a chiral phosphine gave a product with detectable optical activity, and this discovery was claimed to have been made in 1967 [287], one year before the first results on enantioselective hydrogenation with rhodium-chiral phosphine complexes were made public. The codimerization of cyclooctadiene and norbornene with ethene (Fig. 6-27 (a) and (b)) was studied in detail and it was possible to raise the optical purity of the products to a useful level [288]. The maximum ee of the by-product, 3-methylpentene, was 64070, but under different conditions and with another catalyst [287]. Low temperatures improved selectivity in the codimerization of norbornene and norbornadiene with ethene (Fig. 6-27 (b)). Codimerization of cyclohexadiene and ethene gave (S)-3-vinylcyclohexene in 73% ee [289]. Diastereoselective dimerization of butadiene derivatives was also described (Fig. 6-27 (c)) [290]. Despite this encouraging start, further interest in the method seems to be very small.

* For review, see ref.

[287].

252

6.4 Miscellaneous Stereoselective Carbon-Carbon Bond Forming Reactions

ee

+

H2C=CH2

lCH2*-CH-CH2)NiCfl Et,AI2CII +48, PhCI, - 9 7 T

*

Fig. 6-27. Stereoselective codimerization of olefins.

6.4 Miscellaneous Stereoselective CarbonCarbon Bond Forming Reactions This section is a collection of stereoselective carbon-carbon bond-forming reactions which did not fit into any of the types discussed in other sections of Chapters 5 and 6. In the course of their pioneering investigations on polyene cyclizations (cJ:Section 7.3), Johnson and his group exploited stereodirection by chiral2,4-pentanediol in its acetals. Later it was shown that such acetals can be attacked very selectively at one of their diastereotopic C - 0 bonds by silylacetylenes [291], trimethylsilyl cyanide [292, 2931, allyltrimethylsilanes [294] and Grignard reagents [295]. Removal of the chiral auxiliary is performed by oxidation to a ketolether, followed by base- or acidcatalyzed elimination (Fig. 6-28 (a)). Acetals of type 49 can also be cleaved by silylketene acetals [296] or silylenolethers [297] (Fig. 6-28 (b)). Acetals of (R,R)-2,3-butanediol can be cleaved diastereoselectively with R2CuLi-BF3*Et20(de 67- 100%) [298].

6.4 Miscellaneous Stereoselective Carbon-Carbon Bond Forming Reactions

Fig. 6-28. Stereoselective ring opening of chiral acetals.

253

254

6.4 Miscellaneous Stereoselective Carbon-Carbon Bond Forming Reactions

A mechanistic study by Denmark et al. suggested that selective complexation of the Lewis acid with the oxygen adjacent to the equatorial methyl group, rather than relief of strain caused by the same, explains high selectivity of ring opening [299]. An interesting variant of the reaction leading ultimately to chiral monoprotected 1,3-diols exploits selectivity in the formation and cleavage of (-)-menthone acetals. In a mixture of 1,3-disubstituted meso- and (&)-diols only the rneso-diol reacts (Fig. 6-28(c)) [300]. Seebach et al. [301] adapted the reaction to cis-1,3-dioxanones prepared from aldehydes and the readily available (R)-3-hydroxypentanoic acid. Wilson and Cram provided the first example for asymmetric induction by a chiral leaving group in the reaction shown in Fig. 6-29 [302]. Meyers and Lutowski carried out the same coupling with a substrate in which the leaving group was achiral but the oxazoline moiety was chiral. Stereoselectivity was somewhat lower [303].

&\:( \

/

Me

& MgBr

0-(quininyl)

+

Me

Me

/

-THF 2OT

~

/

\

Fig. 6-29. Stereoselective Ullmann coupling.

Complete transfer of chirality was observed in an alkyl migration resembling the pinacol rearrangement by Suzuki et al. (Fig. 6-30(a)) [304, 30.51. If the groups at the quaternary center are methyl and substituted vinyl, it is the latter which migrates (ee 95 - 99V0) [306]. The method works also with a-mesyloxyacetals of aryl- [308] or vinylketones [309]. When rearrangement is accomplished under reducing conditions aldehyde acetals, while under thermolysis esters are obtained (Fig. 6-30(b)). Cyclization of a chiral a-diazoester, in which the nearest chiral center of the auxiliary alcohol was three and seven bonds away from the reacting center, was described by Taber and Raman (Fig. 6-30(c)) [310]. In the presence of a chiral catalyst, moderate enantioselectivity was achieved in the ring closure of achiral diazoesters [3 1 I]. A phenyl group adjacent to the site of insertion resulted in complete asymmetric 1,2-induction [276].

25 5

6.4 Miscellaneous Stereoselective Curbon-Curbon Bond Forniing (a)

R. R

O3SMe

R

R-Et,

RO OR

Bu. Oct. cHe.

R = Me. DlBAH Et,N. 4 2 "C * M c y \ \ / l l u &OMc), de. cc - 100%

dc, cc >%Yo

, I-Naph

..*R

I

iPr

tBu

Pent

-&".

Bn

YO de

I

66

66

14

84

70

Fig.6-30. Transfer of chirality in a pinacol-type rearrangement, (a) and (b). Diastereoselective carbon-insertion reactions (c).

Examples for stereoselective carbon-carbon bond forming radical reactions are rare; two examples are presented below [150, 1511: Me

R=H Ph de 90%

0

0

R = CI

de 80%

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

7 Stereoselective Carbon-Carbon Bond Formation by Pericyclic Reactions According to Woodward and Hoffmann [ I ] pericyclic “is a reaction in which all first order changes in bonding relationships take place in concert on a closed curve”. Because of the constraints inherent to cyclic transition states, pericyclic reactions are good candidates for stereoselective transformations. In fact most of the pericyclic reactions were found to be highly or even totally diastereoselective. Stereoselective pericyclic reactions are adequately covered by reviews [e.g. I , 21. From this vast area two topics were selected for more detailed discussion in this book, namely, asymmetric induction in cycloadditions and transfer of chirality in sigmatropic rearrangements.

7.1 Asymmetric Induction in Cycloadditions Cycloaddition is a process in which two or more reactants combine to form a stable cyclic molecule, during which no small fragments are eliminated and a bonds are formed but not broken [3]. The Diels-Alder reaction is a typical cycloaddition, while cyclizations involving a single molecule (e.g. electrocyclic reactions) are excluded from the definition. Cycloadditions are conveniently classified by the number of ring atoms contributed by each component to the new ring [3], or the number of electrons (usually n electrons) involved in each component [I]. The two notations coincide in the case of the Diels-Alder reaction (4+2 in both cases) but differ, for example, with 1,3-dipolar cycloadditions, in which the dipolar component contains four n electrons over a chain of three atoms. The number of electrons involved may be indexed according to their character (a,n or n) and the mode of approach of the components (supra- or antarafacial, s or a) [ 11. For example, the Diels-Alder reaction is a n:+ nt cycloaddition. In the following sections asymmetric induction in the Diels-Alder reaction, stereoselective 1,3-dipolar cycloadditions and carbene additions will be described.

258

7.1 Asymmetric Induction in Cycloadditions

7.1.1 Asymmetric Induction in the Diels-Alder Reaction * The Diels-Alder cycloaddition was perhaps the first reaction in which complete stereoselectivity was recognized. In Diels-Alder type cycloadditions stereoselectivity is manifested at three levels and is often combined with problems of regioselectivity. (1) The reaction is a concerted process and therefore its basic stereochemistry is dictated by the Woodward-Hoffmann rules: the orbitals of both partners interact in a suprafacial mode, resulting in exclusive syn addition. ** In other words, the initial diastereomeric configuration of the double bonds of the partners is retained in the product. Thus, addition of an (E)-olefin to a diene gives rise to a trans disubstituted exclusively (cJ: Fig. 7-1 (a)). (2) Two different relative orientations of the approaching reactants can be envisaged, and it was found that the one leading to the endo product was generally, though not exclusively, preferred (Fig. 7-1 (b)). This is the s. c. endo rule. The explanation for a preference of the usually more hindered endo product is based on secondary orbital interactions [2, 31. ( 3 ) A further complicating factor is regioselectivity, which arises when both components are unsymmetrically substituted (Fig. 7-1 (c)). The Diels-Alder reaction is often remarkably selective even in this respect. A rationalization of this regioselectivity invoking orbital interactions was proposed by Houk [6]. (4)Unsymmetrically substituted dienes and dienophiles have enantiotopic faces. Therefore, if one of the partners has enantiotopic faces, even in the case of exclusive cis-endo addition and total regioselectivity, the product occurs as a pair of enantiomers. Thus, in the reaction of acrylates with cyclopentadiene (Fig. 7-1 (d)), the ester may react on either its Re or its Si face yielding the (2R)-and (2S)-esters, respectively. Provided that the endo rule is operative and addition is regioselective, no additional stereoisomers emerge even when both partners are unsymmetrically substituted. From the two possible ways to obtain one of the enantiomers in excess, i.e. chiral catalysis and attaching a chiral auxiliary to one of the components, only the latter has met with real success so far. The first asymmetric Diels-Alder reaction was reported in 1948 by Korolev and Mur [7], whose reacted di-( -)-menthy1 fumarate with butadiene, but the observed enantioselectivity was infinitesimal. However, in 1963 Walborsky et a[. [8] discovered that enantioselectivity increased dramatically when the reaction was catalyzed by a Lewis acid. In the presence of a Lewis acid it was possible to carry out the reaction at an acceptable rate at room temperature and even at -70°C.

* For a review see ref. [4]. ** For the characterization of the stereochemistry of addition the terms syn and anti were recommended by IUPAC Rules for Organic Chemical Transformations IS].

7.1.1 Asymmetric Induction in the Diels-Alder Reaction

+ X

)<-

259

a-

cis

trans

endo

ex0

X

(4 ROzC.,.. ( s i

(2s)

6,

H

ROK

Re

0

+\,Ls;

+

0

.COz R Re

(2 R )

Fig. 7-1. Stereo- and regioselectivity in the Diels-Alder reaction.

Addition of chiral acrylates to cyclopentadiene became a popular model for the exploration of asymmetric induction in the Diels-Alder reaction, although incomplete endo selectivity made the evaluation of the results more difficult. Product analysis was usually performed on the mixture of alcohols obtained after removal of the chiral auxiliary group by LiAlH, reduction. The high enantioselectivities (88% ee with 3,3-dimethyl-2-butanol as the chiral alcohol [9]) achieved by this method were almost unprecedented at that time, but faded in the light of later discoveries. Corey and Ensley reported virtually complete (99% de) stereoselectivity when (-)-8-phenylmenthol ( I ) , prepared from pulegone, was used as a chiral aid 110, 113 *.

* Only 90% ee was found later

by Oppolzer et al. [12].

260

7.1 Asymmetric Induction in Cycloadditions

By conversion of natural (R)-pulegone to (S)-pulegone [ 131 both enantiomers of I became available. Unfortunately, this substance is an oil which is not readily available and has to be purified by HPLC. Oppolzer and his coworkers undertook a very extensive study of the reaction to find, on the one hand, the best experimental conditions, and, on the other, chiral alcohols which were both highly stereodirecting and easily available [12, 14, 151. In Table 7-1 selected data for the Diels-Alder reaction of chiral acrylates and cyclopentadiene are presented. Table 7-1. Stereoselectivity in the Diels-Alder Reaction of Cyclopentadiene with Chiral Acrylates (CH, = CH - CO,R*). R*

Catalyst

( - )-Menthy1

none BF, . Et,O TiC1, BF, .Et,O TiCI, TiCl, TiCl,(OiPr), TiCl,(OiPr), TiCl,(OiPr), TiCl,(OiPr), TiCl,(OiPr), Et2A1C1 TiCI, TiC1, AICI, Me,AICl Et,AICl Et2AICl Et,AICl

(S)-Lactoyl (S)-3,3-Dimethyl-2-butyl ( - )-8-Phenylmenthyl 2 4 6 7 9 10 13 15 16

I7 19 22 23 24 a

Temp. "C

35

070

endoa

69

- 70 - 64

97

- 70 - 20 - 20 - 20 - 20 - 20 - 20 - 20 - 78

- 10 - 60 0 - 90 - 100 - 100 - 100

96 95 96 97 96 > 99 91 98 95 > 99 > 99 > 99 > 99

Vo ee of endo (Conf.)

7 85 86 88 90 81 97 91 99 88 93 95 94 99 98 93 86 90 90

(2R) (2R) (2s) (2s) (2R) (2R) (2s) (2R) (2R) (2R) (2R) (2s) (2s) (2R) (2R) (2s) (2R) (2R) (2s)

% ee of ex0 (Conf.)

3 (2s)

Ref.

[I61 [91 1171 191 [I21 [I21 1151 1181 ~191 1141 1201 1241 1251 1261 1231 [221 1-21] 1211 [211

Normalized to % endo + % ex0 = 100.

The alcohol type auxiliaries 2, 4 - 7, 9 and 12 (Fig. 7-2) can be prepared from inexpensive (+)-camphor and mostly offer excellent selectivities, but they decompose on exposure to TiC1,. This difficulty was circumvented by applying, as catalyst, a mild Lewis acid, TiC12(OiPr)2, at low temperature. Intensive search for new chiral auxiliaries to be attached to dienophiles continued in the past decade turning out some efficient compounds such as the alcohols 13 [24], cinchonidine and cinchonine [27], 14 [28], pantolactone (15) [25], 16 [26] (prepared from (S)-malic acid), and cheap ethyl @)-lactate [29]. With the proper choice of catalyst even good old (-)-menthol can occasionally give excellent results, as in

7.1.1 Asymmetric Induction in the Diels-Alder Reaction

b,, ,.CMezPh

Me

(-)-(I)

(-)-

8-phenylmenthol

26 1

R ; & Me

OH

(2) R=Ph

Me

(3) R = P h 2 C H ( 4 ) R=tBuCH, ( 5 ) R=CONHPh

(-)-Mento Me

(6) R=Ph,CH (7) R=tBuCH2 (8) R = B n

SOzNRz

(9) R=iPr (10) R=cHex

Me

Fig. 7-2. Chiral alcohols and amines used as auxiliary groups in enantioselective Diels-Alder synthesis.

262

7.1 Asymmetric Induction in Cycloadditions

the Et2A1C1-catalyzed addition of (-)-menthy1 fumarate to dienes [30]. Thermal cycloaddition of the (-)-menthy1 mixed acetal I1 in Fig. 7-2 to cyclic and acyclic dienes proceeds with excellent endo- and enantioselectivity [3I]. Several amines proved also to be useful: 17 1231, 18 1321 (the latter two give antipodal products), 19 (prepared from saccharine) [22] and Oppolzer’s sultam (20) 1331. A plausible model, supported by photophysical experiments [34], which may explain the stereoselectivity of 8-phenylmenthol and its congeners is 12, in which the dienophile is in the s-trans conformation. For simple acrylates this has been shown to be the most stable arrangement. The fact that enantioselectivity improved on lowering the temperature supported the proposition that selectivity was controlled by a conformational equilibrium involving relatively small energy differences. Although this model, as well as analogous ones constructed for other substrates, was based on the assailable assumption that the preferred ground state conformation was retained in the transition state, it not only predicted the predominant configuration surprisingly well, but was also useful for the design of more selective chiral auxiliaries. Thus it could be anticipated that chiral alcohols which provided for a parallel positioning of the acrylate moiety and had a bulky shielding group would be hopeful candidates as highly selective auxiliaries. The bornane derivatives 5 - I0 fulfilled this requirement particularly well, and those in which an angular methyl group adjacent to the ester function enhanced constraints in the transition state (6 and 7) were especially selective. The pairs 3 - 4 and 6- 7 are quasi-enantiomeric and therefore give rise to enantiomeric products. Addition of chiral acrylates to cyclopentadiene is just one of the many examples in which the Diels-Alder reaction was exploited to produce chiral compounds of high optical purity (Fig. 7-3). (-)-8-Phenylmenthyl acrylate was, in fact, first used as a partner to be reacted with 5-benzyloxymethyl-cyclopentadiene;the product served as a chiral starting material for a prostalglandin synthesis [lo, 111. * With butadiene the same acrylate gave rise to a cyclohexene used for the synthesis of sarcomycin [35] (Fig. 7-3(a)). The neopentyl compound 7 proved to be a powerful directing group when it was linked with allene-carboxylic acid (Fig. 7-3 (b)) [36,37]. In chiral acrylates the directing chiral moiety is rather remote from the reaction center. Indeed, the enone 21 and its congeners, in which this distance is smaller, are very selective dienophiles. The chiral-inducing group can be removed oxidatively (Fig. 7-3 (c)) [38,39]. N-acrylates and crotylates of chiral oxazolines, prepared from (S)-valine (22), (S)-phenylalanine (23) and (1 S, 2R)-norephedrine (25)are used as chiral dienophiles (see Table 7-1 for the acrylates, for the crotylates selectivities were slightly better). Diastereoselectivities with isoprene and 1,4-pentadiene were also high when 22 was used as chiral auxiliary [40]. A range of other 3-substituted acrylamides formed with 23 gave also high endo- and enantioselectivity in their addition to cyclopentadiene 1411.

* Note that the diene component has diastereotopic faces, one of which preferred.

(zd approach) is highly

263

7.1.1 Asyminetric Induction in the Diels-Alder Reaction

R*OH =(-)

I

-8 -phenylrnenthol

*R02C 89% de

-2O'C

4

c

G

R

987; endo * >98% de

iBu

0

(d)

(22) R=iPr (23) R=Bn R'

A.

"NP

cHex

(24)

Me

Ph

(2.51

Fig. 7-3. Examples for the use of chiral auxiliary groups in the Diels-Alder reaction (I).

Analysis of the molecular geometry of chiral auxiliaries led Helmchen and Schmierer to the conclusion that for effective stereodirection the inducing group should form a concave face around the reacting functional group [42]. The monosubstituted diols 5 and 8 (Fig. 7-2), both prepared from (+)-camphor, are examples of such convex directing groups, and in fact the AlC1,-catalyzed Diels-Alder addition of the corresponding methyl fumarates to anthracene at - 30 "C proceeded with complete stereoselectivity [43]. The uncatalyzed hetero Diels-Alder reaction of 2,3-dimethylbutadiene to chiral sulfines produced a single diastereomer [44]. The aza-Diels-Alder reaction of the azomethine 26 is remarkable inasmuch as it is ex0 selective (Fig. 7-4(a)) [41]. An example for an oxaza-Diels-Alder reaction is shown in Fig. 7-4(b) [45]. Very few studies have been reported in which a chiral diene was reacted with an achiral dienophile [43,46]. Selectivity of the enolester 27 was modest towards acrolein (de 60%) [47] but excellent (de >97%) when the dienophile was a naphtho-

264 (a)

7.1 Asymmetric Induction in Cycloadditions

8,

Me

BF3.Et20

C€€$12

Ph &NJ'C02Me

(b)

-78OC

*

de > 92% de > 92%

=OR*

,OMe

O b P h

0

de 84% Fig. 7-4. Examples for the use of chiral auxiliary groups in the Diels-Alder reaction (11).

quinone [48]. The enolether in Fig. 7-4(c) reacted with a variety of dienophiles with almost complete trans selectivity [49]. Diels-Alder reaction under asymmetric catalysis was pioneered by Guseinov et al. [50]. Selectivities are generally poor, but except for the (- )-menthyloxydichloroaluminum-catalyzed addition of methacroleine to cyclopentadiene (98% endo, 57% ee) [51, 521 efficient chiral catalysis has only been an achievement of the past decade. Reacting BINOL (p. 83), (R,R)-1,2-diphenylethanediol [53] or TADDOL (10 on p. 139 in Ch. 5 ) [54] with Ti(OiPr)2C12, or the silylated diol 29 (prepared from Dmannitol) [55] with TiCl, provide catalysts (e.g. 30) which secure good to excellent endo- and enantioselectivities with standard dienophiles or in intramolecular cycloadditions [56] (Fig. 7-5 (a)). One of the several examples is shown in Fig. 7-5 (b) [571. The aluminates 31 and 32 can be used either in the optically active form [58] or as the racemate availing of an interesting trick. Adding namely (+)-3-bromocamphor to the racemate, one of the enantiomers becomes selectively complexed, while the other acts as catalyst (Fig. 7-5(c)) [59].

7.1.1 Asymmetric Induction in the Dieis-Alder Reaction

265

z;3

(31) R = 3,5-Me2-C& (32) R = Ph (33) R = tEbPh,

-Me

de ~ ~ 7 4 % ee of C I S 96% C0,Me

R Me

OMe

Me

+ Me<

(32)+

Me

R C H O r l D - 78 OC

Me,SiO

""R

Me

Me

R I Ph Ph* PhCH=CH-* w1 %de I 8 5 93 96 55 %ee of C I S I92 96 96 86 *With (5')-(32)

hR2

OSiMe,

(S)-BINOL 4A MS, + B(OPh), - 78 OC CH,CI,,

R'

I

Bn R'

I Ph

Ph 2-Pyridyl C&X CHCXl-NVht Me H H Me H O/oeel82 81 90 76 78 85 R21H

Fig. 7-5. Examples for the Diels-Alder addition involving a chiral catalyst.

266

7.1 Asymmetric Induction in Cycloadditions

Complexes prepared in situ from monoesters of tartaric acid with 2,6-dialkoxybenzoic acid and boronic acids [60] or BH3 [61], from BINOL and phenylborate [62], or N-tosyl-D-tryptophane with butylboronic acid [63, 641 efficiently catalyze Diels-Alder cycloadditions, such as the aza-addition in Fig. 7-5 (d) [62]. In the intramolecular version of chirally biased Diels-Alder cycloaddition * the reacting groups are linked by a chain of at least three atoms (e.g. Fig. 7-6(a)) [66, 671.

(a)

Me

Me

PhMe Ph

Me

11ooc

76%de

Ph

Me

I

A

I

(4

I

0

H Me

100% de

C0,Bn -

FON *

: H

CH,CI,; 1) Me,AICI, 2) LiOBn -30 O C

W

\

C

H

Fig. 7-6. Intramolecular stereoselective Diels-Alder reactions.

n

H

N*

20 22 23 24 25

J

% CC

n=l

n=2

conf.

9s 66 90 94 70

80 84 94 88

1 OR 10s 10s 1OR

82

1OR

7.1.1 Asymmetric Induction in the Diels-Alder Reaction

267

Here and in example (b) [68] constraints in the substrate only permit an exo approach, which involves less favorable orbital interactions and leads to a transanellated product which in the case of (a) is rather strained. When it is sterically possible, the endo rule prevails, as in example (c), in which a removable chiral auxiliary group also imparts enantioselectivity to the addition [69]. Enantioselectivity is even higher when amides prepared from 22-25 or Oppolzer’s sultam (20) [71] are cyclized (Fig. 7-6 (d)) [70]. o-Quinodimethanes generated from benzocyclobutanes or 1,3-dihydroisothionaphthene-2,2-dioxides are very useful dienes for the stereoselective construction of complex polycyclic molecules. The general scheme [39, 401 and one of its applications are shown in Fig. 7-7(a) and (b) [74].

100 o c

&de

97%

Fig. 7-7. Intramolecular stereoselective Diels-Alder reactions via o-chinodimethanes.

Experimental [75] and theoretical [76] investigations on the addition of acroleine and methacroleine onto 1-(a-methoxyphenyl-acetoxy)-butadiene,however, evidenced a stretched transition state. Shielding of one of the faces of chiral acrylic esters and amides by coordination with the catlyst in Tic&-mediated additions was supported by chemical evidences [77] and X-ray crystallography of a complex [78]. A rationalization of diastereoselectivity in the Diels-Alder reaction by electrostatic interactions was put forward by Kahn et al. [79].

268

7.1 Asymmetric Induction in Cycloadditions

7.1.2 Asymmetric Induction in [3 + 21 Cycloadditions [3 + 21 Cycloadditions are mainly represented by 1,3-dipolar cycloadditions. These involve, as partners, an olefin and a linear system of three atoms over which four n electrons are distributed. In the Woodward-Hoffmann notation this is a n i + n f cycloaddition, which is allowed in the ground state in the suprafacial-suprafacial mode. Additions of nitrile oxides to acryloyl chiral amines, such as Oppolzer's sultam (20) [80], 34, 35 [81, 821 or 36 [83] give 3-substituted isoxazoline carboxylic amides in high ( > 78%) to complete stereoselectivity (e.g. Fig. 7-8(a)) [83]. Nitriloxide addition to O-dimethyl-tert-butylsilyl-a-triethylsilyl-allylic alcohols is anti selective (de 84-88%) [84]. In the hemiacetal lactone 37 the (-)-menthy1 group is a surprisingly effective inducer also in 1,3-dipolar cycloadditions (Fig. 7-8 (b)) [ 8 5 ] . Me

(34) R = Me (35) R = H

Me

' * p ' R-C=N-wO,

0 (R*H = 36)

R*+

LSelectride

I I 0 C R . R

+

R*H

____+

0

R = Me, Ph, t b , Ph, de>98%

(37)

de 100% 0

R = Pr, Bu, Pent, kkx Fig. 7-8. Stereoselective 1,3-dipolar cycloadditions (I).

de 100%

0

7.1.3 Asymmetric Induction in [2+2] Cycloadditions

269

Dichloroketene adds to vinyl aryl sulfoxides with total diastereoselectivity [86, 871 giving after dehalogenation and desulfuration d-lactones (e.g. Fig. 7-8 (c)) 1871. The (E)-sulfoxide gives the (3s)-lactone in > 98% ee too, suggesting a concerted mechanism. In the addition of olefins to nitrones either partner may be chiral. In example (a) in Fig. 7-9 the nitrone [88], while in example (b) the olefin is chiral [89]. In example (a) total Re/Si discrimination is accompanied by very low endo/exo selectivity. Diastereoselectivity with some nitrones substituted at the nitrogen with carbohydrates was not outstanding either [90 - 921. A rare example of a [3+2] cycloaddition involving only carbon atoms is shown in Fig. 7-9(c) [93].

I

,

dNyH

Ph-

oPo2R* +

Ph

(3R.5s)-C ~ S(38% ) loo"/, de

Ph

+

I

I

Pho*co2R*

CO,(-)-Ment

OJkfH

1 Bu

t Bu

.:J-'"

(3R.SR)-rruns ( 6 2 % ) 100% de Me I

7-,.To1 $\.

------+

90%deatC-3

0 SOTol

R

(R*H= 20) Fig. 7-9. Stereoselective 1,3-dipolar cycloadditions (11).

7.1.3 Asymmetric Induction in [2+ 21 Cycloadditions [2 + 2 ] Cycloadditions can proceed in a concerted manner under photochemical activation or in a non-concerted way by a stepwise ionic mechanism [2]. Both modes are subject to asymmetric induction.

270

7.1 Asymmetric Induction in Cycloadditions

R\/R

0

Ph

R* = (1)

R

R* = (1) Fig. 7-10. Asymmetric induction in photoinduced [2+ 21 cycloadditions.

A classic of [nf + n f ] cycloadditions, the dimerization of cinnamic esters to stereoisomeric diphenylcyclobutane-dicarboxylicesters (truxinates), has also been realized in an enantioselective and regioselective way, by photolysis of a threitol derivative (Fig. 7- 10(a)). Again, enantioselectivity was higher than diastereoselectivity [93]. The complementary reaction, i.e. the cycloaddition of stilbene to a series of chiral fumarates again to yield truxinates, was also studied [94]. Photoaddition of esters of phenylglyoxylic acid with (-)-8-phenylmenthol or the neopentylether 7 to 2,3-dimethyl-2-butene both ensured high stereoselectivity (Fig. 7-10(c)) [95]. [ 2 + 21 Reactions proceeding by an ionic or other non-concerted mechanism have acquired great importance in the preparation of four-membered heterocycles, for example, plactams. Owing to the prominent importance of P-lactam antibiotics, the pertinent literature is very extensive, only a few examples of which are quoted here. Scharf and his group studied very thoroughly asymmetric induction in the Paterno-Buchi reaction, i.e. in the photoinduced [2+ 21 cycloaddition of electron-rich

271

7.1.3 Asymmetric Induction in [2+2/ Cycloadditions

olefins onto a-ketoesters giving oxetanes with the ester group mostly in ex0 disposition (Fig. 7-10(b-d)) [96]. This work is almost unique inasmuch as that not only a wide range of substrates and inducing groups have been covered, but also the temperature dependence of the reaction providing thermodynamic parameters of the reactions has been measured. All this led to the recognition of the isoinversion principle (see Chap. 1, p. 32) [97]. Azomethines and ester enolates undergo [2 + 21 cycloaddition, and, when one of the partners is chiral, a product non-racemic in the /3-lactam ring may be obtained. In example (a) in Fig. 7-1 1 the amine component of the azomethine [98, 991, whereas in example (b) the ester component, is chiral [IOO]. If in the latter reaction instead of tBuMgCl Et,B is added, the cis isomer is obtained in - 90% de. In the TiCl,-cat-

/="

R

+

Me+r

Me

OSiMe3

TiCI4 CHICII, - 7 8 T

*

k*

R+Me

Me

R

R*

% de

iBu Et PI iBu

(S)-PhMeCH(S)-MeO*CiPrCH(S)-Me02CiPrCH(S)-Me02CiPrCH

78 94 90 98

. 'C6&-4-OMe dc > 90%

Fig. 7-11, Stereoselective formation of 4-membered heterocycles by [2 + 21 cycloaddition.

272

7.1 Asymmetric Induction in Cycloadditions

alyzed reaction the favored transition state can be envisaged as in 38. E and Z isomers of the azomethine are at equilibrium at room temperature, while at -78 "C the E form becomes exclusive [99]. Lewis acid catalysts prepared in situ from TADDOL (10, p. 139 in Ch. 5) and Ti(OiPr),Cl, proved also useful in [2+2] cycloadditions (Fig. 7-12(a)) [loll. By comparing a series of cinchona alkaloids as catalysts for the [2+2] cycloaddition in Fig. 7-1 1 (c) [ 102- 1041 it was found that the chirality of the product in excess was determined by the configuration of the catalyst at C-8.

0

n R

'

V

q 0

N

y

TADDOL; Ti(OlPr)2C12 PhMe + petr. ether, 4A MS)

O + R*-C=C-SMe

0

R1 = fl C02Me, R2 = q Me, WI, ckx, de, ee >98%

Me J

w 3

N tBu

H 1) sonication 20 oc; 2 ) y o

MeCO, -78 OC PhCOMe, CY=CH2, CH,Cl,, hv

M e$$

0

1Pr

(39)

M ,e?$

H@

Me0,C

0

lPr

de 85%

Fig. 7-12. Stereoselective formation of 4-membered carbocycles by [2+ 21 cycloaddition.

4 .

Me

7.1.4 Stereoselective Carbene Additions

273

Intramolecular [2+2] addition to a chiral imminium salt is exemplified in Fig. 7-1 2 (b). Photoaddition of ethene to the (S)-valinol-derived oxazolidine 39 seems to defy steric hindrance (Fig. 7-12(c)) [106].

7.1.4 Stereoselective Carbene Additions * Interest in the stereoselective additions of carbenes to olefins to form a cyclopropane ring was stimulated by a demand for chrysanthemic acid and its analogues which are building blocks of environmentally safe pyrethroid-type insecticides. The only industrially viable method to synthesize these compounds is still the addition of a carbene generated from ethyl diazoacetate to hexa-2,4-dienes. This reaction is catalyzed, for example, by salicylaldimine-type copper(I1) complexes, and in order to obtain an optically active product either a chiral ester, or a catalyst prepared from a chiral amine has to be used. The use of 2-phenylethylamine as the amine component in the copper(I1) complex [I081 or of diazoacetates prepared with simple chiral alcohols in the presence of copper(1) chloride [ 1091 gave discouraging results. Selectivities were much better in a study involving various amino sugars and aldehydes [IIO]. The optimum was attained with the Schiff base prepared from 2-pyridine aldehyde and the rather inaccessible 2-amino-4,6-benzylidene-2-deoxy-a-~-allopyranoside, which provided, in the addition of ethyl diazoacetate to 1,l-dichloro-4-methyl-1,3-pentadiene, 80% of the cis product in 44% ee and 20% of the trans product in 38% ee. Following moderately successful studies by several authors Aratani et al. investigated a series of chiral binuclear copper(I1) complexes and selected 40 (prepared from (R)-alanine) as the best catalyst [I 1I]. It was found that the structure of the alcohol component of the diazoester also played an important role in the control of both dia- and enantioselectivity [ 1 121 and the bulky diisopropylmethylcarbinol gave the best result (Fig. 7-1 3 (a)). Application of the method to the synthesis of the commercially important cis-permethric acid (R' = Cl) was less successful [113]. 41, in turn, could be obtained rather selectively (Fig. 7-13(b)) and was smoothly converted to cis-permethric acid by dehydrochlorination [ 1131. Addition of (-)-menthy1 diazoacetate to some simple olefins was also rather selective, especially when (S)-40 or the corrin-imitating copper complex 42 prepared from pyroglutamic acid [114] was used as catalyst [I 131 giving the (IR,2R)-trans products in excess. In an intramolecular version of the reaction Rh(I1)-methylpyroglutamate

* For a review see ref.

[107].

274

7.1 Asymmetric Induction in Cycloudditions

M5=T

Mck4Me CO, Me(iPr),

N,CHC02CMe(tPr)2

Me

Me

~

(A? -( 40)

84% de ee of fruns 8 1%

Me

Me

c13cy

( b)

c'scwMe NZCHICOzEt

(s)- ( 4 0 )

Me

C02Et

,

cislrrans 85 : 15 c i s : 91% ee

Me

(411

R

H 1 B u P h h M e H

Fig. 7-13. Stereoselective carbene additions (I).

was used as catalyst (Fig. 7-1 3 (c)) [ 1 151. Rh2 (OAc),-catalyzed addition of diazoacetatic acid 2,6-t-butyl-4-methylphenyl ester to simple olefins is trans selective (de 70-89%) [116]. Cyclopropane aldehydes of high optical purity ( > 90%) were prepared via the sequence of reactions exemplified in Fig. 7-14(a) [117]. Note that oxazolidine formation was also highly diastereoselective. Addition of N2CHCH(OMe)2 to the same substrate, followed by photodecomposition of the isolable pyrazoline-type intermedi-

275

7.1.4 Stereosefective Carbene Additions

(R*H = 20)

)cCHzBr /co2R* d-

BrH2C

H2C\

COzR* Oct~MeN~Clyhexane 50% NaOH, H20

C02R* 96% de

Me

R* =(-)-

8 - phenylmenthyl

(1H;2H)

82% de

Fig. 7-14, Stereoselective carbene additions (11).

ate and hydrolysis gave the corresponding cyclopropane dialdehyde derivative in more than 90% ee. Palladium-catalyzed methylation with diazomethane is another example for the usefulness of sultam 20 (Fig. 7- 14 (b)) [ 11 81. A rather special reaction leading to vinylcyclopropanes with high stereoselectivity was reported by Quinkert et al. [ 1 19, 1201, who used the products as starting material for the total synthesis of 19-nor-steroids (Fig. 7-14(c)). Phosphoranes can transfer a C1 unit to activated olefins, and relatively high diastereoselectivities could be realized with chiral fumarates (Fig. 7-14(d)) [I 211. In a stereoselective version of the Simmons-Smith cyclopropanation of 2-cycloalkenones of a wide range of ring sizes Mash et al. selected 2,3-di-O-benzylthreitol as an effective ketal forming chiral auxiliary [122, 1231. The fact that 1,2-diphenyl-

7.1 Asymmetric Induction in Cjdoadditions

R

R

i "

b n

R

R

I

C%Oh

I

Ph

R = Me, Pr, iPr, h, ee 88 - 90%

(4 cH212, E t G , cat.* R4*oH CH&12 - h e m e , - 23

\/'''llNHSO,C,H,

OC

*%' OH

4-N02

Fig. 7-15. Stereoselective cyclopropanation.

ethanediol gave even better selectivities [ 124, 1251 contradicted an earlier hypothesis [ 1261 that the reaction is chelation controlled (Fig. 7-1 5 (a)). Selectivity with analogoues 3-enone and aldehyde ketals was poor [122]. Ketals of a,punsaturated aldehydes with (R,R)-diisopropyltartrate can be methylenated with 91 - 94% diastereoselectivity [ 1271. Chiral ketals of cycloalkanones can be subjected to methylenation after conversion to the enolethers (Fig. 7-15 (b)) [128]. Asymmetric induction by the chiral iron center is also effective (Fig. 7-15 (c)) [129]. In contrast to photoaddition of ethene, methylenation of some analogues of 39 (R = H, Ph, C02Me) with CH2 = S(0)Me proceeds with the expected trans selectivity (de 93-98%) [130]. Methylenation of a dianion generated from di-(-)menthylsuccinate with CH2BrCI gave (1 S, 2s)-cyclopropane dicarboxylate ester in 99% de [131].

7.2 Chirality Transfer in Sigmatropic Rearrangements

277

Recently an example for methylenation under chiral catalysis has been reported (Fig. 7-15(d)) [132].

7.2 Chirality Transfer in Sigmatropic Rearrangements The concerted transposition of a (7 bond from an atom next to the end of a 7c system to the other end of the 71 system coupled with the transposition of the 7c bonds is a sigmatropic rearrangement. The rearrangement may involve a single 7c system,

he

[ 1,5] shift

(b)

Me

[ 2,3] rearrangement

(d ) A

Me

0

ene reaction

Fig. 7-16. Examples for different types of sigmatropic rearrangements.

0

278

7.2 Chirality 7ransfer in Sigmatropic Rearrangements

as in example (a) in Fig. 7-16, or two n systems simultaneously (b). The figures also illustrate the nomenclature used to classify sigmatropic rearrangements. The ene reaction, which is a hybrid of a Diels-Alder reaction and a [ 1,5] hydrogen shift (d), will also be discussed here, as well as a family of concerted multiple cyclization reactions involving polyenes. For stereoselective synthesis [2,3] and [2,5] rearrangements and the ene reaction have been exploited. An essential feature of these transformations is the intramolecular transfer of chirality from one center to the other. Except for chiral sulfur centers, the moiety donating chirality is usually not removed from the molecule afterwards.

7.2.1 [3,3] Sigmatropic Rearrangements [3,3] Sigmatropic rearrangements have been shown to be generally concerted, involving a chair-like transition state, as shown in Fig. 7-16(b) [133]. A boat-shaped transition state would lead either to (Z,Z)- or to (E,E)-2,5-heptadiene [134]. A chairlike transition state only provides for diastereoselectivity but not for enantioselectivity, since with chiral substrates two competing diastereomeric chair-like transition states such as A and B can be envisaged. Efficient transfer of chirality from one center to the other was first demonstrated by Hill and Gilman for the Cope rearrangement (Fig. 7-17 (a) and the favored pathway was distinguished by an equatorial phenyl group [235]. Note that if we take the chair-like transition state for granted, an experiment with racemic starting material would yield the same information, since an E configuration of the double bond is a marker for transition state A and Z for B. When at least one of the n systems is part of a ring, only suprafacial transposition of the bonds is possible, and, as far the cyclic part of the system is concerned, complete transfer of chirality can be anticipated when the reaction is concerted. This was shown to be valid for the Claisen rearrangement and its thio and aza analogues (Fig. 7-17(b)) [136]. Among [3,3]sigmatropic changes, the Claisen rearrangement has the most relevance to practical stereoselective synthesis. * In the original version an ally1 vinyl ether was transformed thermally to a y,&unsaturated carbonyl compound. The reaction was shown to be concerted and to proceed through a chair-like transition state. The observed stereoselectivities can usually be adequately rationalized by assuming preference for such transition states which minimize 1,3-diaxial interactions [138]. Efficient conservation of chirality in 0-Claisen rearrangements was demonstrated for simple models by Hill and coworkers. Interestingly, E selectivity was higher than

* For reviews,

CJ

refs. [137-1401.

7.2.1 [3,3] Sigmatropic Rearrungements

r

1

H

"

W

P

Me

219

Ph ,Me

h

* Corrected for the optical purity of the starting material Fig. 7-17. The stereochemistry of [3,3] sigmatropic rearrangements.

enantiomeric purity, which may indicate that the chair-like transition state was not exclusive. When the allylic alcohol was reacted with triethylorthoacetate or N,N-dimethylacetamide dimethylacetal, the rearrangement product was an ester or an amide, respectively (Fig. 7-18(a)) [139]. Using a vinyl ether as the two-carbon fragment, an aldehyde was obtained (Fig. 7-18(b)) [140]. Thermal and catalyzed rearrangement may lead to opposite relative configuration (Fig. 7-18(c)) [141]. The method has been exploited for the synthesis of the side chain of tocopherol (cf. Fig. 1-32) [140, 142- 1451. Under the action of a strong base esters can be converted to ester enolates, and when this is done with an allyl ester a system amenable to Claisen rearrangement is formed. In fact Ireland et al. found [I461 that on treatment with LDA allyl esters readily rearranged to y,&unsaturated acids. The lithium enolates could be stabilized by silylation, and the resulting ketene acetals reacted in the same way. Both double bond formation and single bond formation are stereoselective, and product configurations can be predicted on the basis of a chair-like transition state model (Fig. 7-18 (d)) i.e. (E)-allylesters giving syn products, the Z isomers anti products. Rearrangement of the Z isomer to an anti product is usually less selective (de 50%). Interestingly, on conversion of the lithium enolates to the tBuMezSi ethers selectivity decreased with the E substrate (87% anti) and increased with the Z substrate (89% syn).

280

7.2 Chirality Transfer in Sigmatropic Rearrangements R I

Me

Me

MeC(OEi)a or MeC(0Me)zNMel

*

Me

R

R

de 76%

de 76%

(E)

A -ktOH

(€3)

syn

Fig. 7-18. Stereoselectivity in the Claisen rearrangement (I).

Note that since the allylic double bond is fixed, synlanti selectivity is governed by the configuration of the enolate bond, and this was found to be solvent dependent: by adding to T H F 23% of HMPA, a solvent coordinating with Li', selectivity was reversed and 87% of syn product was obtained from the @)-ally1 ester. Except for the anion derived from the (,!?)-ally1 ester in pure THF, yields are much better with the silylated ketene acetals, as also proved for a-hydroxy and a-benzyloxy esters [147, 1481. The predominant diastereomer can again be changed by adding HMPA to the solvent during the formation of the enolate anion [149].

7.2.1 [3,3] Sigmatropic Rearrangements

28 1

(4 1) ~ ( M ~ ? S I ) ~ N THF. L I , -78 2). Me,SiCk 3) 25 O C

O

Y 0

O

OC ~

OSiMe3

H

98% g n 9 I% ee

( H ) (91% ee)

I) -20 O C 2) H,O

Ho2C+pMe

--CyCI,,

LR' b

(44). 1Pr,NEt 0 -78 O C

(44), Ek,N PhMe + h e m e , -78

OC

1) -20 O C 2) 5 0

R2

syn

anti

Me Fii Me €3 R2 Me Me Ph Ph Ydc 98 96 82 54+ %ec of anfr >97% *Opposite confguration

B' o/udc/80 78 96 96 %ee of syn >96%

r

PhNSiMe3

1

R IMe Ek 1Pr t h Ph Bn %de( 80 86 92 100 72 92 Fig. 7-19. Stereoselectivity in the Claisen rearrangement (11).

282

7.2 Chirality Pansfer in Sigmatropic Rearrangements

a)

qT:,

1) pTosC1, 2) B u 4 THF

- 78 O C , 3) decalme 185 O C

R2

R4 R'

H H H CyPh Me CyPh

R2 R3 H H H Me Me H H H H H Me H

R4 Yde Ref. Me H Me H H H

82 74 70 94 94 94

160 160 160 162 162 161

WI Me,)(,/

Me

A=

1) LDA LDA THF, THF - 78 78 O C 2) Me,SiC, Me,SiCI; 3) J ) xylene, x 135 O C

*

99% de

Me

Fig. 7-20. Stereoselectivity in the aza-Claisen rearrangement.

Excellent to complete transfer of chirality at C-1 was experienced in the rearrangement of the I-silyl [150], and 1-alkyl-substituted ester enolates (Fig. 7-19(a)) [I51 -1531. With allylesters of b-hydroxyacids stereodirection in respect of hydroxy and carboxy groups was complete (like I ) , while at the newly formed bond moderate (de 66-78%) [154]. As a removable and recoverable chiral auxiliary the diazaborolidine 43 introduced by Corey et al. proved to be useful and enabled the generation of both E and Z ester enolates (Fig. 7-19(b)) [155]. Usually ally1 vinyl thioethers derived from thioamides cannot be isolated because they undergo a spontaneous and highly stereoselective thio-Claisen rearrangement (e.g Fig. 7-19 (c)). With tertiary thioamides the N-silylation step drops out, but diastereoselectivity is less complete [ 1561. Alkylation with (Z)-crotyl tosylate gives syn products with 99% selectivity. Tamaru et al. have shown that stereoselectivity corresponds to the diastereomeric purity of the thioamide anion [157]. The reaction was extended to dithioesters (Fig. 7-19(d)) [158]. An aza-Claisen rearrangement proceeding with moderate stereoselectivity was reported by Kurth and Decker (Fig. 7-20) [159- 1621. The silylated enolate of an N-(E)-crotyl-propionamide amide undergoes Claisen rearrangement, but at higher temperature than the ester enolates (Fig. 7-20(b)) [163].

7.2.2 /2,3] Sigmatropic Rearrangements

283

A new feature in stereoselective Claisen rearrangement is activation by a chiral Lewis acid [164].

7.2.2 [2,3] Sigmatropic Rearrangements [2,3] Sigmatropic changes involve six electrons and a five-membered cyclic transition state. Of the six electrons two each are provided by a double bond, a 0 bond, a non-bonding pair of a carbanion or a heteroatom. A [2,3]sigmatropic change involving a carbanion is the Wittig rearrangement * of allyl ethers of the type 44, where R is a group such as an olefinic or acetylenic bond, C02R, or SnR,, facilitating carbanion formation (Fig. 7-21 (a)). In the later case diastereoselectivity becomes only apparent in the presence of an a-substituent at either end of the migrating groups, as e.g. in Fig. 7-22 (a) [I671 or in Fig. 7-22(b) where stereoselectivity is complete regarding both stereogenic elements [ 1681. In Fig. 7-21 (a) the reaction is illustrated for some simple crotyl ethers [167, 169- 1721. Stereoselectivity has been rationalized by considering envelope-shaped five-membered cyclic transition states, of which those leading to the minor products are destabilized by a “pseudo-1,3-diaxial” interaction (Fig. 7-21 (b)). It has also been suggested that in the case of bulky R groups, this preference diminishes due to R +, Me gauche interactions for the E but not for the Z substrates [170]. Selectivity is reversed with the bulky 3-tert.-butyl-dimethylsilyl enolethers which give from Zrich substrates the anti products in excess (de 64-96%) [173]. It has been shown independently by Hoffmann [I741 and Tomioka [I751 and their coworkers that secondary lithiated carbon atoms, being configurationally stable at low temperature, suffer inversion during rearrangement. While the above examples only demonstrated diastereoselectivity in [2,3] sigmatropic rearrangements, moderately efficient transfer of chirality was found in the transformation of a macrocyclic allyl ether (Fig. 7-23 (a)) [I761 and of a chiral ammonium ylide derived from (S)-prolinol [177]. 0-Ally1 glycolic esters or amides readily undergo [2,3] Wittig rearrangement to give 2-hydroxy-3-substituted-bpentenoic esters (see later). ( E ) -or (Z)-crotylethers of

* For

a review see refs. [165, 1661.

284

7.2 Chirality Transfer in Sigmatropic Rearrangements

Me

Me

A WPh

4

R

72 95

Ph

14 12

0

100

Me

GCH

99

m

98 65

J. syn

T

0 Me

H

Jbo

Fig. 7-21. Stereoselectivity in the Wittig rearrangement (I).

285

7.2.2 [2,3] Sigmatropic Reurrungements

4 EuLi, THF

MOMO

HMPq -78

OH

OC

de 96%

MOMO

Me 100% k ee 92%

0

OMOM

R1 )

BULL -100 o c oc

R'

*

cp2zrc1,,-100

1) LDA, -100 o c 2) CpgrCI,, -20 o c

R2

OH 1

*

COzlPr

R

Me

100% Z, de syn >96%, ee .syn >96%

C0,H

mA,m -78 OC

Fig. 7-22. Stereoselectivity in the Wittig rearrangement (11).

*

H

286

(b)

7.2 Chirality Dansfer in Sigmatropic Rearrangements

OH

,NMe2

Me

-

1

0 \\c ,N Me 2

d i Bu

-

-

M

e

9 2 % ee

H

(S)-(E)

Fig. 7-23. Stereoselectivity in the Wittig rearrangement (111).

the chiral amide 45 rearranged on transformation to a zirconium enolate with excellent selectivity (Fig. 7-22 (c)) [ 1781. Syn- and enantioselectivity in the rearrangement of (E)-crotylethers of glycolic acid 8-phenylmenthylester as the lithium enolates was also high, but poor with the allyl and methallyl analogues [ 1791. In contrast to systems discussed before a-alkyl-(E)-crotyl glycolic esters give, via the zirconium enolates, (Z)-syn products (Fig. 7-22(d)) [I 801. With substituents at other positions stereoselectivity is variable [ 1811. An explanation for this dichotomy based on ab initio calculations and con formational arguments was provided by Houk et a/. [I821 and by Buchner et a/. [183]. [2,3] Wittig rearrangement of propinyl glycolates [I841 is a good method to prepare chiral allenes (Fig. 7-22(e)). Using instead of a carbonyl group SnBu3 as activating group, the corresponding hydroxymethyl compounds are obtained. A [2,3] sigmatropic rearrangement which was supposed to proceed through a carbene intermediate was used by Chan and Saucy to prepare chiral isoprenoid synthons (Fig. 7-23(e)) [185]. As shown by Hoffmann et a/. aryl-vinyl sulfoxides with a /3-substituted vinyl group are converted on treatment with a strong base, to allyl aryl sulfoxides which rapidly rearrange to sulfinates of an allyl alcohol with fairly efficient transfer of chirality (Fig. 7-24) [187- 1891.

7.2.3 Ene Reactions

0 :11,,,

To1

R

sI

‘

287

O

U

P

R

DMF 9 0 - 120OC

TolSO

%ee (conf.)

I

87 (S)

84 ( R )

89 (S)

82 (R)

Fig. 7-24. Stereoselectivity in the rearrangement of allylsulfenates.

Transfer of chirality in the thermal rearrangement of sulfinates to sulphones was studied by Hiroi et al. 11901, who established that the preferred transition state was again the one in which the bulkiest group was quasi-equatorially disposed (Fig. 7-24 (b)).

7.2.3 Ene Reactions * The ene reaction is a hybrid between a Diels-Alder reaction and a [I ,5]sigmatropic hydrogen shift comprising the addition of an olefin containing an allylic hydrogen atom to a n bond, as shown in Fig. 7-25 (a). It can be regarded as a (of + 7-r: + n f ) reaction and is thermally allowed in a concerted suprafacial fashion [2]. Asymmetric induction in the ene reaction was first demonstrated by Hill and Rabinovitz [192], and the experiment supported its concerted nature. As the DielsAlder reaction, it can be accelerated by Lewis-acid catalysis. In the uncatalyzed reaction of chloral with (-)-P-pinene the diastereomer with an R configuration in the side chain was obtained in 64% ee. Stereoselectivity was both enhanced and inverted on addition of TiCl, (e.g. Fig. 7-25 (b)) [193]. Due to complexation of the carbonyl group with the catalyst the bulky CCl, group was forced into the unfavorable endo disposition.

* For a review

see ref. [I911

288

7.2 Chirality Transfer in Sigmatropic Rearrangements

TICI,

*

100% de

H CHz

.OH

CI3C

A 100%) de

R*OL

P

0 II

OH

R*0+H

R*- (- -)-8-phenylmenthyl

OH

Fig. 7-25. Stereoselectivity in intermolecular ene reactions.

r

9 8 % de

289

7.2.3 Ene Reactions

The intermolecular ene reaction involving a triple bond was successfully exploited by Dauben and Brookhart for the stereocontrolled construction of steroidal side chains (Fig. 7-25 (c)) [ 1941. The E olefin afforded the 2 0 s diastereomer [ 1951. Appreciable diastereoselectivity has been achieved in the Lewis acid-catalyzed ene reaction of a-substituted acrylates, but asymmetric induction using chiral acrylates is poor (Fig. 7-25(d)) [196]. Note that the reaction was regioselective for attachment to the less substituted terminal of the olefin. With (E)-2-butene as partner, diastereoselectives were similar. Efficient asymmetric induction was reported in the ene reaction of chiral glyoxylate esters with olefins by Whitesell et al. [40, 197, 1981, as exemplified in Fig. 7-25 (e). Lewis acid-catalyzed ene reaction of a-chiral aldehydes (a-OBn [ 1991, a-Bn2N [200]) with simple olefins, as well as of silylcrotylalcohol with methylglyoxylate I2011 proceed, in turn, with high syn selectivity.

x

D

’0

O-k-OMe

EtlAlCl

-7vc

(3S.4R)

Fig. 7-26. Stereoselectivity in intramolecular ene reactions (I).

or&

290

7.2 Chirality Transfer in Sigmatropic Rearrangements

One of the early examples for the utilization of an intramolecular ene reaction was the ingenious synthesis of chiral acetic acid by Arigoni and his coworkers (Fig. 7-26(a)) [202]. Here, in the first step, only the Z diastereoselectivity of the ene reaction was exploited, the configuration at C-3 was not checked and was in fact irrelevant. Chirality transfer in the second step, a retro-ene reaction involving suprafacial deuterium migration, completed the formation of a chiral methyl group of predictable configuration. Intramolecular ene reactions became a powerful tool in the synthesis of natural products in the hands of Oppolzer and his coworkers [15, 731. The basic idea of this approach is to link the ally1 group and the acceptor double bond by a chain which should be long enough to allow the proper mutual disposition of the interacting groups, but not too long in order to lend sufficient constraint to the system. When the chain contained three atoms and a (Z)-ally1 enophile, the reaction was completely diastereoselective, while with (E)-ally1 enophiles the cis product was not exclusive due to competition of two diastereomeric transition states (Fig. 7-26(b)) 1731. The method was employed for the synthesis of cyclopentanoid sesquiterpenes [203, 2041. With 1,2-disubstituted ene groups the stereochemical result was indifferent to olefin configuration (E or Z ) [73].

C02Me

60,Me

R = Me, de (1,5-cis) 93%

(46)

70°C

+?

IlO'C

MgBr

(4 Rl

"qR2 & >95%

p ~ p ~ ~ p 1 4

\$"

\

RO Fig. 7-27. Stereoselectivity in intramolecular ene reactions (11).

x=o,cH2 R1 = Q R2 = Bx,CYOW

Rl = Hey CYOW, R2 = H

291

7.2.3 Ene Reactions

Lewis acid catalysis opened up the way to the trans series (Fig. 7-26(c)) [205, 2061. The correspinding (E)-1,6-hexadiene also gave mainly the trans product (trans/cis 89: 1 I). Finally, the problem of enantioselectivity was solved by subjecting chiral (e.g. (-)-8-phenylmentyl) esters to Lewis acid-catalyzed cyclization [207,208]. A chiral center adjacent or in ,!?-position to the ene function can also effectively control enantioselection, as demonstrated by the cyclization of 46 in which only 1,t-trans isomers were formed (Fig. 7-27 (a)) [209]. The scope of the stereoselective intramolecular ene reaction can be extended to substrates in which, instead of a hydrogen atom, magnesium halide is transferred from one end of the system to the other. The intermolecular version of this reaction suffers from poor regioselectivity, but Felkin et a!. found surprising selectivity for

19, Me

Me

Me

2,2-bipyridme-Fe( C&, 25 O C 0) w

O+HPh

M Me

e

n

o

f

i

P

h

0

Me

de > 90%

0

(1pr0),TiClz + (R)-BINOL CHzCl, -70 -30 OC

H I I \ ~ ~ z ~ e-

C0,Me

R ’ H H M e Rz Me Ph Et %ee 95 97 98 Fig. 7-28. Stereoselectivity in intramolecular ene reactions (111).

(W4 (

8

0 5 98

292

7.2 Chirality Transfer in Sigmatropic Rearrangements

an intramolecular magnesium-ene reaction (Fig. 7-27 (b)) [210]. A formal analogue of the latter reaction is the Pd(0)- [211] or Ni(0)-catalyzed cyclization of 1-acetoxy-2,7-dienes (Fig. 7-27(c)) [212]. So far only such ene reactions have been considered in which the ene component is linked to the migrating group in y-position. Compounds in which the same is attached to the P-position can also undergo an ene reaction, and several examples for this involving the transposition of magnesium have been reported. The reaction is regioselective inasmuch as magnesium always migrates to the remote end of the double bond, and the smaller of the two possible rings is formed (Fig. 7-28 (a)). Further C -C coupling occurs with the more substituted end of the allylmagnesium halide system. (The allylmagnesium moiety undergoes rapid 1,3-metal migration.) The reaction was found to be remarkably stereoselective, which is in agreement with a concerted reaction, as shown in Fig. 7-28(b) [213]. A special ene reaction between a diene and an unreactive dienophile involving six n-electrons is shown in Fig. 7-28(c) [214]. Chiral Lewis acid catalysts, such as BINOL-TiCl, [215, 2161, BINOL-Zn [217] and the aluminate 33 [218] (e.g. Fig. 7-28(d)) [215], proved to be useful also in ene reactions involving aldehydes.

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

8 Stereoselective Formation of Carbon-Heteroatom Bonds

In this chapter the stereoselective, mainly enantioselective, formation of C - N, C-P, C - 0 and C-S bonds, as well as stereoselective protonation will be briefly discussed. Although the stereoselective formation of carbon-metal bonds is an important field of research, owing to its special methodology this topic will not be covered. Some aspects of the stereochemistry of organic silicon, tin and titanium compounds have already been mentioned in Chapters 5 and 6. Despite their importance, stereoselective formation of halides, as well as the ring opening reactions of epoxides will be omitted also, since these can be readily interpreted and devised on the basis of the well known principles of classical mechanistic organic chemistry.

8.1 Stereoselective Formation of Carbon-Nitrogen Bonds Except for the trivial case of SN2 substitution, general methods for the stereoselective linking of nitrogen to carbon have only been elaborated in recent years. Early attempts for the addition of achiral amines to chiral activated olefins or of chiral amines to achiral activated olefins resulted in poor selectivities [I -41. By now, however, some examples for highly diastereoselective P-additions of amines to a,P unsaturated esters [ 5 ] , or lactones [6] emerged, e.g. [7]:

de - 100%

294

8.1 Stereoselective Formation of Carbon-Nitrogen Bonds

Addition of a chiral lithium base was completely enantioselective [8]: Me Me

R*

'N'

C0,tEU

+ b

+

Ph

I

IJ

- 78

OC

Me

Bn H,. Pd(OH)* EtOH

j\/C02th

Base-catalyzed intramolecular Michael addition of the carbamates 1 leads, depending on double bond configuration 191 and the position of the carbamate group to 3,4-syn- or 3,4-anti products [lo- 121: OR Me

Me

C0,Et

de > YO%

0

de 94%

Hayashi et al. and later others [I31 found that the palladium complex of diphosphine ligands (e.g. 2) catalyzed both intramolecular [ 141 and intermolecular enantioselective /3-addition of amines to allylic esters with concomitant allylic rearrangement 1151: Me

Simple olefins can be activated by complexation. Thus Panunzi et al. found that amine addition to olefins activated by a chiral platinum complex was highly enantioselective (Fig. 8- 1 (a)). The related palladium-assisted oxyamination was shown by Backvall and Bjorkman to be highly diastereoselective with (E)- and (2)-butene and moderately so with I-phenyl-I-propene (Fig. 8-1 (b)) [26]. Stereoselective hydroxyamination of enolates with a-chloronitroso compounds can be performed in two ways: either the enolate [I71 or the reagent [18] can be chiral:

295

8.1 Stereoseiective Formation of Carbon-Nitrogen Bonds

HO~C-R

R = Me. iPr. alhl. 1w1 Ph. Bn. de >94%

Enolates of chiral amides [19-211 and silylketene acetals of chiral alcohols [22, 231 add with high selectivity to azodicarboxylate esters giving the a-hydrazino derivatives and ultimately a-aminoacids, e.g. [19]. 0

0

I

11

‘AN/\, &‘.\

u

1) LDA, -78 O C 2) tho,-N=N-C0,tBu

*

Y

R)+R** N o w ‘Boc

B

BocNH/

( 3)

N

co2wI

‘Boc

R = Me, IPr, t h , ally\ Ph, Bn, de 94-98% Transformation of an allylic acetate into an amine proceeded with complete retention of configuration when it was catalyzed by a Pd(0) complex bound to silica gel (Fig. 8-1 (c)) [24].

A--H

CI

Me

R2NH

CI

Et H

R2N

Ph

R=Et, 95% ee R=H, 97%) ee

Et

(b) 1) (PhCN)2PdCI2, THF 2) MezNH, -5OoC, 3 ) AcOH, NBS

Me-Me

(c)

*

C02Me

C02Me

> 9 8 % syri

OAc

Et2NH, T H F Pd(0)- Si 0 2

(4

c, Me

,I

I

IK1,

+ CRXN,

H,NOSO,H

Fig. 8-1. Stereoselective formation of carbon-nitrogen bonds (I).

,, iN\itlCEkx

296

8.2 Stereoselective Formation of Carbon-Phosphorus Bonds

Boronic esters [25] and dichloroboranes [25, 261 can be transformed to amines with the preservation of their configuration and enantiomeric purity by reaction with H,NOS03H and alkyl azides resp. (Fig. 8-1 (d)).

8.2 Stereoselective Formation of Carbon-Phosphorus Bonds Optically active phosphorus compounds are usually obtained either by resolution or from chiral starting materials using transformations proceeding with complete or nearly complete retention and inversion, respectively [27]. Of the few examples, in which asymmetric induction and/or transfer of chirality have been exploited some will be discussed here. Abbott et al. used ephedrine as a chiral auxiliary in the preparation of a monophosphate ester chiral at phosphorus due to the presence of three different oxygen isotopes (Fig. 8-2) [28]. In a similar way, chiral esters of thiophosphoric acid of the type 4 were prepared by Hall and Inch [29]. 2-Phenylethylamine was used as a chiral auxiliary in the synthesis of aminophosphonic acids [30] and cyclophosphamides 13 11 with moderate success. Mc I

NHMe

P'70cI, + P h d M e OH

Et,N

"<~,"~~

Ow1

M

~

&

retention

0 4 -0

Me I

1;

-

OBn

0

II

P.

/ \''''10RZ

OR'

(4 Fig. 8-2. Synthesis of monophosphate ester chiral at phosphorus.

~Me /\l"O\,/" ~ 170//

\ 16

0

8.3 Stereoselective Formation of Carbon-Oxygen Bonds

297

8.3 Stereoselective Formation of Carbon-Oxygen Bonds One of the most important methods for the enantioselective formation of secondary alcohols is the hydroboration of non-terminal olefins followed by oxidation. The method should have been classified as a stereoselective carbon-boron bond-forming reaction, since stereoselection takes place in the first step, but the intermediate boranes are of little practical interest as compared with the end products, i.e. the alcohols. In 1961 Brown and Zweifel discovered that on reacting diborane with an excess of a-pinene a dialkylborane, diisopinocampheylborane (IPC)2BH (5) was obtained in a completely diastereoselective reaction. 5 proved to be a highly enantioselective hydroborating reagent [32] *. (IPC),BH was found to be most useful with unhindered (Z)-olefins. Reaction with (E)-olefins was sluggish and less selective [33]. The primary products were usually not isolated but oxidized to secondary alcohols with complete retention of configuration. Typical results obtained with optically pure [(-)-IPC],BH [34] or recalculated to such conditions [35-381 are shown in Fig. 8-3. ** Selectivity with trisubstituted olefins is poor.

1 ) [(-)-IPC],BH

OH

P

R~

2) H 2 0 2 , 0 H -

-b

..--OH

R'=R*=Me, Y8% ee R'=RZ=Et, 95% ee R'=iPr, R2=Me, 82% ee

100% trans

Fig. 8-3. Stereoselective hydroboration with diisopinocampheylborane.

* **

For a review, see ref. [39]. The optical purity of natural a-pinene is -95%.

R=CH2C02Me. > 92% ee R=Me, 95% ee

298

8.3 Stereoselective Formation of Carbon-Oxygen Bonds

Monoisopinocampheylborane (IPCBH2) is more reactive but difficult to prepare pure [40]. It is not competitive with (IPC),BH in the case of reactive olefins but exhibits moderate to excellent enantioselection with (E)-olefins (Fig. 8-4 (a)) [41 - 431. With trisubstituted olefins diastereoselectivity is complete in our examples (syn from E, anti from Z ) .

R' /Me tBu R2 H R3 Me tBu R 4 H H %ee 73 92

Ph H Ph H 65

Me Me Me H 53

Me Me Me Ph Ph H H M e 72 81

Me Ph Me H 82

Me Me - ( C q 3 - -(CHa,-

H 68

H 72

Ph

Ph

-(cq3- - ( C q , H 100

H 88

100% ee

Fig. 8-4. Stereoselective hydroboration with monoisopinocampheylborane.

When, instead of being oxidized, the primary adducts were reacted with acetaldehyde, chiral boronic acids were obtained in 70- 100% ee (Fig. 8-4(b)) [44,45]. Boranes prepared from less accessible terpenes, like longifolene [46], (+)-2- or (+)-3-carene 1471 offer no significant advantages over ( -)-IPC2BH. (R,R)- and (S,S)-2,5-dimethylborolane (6)is laborious to prepare but gives excellent selectivities (ee > 97%) even with trisubstituted olefins [48]. Asymmetric induction in hydroboration is only efficient when hindered boranes, such as 1,I ,2-trimethylpropylborane (thexylborane), or 9-BBNH are used (Fig. 8-6 (a)). With free allylic alcohols and thexylborane anti selectivity relative to the hydroxyl group (i.e. syn attack of hydride) [49] was observed, while protection of the hydroxyl group resulted in inverted selectivity [50](Fig. 8-5 (a)). Unexpectedly hydroboration of (E)-2-ethylidenecyclohexanone with Thex2BH is nonselective [5I]. 1-Substituted (E)-crotylsilanes are hydroborated with 9-BBNH with anti selectivity, the Z olefins give the syn products (Fig. 8-5(b)) [52]. Hydroboration with catecholborane can be catalyzed by rhodium-diphosphine complexes, whereby regioselectivity may be dramatically changed (Fig. 8-5 (c)) [53].

8.3 Stereoselective Formation of Carbon-Oxygen Bonds

Me

(a)

299

Mt:

y ' M e 92 90 >94

SiMe2Ph R

A

=

1) 9-BBNH

M

e

2, 4'2,

PhMe,Si

OH'

R

Me

Me

OH

OH 1) 9-BBNq THF ' ' IOH

Rh[Ph,P(CHJ,PPh2] $BF, 8

72% (+ isomers)

2) [OI 83% (+ isomers)

Fig. 8-5. Asymmetric induction in hydroboration (I).

With BINAP as ligand and catecholborane high (82 - 96010) enantioselectivities could be realized in the hydroboration of monosubstituted ethenes [54, 551. Hydroboration of olefins by diborane is generally assumed to involve syn addition and to proceed with high regiospecificity for the less substituted carbon atom. A study by Zioudrou et al. on a series of simple olefins has shown that regioselectivity is 80-100010, and the ratio of syn to anti addition is 87: 3-97: 3 [56]. Hydroboration of a 1,5-diene is illustrated in (Fig. 8-6(b)) [57,58]. The concept of attaining acyclic stereoselection with the aid of a cyclic intermediate was successfully extended to l,Cdienes, but failed with 1,6-dienes [59]. A computational study of stereoselective hydroboration was published by Houk et al. [60]. Palladium-catalyzed oxidative 1,4-acetoxy addition to 1,3-dienes is highly diastereoselective and can be controlled either by reaction conditions [61] or substrate configuration ( E or 2 ) [62]:

300

8.3 Stereoselective Formation of Carbon-Oxygen Bonds

CO,H

LiCI, 4OoC

de >94%

de >94%

OAc

The E, 2-diene is transformed to the (E)-anti product with the same selectivity.

Me

R;I"

(4 OH

R2 -

:

Them2 THF

R2

Bu

W

R

OH

OH

1,2-anh

eMe

Me

H

1 R1 R* % 1.2-anti

1 I

I

Bu

H

H

Bu

92

YO

>Y4

OH SY f l

Fig. 8-6. Asymmetric induction in hydroboration (11).

In halolactonization reactions a halogen atom and an intramolecular carboxy group are added to a double bond in an anti manner to give an a-halogenolactone. Koga et al. recognized that application of this transformation to the proline amides of a,punsaturated acids permitted the highly enantioselective preparation of certain a-hydroxycarboxylic acids [63 - 6.51 (Fig. 8-7 (a)). Unfortunately, no reaction takes

30 1

8.3 Stereoselective Formation of Carbon-Oxygen Bonds (a)

O C O * K

NBS D W , -20

I ) 13u3snH 2) He

OC

RIRZCfl

I

R'

R3

R'

R2

H H H

Me Me Ph Me H Me -(CH2)4Hex Me

22 98 84

H

H

Me

Me H

R'

l a z c ~ o H

D

Me

%ee

90 98

Me

\

C02H 0 70 Y" cis

94% fratis

Me

0

NMe2 R'

Me

H

Me

OH

H

OAc

H

R2

H

Me

Me

H

OH

H

OAc

% 2.4-irans

97

97

>98

20

99

45

99

Fig. 8-7. Enantioselective synthesis of a-hydroxycarboxylic acids by halolactonization.

302

8.3 Stereoselective Formation of Curbon-Oxygen Bonds

place with the crotyl amide, and a 7-membered bromolactone is formed from the 3,3-dimethylacrylamide [65]. Treatment of the bromolactones with sodium methoxide and reduction with bis-(2-methoxyethoxy)-aluminum hydride yields chiral a,pepoxyaldehydes [65]. Stereoselectivity of the halolactonization of chiral 4 3 - and 5,6-unsaturated carboxylic acids is sensitive both to reaction conditions and the constitution of the substrate. From acids having a substituent adjacent to the double bond the transhalolactones are formed in excess under equilibrium conditions (no base added), while the cis-products arise under kinetic control in the presence of KHCO, or NaHC0, (e.g. Fig. 8-7(b)) [66, 671. When in the 5,6-unsaturated acid 7 the methyl group was moved to C-3 and C-2, the predominant product became the cis- and the trans-lactone, respectively [67]. Similarly, from 2-substituted 4,5-unsaturated amides and thioamides, trans-lactones and -thiolactones were obtained after reduction (Fig. 8-7 (c)) [68]. The reaction was extended to 4,5-unsaturated alcohols (haloetherification) (Fig. 8-8 (a)) [69]. Asymmetric induction by a hydroxyl group of allylic alcohols was found to be surprisingly effective in iodohydrin formation (Fig. 8-8 (b)) [70].

(b)

OH

R3 H 1120

>

----Rz 1

~

R' Me

R' J syn, syn

Bu tBu

R' Bu g

R3 H

H

Bu

Me H Me

H Me H

%syn, syn

93

~

97 99 77 97

Fig. 8-8. Asymmetric induction in iodohydrine formation.

Recently, Mukaiyama and his coworkers discovered that tin(I1) and tin(1V) esters derived from diols having enantiotopic hydroxy groups can be acylated with considerable selectivity, either with benzoyl chloride in the presence of a chiral base [71] or with a chiral acid chloride [72], but the reactions are of rather narrow scope. More general is the group-selective hydrolysis of symmetrical anhydrides with chiral alcohols [73,74]:

~

8.4 Stereoselective Formation of Carbon-Sulfur Bonds

303 R

'0 %de 90 90

76 66

X -(CH2)4- -CyCH=CHcH, -(CHJ,90 I0 Ydc 99

8.4 Stereoselective Formation of Carbon-Sulfur Bonds Enantioselective Michael addition of thiols in the presence of chiral catalyst has been extensively studied, not so much for the sake of the products, but rather to collect information about chiral catalyst systems. Up to 67% ee were realized by Wynberg and his coworkers in the addition of thiolesters [75]and thiophenols [76]to cyclohexen-2-one and 81 Yo ee in the addition of thiophenol to diisopropyl maleinate [77],in the presence of alkaloids. In the first reaction, as shown by Mukaiyama et al., hydroxyproline derivatives were more efficient [7 81. Thiol addition to open-chain enones and nitroolefins was much investigated but remained only of theoretical interest (e.g. [79, SO]). Since chiral sulfur compounds, primarily sulfoxides, are useful as chiral auxiliaries [81], their enantioselective synthesis is important. The p-tolylsulfinylester of (-)-menthol, one diastereomer of which can be obtained pure relatively easily [82], is a useful starting material since it can be transformed with complete inversion to compounds amenable to versatile transformations (Fig. 8-9(a)) [83]. The racemic sulfinylester 8 underwent enantioselective 1,4-rearrangement to an aalkyl sulfone under chiral catalysis (Fig. 8-9(b)) [84]. Lewis acid-catalyzed ene reaction of a chiral N-sulfinylcarbamate with ( E ) - and (Z)-2-butenes gave pure syn- and anti-sulfinamides as shown next for the E isomer

66

-CH, 40

304

(c)

8.4 Stereoselective Formation of Carbon-Sulfur Bonds

0 -

Ph Me/\\/Me

SKI, -78OC

R*OOCNH Me

de 100%

.t

(9)

R' /Me vinyl %deI 99 96

aUyl 97

Bu tBu 75

98

Fig. 8-9. Stereoselective formation of carbon sulfur bonds.

(Fig. 8-9(c)) 1851. The thiaoxazolidine 9 prepared in good yield from ephedrine can be converted in two steps to sulfoxides. By recrystallization of the intermediate (10) optically pure sulfoxides can be obtained (Fig. 8-9(d)) [86]. In a similar addition of Grignard reagents to sulfinamides of an oxazolidin-2-one derived from norephedrine give sulfoxides in 97 - 99% ee [87].

305

8.5 Stereoselective Formation of Carbon-Halogen Bonds

8.5 Stereoselective Formation of Carbon-Halogen Bonds * While fluorination of enolates with the N-fluoro derivative of Oppolzer's sultam (20, p. 261) gives generally poor enantioselectivities [88], fluorination of chiral amide enolates was satisfactory (Fig. 8-l0(a)) [8Y]. Bromination of silylketene acetals formed with the sulfonamide ZZ (available as both enantiomers) with NBS is also highly diastereoselective (Fig. 8-10(b)) [YO].

R*

L F

-

F

Me

Ph

R*O

,$,

R*O

NEB,M e 0 O-M: H

- 70 O C

,&

R IWI tb W Ph Ydel97 96 89 86

YO

-H

Ek

0

OSMe,

M e ' b

2)NaOMe)

R *

0

ee>96%

R = Me, b,&I, Oct, Ph

R*OH=

Me0,C

C0,Me L.-=

de YO% Fig. 8-10. Stereoselective formation of carbon-halogen bonds.

* For halolactonization see Section 8.3.

de 98%

306

8.6 Stereoselective Protonation and Hydrogen Migration

A similar bromination of the boron enolate of amide 3 (p. 295) was also of useful selectivity (de >90% for R = alkyl) 1911. a-Brornination of ketals with methyl tartrate is diastereoselective (de 80 - 90%) [92, 931 and with l-tetralone the prevailing configuration depends on the reagent (Fig. 8-1O(c)) [94]. Finally, an example for deracemization by HCl addition and enantioselective HCl elimination should be mentioned [95]:

8.6 Stereoselective Protonation and Hydrogen Migration * Although hydrogen is not regarded as a heteroatom, the discussion of protonation and hydrogen migration seems to be more in place in this chapter than in those dealing with reductions. Chiral centers in which one of the ligands is an acidic hydrogen can be racemized by deprotonation-protonation, since carbanions are configurationally unstable. If this process interconverts diastereomers (epimerization), product distribution is almost invariably under thermodynamic control, since even a hydrated proton is very small, and therefore it is difficult to direct its approach to a specific face of the molecule. Despite their importance, such protonation reactions will not be discussed here. Protonation of enamines [97, 981 and enolates [99] with di-O-acyltartaric acids is of poor enantioselectivity, but can be much improved by performing deprotonation with a chiral lithium base. Remarkable enantioselectivities can be achieved in the rearrangement by deprotonation-protonation of symmetrical epoxides to allylic alcohols mediated by prolinederived lithiated amines (Fig. 8-1 1 (a)) [loo- 102). ~

*

For a review on asymmetric deprotonation see ref. [96].

8.6 Stereoselective Protonntion und Hydrogen Migrution

L

" 1

307

1

0

{')Me H2 N

x,,,, C02H

I ) [COCI~L*J@CI" 2 ) MeOH, 65'C, --Co2

. H

Fig. 8-11. Enantioselective protonation.

Deuterium-hydrogen exchange in a-deuterated cyclopentanone affected the pro-S deuterium preferentially when catalyzed by a chiral diamine (Fig. 8-1 1 (b)) [103]. Decarboxylation of a prochiral malonate was found to be enantioselective in the in the presence of a chiral Co(II1) complex. In fact enantioselectivity was due to preferential coordination of one of the enantiotopic carboxyl groups (Fig. 8-12(c)) [104]. Of technological importance may be the asymmetric transformation of ( k)-Nmethylphenylglycine by (S)-camphorsulfonic acid in butyric acid. The (R)-acid of crystallizes out in 85-89% ee [IOS]. The concerted [1,2n+l] sigmatropic shift of a hydrogen atom from one end of a 71 electron system to the other follows the Woodward-Hoffmann rules and is completely diastereoselective, provided that the formation of only one of the diastereomers is symmetry allowed (cJ:Chapter 7). Some interesting examples of hydrogen migration in allylic systems, the mechanism of which has not yet been clarified, will be discussed below. With allylsilanes, Wilson and Price observed effective 1,3- and 1,4-induction in thermally induced or Lewis acid-catalyzed hydrogen transfer (Fig. 8-12 (a)) [ 106, 1071. Photoenolization followed by proton migration of a-substituted a,/hnsaturated esters can be conducted enantioselectively in the presence of a chiral additive [108, 1091, or when the esterifying alcohol is chiral [I 10, 11 I], e.g.:

308

8.6 Stereoselective Protonation and Hydrogen Migration

r

HO

OW?

ec 91%

Lewis acid-catalyzed cleavage of chiral4-alkylcyclohexanone ketals [ 1 121 or enolization of the ketone with a chiral lithium base [ 1131 both lead to chiral enolethers, which can be then transformed to P-substituted dicarboxylic acids.

i

1) B*Li

K

2) Me3SiCI

0

0

OSIMe,

CH2Cl2,-78

OC

OR*OH

Mc

Enantioselective hydrogen migration in an allylamine-enamine transformation catalyzed by a Co-DIOP complex was first reported by Kumobayashi and Akutagawa (ee up to 33%) [I 141. A dramatic increase in selectivity when the rhodium complex of a chiral biphenyl or binaphthyl was applied as catalyst was simultaneously reported by Hansen et a[. [ 1 15, 1161 and by Noyori and his coworkers [ 117, 1181 (Fig. 8-12(b)) *. The (E)-allylamine gave the (S)-enamine in excess [I 171.

* For a rewiev see ref.

[119].

8.6 Stereoselective Protonation and Hydrogen Migration

(a)

Me

309

Me

anti (b)

CHO

Me

RZ

P* = (S)-(Li’), 99% ee P* = (R)-BINAP, 96% ee

(SJ-(i2) R1= Me, R2 = H (&’)-(I$ R1= Me, Rz = NMc, (S’)-(/4) RA = CH,OCH,, R2 = H

Fig.8-12. Stereoselective intramolecular hydrogen transfer.

The method has been extended to other enamines [I201 and additional efficient catalysts (13, 14 [121]) have been discovered. An industrial synthesis of (-)-menthol from citronella] based on the above rearrangement has been realized on a multiton scale [119]. The mechanism of the reaction was studied by Noyori et af. By deuteration experiments suprafacial migration of a hydrogen atom in the S-trans configuration of the CH2NR2 unit was proved. The rate-determining step was hydrogen abstraction in the tricoordinate species [Rh(biphosphine)(allylamine)] [ 1221. +

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

Appendix

Relative prices of some chiral reagents per mole as compared with that of (R,R)-tartaric acid (taken as 1.OO). * D-Alanine L-Alanine ( R)-2-Aminobutane (S)-2-Aminobutane (R)-2-Amino-l -butanol (S)-2-Amino-l -butanol (S)-2-Amino-3-phenyl- 1-propano1 (phenylalaninol) (R)-l -Amino-2-propanol ( S ) -1-Amino-2-propanol (S)-2-Amino- 1-propano1 (alaninol) (R)-I ,I ‘-Bi-2-naphthol (BINOL) (S)-I,l’-Bi-2-naphthol (BINOL) ( R ) -and (S)-2,2’-Bis(diphenylphosphino)-l,l-binaphthyl(BINAP) (-)-Borneo1 (R,R)-2,3-Butanediol (S,S)-2,3-Butanediol (+)-Camphor (-)-Camphor (+)-Camphoric acid D-Camphorsulfonic acid L-Camphorsulfonic acid (+)-Carvone (-)-Carvone (-)-Cinchonidine (+)-Cinchonine L-Cysteine (R,R)-2,3-Dimethoxy-I ,4-bis(dimethylamino)butane (DDB) (S,S)-2,3-Dimethoxy-1,4-bis(dimethylamino)butane (DDB) ( R ) -and (S)-N,N-Dimethyl-I -phenylethylamine (S)-a,a-Diphenyl-2-pyrrolidinemethanol (+)-Ephedrine (-)-Ephedrine

83 12 6 000 5 400 6021 5 409 83 1 356 445 235 7 650 12 440 41 060 12 888 1760 6 130 19 107 43 29 8 52 50 15 1497 5 600 2251 1988 20 18

312

Appendix

Ethyl L-lactate ( R)-Glyceraldehyde (S)-Glyceraldehyde cis-4-Hydroxy-~-proline cis-4-Hydroxy-D-proline trans-4-Hydroxy-~-proline (+)-Isomethol ( +)- and ( -)-Tsopinocampheyl-9-borabycyclo[3.3.l]nonane (Alpine-borane@) (R,R)- and (S,S)-2,3-Isopropylidene-2,3-dihydroxy-l,4-bis(diphenylphosphino)butane (DIOP) (S)-Lactic acid (+)-Longifolene (R)-Malic acid (S)-Malic acid (R)- and (S)-Mandelic acid D-Mannitol (-)-Menthol (-)-Menthone (-)-Menthy1 chloroformate

(4S,5S)-4-Methoxymethyl-2-methyl-5-phenyl-2-oxazoline (R)- and (S)-a-Methoxy-a-trifluoromethylphenylacetyl chloride (- )-N-Methylephedrine (- )-cis-Myrtanylamine

( R ) - and (S)-l-(l-Naphthy1)ethylamine (-)-Neomenthol (R,R )-2,4-Pentanediol D-Phenylalanine L-Phenylalanine ( R ) -and (S)-I-Phenylethanol ( R ) -and (S)-1 -Phenylethylamine (I? )-Phenylglycine (S)-Phenylgl ycine (S)-3-Phenyllactic acid (+)- and (-)-Pinanemethylamine- HCI (+)-a-Pinene (-)-a-Pinene D-Proline L-Proline (+)-Pulegone (S)-2-Pyrrolidinemethanol (prolinol) (-)-Quinine Quinidine D-Serine

0.88 1280 3 362 15035 1773 67 11 111

8 000 0.54

395 236 20 42 1.1 7 24 287 820 16200 256 950 1180 236 1725 110 13 4 026 21 22 48 553 57 11 1.2 522 2.8 7 116 151 328 75

Appendix

L-Serine (R,R)-Tartaric acid (S,S)-Tartaric acid L-Threonine D-Valine L-Valine

* Average price per g in 1993: US$ 0.1 1 (US$ 18.8 per mol).

313 14

1

.oo

35 18 119 9

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

References

Chapter 1 [ I ] Nomenclature of Organic Chemistry, Pari E. Pure Appl. Chem. 45, 11 (1976). [2] a) R.S. Cahn, C. Ingold and V. Prelog. Angew. Chem. 78, 413 (1966); Int. Ed. Engl. 5, 385 (1 966). b) V. Prelog and G. Helmchen, Angew. Chem. 94, 614 (1982); Int. Ed. Engl. 21, 567 (1982). [3] Nomenclature for Straightforward Transformations, Pure Appl. Chem. 53, 305 (1 981). [4] C. H. Heathcock, C.T. Buse, W. A. Kleschick, M. C. Pirrung, J.E. Sohn and J. Lampe, J. Org. Chem. 45, 1066 (1980). [5] S. Masamune, S.A. Ali, D.L. Snitman and D.S. Garrey, Angew. Chem. 92, 573 (1980); Int. Ed. Engl. 19, 557 (1980). [6] F.A. Carey and M.E. Kuehne, J. Org. Chem. 17, 3811 (1992). [7] R. Noyori, I. Nishida and J. Sakata, J. Am. Chem. SOC.103, 2108 (1981). [8] D. Seebach and V. Prelog, Angew. Chem. 94, 696 (1982); Int. Ed. 21, 654 (1982). [9] C. A. Heathcock, in Asymmetric Synthesis, (J. D. Morrison, ed.). Academic Press, New York, 1984, p. 111. [lo] K. Mislow and M. Raban, Top. Stereochem. 1, 1 (1967). [ I 11 M. Raban, Tetrahedron Lett. 3105 (1966). [12] K.R. Hanson, J. Am. Chem. SOC.88, 2771 (1966). [I31 V. Prelog and S. Helmchen, Helv. Chim. Acta 55, 2581 (1972). [14] P. Muller and J.C. Perlberger, Helv. Chim. Acta 59, 2335 (1976). [15] S. Winstein D, Pressman and W.G. Young, J. Am. Chem. SOC.61, 1645 (1939). [16] L. Poppe and L. Novak, Selective Biocatalysis, a Synthetic Approach. Weinheim, VCH, 1992. [I71 J. Tramper, H.C. Plas and P. Linko (Eds.), Biocatalysis in Organic Synthesis (Stud. Org. Chem., Vol. 22), Amsterdam: Elsevier, 1985. [I 81 M. P. Schneider (Ed.) Enzymes as Catalysts in Organic Synthesis, Dorndrecht: D. Reidel Publishing, 1986. [I91 H. G. Davies, R. H. Green, D. R. Kelly and S. M. Roberts (Eds.) Biotransformations in Preparative Organic Chemistry, London: Academic Press, 1989. [20] H. G. Crout and M. Christen, “Biotransformations in Organic Synthesis” in: Modern Synthetic Methods, Vol. 5 , R. Scheffold (Ed.), New York: Wiley, 1989, p. 1. [21] W.H. Pirkle and P.L. Rinaldi, J. Am. Chem. SOC.99, 3510 (1977). [22] S. Miyano, L. D. L. Lu, S. M. Viti and K. B. Sharpless, J. Org. Chem. 48, 3068 (1983). [23] T. Shiraiwa, S. Sakata, K. Fujishima and H. Kurokawa, Bull. Chem. SOC.Jap. 64, 191 (1991). [24] W. Marckwald and A. McKenzie, Ber. dtsch. chem. Ges. 32, 2130 (1899). [25] V.S. Martin, S.S. Woodard, T. Katsuki, Y. Yamada, M. Ikeda and K.B. Sharpless, J. Am. Chem. SOC.103, 6237 (1981). [26] J. Jacques, A. Collet and S. H. Wilen, Enantiomers, Racemates and Resolution. Wiley-Interscience, New York, 1981, p. 81.

3 16 [27] [28] [29] [30] [31] [32] [33] [34] [35] (361 [37] [38] [39] [40] [4t] 1421 [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] I561 [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71]

References

K. Mislow, Zntroduction to Stereochemistry. Benjamin, New York, 1966, p. 13 1 . C. Quannes, Beugelmans and G. Rossi, J. Am. Chem. SOC.95, 8272 (1973). L. Duhamel, A. Ravard and J.-C. Plaquevent, Tetrahedron Asym. I, 347 (1990). J. Hiratake, Y. Yamamoto and J. Oda, J.C.S. Chem. Comm. 1717 (1985). R. Noyori, I. Tornino and Y. Tanimoto, J. Am. Chem. SOC. 101, 3129 (1979). Y. Izumi, Angew. Chem. 83, 956 (1971); Int. End. Engl. 10, 871 (1971). Y. Izumi and A. Tai, Stereodifferentiating Reactions. Academic Press, New York, 1977. D. Seebach, R. Dorr, B. Bastani and V. Ehrig, Angew. Chem. 81, 1002 (1969). M. Shindo, K. Koga and K. Tomioka, J. Am. Chem. SOC. 114, 8732 (1992). Y. Inoue, N. Yamasaki, T. Yokoyama and A. Tai, J. Org. Chem. 57, 1332 (1992). R. Hazard, S.Jaouannat and A. Tallec, Tetrahedron 38, 93 (1982). Z.G. Hajos and D.R. Parrish, J. Org. Chem. 39, 1615 (1974). I. Ojima, T. Kogure and T. Terasaki, J. Org. Chem. 43, 3444 (1978). F.A. Davis and B.C. Chen, Tetrahedron Lett. 1001 (1977). S. Hashimoto and K. Koga, Chem. Pharm. Bull. 27, 2760 (1979). R. Meric and J.-P. Vigneron, Tetrahedron Lett. 2059 (1974). K. Weinges, K.-L. Klotz and H. Droste, Chem. Ber. 113, 710 (1980). J.C. Fiaud and H. Horeau, Tetrahedron Lett. 2565 (1972). N. Oguni, Y. Matsuda and T. Kaneko, J. Am. Chem. SOC. 110, 7877 (1988). K. Tanaka, J. Matsui, Y. Kawabata, H. Suzuki and A. Watanabe, J. C. S. Chem. Comm. 1632 (1991). K. Mikami and M. Terada, Tetrahedron 48, 5671 (1992). A. McKenzie, J. Chem. SOC.101, 1196 (1912). E. L. Eliel and Y. Senda, Tetrahedron 26, 24 11 (1 970). E. Fischer, Ber. dtsch. chern. Ges. 27, 3189 (1894). V. Prelog, Helv. Chim. Acta 36, 308 (1953). D. J. Cram and F.A. Abd Elhafez, J. Am. Chem. SOC. 74, 5828 and 5951 (1952). E. Ruch and I. Ugi, Top. Stereochem. 4, 99 (1969). W.T. Wipke and P. Gund, J. Am. Chem. SOC. 98, 8107 (1976). M. Hirota, K. Abe, H. Tashiro and M. Nishio, Chern. Lett. 777 (1982). H. Buschmann, H.-D. Scharf, N. Hoffmann, M. W. Plath and J. Runsink, J. Am. Chem. SOC. 111, 5367 (1989). W. Marckwald, Ber. dtsch. chern. Ges. 37, 1368 (1904). J. D. Morrison and H. S. Mosher, Asymmetric Organic Reactions. Prentice-Hall, Engelwood Cliffs, N. J. U. S. A,, 1971. H. Handel and J.-L. Pierre, Tetrahedron Lett. 2029 (1976). J.-P. Guette and A. Horeau, Bull. SOC.Chim. France, 1747 (1967). S. Masarnune, W. Choy, J. S. Petersen and L. R. Sita, Angew. Chem. 97, I (1985); Int. Ed. Engl. 24, 1 (1985). K. Mislow, Zntroduction to Stereochemistry. Benjamin, New York, p. 131. R. K. Hill and N. W. Gilman, J. C. S. Chern. Comm. 619 (1967). T. Nakai, K. Mikami and S. Taya, J. Am. Chem. SOC. 103, 6492 (1981). A. Fischli, M. Klaus, H. Mayer, P. Schonholzer and R. Riiegg, Helv. Chim. Acta 58, 564 (1975). S. Terashima and S. Yamada, Tetrahedron Lett. 1001 (1977). K.-K. Chan, N. Cohen, J. P. DeNoble, A.C. Specian, Jr. and G. Saucy, J. Org. Chem. 41, 3505 (1976). T. Koizumi, Y. Kobayashi, H. Amitani and E. Yoshii, J. Org. Chem. 42, 3459 (1977). A. I. Meyers, G. Knaus, K. Kamata and M.E. Ford, J. Am. Chem. SOC. 98, 567 (1976). For details see p. 237 in E. L. Eliel, Stereochemistry of Carbon Compounds. McGraw-Hill, 1962. W.C. Still and I. Galynker, Tetrahedron 23, 3981 (1981).

References

3 17

Chapter 2 R. S. Dickson, Homogeneous Catalysis with Compounds of Rhodium and Iridium. D. Reidel Publ. Co., Dordrecht, 1985. J.A. Osborne, F.H. Jardine, J.F. Young and G. Wilkinson, J. Chem. SOC.A, 1711 (1966). L. Horner, H., Winkler, A. Rapp, A. Mentrup, H. Hoffmann and H. Beck, Tetrahedron Lett. 161 (1961). For a review, see N. J. Gallagher and I. D. Jenkins, in Topics in Stereochem. Vol. 3, Chap. 1, (N. L. Allinger and E. L. Eliel, eds.). Wiley, New York, 1968. 0. Korpium, R.A. Lewis, J. Chickos and K. Mislow, J. Am. Chem. SOC.90, 4842 (1968). W. S. Knowles and M. J. Sabacky, J. C. S. Chem. Comm. 1445 (1968). B.D. Vineyard, W.S. Knowles and M. J. Sabacky, J. Mol. Catal. 19, 159 (1983). L. Horner, H. Siege1 and H. Biithe, Angew. Chem. 80, 1034 (1968); Int. Ed. Engl. 7, 942 (1968). P. Abley and F. J. McQuillin, J.C.S. Chem. Comm. 477 (1969). P. Abley and F. J. McQuillin, J. Chem. SOC.C. 844 (1971). J.D. Morrison, R.E. Burnett, A.M. Aguiar, C. J. Morrow and C. Phillips, J. Am. Chem. SOC. 93, 1301 (1971). W. S. Knowles, M. J. Sabacky and B. D. Vineyard, J. C. S. Chem. Comm. 10 (1972); U. S. Pat. 4.005.127 (25.1.1977). C.A. 86, 190463 (1977). B.D. Vineyard, W.S. Knowles, M.J. Sabacky, G.L. Bachman and D.J. Weinkauff, J.Am. Chem. SOC.99, 5946 (1977). H.B. Kagan and T.-P. Dang, J.C.S. Chem. Comm. 481 (1971). H.B. Kagan and T.-P. Dang, J. Am. Chem. SOC.94, 6429 (1972). M.J. Burk, J. Am. Chem. SOC.113, 8518 (1991). M. J. Burk and J.E. Feaster, J. Am. Chem. SOC.114, 6266 (1992). T. Hayashi, N. Kawamura and Y. Ito, Tetrahedron Lett. 29, 5969 (1988). T. Morimoto, M. Chiba and K. Achiwa, Tetrahedron Lett. 30, 735 (1989). H. Takahashi and K. Achiwa, Chem. Lett. 305 (1989). T. Morimoto, H. Takahashi, K. Fuji, M. Chiba and K. Achiwa, Chem. Lett. 2061. H. Takahashi, T. Morimoto and K. Achiwa, Chem. Lett. 855 (1987). H. Takeda, T. Tachinami and M. Aburatani, Tetrahedron Lett. 30, 367 (1989). T. Chiba, A. Miyashita and H. Nohira, Tetrahedron Lett. 32, 4745 (1991). V. Caplar, G. Comisso and V. b j i c , Synthesis 85 (1981). B.D. Vineyard, W. S. Knowles, M. J. Sabacky, G.L. Bachman and D. J. Weinkauff, J. Am. Chem. SOC.99, 5946 (1977). W. S. Knowles, M. J. Sabacky, B. D. Vineyard and D. J. Weinkauff, J. Am. Chem. SOC.97, 2567 (1975). M.D. Fryzuk and B. Bosnich, J. Am. Chem. SOC. 99, 6262 (1977). J. Kottner and G. Greber, Chem. Ber. 113, 2323 (1980). P.A. MacNeil, N.K. Roberts and B. Bosnich, J. Am. Chem. SOC.103, 2273 (1981). A. Miyashita, A. Yasuda, H. Takaya, K. Toriumi, T. Ito, T. Souchi and R. Noyori, J. Am. Chem. SOC.102, 7932 (1980). K. J. Brown, M.S. Berg, K.K. Waterman, D. Lingenfelter and J.R. Murdoch, J. Am. Chem. SOC.106, 4717 (1984). A. Miyashita, H. Takaya, T. Souchi and R. Noyori, Tetrahedron 40, 1245 (1984). H. Takaya, K. Mashima, K. Koyano, M. Yagi, H. Kumobayashi, T. Taketomi and S. Akutagawa, J. Org. Chem. 51, 629 (1986). T.-P. Dang and H. Kagan, J.C.S. Chem. Comm. 481 (1971).

3 18

References

[36] M.D. Fryzuk and B. Bosnich, J. Am. Chem. SOC.101, 3043 (1979). [37] B.A. Murrer, J.M. Brown, P.A. Chaloner, P.N. Nicholson and D. Parker, Synthesis 350 (1 979). [38] 0. Cervinka and K. Gajewski, Czechosl. Pat. 194.386 (1979). [39] V. Caplar, G. Comisso and V. Sunjic, Synthesis 85 (1981). [40] S. Y. Zhang, S. Yemul, H. B. Kagan, R. Stern, D. Commereuc and Y. Chauvin, Tetrahedron Lett. 3955 (1981). [41] D. Lafont, D. Sinou and G. Descotes, J. Organomet. Chem. 169, 87 (1979). 1421 D. Lafont, D. Sinou, G. Descotes, R. Glaser and R. Geresh, J. Mol. Catal. 10, 305 (1981). [43] D. Sinou, Tetrahedron Lett. 2987 (1981). [44] T. Hayashi, T. Mise, M. Fukushima, M. Kogatani, N. Nagashima, Y. Hamada, A. Matsumoto, S. Kawarami, M. Konishi, K. Yamamoto and M. Kumada, Bull. Chem. SOC.Jap. 53, 1138 (1980). [45] T. Hayashi, T. Mise, S. Mitachi, K. Yamamoto and M. Kumada, Tetrahedron Lett. 1133 (1976). [46] H. Brunner, B. Schonhammer and C. Steinberger, Chem. Ber. 116, 3529 (1983). [47] B. L. Baker, S. J. Fritschel, J. R. Stille and J. K. Stille, J. Org. Chem. 47, 2954 (1981). [48] K. Achiwa, J. Am. Chem. SOC.98, 8265 (1976). [49] K. Achiwa, Chem. Lett. 561 (1978). [50] I. Ojima and T. Kogure, Chem. Lett. 567 (1978). (511 I. Ojima, T. Kogure and N. Yoda, J. Org. Chem. 45, 4728 (1980). (521 H.W. Krause, React. Kinet. Catal. Lett. 10, 243 (1979). [53] K. Ohkubo, K. Fujimori and K. Yoshinaga, Inorg. Nucl. Chem. Letters 15, 231 (1979). (541 W. Dumont, J.-C. Poulin, T.-P. Dang and H.B. Kagan, J. Am. Chem. SOC.95, 8295 (1973). [55] K. Ohkubo, M. Haga, K. Yoshinaga and Y. Motazato, Inorg. Nucl. Chem. Letters 17, 215 (1981). [56] T. Masuda and J. K. Stille, J. Am. Chem. SOC.100, 268 (1978). [57] N. Takaishi, H. Imai, C.A. Bertelo and J.K. Stille, J. Am. Chem. SOC.100, 264 (1978). (581 G.L. Baker, S. J. Fritschel and J.K. Stille, J. Org. Chem. 46, 2960 (1981). 1591 K. Achiwa, Heterocycles 9, 1539 (1978). [60] U. Nagel, H. Menzel, P. W. Lednor, W. Beck, A. Guyot and M. Barholin, Naturf. 36, 578 (1981). [61] M. Inoue, K. Ohta, N. Ishizuka and S. Enomoto, Chem. Pharm. Bull. 31, 3371 (1983). (621 I. Lazar and A.D. Sherry, J. Org. Chem. 56, 1253 (1991). 1631 U. Nagel and E. Kinzel, J. C. S. Chem. Comm. 1098 (1986). [64] U. Matteoli, P. Frediani, M. Bianchi, C. Botteghi and S. Gladiali, J. Mol. Catal. 12,265 (1981). [65] R. Noyori, M. Ohta, Y. Hsiao, M. Kitamura, T. Ohta and H. Rakaya, J. Am. Chem. SOC.,108, 7117 (1986). 1661 M. Kitamura, M. Tokunaga and R. Noyori, J. Org. Chem. 57, 4053 (1992). (671 B. Heiser, E.A. Broger and Y. Crameri, Tetrahedron Asym. 2, 51 (1991). [68] H. Kawano, T. Ikariya, Y. Ishii, M. Saburi, S. Yoshikawa, Y. Uchida and H. Kumobayashi, J.C.S. Perkin Trans. 1 1571 (1990). (691 K. Mashima, K. Kusano, T. Ohta, R. Noyori and H. Takaya, J.C.S. Chem. Comm. 1208 (1989). [70] K. Mashima, Y. Matsumura, K. Kusano, H. Kumobayashi, N. Sayo, Y. Hori, T. Ishizaki, S. Akutagawa and H. Takaya, J.C.S. Chem. Comm. 609 (1991). (711 B.D. Zwich, A.M. Arif, A.T. Patton and J.A. Gladysz, Angew. Chem. 99, 921 (1986); Int. Ed. Engl. 26, 910 (1986). [72] J.M. Brown, P.L. Evans and A.R. Lucy, J.C.S. Perkin Trans. 2 1589 (1987). (731 R. Glaser, S. Geresh and J. Blumenfeld, J. Organomet. Chem. 112, 355 (1976). (741 R. B. King, J. Bakos, C. D. Hoff and L. Marko, J. Org. Chem. 44, 1729 (1979). (751 I. Ojima and T. Kogure, Chem. Lett. 567 (1978). [76] J.M. Brown, P.A. Chaloner, B.A. Murrer and D. Parker, A.C.S. Symposium Ser. 119, 1731 ( I 980).

References

3 19

[77] K. Achiwa, Tetrahedron Lett. 3735 (1977). [78] W.S. Knowles, ACC.Chem. Res. 16, 106 (1983). [79] M. Capka, J. Hetflejs and R. Selke, React. Kinet. Catal. Lett. 10, 225 (1979). [80] W.R. Cullen and Y. Sugi, Tetrahedron Lett. 1635 (1978). [Sl] A. Uehara, T. Kubota and R. Tsuchiya, Chem. Lett 441 (1983). [82] J. Halpern, Pure Appl. Chem. 55, 99 (1983). [83] D. Sinou, Tetrahedron Lett. 2987 (1981). [84] J.M. Brown and D. Parker, J.C.S. Chem. Comm. 342 (1980). [85] K. Achiwa, Tetrahedron Lett. 2583 (1978). [86] K. Achiwa and T. Soga, Tetrahedron Lett. 1 1 19 (1978). [87] T. Hayashi, M. Tanaka and I. Ogata, Tetrahedron Lett. 295 (1977). [88] K. E. Koenig, G. L. Bachmann and B. D. Vineyard, J. Org. Chem. 45, 2362 (1980). [89] W.C. Christophel and B.D. Vineyard, J. Am. Chem. SOC.101, 4406 (1979). [90] T.H. Johnson and G. Rangarajan, J. Org. Chem. 45, 62 (1980). [91] H. Kawano, Y. Ishii, T. Ikariya, M. Saburi, S. Yoshikawa, Y. Uchida and H. Kumobayashi, Tetrahedron Lett. 28, 1905 (1987). [92] M. Saburi, L. Shao, T. Sakurai and Y. Uchida, Tetrahedron Lett. 33, 7877 (1992). [93] T. Ohta, H. Takaya, M. Kitamura, K. Nagai, R. Noyori, J. Org. Chem. 52, 3174 (1987). [94] W.H. Pearson and M.-C. Cheng, J. Org. Chem. 52, 3176 (1987). 1951 T. Morimoto, M. Chiba and K. Achiwa, Tetrahedron Lett. 30, 735 (1989). I961 U.S. Pat. 4.409.397 (11.10.1983). C.A. 100, 51098 (1984). [97] H. Takaya, T. Ohta, N. Sayo, H. Kumobayashi, S. Akutagawa, S. Inoue, I. Kasahara and R. Noyori, J. Am. Chem. SOC.109, 1596 (1987). [98] T. Yamagishi, S. Ikeda, M. Yatagai, M. Yamaguchi and M. Hida, J. Chem. SOC.Perkin Trans. 1 1787 (1988). [99] W. S. Knowles, M. J. Sabacky and B. D. Vineyard, Advan. Chem. Ser. 132, 274 (1 974). [loo] H.B. Kagan, N. Langlois and T.-P. Dang, J. Organomet. Chem. 90, 353 (1975). [I011 R. Glaser, S. Geresh, J. Blumenfeld and M. Twaik, Tetrahedron 34, 2405 (1978). 11021 R. Glaser and S. Geresh, Tetrahedron Lett. 29, 2527 (1977). [I031 R. Glaser and B. Vainas, J. Organomet. 121, 249 (1976). [I041 R. Glaser, S. Geresh, J. Blumenfeld, B. Vainas and M. Twaik, Israel J. Chem. 15, 17 (1976177). [I051 J. W. Scott, D.D. Keith, G. Nix, Jr., D. R. Parrish, S. Remington, G. P. Roth, J. M. Townsend, D. Valentine, Jr. and R. Yang, J. Org. Chem. 46, 5086 (1981). [lo61 J. M. Brown and P.A. Chaloner, J. C. S. Chem. Comm. 613 (1 979). [lo71 K.E. Koenig and W.S. Knowles, J. Am. Chem. SOC.100, 7561 (1978). [I 081 J. M. Brown and P. A. Chaloner, Tetrahedron Lett. 21, 1877 (1 978). [I091 A.S.C. Chan, J. J. Pluth and J. Halpern, J. Am. Chem. SOC.102, 5952 (1980). [I 101 P. S. Chua, N.K. Roberts, B. Bosnich, S. J. Okrasinski and J. Halpern, J. C. S. Chem. Comm. 1278 (1981). [ i 111 J. M. Brown and P. A. Chaloner, J. C. S. Chem. Comm. 344 (1980). [112] J.M. Brown and P.A. Chaloner, J.C.S. Chem. Comm. 321 (1978). [113] J. M. Brown, P.A. Chaloner, G . Descotes, R. Glaser, D. Lafont and D. Sinou. J. C.S. Chem. Comm. 611 (1979). 11141 K. Achiwa, P.A. Chaloner and D. Parker, Organomet. Chem. 218, 249 (1981). 11151 K. Achiwa, Y. Ohga, Y. Itaka and H. Saito, Tetrahedron Lett. 4683 (1978). [I161 Y. Izumi, Angew. Chem. 83, 956 (1971); Int. Ed. Eng. 10, 871 (1971). [I 171 J. Halpern, D.P. Riley, A.S.C. Chan and J. J. Pluth, J. Am. Chem. SOC.99, 8055 (1977). [I 181 A. S. C. Chan and J. Halpern, J. Am. Chem. SOC.102, 838 (1980). [I191 A.S.C. Chan, J. J. Pluth and J. Halpern, Inorg. Chim. Acta 37, 2477 (1979). [120] D.P. Riley, J. Organomet. Chem. 234, 85 (1982). [121] K. Achiwa, P.A. Chaloner and D. Parker, J. Organomet. Chem. 218, 249 (1981). [I221 J.M. Brown and B.A. Murrer, Tetrahedron Lett. 4859 (1979).

320

References

J.M. Brown and P.A. Chaloner, J. Am. Chem. SOC. 102, 3040 (1980). J.M. Brown and D. Parker, J. Org. Chem. 47, 2722 (1982). N. W. Alcock, J.N. Brown, A. E. Derome and A. R. Lucy, J. C. S. Chem. Comm. 575 (1985). C. R. Landis and J. Halpern, J. Am. Chem. SOC.109, 1746 (1987). J.M. Brown and P.L. Evans, Tetrahedron 44, 4905 (1988). J. M. Brown, P. A. Chaloner and G.A. Morris, J. C. S. Perkin Trans. 2 1583 (1987). T. Ohta and T. Takaya, Tetrahedron Lett. 49, 7189 (1990). A. Fischli and D. Suss, Helv. Chim. Acta 62, 2361 (1979). A. Fischli and J. J. Daly, Helv. Chim. Acta 63, 1628 (1980). Y. Ohgo, K. Kobayashi, S. Takeuchi and J. Yoshimura, Bull. Chem. SOC.Jap. 45, 933 (1972). S. Takeuchi and Y. Ohgo, Bull. Chem. SOC.Jap. 54, 2136 (1981). S. Takeuchi and Y. Ohgo, Bull. Chem. SOC.Jap. 60, 1449 (1987). E. Cesarotti, R. Ugo and H.B. Kagan, Angew. Chem. 91, 842 (1979); Int. End. Engl. 18, 779 (1979). [I361 C.A. Willoughby and S.L. Buchwald, J. Am. Chem. SOC.114, 7563 (1992). 11371 J. Solodar, Chem. Techn. 421 (1975). [I381 T, Hayashi, K. Kanehira and M. Kumada, Tetrahedron Lett. 4417 (1981). [139] I. Ojima, T. Kogure and A. Achiwa, J. C. S. Chem. Comm. 428 (1977). [140] Sz. Ttlros, B. Heil, L. Kollar and L. Marko, J. Organomet. Chem. 197, 85 (1980). [141] I. Ojima and T. Kogure, J. Organomet. Chem. 195, 239 (1980). [142] K. Achiwa, T. Kogure and and I. Ojima, Chem. Lett. 297 (1978). [143] Ger. Offen. 3.302.697 (18.8.1983). C.A. 100, 22797 (1984). [144] T. Hayashi, A. Katsumura, M. Konishi and M. Kumada, Tetrahedron Lett. 425 (1979). [145] G. Zassinovich and F. Grisoni, J. Organomet. Chem. 247, C24 (1983). [146] H. Takahashi, S. Sakuraba, H. Takeda and K. Achiwa, J. Am. Chem. SOC.112, 5876 (1990). [147] M. Kitamura, T. Ohkuma, S. Inoue, N. Sayo, H. Kumobayashi, S. Akutagawa, T. Ohta, H. Takaya and R. Noyori, J. Am. Chem. SOC.110, 629 (1988). [148] M. Kitamura, M. Tokunaga, T. Ohkuma and R. Noyori, Tetrahedron Lett. 32, 4163 (1991). [149] R. Noyori, T. Ohkuma and M. Kitamura, J. Am. Chem. SOC. 109, 5856 (1987). [150] R. Noyori, T. Ikeda, T. Ohkuma, M. Widhalm, M. Kitamura, H. Takaya, S. Akutagawa, N. Sayo, T. Saito, T. Taketomi and H. Kumobayashi, J. Am. Chem. SOC.111, 9134 (1989). [151] R.A. Aitken, J. Gopal and J.A. Hirst, J.C.S. Chem. Comm. 632 (1988). [152] L. Shao, T. Seki, H. Kawano and M. Saburi, Tetrahedron Lett. 32, 7699 (1991). [153] M. Kitamura, T. Ohkuma, M. Tokunaga and R. Noyori, Tetrahedron Asym. 1, 1 (1990). [154] H. Kawano, Y. Ishii, M. Saburi and Y. Uchida, J.C.S. Chem. Comm. 87 (1988). [155] I. Ojima and T. Suzuki, Tetrahedron Lett. 1239 (1980). [I561 D. Sinou, D. Lafont, G. Descotes and A.G. Kent, J. Organomet. Chem. 217, 119 (1981). [I571 D. Meyer, J.-C. Poulin and H. B. Kagan, J. Org. Chem. 45, 4680 (1980). [158] K. Onuma, T. Ito and A. Nakamura, Chem. Lett. 481 (1980). [159] M. Yatagai, M. Zama, T. Yamagishi and M. Hida, Chem. Lett. 1203 (1983). [I601 S. El-Baba, J.M. Nuzillard, J.C. Poulin and H.P. Kagan, Tetrahedron 42, 3851 (1986). [161] J.M. Brown, Angew. Chem. 99, 169 (1987); Int. Ed. Engl. 26, 190 (1987). [I621 D.A. Evans and M.M. Morrisey, J. Am. Chem. SOC.106, 3866 (1984). [163] G. Stork and D.E. Kahne, J. Am. Chem. SOC.106, 1071 (1984). [164] R.H. Crabtree and M. W. Davis, J. Org. Chem. 51, 2655 (1986). [165] J.M. Brown and I. Cutting. Comm. 578 (1985). [166] J. M. Brown and I. Cutting, J. C. S. Chem. Comm. 578 (1985). [167] K. Yamamoto, M. Takagi and J. Tsuji, Bull. Chem. SOC.Jap. 61, 319 (1988). [168] T. Nishi, M. Kitamura, T. Ohkuma and R. Noyori, Tetrahedron Lett. 29, 6327 (1988). [169] J. M. Brown and A.P. James, J.C.S. Chem. Comm. 181 (1987). [170] J. M. Brown and S.A. Hall, Tetrahedron Lett. 25, 1393 (1984). [171] D.A. Evans and M.M. Morrisey, J. Am. Chem. SOC.106, 3866 (1984). [172] R.N. Meals, Pure Appl. Chem. 13, 148 (1966). [123] [124] [ I 251 [126] [127] [128] [129] [I301 [131] [I321 [133] [134] [135]

References

32 1

11731 C. Eaborn and R. W. Bott, in Organometallic Compounds of the Group IVElements, Vol. I . Part 1, (A. G. MacDiarmid ed.). Marcel Dekker, New York, 1968. [I741 I. Ojima, M. Nihonyanagi and Y. Nagai, Chem. Commun. 938 (1972). [I751 I. Ojima, T. Kogure, M. Nihonyanagi and Y. Nagai, Bull. Chem. SOC.Jap. 45, 3506 (1972). [I761 K. Yamamoto, Y. Uramoto and M. Kumada, J. Organomet. Chem. 31, C9 (1971). [177] T. Hayashi and M. Kumada, ACC.Chem. Res. 15, 395 (1982). [I781 I. Ojima, K. Yamamoto and M. Kumada, in Aspects of Homogeneous Catalysis, Vol. 3. D. Reidel. Dorndrecht, 1977, p. 186. [I791 H. Brunner, Angew. Chem. 95, 921 (1983); Int. Ed. Engl. 22, 897 (1983). [I801 B. Bosnich and M.D. Fryzuk, Top. Stereochem. 12, 119 (1980). [I811 T. Hayashi, K. Tamao, Y. Katsuro, I. Nakae and M. Kumada, Tetrahedron Lett. 1871 (1980). [182] Y. Uozumi and T. Hayashi, J. Am. Chem. SOC.113, 9887 (1991). [183] Y. Uozumi, S.-Y. Lee and T. Hayashi, Tetrahedron Lett. 33, 7185 (1992). [I841 S.H. Bergens, P. Noheda, J. Whelan and B. Bosnich, J. Am. Chem. SOC.114, 2121 (1992). [185] S.H. Bergens, P. Noheda, J. Whelan and B. Bosnich, J. Am. Chem. SOC.114, 2128 (1992). [186] T. Hayashi and K. Kabeta, Tetrahedron Lett. 26, 3023 (1985). [187] T. Hayashi and Y. Matsumoto, Tetrahedron Asym. 1, 151 (1990). [lSS] T. Hayashi, K. Yamamoto and M. Kumada, Tetrahedron Lett. 4405 (1974). [189] H. Brunner and G. Riepl, Angew. Chem. 94, 369 (1982); Int. Ed. Engl. 21, 377 (1982). [190] H. Brunner, G. Riepl and H. Weitzer, Angew. Chem. 95, 326 (1983); Int. Ed. Engl. 22, 331 (1983). [I911 H. Brunner, R. Becker and G. Riepl, Organometallics 3, 1354 (1983). [I921 H. Brunner and U. Obermann, Chem. Ber. 122, 499 (1989). [I931 G. Balavoine, J.C. Clinet and I. Lellouche, Tetrahedron Lett. 30, 5141 (1989). [I941 M. J. Burk and J.E. Feaster, Tetrahedron Lett. 33, 2099 (1992). [I951 R. Glaser, Tetrahedron Lett. 25, 2127 (1975). [196] I. Ojima, T. Kogure and M. Kumagai, J. Org. Chem. 42, 1671 (1977). [197] I. Ojima, T. Kogure and Y. Nagai, Tetrahedron Lett. 1889 (1974). [I981 I. Ojima, T. Kogure and Y. Nagai, Tetrahedron Lett. 2475 (1973). [199] R. Becker, H. Brunner, S. Mahboobi and W. Wiegrebe, Angew. Chem. 969 (1985); Int. Ed. Engl. 24, 955 (1985). [200] N. Langlois, T.-P. Dang and H.B. Kagan, Tetrahedron Lett. 4865 (1973). 12011 K. Tamao, T. Nakajima, R. Sumiya, H. Arai, N. Higuchi and Y. Ito, J. Am. Chem. SOC.108, 6090 (1986). [202] M. Fujita and T. Hiyama, J. Am. Chem. SOC.107, 8294 (1985). [203] S. Anwar, G. Bradley and A.P. Davis, J.C.S. Perkin Trans. 1 1383 (1991). [204] R. J.P. Corrin and J. J.E. Moreau, J. Organomet. Chem. 64, C51 (1974); ibid. 85, 19 (1975). I2051 I. Ojima, T. Kogure and Y. Nagai, Chem. Lett. 541 (1973); I. Ojima and Y. Nagai, ibid. 223 (1974). [206] H. Brunner, B. Reiter and G. Riepl, Chem. Ber. 117, 1130 (1984). [207] J. Peyronel, J. Fiaud and H. B. Kagan, J. Chem. Res. (S) 320, (M) 4057 (1 980). [208] T. Onoda and S. Tomita, Japan Kokai 74, 110,631, quoted in ref. [239]. [209] K. Yamamoto, T. Hayashi and M. Kumada, J. Organomet. Chem. 54, C45 (1973). [210] I. Kolb and J. HetflejS, Coll. Czechoslov. Chem. Comm. 45, 2224 and 2808 (1980). [21 I] I. Ojima, M. Nihonyanagi and Y. Nagai, Bull. Chem. SOC.Jap. 45, 3722 (1972). [212] T. Kogure and I. Ojima, J. Organomet. Chem. 234, 249 (1982). [213] I. Ojirna, T. Tanaka and T. Kogure, Chem. Lett. 823 (1981). [214] M. Bartok, Stereochemistry of Heterogeneous Metal Catalysis, Wiley, New York, 1985. [215] S. Akabori, S. Sakurai, Y. Izumi and Y. Fujii, Nature 178, 323 (1956). [216] E.I. Klabunovskii, Izv. Akad. Nauk SSSR, Ser. Khim. 505 (1984). [217] M. Bartok, Gy. Witmann, Gy. Condos and G.V. Smith, J. Org. Chem. 52, 1139 (1987). [218] T. Osawa, Chem. Lett. 1609 (1985). 12191 H. Brunner, M. Muschiol, T. Wischert and J. Wiehl, Tetrahedron Asym. 1, 159 (1990).

322

References

[220] T. Harada and Y. Izumi, Chem. Lett. 1195 (1978). [221] A. Tai, M. Nakahata, T. Harada and Y. Izumi, Chem. Lett. 1125 (1980). [222] A. Tai, T. Harada, Y. Hiraki and S. Murakami, Bull. Chem. SOC.Jap. 56, 1414 (1983). [223] S. Murakami, T. Harada and A. Tai, Bull. Chem. SOC.Jap. 53, 1356 (1980). [224] T. Osawa and T. Harada, Bull. Chem. SOC.Jap. 57, 15 18 (1 984). [225] Y. Hiraki, K. Ito, T. Harada and A. Tai, Chem. Lett. 131 (1981). [226] K. Ito, T. Harada, A. Tai and Y. Izumi, Chem. Lett. 1049 (1979). [227] K. Ito, T. Harada, and A. Tai, Bull. Chem. SOC.Jap. 53, 3367 (1980). [228] A. Tai, H. Watanabe and T. Harada, Bull. Chem. SOC.Jap. 52, 1468 (1979). [229] H. Poise1 and U. Schmidt, Chem. Ber. 106, 3408 (1973). [230] B.W. Bycroft and G.R. Lee, J.C.S. Chem. Comm. 988 (1975). [231] T. Kanmera, H. Aoyagi, S. Lee and N. Izumiya, Pept. Chem. 59 (1979). [232] J. Vasilevkis, J.A. Gualtieri, S. D. Hutsching, R. C. West, J. W. Scott, D. R. Parrish, F.T. Bizzaro and G. F. Field, J. Am. Chem. SOC.100, 7423 (1978). [233] N. Izumiya, S. Lee, T. Kanmera and H. Aoyagi, J. Am. Chem. SOC.99, 8346 (1977). [234] K. Harada, T. Munegumi and S. Nomoto, Tetrahedron Lett. 22, I 1 1 (1981). [235] R. T. Standridge, H. G. Howell, J.A. Gylys, R. A. Partyka and A.T. Shulgin, J. Med. Chem. 19, 1400 (1976). [236] D.E. Nichols, C. F. Barfknecht, D. R. Rusterholz, F. Benington and R. D. Morin, J. Med. Chem. 16, 480 (1973). [237] G. Bringmann, J.-P. Geisler, T. Geuder, G. Kiinkel and L. Kinzinger, Liebigs Ann. Chem. 795 (1990). [238] S. Yamada, N. Ikota and K. Achiwa, Tetrahedron Lett. 1001 (1976). [239] A. W. Frahm and G. Knupp, Tetrahedron Lett. 2633 (1981). [240] G. Knupp and A. W. Rahm, Chem. Ber. 117, 2076 (1984). [241] W. Wiehl and A. W. Frahm, Chem. Ber. 119, 2668 (1986). [242] K. Harada and K. Matsumoto, J. Org. Chem. 32, 1794 (1967). [243] K. Harada and K. Matsumoto, J. Org. Chem. 33, 4467 (1968). [244] K. Harada and Y. Kataoka, Chem. Lett. 791 (1978). [245] K. Harada, T. Iwasaki and T. Okawara, Bull. Chem. SOC.Jap. 46, 1901 (2973).

Chapter 3 [I] H. Haubenstock, Top. Stereochem. 14, 231 (1982). J.D. Morrison; H.C. Brown, W.S. Park, B.T. Cho and P.V. Ramachandran, J. Org. Chem. 52, 5406 (1987). [2] A.A. Bothner-By, J. Am. Chem. Soc. 73, 846 (1951). [3] P.S. Portoghese, J. Org. Chem. 27, 3359 (1962). [4] S. R. Landor, B. J. Miller and A. R. Tatchell, J. Chem. SOC.C, 1922 (1966). [5] S. Yamaguchi and H.S. Mosher, J. Org. Chem. 38, 1870 (1973). [6] Y. Minoura and H. Yamaguchi, J. Polym. Sc. 6, A-I, 2012 (1968). [7] A. Horeau, H. Kagan and J. Vigneron, Bull. Soc. Chim. France 3795 (1968). [8] 0 . Cervinka, Coll. Czech. Chem. Commun. 26, 673 (1961). [9] 0. Cervinka, COIL Czech. Chem. Commun. 30, 2403 (1965). [lo] 0. Cervinka, V. Suchan, 0. Kotynek and V. Dudek, Coll. Czech. Chem. Commun. 30, 2484 (1965).

References [ I 11 [I21 [I31 [ 141 [15] [I61 [I71 [I81 [I91 (201 [21] (221 1231 [24] [25] [26] [27] [28]

[29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] (501 [51] [52] [53] [54] [55] [56]

[57] [SS]

323

S. Yamaguchi and K. Kabuto, Bull. Chem. SOC.Jap. 50, 3033 (1977). R. J. D. Evans, S. R. Landor and J. P. Regan, J. C. S. Perkin 1 552 (1974). R. Andrisano, S.R. Angeloni and S. Marzocchi, Tetrahedron 29, 913 (1973). S. R. Landor, B. J. Miller and A. R. Tatchell, J. Chem. SOC.C 2280 (1966). S. R. Landor, B. J. Miller and A. R. Tatchell, J. Chem. SOC.C 197 (1967). J. Hutton, M. Senior and N.C.A. Wright, Synth. Comm. 9, 799 (1979). R. Noyori, I. Tomino and Y. Tanimoto, J. Am. Chem. SOC.101, 3129 (1979). R. Noyori, Pure Appl. Chem. 53, 2316 (1981). R. Noyori, I. Tomino, Y. Tanimoto and M. Nishizawa, J. Am. Chem. SOC.106, 6709 (1984). M. Nishizawa, M. Yamada and R. Noyori, Tetrahedron Lett. 247 (1981). R. Noyori, I. Tomino, M. Yamada and M. Nishizawa, J. Am. Chem. SOC.106, 6717 (1984). M. Nishizawa and R. Noyori, Tetrahedron Lett. 247 (1981). R. Noyori, I. Tomino and M. Nishizawa, J. Am. Chem. SOC.101, 5843 (1979). H. Suda, S. Kanoh, N. Umeda, M. Ikka and M. Motoi, Chem. Lett. 899, (1984). K. Yamamoto, H. Fukushima and M. Nakazaki, J.C.S. Chem. Comm. 1490 (1984). N. Srivastava, A. Mital and A . Kumar, J.C.S. Chem. Comm. 493 (1992). K. Yamamoto, K. Ueno and K. Naemura, J.C.S. Perkin Trans. 1 2607 (1991). K. Kabuto, T. Yoshida, S. Yamaguchi, S. Miyano and H. Hashimoto, J. Org. Chem. 50, 3013 (1985). 0. Cervinka and 0. BElovsky, Coll. Czech. Chem. Comm. 32, 3897 (1965). J. D. Morrison and H. S. Mosher, Asymmetric Organic Reactions. Prentice-Hall, Englewood Cliffs, New Jersey, 1971. I. Jacquet and J.-P. Vigneron, Tetrahedron Lett. 2065 (1974). J.-P. Vigneron and I. Jacquet, Tetrahedron 32, 939 (1976). N. Cohen, R. J. Lopresti, C. Neukon and G. Saucy, J. Org. Chem. 45, 582 (1980). J.-P. Vigneron and V. Bloy, Tetrahedron Lett. 2683 (1979). S. Yamada, M. Kitamoto and S. Terashima, Tetrahedron Lett. 3265 (1976). M. Kitamoto, K. Kameo, S. Terasjo and S. Yamada, Chem. Pharm. Bull. 25, 1273 (1977). S. Terashima, N. Tanno and K. Koga, Chem. Lett. 981 (1980). M. Kawasaki, Y. Suzuki and S. Terashima, Chem. Lett. 239 (1984). A. Pohland and H.R. Sullivan, J. Am. Chem. SOC.75, 4453 (1953). S. Yamaguchi, H.S. Mosher and A. Pohland, J. Am. Chem. SOC.94, 9254 (1972). R.S. Brinkmeyer and V.M. Kapoor, J. Am. Chem. SOC.99, 8341 (1977). S. Yamaguchi, F. Yasuhara and K. Kabuto, J. Org. Chem. 42, 1578 (1977). R . A . Kretchmer, J. Org. Chem. 37, 801 (1972). M. Asami, H. Ohno, S. Kobayashi and T. Mukaiyama, Bull. Chem. SOC.Jap. 51, 1869 (1978). M. Asami and T. Mukaiyama, Heterocycles 499 (1979). T. Sato, Y. Goto and T. Fujisawa, Tetrahedron Lett. 41 11 (1982). T. Sato, Y. Goto, M. Watanabe and T. Fujisawa, Chem. Lett. 1533 (1983). D. Seebach and H. Daum, Chem. Ber. 107, 1748 (1974). M. Schmidt, R. Amstutz, G. Grass and D. Seebach, Chem. Ber. 113, 1691 (1980). 0. Cervinka, Coll. Czech. Chem. Comm. 26, 673 (1961). S. R. Landor, 0.0. Sonola and A. R. Tatchell, J.C.S. Perkin 1 605 (1978). S. R. Landor, Y.M. Chan, 0.0. Sonola and A. R. Tatchell, J.C. S. Perkin 1 493 (1983). S. R. Landor, 0.0. Sonola and A. R. Tatchell, J. C. S. Perkin 1 1902 (1974). M. M. Midland, in Asymmetric Synthesis, Vol. 2, (J. D. Morrison, ed.). Academic Press, New York, 1984, p. 45. A. Hirao, H. Mochizuki, S. Nakahama and N. Yamazaki, J. Org. Chem. 44, 1720 (1979). A. Hirao, S. Nakahama, D. Mochizuki, S. Itsuno, M. Ohowa and N. Yamazaki, J. C. S. Chem. Comm. 807 (1979). A. Hirao, M. Ohwa, S. Itsuno, H. Mochizuki, S. Nakahara and N. Yamazaki, Bull. Chem. SOC. Jap. 54, 1424 (1981). K. Soai, H. Oyamada and T. Yamanoi, Chem. Lett. 251 (1984).

324

References

[59] K. Soai, T. Yamanoi, H. Hikima and H. Oyamada, J. C. S. Chern. Comm. 138 (1985). [60] K. Soai, S. Niwa, T. Yamanoi, H. Hikirna and M. Ishizaki, J.C.S. Chern. Comm. 1018 (1 986). [61] K. Soai, S. Niwa and T. Kobayashi, J.C.S. Chem. Comrn. 801 (1987). [62] J. P. Masse and E. R. Paryre, J. C. S. Chem. Comm. 438 (1976). [63] R. Kinishi, N. Uchida, Y. Yarnamoto, J. Oda and Y. Inouye, Agric. Biol. Chem. 44, 643 (1980). [64] T. Doiuchi and Y. Minoura, Isr. J. Chem. 15, 84 (1976/77). [65] T. Sugimoto, Y. Matsumura, S. Tanimoto and M. Okano, J. C. S. Chem. Comm. 926 (1 978). [66] M. Utaka, H. Watabu and A. Takeda, J. Org. Chem. 51, 5423 (1986). [67] U. Leutenegger, A. Madin and A. Pfaltz, Angew. Chem. 101, 61 (1989); Int. Ed. Engl. 28, 60 (1989). [6X] S. Itsuno, Y. Sakurai, K. Shimizu and K. Ito, J.C.S. Perkin Trans. 1, 1859 (1990). [69] S. Krishnamurthy, F. Vogel and H.C. Brown, J. Org. Chem. 42, 2534 (1977). [70] H. C. Brown, B. T. Cho and W. S. Park, J. Org. Chem. 51, 3396 (1986). [71] H.C. Brown, W.S. Park and B.T. Cho, J. Org. Chem. 56, 1934 (1991). [72] H.C. Brown, B.T. Cho and W.S. Park, J. Org. Chem. 53, 1231 (1988). 1731 M. M. Midland and A. Kazubski, J. Org. Chem. 47, 2495 (1982); M. M. Midland, A. Kazubski and R.E. Woodling, J. Org. Chem. 56, 1068 (1991). [74] H. C. Brown, B. T. Cho and W. S. Park, J. Org. Chem. 52, 4020 (1987). [75] H.C. Brown and D.B. Bigley, J. Am. Chem. SOC. 83, 3166 (1961). [76] H.C. Brown and A.K. Mandal, J. Org. Chem. 42, 2996 (1977). [77] E. Caspi and K.R. Varma, J. Org. Chem. 33, 2181 (1968). [78] H. C. Brown, P. K. Jadhav and A. K. Mandal, Tetrahedron 37, 3547 (1 981). [79] H.C. Brown and A.K. Mandal, J. Org. Chem. 49, 2558 (1984). (801 J. C. Fiaud and H. B. Kagan, Bull. SOC.Chim. France 2742 (1 969). [81] M.F. Grundon, W.A. Khan, D.R. Boyd and W.R. Jackson, J. Chem. SOC.C 2557 (1971). [82] R.F. Borch and S.R. Leviton, J. Org. Chem. 37, 2347 (1972). [83] M.F. Grundon, D.G. McGleery and J.W. Wilson, Tetrahedron Lett. 295 (1976). [84] M.F. Grundon, D.G. McGleery and J.W. Wilson, J.C.S. Perkin 1 231 (1981). [85] N. Umino, T. Iwakuma and N. Itoh, Chem. Pharm. Bull. 27, 1479 (1979). [86] A. Hirao, S. Itsuno, S. Nakahama and N. Yamazaki, J.C.S. Chem. Comm. 315 (1981). [87] S. Itsuno, A. Hirao, S. Nakahama and N. Yamazaki, J. C. S. Perkin 1 1673 (1983). [88] S. Itsuno, K. Ito, A. Hirao and S. Nakahama, J. Org. Chem. 49, 555 (1984). [89] S. Itsuno, M. Nakano, K. Miyazaki, H. Masuda and K. Ito, J.C.S. Perkin Trans. 1 2039 (1985). [90] E. J. Corey, R.K. Bakshi and S. Shibata, J. Am. Chem. SOC. 109, 5551 (1987). [91] S. Itsuno, Y. Sakurai, K. Ito, A. Hirao and S. Nakahama, Bull. Chem. SOC. Jap. 60, 395 (1987). [92] B.T. Cho and Y.S. Chun, J.C. S. Perkin Trans. 1 3200 (1990). [93] B. T. Cho and Y.S. Chun, Tetrahedron Asymm. 3, 1583 (1992). [94] Y. Sakito, Y. Yoneyoshi and G. Suzukamo, Tetrahedron Lett. 29, 233 (f9X8). [95] E. J. Corey and R.K. Bakshi, Tetrahedron Lett. 31, 614 (1990). [96] E. J. Corey, M. Azimioara and S. Sarshar, Tetrahedron Lett. 33, 3429 (1992). [97] E. J. Corey and J.O. Link, Tetrahedron Lett. 33, 4141 (1992). [9X] S. Wallbaum and J. Martens, Asymetric synthesis with chiral oxazaborolidines. Tetrahedron Asymm. 3, 1475 (1992). [99] E. J. Corey and J.O. Link, Tetrahedron Lett. 33, 3431 (1992). [IOO] E. J. Corey, X.-M. Cheng, K.A. Cimprich and S. Sarshar, Tetrahedron Lett. 32, 6835 (1991). [I011 S. Itsuno, M. Nakano, K. Ito, A. Hirao, M. Owa, N. Kanda and S. Nakahama, J. C. S. Perkin Trans. 1 2615 (1985). [I021 S. Itsuno, K. Ito, T. Mariyama, N. Kanda, A. Hirao and S. Nakaharna, Bull. Chem. SOC.Jap. 59, 3329 (1986).

References

325

[I031 V. Nevalainen, Tetrahedron Asymm. 3, 921 (1992). [I041 V. Nevalainen, Tetrahedron Asymm. 3, 933 (1992). [ 1051 V. Nevalainen, Tetrahedron Asymm. 3, 1441 (1992). [ 1061 V. Nevalainen, Tetrahedron Asymm. 3, 1563 (1992). [lo71 E. J. Corey and J.O. Link, Tetrahedron Lett. 30, 6275 (1989). [lo81 T. Imai, T. Tamura and A. Yamamuro, J. Am. Chem. SOC.108, 7402 (1986). [lo91 S. Masamune, R. M. Kennedy and J. S. Petersen, J. Am. Chem. SOC.108, 7404 ( 1 986). [I 101 A. V. R. Rao, M. K. Gurjar, P. A. Sharma and V. Kawiar, Tetrahedron Lett. 31, 2341 (1 990). [ill] J.-M. Brunnel, 0. Pardigon, B. Faure and G. Buono, J.C.S. Chem. Comm. 287 (1992). [112] M. Falorni and L. Lardicci, Tetrahedron Lett. 30, 3551 (1989). [ l 131 T. Mukaiyama, K. Tomimori and T. Oriyama, Chem. Lett. 1359 (1985). [ I 141 T. Kawate, M. Nakagawa, T. Kakikawa and T. Hino, Tetrahedron Asymm. 3, 227 (1992). [ 1 151 M. M. Midland, in Asymmetric Synthesis, Vol. 2., (J. D. Morrison, ed.) Academic Press, New York, 1984, p. 45. [ I 161 H. C. Brown, W. S. Park, B. T. Cho and P. V. Ramachandran, J. Org. Chem. 52, 5406 (1987). [I 171 H.C. Brown, M. Srebnik and P.V. Ramachandran, J. Org. Chem. 54, 1577 (1989). [I181 M.M. Midland, A. Tramontano and S.A. Zderic, J. Am. Chem. SOC.99, 521 1 (1977). [119] M.M. Midland and N.N. Nguyen, J. Org. Chem. 46, 4107 (1981). [120] M.M. Midland, S. Greer, A. Tramontano and S.A. Zderic, J. Am. Chem. SOC.101, 2352 (1979). [I211 M.M. Midland and J.I. McLoughlin, J. Org. Chem. 49, 4102 (1984). [I221 M.M. Midland and A. Kazubski, J. Org. Chem. 47, 2814 (1982). [I231 M. M. Midland, A. Tramontano, A. Kazubski, R. S. Graham, D. J. S. Tsai and D. Cardin, Tetrahedron 40, 1371 (1984). 11241 M.M. Midland and A. Tramontano, Tetrahedron Lett. 3549 (1980). [I251 M.M. Midland and J.I. McLoughlin, J. Org. Chem. 49, 4101 (1984). [126] M.M. Midland, J. I. McLoughlin and J. Gabriel, J. Org. Chem. 54, 159 (1989). [127] H.C. Brown and G.G. Pai, J. Org. Chem. 47, 1606 (1982). [I281 H. C. Brown, G. G. Pai and P.K. Jadhav, J. Am. Chem. SOC.106, 3531 (1984). [129] M. M. Midland and P.E. Lee, J. Org. Chem. 50, 3237 (1985). [130] H.C. Brown, P.V. Ramachandran, S.A. Weissman and S. Swaminathan, J. Org. Chem. 55, 6328 (1990). [131] H.C. Brown, J. Chandrasekharan and P.V. Ramachandran, J. Am. Chem. SOC.110, 1539 (1 988). [I321 M. Srebnik, P.V. Ramachandran and H.C. Brown, J. Org. Chem. 53, 2916 (1988). [133] J. Chandrasekharan, P.V. Ramachandran and H.C. Brown, J. Org. Chem. 50, 5446 (1985). [I341 H.C. Brown, J. Chandrasekharan and P.V. Ramachandran, J. Org. Chem. 51, 3394 (1986). [I351 J.A. Soderguist, C.L. Anderson, E.I. Miranda and I. Rivera, Tetrahedron Lett. 31, 4677 (1 990). [I361 P. V. Ramachandran, A. V. Teodorovic, M. V. Rangaishenvi and H. C. Brown, J. Org. Chem. 57, 2379 (1992). [137] H.C. Brown and V. Ramachandran, J. Org. Chem. 54, 4504 (1989). 11381 For details, see pp. 160- 165 in ref. [30]. [I391 D. Nasipuri, C.K. Grosh and R. J.L. Martin, J. Org. Chem. 35, 657 (1970). [140] For details, see pp. 171 - 172 in ref. [30]. [141] D. Nasipuri and P.K. Bhattacharya, J.C.S. Perkin 1 576 (1977). [I421 A.K. Samaddar, S.K. Konar and D. Nasipuri, J.C.S. Perkin 1 1449 (1983). [I431 J.D. Morrison and R. W. Ridgway, J. Org. Chem. 39, 3107 (1974). [144] A. Streitwieser and M.R. Granger, J. Org. Chem. 32, 1528 (1967). [I451 P.M.T. de Kok, M.C.A. Donkersloot, P.M. van Lier, G.H. W.M. Meulendijka, L.A.M. Bastiaansen, H. J.G. van Hooff, J.A. Kanters and H.M. Buck, Tetrahedron 42, 941 (1986). [I461 A. Ohno, M. Ikeguchi, T. Kimura and S. Oka, J. Am. Chem. SOC.101, 7036 (1979). [I471 0. Cervinka, P. Malon and H. Prochazkova, Coll. Czech. Chem. Comm. 39, 1869 (1974).

326 [I481 [I491 [I501 [I511 11521

References

0. Cervinka and 0. Bglovsky, Coll. Czech. Chem. Comm. 32, 3897 (1967). G.M. Giongo, F. Di Gregorio, N. Palladino and W. Marconi, Tetrahedron Lett. 3195 (1973). A.I. Meyers and P.M. Kenda, Tetrahedron Lett. 1357 (1974). R.J. McMahon, K.E. Wiegers and S.G. Smith, J. Org. Chem. 46, 99 (1981). A. Hirao, S. Nakahama, H. Mochizuki, S. Itsuno and N. Yamazaki, J. Org. Chem. 45, 4231

( 1 980). [153] M.M. Midland, D. C. McDowell, R. L. Hatch and A. Tramontano, J. Am. Chem. SOC.102, 867 (1980). [I541 G. Giacomelli, L. Lardicci and A.M. Caporusso, J.C.S. Perkin 1 1795 (1975). [I551 Unpublished results quoted in ref. [30]. [I561 G. Giacomelli, R. Menicagli and L. Lardicci, J. Org. Chem. 38, 2370 (1973). [I571 G.P. Giacomelli, R. Menicagli and L. Lardicci, J. Am. Chem. SOC.97, 4009 (1975). [158] G. Giacomelli, L. Lardicci and F. Palla, J. Org. Chem. 49, 310 (1984). [159] The literature up to 1968 has been reviewed in ref. [30] pp. 84- 132. [I601 M.T. Reetz, S. Stanchev and H. Haning, Tetrahedron 48, 6813 (1992). [I611 J. Canceill and J. Jacques, Bull. SOC.SOC.Chim. France 2180 (1970). [162] M. Cherest, H. Felkin and N. Prudent, Tetrahedron Lett. 2199 (1968). [I 631 C. Zioudrou, I. Moustakali-Mavridis, P. Chrysochou and G. Karabatsos, Tetrahedron 34, 3181 (1978). [I641 M.T. Reetz, S. Stanchev and H. Haning, Tetrahedron 48, 6813 (1992). [I651 Data are taken from refs. [166, 167, 1971. [I661 J. R. Boone and A.C. Ashby, Top. Ster. 11, 53 (1979). [I671 D. J. Cram and F.A. Abd Elhafez, J. Am. Chem. SOC.74, 5828 (1952). [I681 V. Prelog, Helv. Chim. Acta 36, 308 (1953). [I691 J.W. Cornforth, R.H. Cornforth and K.K. Mathew, J. Chem. SOC.112 (195). [I701 G. J. Karabatsos, J. Am. Chem. SOC.89, 1367 (1967). [ I 711 G. J. Karabatsos and D. J. Fenoglio, Top. Stereochem. 5, 167 (1970). [172] D. J. Cram and K.R. Kopecky, J. Am. Chern. SOC.81, 2748 (1959). [I731 M.T. Reetz, M. Hiillmann und T. Seitz, Angew. Chem. 99, 478 (1987); Int. Ed. Engl. 26, 477 (1 987). [I741 G.E. Keck and S. Castellino, J. Am. Chem. SOC.108, 3847 (1986). [I751 Chen, E.R. Hortelano, E.L. Eliel and S. V. Frye, J. Am. Chem. SOC.114, 1778 (1992). [I761 S.D. Kahn, G.E. Keck and W. J. Hehre, Tetrahedron Lett. 28, 279 (1987). [I771 S.S. Wong and M.N. Paddon-Row, J.C.S. Chem. Comm. 327 (1991). [I781 A. R. Bassindale, R. J. Ellis, J. C. Y. Lau and P. G. Taylor, J. C. S. Chem. Comm. 98 (1986). [I791 Y. Yamamoto, K. Matsuoka and H. Nemoto, J. Am. Chem. SOC.110, 4475 (1988). [I801 M. ChCrest and H. Felkin, Tetrahedron Lett. 2205 (1968). [I811 G. J. Karabatsos and D. J. Fenoglio, J. Am. SOC.91, 1124 (1969). [I821 N.T. Anh and 0. Eisenstein, Nouv. J. Chim. 1, 61 (1976). [I831 N.T. Anh, Top. Curr. Chem. 88, 146 (1980). [I841 K.N. Houk and Y.-D. Wu, J. Am. Chern. SOC.109, 907 (1987). [I851 H.B. Biirgi, J.D. Dunitz, J.M. Lehn and G. Wipf, Tetrahedron 30, 1563 (1974). [I861 H.B. Biirgi, J.M. Lehn and G. Wipf, J. Am. Chem. SOC.96, 1956 (1974). [I871 E.P. Lodge and C.H. Heathcock, J. Am. Chem. SOC.109, 2819 (1987). [I881 For leading reviews about this topic cJ: refs. [166, 190, and 1971. [I891 D. H. R. Barton, J. Chem. SOC. 1027 (1953). [I901 D.C. Wigfield, Tetrahedron 35, 449 (1979). [I911 W.G. Dauben, G. J. Tonken and D.S. Noyce, J. Am. Chem. SOC.78, 2579 (1956). [I921 E.L. Eliel and Y. Senda, Tetrahedron 26, 2411 (1970). [I931 J.X. Richer, J. Org. Chem. 30, 324 (1965). [I941 J. Huet, Y. Maroni-Barnaud, N. Tron Anh and J. Seyden-Penne, Tetrahedron Lett. 159 (1 976). [I951 W.T. Wipke and P. Gund, J. Am. Chem. SOC.98, 8107 (1976).

References

321

M.-H. Rei, J. Org. Chem. 48, 5386 (1983). E. C. Ashby and J. T. Laemmle, Chem. Rev. 75, 521 ( 1 975). E. C. Ashby and S. A. Noding, J. Org. Chem. 42, 264 ( I 977). A. S. Cieplak, B. D. Tait and C. R. Johnson, J. Am. Chem. SOC.111, 8447 (1989). T. Oishi and T. Nakata, Acc. Chem. Res. 17, 338 (1984). T. Nakata, T. Tanaka and T. Oishi, Tetrahedron Lett. 2653 (1983). T. Nakata, Y. Tani, M. Hatozaki and T. Oishi, Chem. Pharm. Bull. 32, 1411 (1984). Y. Ito and M. Yamaguchi, Tetrahedron Lett. 5385 (1983). M. Shimagaki, T. Maeda, Y. Matsuzaki, I. Hori, T. Nakata and T. Oishi, Tetrahedron Lett. 25, 4775 (1984). [205] K. Suzuki, E. Katayama and G. Tsuchihashi, Tetrahedron Lett. 25, 2479 (1984). [206] M. Shimagaki, Y. Matsuzaki, I. Hori, T. Nakata and T. Oishi, Tetrahedron Lett. 25, 4775 (1984). [207] Y. Ito, T. Katsuki and M. Yamaguchi, Tetrahedron Lett. 25, 6015 (1984). [208] Y. Gault and H. Felkin, Bull. SOC.Chim. France 1342 (1960). [209] D. J. Cram and J. Allinger, J. Am. Chem. SOC.76, 4516 (1954). [210] D.J. Cram, F.A. Abd Elhafez and H. Weingartner, J. Am. Chem. SOC.75, 2293 (1953). [211] D.J. Cram and F.D. Greene, J. Am. Chem. SOC.75, 6005 (1953). [212] R. Guyon and P. Villa, Bull. SOC.Chim. France 152 (1977). [213] D. Kruger, A.E. Sopchik and C.A. Kingbury, J. Org. Chem. 49, 778 (1984). [214] S. Yamada and K. Koga, Tetrahedron Lett. 1711 (1967). [215] M. ProStenik and B. AlanpoviE, Croat. Chem. Acta 29, 393 (1957); C.A. 53, 113 1 (1959). [216] H.K. Muller, I. Jarchow and G. Rieck, Liebigs Ann. Chem. 613, 103 (1958). [217] K. Koga and S. Yamada, Chem. Pharm. Bull. 20, 526 (1972). [218] M. Tramontini, Synthesis 605 (1982). [219] S. Yamada and K. Koga, Tetrahedron Lett. 1711 (1967). [220] T. Kametani, K. Kigasawa, M. Hiiragi, N. Wagatsuma, T. Kohagizawa and H. Inone, J. Pharm. SOC. Jap. 100, 839 (1980). [221] G. Guanti, L. Banfi, E. Narisano and C. Scolastico, Tetrahedron Lett. 25, 4693 (1984). [222] K. Soai and A. Ookawa, J.C.S. Chem. Comm. 412 (1986). [223] D. Enders and S. Nakai, Helv. Chim. Acta 73, 1833 (1990). [224] H. Bodot, E. Dieuzeide and J. Julien, Bull. SOC.Chim. France 1086 (1960). [225] J.H. Stocker, P. Sidismuthorn, B. M. Benjamin and C. J. Collins, J. Am. Chem. SOC. 82, 3913 (1960). [226] C. Ziodrou and P. Chrysochou. Tetrahedron 33, 2103 (1977). [227] J.A. Katzenellenbogen and S.B. Bowlus, J. Org. Chem. 38, 627 (1973). [228] S. B. Bowlus and J. A. Katzenellenbogen, J. Org. Chem. 39, 3309 (1 974). [229] W. J. Gensler, F. Johnson and A. D. B. Sloan, J. Am. Chem. SOC. 82, 6074 (1960). [230] T. Nakata and T. Oishi, Tetrahedron Lett. 1641 (1980). [231] R. Frenette, M. Kakushima, R. Zamboni, R.N. Young and T.R. Verhoeven, J. Org. Chem. 52, 304 (1987). [232] P. W. Baures, D.S. Eggleston, J. R. Flisak, K. Gombatz, I. Lantos, W. Mendelson and J. J. Remich, Tetrahedron Lett. 31, 6501 (1990). [233] H. Chikashita, T. Nikaya, H. Uemura and K. Itoh, Bull. Chem. SOC.Jap. 62, 2121 (1989). [234] M. Fujita and T. Hiyama, J. Am. Chem. SOC. 107, 8294 (1985). [235] D. Hartley, Chem. Ind. 551 (1981). [236] W. Oppolzer, I. Rodriguez, C. Starkemann and E. Walther, Tetrahedron Lett. 31, 5019 (1990). [237] J. Barluenga, J.G. Resa, B. Olanoand, S. Fustero, J. Org. Chem. 52, 1425 (1987). [238] M. Brienne, C. Ouannis and J. Jacques, Bull. SOC.Chim. France 1036 (1968). [239] R. Annunziata, M. Cinquini and F. Cozzi, J.C.S. Perkin 1 1109 (1981). [240] C.R. Johnson and C. J. Stark, Jr., J. Org. Chem. 47, 1196 (1982). [241] G. Solladie, C. Greck, G. Demailly and A. Solladie-Cavallo, Tetrahedron Lett. 5047 (1982). [242] G. Solladie, C. Frechou, G. Demailly and C. Greck, J. Org. Chem. 56, 1912 (1991). [I961 [ I971 [I981 [I991 [200] [201] [202] [203] [204]

328 [243] [244] [245] [246] [247] [248] [249] [250] [251] [252] [253] [254] [255] [256] [257] [258] I2591 [260] [261] [262] [263] [264] [265] [266] [267] [268] [269] [270] [271] [272] [273] [274] [275] I2761 [277] [278] [279] [280] [281] [282]

References H. Kosugi, H. Konta and H. Uda, J. C. S. Chem. Comm. 21 1 (1985). G. Solladie, J. Hutt and F. Frechou, Tetrahedron Lett. 28, 61 (1987). G. Solladie, C. Frechou and G. Demailly, Tetrahedron Lett. 27, 2867 (1986). M. C. Careno, J.L.G. Ruano, A.M. Martin, C. Pedregal, J. H. Rodriguez, A., Rubio, J. Sanchez and G. Solladie, J. Org. Chem. 55, 2120 (1990). G. Solladie, C. Hamdouchi and M. Vicente, Tetrahedron Lett. 29, 5929 (1988). G. Solladie and N. Ghiatou, Tetrahedron Lett. 33, 1065 (1992). G.A. Molander, K.L. Bobbit and C.K. Murray, J. Am. Chem. SOC.114, 2759 (1992). K.-M. Chen, G.E. Hartmann, K. Prasad, 0. Replic and M. J. Shapiro, Tetrahedron Lett. 28, 155 (1987). F. G. Kathawala, B. Prager, K. Prasad, 0. Repic, M. J. Shapiro, R.S. Stabler and L. Widler, Helv. Chim. Acta 69, 803 (1986). S. Kiyooka, H. Kuroda and Y. Shimasaki, Tetrahedron Lett. 27, 3009 (1986). R. Haller and H.J. Schneider, Chem. Ber. 106, 1312 (1973). R. Bloch, L. Gilbert and C. Girard, Tetrahedron Lett. 29, 1021 (1986). K. Ishihara, A. Mori, I. Arai and H. Yamamoto, Tetrahedron Lett. 26, 983 (1986). A. Mori, K. Ishihara, I. Arai and H. Yamamoto, Tetrahedron 43, 755 (1987). D. A. Evans and A. H. Hoveyda, J. Am. Chem. SOC.112, 6447 (1990). A. McKenzie, J. Chem. SOC.85, 1249 (1904). A. Solladie-Cavallo and J. Suffert, Tetrahedron Lett. 26, 429 (1985). K. Soai, T. Isoda, H. Hasegawa and M. Ishizaki, Chem. Lett. 1897 (1986). G.B. Reddy, T. Minami, T. Hanamoto and T. Hiyama, J. Org. Chem. 56, 5752 (1991). D.F. Taber, P.B. Deker and M.D. Gaul, J. Am. Chem. SOC.109, 7488 (1987). G. Maier, C. Roth and R.K. Schmitt, Chem. Ber. 118, 704 (1985). G. Maier, R.K. Schmitt and U. Seipp, Chem. Ber. 118, 722 (1985). B. P. Giovanni, M. Fabio, P.G. Piero, S. Daniele, B. Achille and B. Simonetta, J. C. S. Perkin Trans 1 2983 (1982). R. Annunziata, M. Benaglia, M. Cinquini, F. Cozzi and P. Giaroni, J. Org. Chem. 57, 782 (1 992). D. C. Wigfield, Can. J. Chem. 55, 646 (1977). E. L. Eliel and S. H. Schroeter, J. Am. Chem. SOC.87, 5031 (1965). R.O. Hutchins, J. Org. Chem. 42, 920 (1977). H.C. Brown and S. Krishnamurthy, J. Am. Chem. SOC.94, 7159 (1972). J. Hooz, S. Akiyama, F. J. Cedar, M. J. Bennett and R. M. Tuggle, J. Am. Chem. SOC.96, 274 (1 974). S. Krishnamurthy and H. C. Brown, J. Am. Chem. SOC.98, 3383 (1 976). H.C. Brown, J.L. Hubbade and B. Singaram, Tetrahedron 37, 2359 (1981). S. Kim, K.H. Ahn and Y.W. Chung, J. Org. Chem. 47, 4581 (1982). G. Kovacs, G. Galambos and Z. Juvancz, Synthesis 171 (1977). E. C. Ashby, S. A. Noding and A. B. Goel, J. Org. Chem. 45, 1028 (1980). E.C. Ashby and A.B. Goel, Inorg. Chem. 17, 1862 (1978). H.C. Brown, J.S. Cha and B. Nazer, J. Org. Chem. 49, 2073 (1984). L. Gruber, I. Tomoskozy and L. Otvos, Tetrahedron Lett. 81 1 (1973). E.C. Ashley and J. R. Boone, J. Am. Chem. SOC.98, 5524 (1976). H.C. Brown and H.R. Deck, J. Am. Chem. SOC.87, 5620 (1965). J. S. Cha, M. S. Yoon, Y. S. Kim and K. W. Lee, Tetrahedron Lett. 29, 1069 (1988).

References

329

Chapter 4 H.B. Henbest, Chem. SOC.Special Publ. No. 19, 83 (1965). W. H. Pirkle and P. L. Rinaldi, J. Org. Chem. 42, 2080 (1977). D. Boschelli and A.B. Smith, Tetrahedron Lett. 4385 (1981). F.A. Davis, G. Ulatowski and M. S. Haque, J. Org. Chem. 52, 5288 (1987). I. Mergelsberg, D. Gala, D. Scherer, D. DiBenedetto and M. Tanner, Tetrahedron Lett. 33, 161 (1 992). [6] F. A. Davis, M. S. Haque, T. G. Ulatowski and J. C. Towson, J. Org. Chem. 56, 2402 (1986). [7] G. Glahsl and R. Herrmann, J. Chem. SOC.Perkin Trans. I 1753 (1988). [8] F.A. Davis and M.C. Weismiller, J. Org. Chem. 55, 3715 (1990). [9] F,A. Davis, R.T. Reddy and R.E. Reddy, J. Org. Chem. 57, 6387 (1992). [lo] R. Helder, J.C. Hummelen, R. W.P.M. Laane, J.S. Wiering and H. Wynberg, Tetrahedron Lett. 1831 (1976). [I 11 S. Julia, J. Guixer, J. Masana, J. Rocas, S. Colonna, R. Annunziata and H. Molinari, J.C.S. Perkin 1 1317 (1982). [I21 S. Yamada, T. Mashiko and S. Terashima, J. Am. Chem. SOC.99, 1988 (1977). [13] R.C. Michaelson, R.E. Palermo and K.B. Sharpless, J. Am. Chem. SOC.99, 1990 (1977). [I 41 K. B. Sharpless and T. R. Verhoeven, Aldrichimica Acta 12, 63 (1 979). [I51 T. Katsuki and K.B. Sharpless, J. Am. Chem. SOC.102, 5974 (1980). [16] B.E. Rossiter, T. Katsuki and K.B. Sharpless, J. Am. Chem. SOC.103, 464 (1981). [I71 M. J. Schweiter and K.B. Sharpless, Tetrahedron Lett. 26, 2543 (1985). [18] J. G. Hill, B. E. Rossiter and K. B. Sharpless, J. Org. Chem. 48, 3607 (1983). [19] M. Asami and N. Kanemake, Tetrahedron Lett. 30, 2125 (1989). [20] Y.-M. Wang and W.-S. Zhou, Tetrahedron 43, 2935 (1987). [21] B.E. Rossiter and K.B. Sharpless, J. Org. Chem. 49, 3707 (1984). [22] S. Ikegamu, T. Katsuki and M. Yamaguchi, Chem. Lett. 83 (1987). [23] C. J. Burns, C.A. Martin and K.B. Sharpless, J. Org. Chem. 54, 2826 (1989). [24] K.B. Sharpless, S.S. Woodard and M.G. Finn, Pure Appl. Chem. 55, 1823 (1983). [25] S.S. Woodard, M.G. Finn and B. Sharpless, J. Am. Chem. SOC.113, 106 (1991). [26] S.F. Pedersen, J.C. Dewan, R.R. Eckmann, K.B. Sharpless, J. Am. Chem. SOC.109, 1279 (1987). 1271 P.R. Carlier and K.B. Sharpless, J. Org. Chem. 54, 4016 (1989). [28] J.M. Hawkins and K.B. Sharpless, Tetrahedron Lett. 28, 2825 (1987). [29] L.D.-L. Lu, R.A. Johnson, M.G. Finn and K.B. Sharpless, J. Org. Chem. 49, 728 (1984). [30] E.J. Corey, J. Org. Chem. 55, 1693 (1990). [31] E.N. Jacobsen, W. Zhang, A. R. Muci, J.R. Ecker and L. Deng, J. Am. Chem. SOC.113,7063 (1991). [32] W. Zhang and E.N. Jacobsen, J. Org. Chem. 56, 2296 (1991). [33] T. Yamada, K. Imagawa, T. Nagata and T. Mukaiyama, Chem. Lett. 2231 (1992). [34] S. H. Zhao, 0. Samuel and H. B. Kagan, Tetrahedron 43, 5135 (1987). [35] N. Komatsu, Y. Nishibayashi, T. Sugita and S. Uemura, Tetrahedron Lett. 33, 5391 (1992). 1361 T.Sugimoto, T. Kokubo, J. Miyazaki, S. Tanimoto and M. Okano, J. C. S. Chem. Comm. 1052 (1 979). [37] T. Sugimoto, T. Kokubo, J. Miyazaki, S. Tanimoto and M. Okano, J. C. S. Chem. Comm. 402 (1979). [38] S. Colonna, N. Gagger0 and M. Leone, Tetrahedron 47, 8385 (1991). [39] S. Colonna, S. Banfi, F. Fontana and M. Sommaruga, J. Org. Chem. 50, 769 (1985). [40] S. Colonna, S. Banfi and R. Annunziata, J. Org. Chem. 51, 891 (1986). [41] B.B. Lohray, Tetrahedron Asym. 3, 1317 (1992). [I] [2] 131 [4] [S]

330

References

[42] S.G. Hentges and K.B. Sharpless, J. Am. Chem. SOC.102, 4263 (1980). [43] E. N. Jacobsen, I. Marko, W. S. Mungall, G. Schroder and K.B. Sharpless, J. Am. Chem. SOC. 110, 1968 (1988). [44] J.S.M. Wai, I. Marko, J.S. Svendsen, M. G. Finn, E.N. Jacobsen and K.B. Sharpless, J. Am. Chem. SOC.111, 1123 (1989). [45] H.-L. Kwong, C. Sorato, Y. Ogino, H. Chen and K.B. Sharpless, Tetrahedron Lett. 31, 2999 (1 991). [46] Y. Ogino, H. Chen, H.-L. Kwong and K.B. Sharpless, Tetrahedron Lett. 32, 3965 (1991). [47] Y. Ogino, H. Chen, E. Manoury, T. Shibata, M. Beller, D. Lubben and B. Sharpless, Tetrahedron Lett. 32, 5761 (1991). [48] K.B. Sharpless, W. Amberg, M. Beller, H. Chen, J Hartung, Y. Kawanami, D. Lubben, E. Manoury, Y. Ogino, T. Shibata and T. Ukita, J. Org. Chem. 56, 4585 (1991). [49] L. Wang and K.B. Sharpless, J. Am. Chem. SOC.114, 7568 (1992). [SO] K.B. Sharpless, W. Amberg, Y.L. Bennani, G.A. Crispino, J. Hartung, K.-S. Jeong, H.-L. Kwong, K. Morikawa, Z.-M. Wang, D. Xu and X.-L. Zhang, J. Org. Chem. 57, 2768 (1992). [51] T. Hashiyama, K. Morikawa and B. Sharpless, J. Org. Chem. 57, 5067 (1992). [52] K . 3 . Jeong, P. Sjo and K.B. Sharpless, Tetrahedron Lett. 33, 3833 (1992). [53] D. Xu, G.A. Crispino and K.B. Sharpless, J. Am. Chem. SOC.114, 7570 (1992). [54] Z.-M. Wang, X.-L. Zhang, K. B. Sharpless, S. C. Sinha, A. Sinha-Bagchi and E. Keinan, Tetrahedron Lett. 33, 6407 (1992). [55] E. J. Corey and G. I. Lotto, Tetrahedron Lett. 30, 2665 (1990). [56] K. Tomioka, M. Nakajima and K. Koga, J. Am. Chem. SOC.109, 6213 (1987). [57] T. Oishi and M. Hirama, J. Org. Chem. 54, 5834 (1989). I581 M. Hirama, T. Oishi and S. Ito, J. C. S. Chem. Comm. 665 (1989). 1591 K. Fuji, K. Tanaka and H. Miyamoto, Tetrahedron Lett. 33, 4021 (1992). [60] J.S. Svendsen, I. Marko, E.N. Jacobsen, Ch. P. Rao, S. Bott and B.N. Sharpless, J. Org. Chem. 54, 2265 (1989). [61] P.E. O’Bannon and W.P. Dailey, J. Am. Chem. SOC.111, 9243 (1989). [62] K. Tomioka, M. Nakajima, Y. Iitaka and K. Koga, Tetrahedron Lett. 29, 573 (1988). 1631 Y.-D. Wu, Y. Wang and K.N. Houk, J. Org. Chem. 57, 1362 (1992). [64] K.A. Jorgensen, Tetrahedron Lett. 31, 6417 (1990). [65] H.B. Henbest and R.A.L. Wilson, J. Chem. SOC. 1958 (1957). [66] H.B. Henbest, Proc. Chem. SOC.159 (1963). [67] P. Chamberlain, M.L. Roberts and G.H. Whitham, J. Chem. SOC. B 1374 (1970); R.B. Dehmel and G.H. Whitham, J.C.S. Perkin Trans 1 953 (1979). 1681 T. Itoh, K. Jitsukawa, K. Kaneda and S. Terenishi, J. Am. Chem. SOC.101, 159 (1979). [69] G. Cicala, R. Curci, M. Fiorentino and 0. Laricchiuta, J. Org. Chem. 47, 2670 (1982). [70] B..A. McKittrick and B. Ganem, Tetrahedron Lett. 26, 4895 (1985). [71] J.-L. Pierre, P. Chautemps and P. Arnaud, Bull. SOC.Chim. France 1317 (1969). [72] P. Chautemps and J.-L. Pierre, Tetrahedron 32, 549 (1976). 1731 H. Kogen and T. Nishi, J.C.S. Chem. Comm. 311 (1987). [74] P. Mohr, Tetrahedron Lett. 33, 2455 (1992). [75] S. Tanaka, H. Yamamoto, H. Nozaki, K. B. Sharpless, R. C. Michaelson and J. D. Cutting, J. Am. Chem. SOC.96, 5254 (1974). [76] B.E. Rossiter, T. R. Verhoeven and K. B. Sharpless, Tetrahedron Lett. 4733 (1979). 1771 E. D. Mihelich, Tetrahedron Lett. 4729 (1979). [78] K. Takai, K. Oshima and H. Nozaki, Bull. Chem. SOC.Jap. 56, 3791 (1983). [79] A.S. Narula, Tetrahedron Lett. 2017 (1981) and 5579 (1982). [80] A.S. Narula, Tetrahedron Lett. 5421 (1983). [81] N. Minami, S. S. KO and Y. Kishi, J. Am. Chem. SOC.104, 1109 (1982). [82] M. Isobe, M. Kitamura, S. Mio and T. Goto, Tetrahedron Lett. 221 (1982). [83] Y. Honda, A. Ori and G. Tsuchihashi, Chem. Lett. 1417 (1986).

References

33 1

[84] Y. Gao, R.M. Hanson, J.M. Klunder, S.Y. KO, H. Masamune and K.B. Sharpless, J. Am. Chem. SOC.109, 5765 (1987). [85] S. Okamoto, T. Shimazaki, Y. Kobayashi, F. Sato and K. B. Sharpless, Tetrahedron Lett. 28, 2033 (1987). [86] V.S. Martin, S. S. Woodard, T. Katsuki, Y. Yamada, M. Ikeda and K.B. Sharpless, J. Am. Chem. SOC. 103, 6237 (1981). [87] Y. Kitano, T. Matsumoto and F. Sato, Tetrahedron 44, 4073 (1988). [88] M. Kusakabe, Y. Kitano, Y. Kobayashi and F. Sato, J. Org. Chem. 54, 2085 (1989). [89] K.B. Sharpless and R.C. Michaelson, J. Am. Chem. SOC. 95, 6136 (1973). [90] T. Fukuyama, B. Vranesic, D.P. Negri and Y. Kishi, Tetrahedron Lett. 2741 (1978). [91] M.R. Johnson and Y. Kishi, Tetrahedron Lett. 4347 (1979); I. Hasan and Y. Kishi, Tetrahedron Lett. 4229 (1981). [92] K. Nishihata and M. Nishio, Tetrahedron Lett. 1041 (1977). [93] F. A. Carey, 0.D. Dailey, Jr., 0. Hernandez and J. R. Tucker, J. Org. Chem. 41, 3975 (1976). [94] I. Jalsovszky, F. Ruff, M. Kajtar-Peredy, I. Kovesdi and A. Kucsman, Tetrahedron 42, 5649 (1 986). [95] J. K. Cha, W. J. Christ and Y. Kishi, Tetrahedron 40, 2247 ( 1 984). [96] G. SolladiC, J. Hutt and C. Frechou, Tetrahedron Lett. 28, 61 (1987). [97] J. S. Panek and P.F. Cirillo, J. Am. Chem. SOC.112, 4873 (1990). [98] D.A. Evans and S. W. Kaldor, J. Org. Chem. 5, 1698 (1990). [99] R. Annunziata, M. Cinquini, F. Cozzi and L. Raimondi, Tetrahedron 44, 6897 (1988). [IOO] S. Saito, Y. Morikawa and T. Moriwake, J. Org. Chem. 55, 5424 (1990). [loll R. Gamboni and C. Tamm, Helv. Chim. Acta 69, 615 (1986). [lo21 W. Oppolzer and P. Dudfield, Helv. Chim. Acta 68, 216 (1985). [I031 D. A. Evans, M. M. Morissery and R. L. Dorow, J. Am. Chem. SOC.107, 4346 (1985). [lo41 E A . Davis and L.C. Vishwakarma, Tetrahedron Lett. 26, 3539 (1985). [105] J. Lee, S. Oya and J.K. Snyder, Tetrahedron Lett. 32, 5899 (1991). [I061 D. Enders and V. Bhushan, Tetrahedron Lett. 29, 2437 (1988).

Chapter 5 [I] H.L. Cohen and G.F. Wright, J. Org. Chem. 18, 432 (1953). [2] D. Seebach and W. Langer, Helv. Chim. Acta 62, 1701 (1979). [3] D. Seebach, H.O. Kalinowski, B. Bastani, G. Crass, H. Daum, H. Dorr, N.P. DuPreez, V. Ehrig, W. Langer, C. Niissler, H. Oei and M. Schmidt, Helv. Chim. Acta 60, 301 (1977). [4] D. Seebach, G. Crass, E. Wilka, D. Hilvert and E. Brunner, Helv. Chim. Acta 62, 2695 (1979). [5] J.-P. Mazaleyrat and D.J. Cram, J. Am. Chem. SOC.103, 4585 (1981). [6] T. Mukaiyama, K. Soai and S. Kobayashi, Chem. Lett. 219 (1978). [7] K. Soai and T. Mukaiyama, Bull. Chem. SOC.Jap. 52, 3371 (1979). [8] T. Mukaiyama, K. Soai, T. Sato, H. Shimizu and K. Suzuki, J. Am. Chem. SOC. 101, 1455 (1 979). [9] T. Akiyama, M. Shimizu and T. Mukaiyama, Chem. Lett. 611 (1984). [lo] T. Mukaiyama, K. Suzuki, K. Soai and T. Sato, Chem. Lett. 447 (1979). [Ill T. Mukaiyama and K. Suzuki, 255 (1980). [12] T. D. inch, G. J. Lewis, G. L. Sainsbury and D. J. Sellers, Tetrahedron Lett. 3657 (1969).

332

References

[I31 B. Weber and D. Seebach, Angew. Chem. 104, 194 (1992); Int. Ed. Engl. 31, 228 (1992). [I41 D. Seebach, D.A. Plattner, A. K. Beck, Y.M. Wang. D. Hunziker and W. Petter, Helv. Chim. Acta 75, 2 171 (1 992). 1151 B. Schmidt and D. Seebach, Angew. Chem. 103, 1383 (1991); Int. Ed. Engl. 30, 1321 (1991). [I 61 J. L. v. d. Busche-Hunnefeld and D. Seebach, Tetrahedron 48, 571 9 (1 992). 1171 D. Seebach, L. Behrendt and D. Felix, Angew. Chem. 103, 991 (1991); Int. Ed. Engl. 30, 1008 (1991). 1181 K. Soai, S. Yokoyama and T. Hayasaka, J. Org. Chem. 56, 4264 (1991). [I91 M. Terada, J.H. Gu, D.C. Deka, K. Mikami and T. Nakai, Chem. Lett. 29 (1992). [20] K. Soai, S. Yokoyama, T. Hayasaka and K. Ebihara, Chem. Lett. 843 (1988). [21] K. Soai, A. Ookawa, T. Kaba and K. Ogawa, J. Am. Chem. SOC.109, 71 1 1 (1987). [22] K. Soai, Y. Hirose and S. Sakata, Tetrahedron Asym. 3, 677 (1992). [23] M. Kitamura, S. Suga, K. Kawai and R. Noyori, J. Am. Chem. SOC.108, 6071 (1986). [24] M. Kitamura, S. Okada, S. Suga and R. Noyori, J. Am. Chem. SOC.111, 4028 (1989). 1251 M. Watanabe, S. Araki, Y. Butsugan and M. Uemura, J. Org. Chem. 56, 2218 (1991). [26] H. Takahashi, T. Kawakita, M. Ohno, M. Yoshioka and S. Kobayashi. Tetrahedron 48, 5691 (1 992). [27] S. Niwa and K. Soai, J.C.S. Perkin Trans. 1 2717 (1991). 1281 G. Chelucci, S. Conti, M. Falorni and G. Giacomelli, Tetrahedron 47, 8251 (1991). 1291 S. Conti, M. Falorni, G. Giacomelli and F. Soccolini, Tetrahedron 48, 8993 ( 1 992). [30] F. Bigi, G. Casiraghi, G. Casnati, G. Sartori, G.G. Fava and M. F. Berlicchi, J. Org. Chem. 50, 5018 (1985). [31] D. Abenhaim, G. Boireau and B. Sabourault, Tetrahedron Lett. 3043 (1980). [32] A.G. Olivero, B. Weidmann and D. Seebach, Helv. Chim. Acta. 64, 2485 (1981). [33] B. Weidmann, L. Wilder, A. G. Olivero, C. D. Maycock and D. Seebach, Helv. Chim. Acta. 64, 357 (1981). 1341 G. Bredig and P.S. Fiske, Biochem. Z. 46, 7 (1912). [35] V. Prelog and M. Wilhelm, Helv. Chim. Acta 37, 1634 (1954). 1361 M. Hayashi, T. Matsuda and N. Oguni, J.C.S. Chem. Comm. 1364 (1990). [37] M. Hayashi, T. Matsuda and N. Oguni, J.C.S. Perkin Trans. 1 3135 (1992). [38] S. Kobayashi, Y. Tsuchiya and T. Mukaiyama Chem. Lett. 541 (1991). [39] M.T. Reetz, T. Kukenhohner and P. Weinig, Tetrahedron Lett. 27, 5711 (1986). [40] J. D. Morrison and H. S. Mosher, Asymmetric Organic Reactions. Prentice-Hall, Englewood Cliffs, N.Y., 1971. 1411 D.J. Cram and F.A. Abd Elhafez, J. Am. Chem. SOC.74, 5828 (1952). 1421 M.T. Reetz, R. Steinbach, J. Westermann, R. Peter and B. Wenderoth, Chem. Ber. 118, 1441 (1985). [43] Y. Yamamoto, J. Yamada and T. Asano, Tetrahedron 48, 5587 (1992). [44] D. J. Cram and F.A. Abd Elhafez and H. Weingartner, J. Am. Chem. SOC.75, 2293 (1953). [45] D. J. Cram and F.A. Abd Elhafez, J. Am. Chem. SOC.76, 22 (1954). [46] D.J. Cram and J. Allinger, J. Am. Chem. 76, 4516 (1954). [47] D. J. Cram and J.D. Knight, J. Am. Chem. SOC.74, 5835 (1952). [48] D.J. Cram and E D . Greene, J. Am. Chem. SOC.75, 6005 (1953). [49] J. W. Cornforth, R.H. Cornforth and K.K. Mathew, J. Chem. SOC.112 (1959). 1501 J. H. Stocker, P. Sidisunthorn, B. M. Benjamin and C. J. Collins, J. Am. Chem. SOC.82, 3913 (1960). 1511 D. J. Cram and D.R. Wilson, J. Am. Chem. SOC.85, 1245 (1963). [52] J.H. Stocker, J. Org. Chem. 29, 3593 (1964). 1531 P. Duhamel, L. Duhamel and J. Gralak, Tetrahedron Lett. 2013 (1977). [54] A. Gaset, P. Andoye and A. Lattes, J. Appl. Chem. Biotechnol. 25, 13 (1975). 1551 A. Gaset, M.T. Maurette and A. Lattes, J. Appl. Chem. Biotechnol. 25, 19 (1975). [56] A. Gaset, M.T. Maurette and A. Lattes, C.R. Acad. Sci. Sec. C 270, 72 and 2002 (1970). [57] M. Taniguchi, H. Fujii, K. Oshima and K. Utimoto, Tetrahedron Lett. 33, 4353 (1992).

References

33 3

H. Fujii, M. Taniguchi, K. Oshima and K. Utimoto, Tetrahedron Lett. 33, 4579 (1992). M.F. Loewe, M. Boes and A.I. Meyers, Tetrahedron Lett. 26, 3295 (1985). M. Asami and R. Kimura, Chem. Lett. 1221 (1985). W.C. Still and J.H. McDonald, Tetrahedron Lett. 1031 (1980). M. Reetz, M. W. Drewes and A. Schmitz, Angew. Chem. 99, 1186 (1986); Int. Ed. Engl. 26, 1141 (1986). 1631 R. Amouroux, S. Ejjiyar and M. Chastrette, Tetrahedron Lett. 27, 1035 (1986). [64] M. T. Reetz and K. Kesseler, J. Org. Chem. 50, 5434 (1 985). [65] W.C. Still and J.A. Schneider, Tetrahedron Lett. 1035 (1980). [66] M. Nakada, Y. Urano, S. Kobayashi and M. Ohno, J. Am. Chem. SOC.110, 4826 (1988). [67] H. Chikashita, Y. Nakamura, H. Uemura and K. Itoh, Chem. Lett. 439 (1992). [68] Y. Takeda, T. Matsumoto and F. Sato, J. Org. Chem. 51, 4728 (1986). [69] H. Tomioka, T. Okinaga, K. Suzuki and G. Tsuchihashi, Tetrahedron Lett. 30, 1563 (1989). [70] D. Y. Curtin, E.E. Harris and E.K. Meislich, J. Am. Chem. SOC.74, 2901 (1952). [7 1 ] E. L. Eliel, in Asymmetric Synfhesis. Vol. 2 (J. D. Morrison, ed.). Academic Press, N. Y., 1983, p. 125. [72] M. Tramontini, Synthesis 605 (1 982). [73] M. T. Reetz, R. Steinbach, J. Westermann, R. Urz, B. Wenderoth and R. Peter, Angew. Chem. 94, 133 (1982); Int. Ed. Engl. 21, 135 (1982). [74] M.T. Reetz, R. Steinbach, J. Westermann and R. Peter, Angew. Chem. 92, 1044 (1980); Int. Ed. Engl. 19, 1011 (1980). [75] M.T. Reetz, K. Kesseler, S. Schmidtberger, B. Wenderoth and R. Steinbach, Angew. Chem. 95, 1007 (1983); Int. Ed. Engl. 22, 989 (1983). [76] R. Meric and J. Vigneron, Bull. SOC.Chim. France 327 (1973). [77] E.L. Eliel, J.K. Koksimies and B. Lohri, J. Am. Chem. SOC.100, 1614 (1978). [78] E.L. Eliel and W. J. Frazee, J. Org. Chem. 44, 398 (1979). 1791 J.E. Lynch and E.L. Eliel, J. Am. Chem. SOC.106, 2943 (1984). [80] X.-C. He and E. L. E M , Tetrahedron 43, 4979 (1987). [81] K. KO, W. J. Frazee and E.L. Eliel, Tetrahedron 40, 1333 (1984). [82] E. L. Eliel and S. Morris-Natschke, J. Am. Chem. SOC.106, 2937 (1 984). [83] S. G. Davies and C. L. Goodfelow, J. C. S. Perkin Trans. 1 192 (1 989). [84] T. Mukaiyama, Y. Sakito and M. Asami, Chem. Lett. 1253 (1978). [85] T. Mukaiyama, Y. Sakito and M. Asami, Chem. Lett. 705 (1979). [86] Y. Sakito and T. Mukaiyama, Chem. Lett. 1027 (1979). [87] M. Asami and T. Mukaiyama, Chem. Lett. 93 (1983). [88] Y. Ukaji, K. Yamamoto, M. Fukui and T. Fujisawa, Tetrahedron Lett. 47, 2919 (1991). [89] M.-J. Wu and L.N. Pridgen, J. Org. Chem. 56, 1340 (1991). I901 Y. Ukaji, H. Kanda, K. Yamamoto and T. Fujisawa, Chem. Lett. 597 (1990). [91] E. Brunet, M.S. Batra, F. J. Aguilar and J.L.G. Ruano, Tetrahedron Lett. 32, 5423 (1991). [92] J.L.G. Ruano, A.M.M. Castro and J.H. Rodriguez, Tetrahedron Lett. 47, 3195 (1991). [93] M.T. Reetz and A. Jung, J. Am. Chem. SOC.105, 4833 (1983). [94] A. McKenzie, J. Chem. SOC.85, 1249 (1904). [95] V. Prelog, Helv. Chim. Acta. 36, 308 (1953). [96] J.K. Whitesell, A. Bhattacharya and K. Henke, J.C.S. Chem. Comm. 988 (1982). [97] J.K. Whitesell, D. Deyo and A. Bhattacharya, J.C.S. Chem. Comm. 802 (1983). [98] A.I. Meyers and J. Slade, J. Org. Chem. 45, 2785 (1980). [99] Y. Kawanami, I. Fujita, Y. Taniguchi, Chem. Lett. 2021 (1987). [IOO] R. Baker, M.A. Brimble and J.A. Robinson, Tetrahedron Lett. 26, 2115 (1985). [loll A. Alexakis, R. Sedrani, J. F. Normant and P. Mangeney, Tetrahedron Asym. 1, 283 (1990). [I021 M. Eyer and D. Seebach, J. Am. Chem. SOC.107, 3601 (1985). [lo31 E.C. Ashby and J.T. Laemmle, Chem. Rev. 75, 521 (1975). 11041 G. Stork and J.M. Stryker, Tetrahedron Lett. 4887 (1983). [I051 E. C. Ashby, S. H. Yu and P. V. Roling, J. Org. Chem. 37, 1918 (I 972). [58] [59] [60] [61] [62]

334

References

(1061 T. L. MacDonald and W.C. Still, J. Am. Chem. SOC. 97, 5280 (1975). [lo71 E.C. Ashby, J.J. Lin and J.J. Watkins, Tetrahedron Lett. 1709 (1977). [I081 M.T. Reetz, R. Steinbach, J. Westermann, R. Peter and B. Wenderoth, Chem. Ber. 118, 1441 (1 985). [lo91 K. Maruoka, T. Itoh, M. Sakurai, K. Nonoshita and H. Yamamoto, J. Am. Chem. SOC.110, 3588 (1988). [I101 N. Kuneida, M. Kinoshita and J. Nokami, Chem. Lett. 289 (1977). [I 111 H. Sakuraba and S. Ushiki, Tetrahedron Lett. 31, 5349 (1990). [I121 M. Braun and W. Hild, Chem. Ber. 117, 413 (1984). [113] C. Mioskowski and G. Solladie, J. C.S. Chem. Comm. 162 (1977). [I141 G. Solladie, F. Matloubi-Moghadam, C. Luttmann and C. Mioskowski, Helv. Chim. Acta. 65, 1602 (1982). [ l 151 M. Asami and T. Mukaiyma, Chem. Lett. 17 (1980). [116] E. Nakamura and I. Kuwajiama, Tetrahedron Lett. 3347 (1983). [117] G. J. McGarvey and M. Kimura, J. Org. Chem. 47, 5420 (1982). [118] A.H. McNell and E. J. Thomas, Tetrahedron Lett. 31, 6239 (1990). [I191 R.B. Gammill, Tetrahedron Lett. 26, 1335 (1985). [I201 R.W. Hoffmann, T. Ruhl and J. Harbach, Liebigs Ann. Chem. 725 (1992). [121] R. W. Hoffmann, M. Bewersdorf, K. Ditrich, M. Kruger and R. Sturmer, Angew. Chem. 100, 1232 (1988); Int. Ed. Engl. 27, 1176 (1988). [I221 R. W. Hoffmann, K. Brumm, M. Bewersdorf, W. Mikolaiski and A. Kusche, Chem. Ber. 125, 2741 (1992). [123] K. Harada and T. Okawara, J. Org. Chem. 38, 707 (1973). [I241 D.M. Stout, L.A. Black and W.L. Matier, J. Org. Chem. 48, 5369 (1983). [125] K. Weinges, K. Cries, B. Stemmle and W. Schrank, Chem. Ber. 110, 2098 (1977). [126] K. Weinges, G. Graab, D. Nagel and B. Stemmle, Chem. Ber. 104, 3594 (1971). [127] K. Weinges, G. Brune and H. Droste, Liebigs Ann. Chem. 212 (1980). [I281 H. Takahashi, K. Tomita and H. Otamasu, J.C.S. Chem. Comm. 668 (1979). [129] H. Takahashi and H. Inagaki, Chem. Pharm. Bull. 30, 922 (1982). [I301 G. Cainelli, D. Giacomini, E. Mezzina, M. Panunzio and P. Zarantonello, Tetrahedron Lett. 47, 2967 (1991). [131] G. Cainelli, D. Giacomini, M. Panunzio, P. Zarantonello, Tetrahedron Lett. 33, 7783 (1992). [132] Y. Ukaji, T. Watai, T. Sumi and T. Fujisawa, Chem. Lett. 1555 (1991). [133] S.G. Pyne and B. Dikic, J.C.S. Chem. Comm. 826 (1989). [134] M. Rahman and P.B. Shevlin, Tetrahedron Lett. 26, 3259 (1985) [135] Y. Yamamoto, T. Komatsu and K. Marayama, J.C.S. Chem. Comm. 814 (1985). [136] M.T. Reetz, R. Jaeger, R. Drewlies and M. Hubel, Angew. Chem. 103, 76 (1991); Int. Ed. Engl. 30, 103 (1991). [137] L. Widler and D. Seebach, Helv. Chim. Acta 65, 1085 (1982). [I381 R. Kober, K. Papadopulos, W. Miltz, D. Enders and W. Steglich, Tetrahedron 41, 1693 (1985). [I391 R. W. Hoffmann, H. J. Zeiss, W. Ladner and S. Tabche, Chem. Ber. 115, 2357 (1982). [140] D. Hoppe, Angew. Chem. 96, 930 (1984); Int. Ed. Engl. 23, 932 (1984). [141] R.W. Hoffmann, Angew. Chem. 94, 569 (1982); Int. Ed. Engl. 21, 555 (1982). [I421 R. W. Hoffmann, Angew. Chem. 99, 503 (1987); Int. Ed. Engl. 26, 489 (1987). [I431 H.C. Brown and P.K. Jadhav, J. Am. Chem. SOC.105, 2092 (1983). [144] G.E. Keck and E.P. Boden, Tetrahedron Lett. 25, 265 (1984). [145] R. W. Hoffmann and H.-J. Zeiss, Angew. Chem. 91, 329 (1979); Int. Ed. Engl. 18, 306 (1979). [146] R.W. Hoffmann and W. Ladner, Tetrahedron Lett. 4653 (1979). [I471 T. Hayashi, M. Konishi and M. Kumada, J. Am. Chem. SOC.104, 4963 (1982). [148] C.T. Buse and C.H. Heathcock, Tetrahedron Lett. 1685 (1978). [149] H.E. Zimmermann and M.D. Traxler, J. Am. Chem. SOC.79, 1920 (1957). [150] D.A. Evans, J.V. Nelson and T.R. Taber, Top. Stereochem. 13, 1 (1981). [151] D. Seebach and L. Widler, Helv. Chim. Acta 65, 1972 (1982).

References [ 1521 [I531 [I541 [I551 [156] [I571 [ 1581 [I591 [I601

[I611 [162] [I631 [I641 [I651 [I661 [I671 [I681 [I691 [I701 [I711 [172] [I731 [I741 [175] [176] [I771 [178] [I791 [I801 [I811 [182] [183] [I841 [185] [186] [I871 [I881 [I891 [I901 [I911 [I921 [193] [ 1941 [I951 [196] [I971

3 35

S. E. Denmark and E. J. Weber, Helv. Chim. Acta 66, 1655 (1 983). S. E. Denmark and E. J. Weber, J. Am. Chem. SOC.106, 7970 (1984). M.T. Reetz and B. Wenderoth, Tetrahedron Lett. 5259 (1982). S. D. Kahn, C. F. Pau and W. J. Hehre, J. Am. Chem. SOC.108, 7396 (1986). S. D. Kahn, C. F. Pau, A. R. Chamberlin and W. J. Hehre, J. Am. Chem. SOC.109,650 (1987). J. Yamada, H. Abe and Y. Yamamoto, J. Am. Chem. SOC.112, 61 18 (1990). D. S. Matteson, Tetrahedron 45, 1859 (1989). U.S. Racherla and H.C. Brown, J. Org. Chem. 56, 401 (1991). H. C. Brown, R. S. Randad, K.S. Bhat, M. Zaidlewicz and U. S. Racherla, J. Am. Chem. SOC. 112, 2389 (1990). R. W. Hoffmann and U. Weidmann, Chem. Ber. 118, 3966 (1985). H. C. Brown, K. S. Bhat and R. S. Randad, J. Org. Chem. 52, 319 (1987). H.C. Brown, K.S. Bhat and R.S. Randad, J. Org. Chem. 52, 3701 (1987). H. C. Brown and K. S. Bhat, J. Am. Chem. SOC.108, 293 (1986). H. C. Brown, K. S. Bhat and P. K. Jadhav, J. C. S. Perkin Trans. 1 2633 (1991). H.C. Brown, P.K. Jadhav and K.S. Bhat, J. Am. Chem. SOC.107, 2564 (1985). H.C. Brown, P.K. Jadhav and K.S. Bhat, J. Am. Chem. SOC.110, 1535 (1988). H.C. Brown, P.K. Jadhav and P.T. Perumal, Tetrahedron Lett. 25, 511 1 (1984). Y. Yamamoto, T. Komatsu and K. Maruyama, J. Am. Chem. SOC.106, 5031 (1984). Y. Yamamoto, S. Nishii, K. Mauyama, T. Komatsu and W. Ito, J. Am. Chem. SOC.108, 7778 (1986). Y. Yamamoto. W. Ito and K. Maruyama, J.C.S. Chem. Comm. 1131 (1985). Y. Yamamoto, T. Komatsu and K. Maruyama, J. Org. Chem. 50, 3115 (1985). M. Yamaguchi and T. Mukaiyama, Chem. Lett. 993 (1980). Y. Yamamoto, T. Komatsu and K. Maruyama, J.C.S. Chem. Comm. 191 (1983). R.P. Short and S. Masamune, J. Am. Chem. SOC.111, 1892 (1989). M.M. Midland and S.B. Preston, J. Am. Chem. SOC.104, 2230 (1982). R. W. Hoffmann, Angew. Chem. 99, 503 (1987); Int. Ed. Engl. 26, 489 (1987). R.W. Hoffmann and H.-J. Zeiss, J. Org. Chem. 46, 1309 (1981). R.W. Hoffmann, B. Kemper, R. Metternich and T. Lehmeier, Liebigs Ann. Chem. 2246 (1985). T. Herold, U. Schrott and R. W. Hoffmann, Chem. Ber. 114, 359 (1981). W.R. Roush, A.E. Walts and L.K. Hoong, J. Am. Chem. SOC.107, 8186 (1985). W.R. Roush, L.K. Hoong, M.A.J. Palmer and J.C. Park, J. Org. Chem. 55, 4109 (1990). T. Herold and R. W. Hoffmann, Angew. Chem. 90, 822 (1978); Int. Ed. Engl. 17, 768 (1978). R.W. Hoffmann and T. Herold, Chem. Ber. 114, 375 (1981). W.R. Roush, L. Banfi, J.C. Park and L.K. Hoong, Tetrahedron Lett. 30, 6457 (1989). W.R. Roush, K. Ando, D.B. Powers, A.D. Palkowitz and R.L. Halterman, J. Am. Chem. SOC.112, 6339 (1990). W. R. Roush, L.K. Hoong, M.A. J. Palmer, J. A. Straub and A.D. Palkowitz, J. Org. Chem. 55, 41 17 (1990). W.R. Roush and J.C. Park, J. Org. Chem. 55, 1143 (1990). R.W. Hoffmann, H. Brinkmann and G. Frenking, Chem. Ber. 123, 2387 (1990). R.W. Hoffmann and B. Landmann, Angew. Chem. 96, 427 (1984); Int. Ed. Engl. 23, 437 (1984). R. W. Hoffmann and B. Landmann, Chem. Ber. 119, 2013 (1986). R. W. Hoffmann and B. Landmann, Chem. Ber. 119, 1039 (1986). R. W. Hoffmann and H.J. Zeiss, Angew. Chem. 92, 218 (1980); Int. Ed. Engl. 19, 218 (1980). W. R. Roush, K. Ando, D. B. Powers, R. L. Halterman and A. D. Palkowitz, Tetrahedron Lett. 29, 5579 (1988). W.R. Roush, P.T. Grover and X. Lin, Tetrahedron Lett. 31, 7563 (1990). W. R. Roush and P.T. Grover, Tetrahedron 48, 1981 (1992). W.R. Roush and R.L. Halterman, J. Am. Chem. SOC.108, 294 (1986).

336

References

[198] W. R. Roush, A.D. Palkowitz and K. Ando, J. Am. Chem. SOC.112, 6348 (1990). [199] W.R. Roush, A.D. Palkowitz and M.A. J. Palmer, J. Org. Chem. 52, 316 (1987). [200] R. W. Hoffmann and S. Dresely, Angew. Chem. 98, 186 (1986); Int. Ed. Engl. 25, 189 ( 1 986). [201] R.W. Hoffmann, S. Dresely and J.W. Lanz, Chem. Ber. 121, 1501 (1988). [202] R.W. Hoffmann, K. Ditrich, G. Koster and R. Sturmer, Chem. Ber. 122, 1783 (1989). [203] R. W. Hoffmann and S. Dresely, Chem. Ber. 122, 903 (1989). [204] R. W. Hoffmann and S. Dresely, Tetrahedron Lett. 28, 5303 (1987). [205] R. W. Hoffmann, S. Dresely and B. Hildebrandt, Chem. Ber. 121, 2225 (1988). [206] W.R. Roush, M.A. Adam, A.E. Walts and D. J. Harris, J. Am. Chem. SOC.108, 3422 (1986). [207] R. W. Hoffmann and A. Endesfelder, Liebigs Ann. Chem. 215 (1987). [208] E. Favre and M. Gaudemar, J. Organomet, Chem. 92, 17 (1975). 12091 E. J. Corey, C.-M. Yu and D.-H. Lee, J. Am. Chem. SOC.112, 878 (1990). [210] M.T. Reetz, Angew. Chem. 96, 542 (1984); Int. Ed. Engl. 23, 556 (1984). [211] M.T. Reetz, R. Steinbach, B. Wenderoth and J. Westermann, Chem. and Ind. 541 (1981). [212] R.W. Hoffmann and T. Sander, Chem. Ber. 123, 145 (1990). [213] E. Sato, K. Iida, K. Ijima, S. Moriyarand and M. Sato, J.C.S. Chem. Comm. 1140 (1981). [214] M.T. Reetz and M. Sauerwald, J. Org. Chem. 49, 2292 (1984). [215] F. Sato, H. Uchiyama, K. Iida, Y. Kobayashi and M. Sato, J. C. S. Chem. Comm. 921 (1983). [216] D. Hoppe and T. Kramer, Angew. Chem. 98, 171 (1986); Int. Ed. Engl. 25, 160 (1986). [217] T. Kramer and D. Hoppe, Tetrahedron Lett. 28, 5149 (1987). [218] D. Hoppe and 0. Zschage, Angew. Chem. 101, 67 (1989); Int. Ed. Engl. 28, 69 (1989). [219] 0. Zschage and D. Hoppe, Tetrahedron 48, 5657 (1992). [220] A. Hafner, R.O. Duthaler, R. Marti, G. Rihs, R. Rothe-Streit and F. Schwarzenbach, J. Am. Chem. SOC.114, 2321 (1992). [221] K. Furuta, M. Ishiguro, R. Haruta, N. Ikeda and H. Yamamoto, Bull. Chem. SOC.Jap. 57, 2768 (1984). [222] Y. Yamamoto, W. Ito and K. Maruyama, J.C.S. Chem. Commn. 1004 (1984). [223] T. Hayashi, M. Konishi and M. Kumada, J. Org. Chem. 48, 281 (1983). [224] K. Mikami, T. Maeda, N. Kishi and T. Nakai, Tetrahedron Lett. 25, 5151 (1984). [225] M.T. Reetz, A. Jung and C. Bolm, Tetrahedron 44, 3889 (1988). [226] T. Hayashi, K. Kabeta, I. Hamachi and M. Kumada, Tetrahedron Lett. 2865 (1983). [227] M. Kira, K. Sat0 and H. Sakurai, J. Am. Chem. SOC.112, 257 (1990). [228] S. Kano, T. Yokomatsu, H. Iwasawa and S. Shibuya, Chem. Lett. 1531 (1987). [229] H. Yatagai, Y. Yamamoto and K. Maruyama, J. Am. Chem. SOC.102, 4548 (1980). [230] A. Itoh, K. Oshima and H. Nozaki, Tetrahedron Lett. 1783 (1979). [231] I. Ojima, Y. Miyazawa and M. Kumagai, J.C.S. Chem. Comm. 927 (1976). [232] S. Kiyoka and C.H. Heathcock, Tetrahedron Lett. 4765 (1983). [233] C.H. Heathcock, S. Kiyoka and T.A. Blumenkopf, J. Org. Chem. 49, 4214 (1984). [234] M.T. Reetz, K. Kesseler and A. Jung, Tetrahedron Lett. 25, 729 (1984). [235] K. Soai, M. Ishizaki and S. Yokoyama, Chem. Lett. 341 (1987). [236] G.A. Molander and S.W. Andrews, Tetrahedron 44, 3869 (1988). [237] T. Mukaiyama, M. Ohshima and N. Miyoshi, Chem. Lett. 1121 (1987). [238] R. Imwinkelried and D. Seebach, Angew. Chem. 781 (1985); Int. Ed. Engl. 24, 765 (1985). [239] G.E. Keck and E.P. Boden, Tetrahedron Lett. 25, 1879 (1984). [240] K. J. Henry, Jr., P.A. Grieco and C.T. Jage, Tetrahedron Lett. 33, 1817 (1992). [241] G.E. Keck, S. Castellino and M.R. Wiley, J. Org. Chem. 51, 5478 (1986). I2421 S.E. Denmark and N.G. Almstead, J. Org. Chem. 56, 6485 (1991). [243] Y. Yamamoto, H. Yatagai, Y. Naruta and K. Maruyama, J. Am. Chem. SOC.102, 7107 (1980). [244] M. Koreeda and Y. Tanaka, Chem. Lett. 1297 (1982). [245] Y. Yamamoto, H. Yatagai, Y. Ishihara, N. Maeda and K. Maruyama, Tetrahedron 40, 2239 (1984). [246] M. Koreeda and Y. Tanaka, Chem. Lett. 1299 (1982). [247] M. Koreeda and Y. Tanaka, Tetrahedron Lett. 28, 143 (1987).

References [248] [249] [250] [251] [252] [253] [254] [2.55] [256] [257] [258] [259] [260] [261] [262] [263] [264] [265] [266] [267] [268] [269] [270] [271] [272] [273] [274] [275] [276] [277] [278] [279] [280] [281] [282] [283] [284] [285] [286] [287] [288] [289] [290] [291] [292] [293]

331

G.E. Keck, D.E. Abbot and M.R. Wiley, Tetrahedron Lett. 28, 139 (1987). Y. Yamamoto, K. Kobayashi, H. Okano and I. Kadota, J. Org. Chern. 57, 7003 (1992). A. J. Pratt and E. J. Thomas, J.C.S. Chem. Comm. 1 I15 (1982). A.M. Touzin, Tetrahedron Lett. 1477 (1975). Y. Yamamoto and K. Saito, J.C.S. Chern. Comrn. 1676 (1989). Y. Yarnamoto, N. Maeda and K. Maruyama, J.C.S. Chem. Comm. 774 (1983). B. W. Gung, D.T. Smith, M.A. Wolf, Tetrahedron 48, 5455 (1992). J.A. Marshall and X.-J. Wang, J. Org. Chem. 55, 6246 (1990). Y. Yarnamoto and K. Maruyama, Heterocycles 18, 357 (1982). J. Mulzer, M. Kappert, G. Huttner and I. Jibril, Angew. Chem. 96, 726 (1984); Int. Ed. Engl. 23, 704 ( 1 984). G. Zweifel and G. Hahn, J. Org. Chern. 49, 4565 (1984). D. Hoppe and F. Lichtenberg, Angew. Chem. 94, 378 (1982); Int. Ed. Engl. 21, 372 (1982). R. Hanko and D. Hoppe, Angew. Chem. 94, 378 (1982); Int. Ed. Engl. 21, 372 (1982). B.N. Minowa and T. Mukaiyama, Bull. Chem. SOC.Jap. 60, 3697 (1987). Y. Okuda, S. Hirano, T. Hiyama and H. Nozaki, J. Am. Chem. SOC.99, 3179 (1977). M.D. Lewis and Y. Kishi, Tetrahedron Lett. 2343 (1982). C. Jubert, S. Nowotny, D. Kornemann, I. Antes, C.E. Tucker and P. Knochel, J. Org. Chem. 57, 6384 (1992). K. Takai, Y. Kataoka and K. Utimoto, Tetrahedron Lett. 30, 4389 (1989). Y. Yamamoto and K. Maruyama, Tetrahedron Lett. 2895 (1981). J. W. Faller, J.A. John and M.R. Mazzieri, Tetrahedron Lett. 30, 1769 (1989). A. J. Downs, M. J. Goode and C.R. Pulham, J. Am. Chem. SOC.111, 1936 (1989). M.T. Reetz, Angew. Chem. 96, 542 (1984); Int. Ed. Engl. 23, 556 (1984). C. H. Heathcock, in Asymmetric Synthesis, Vol. 3, (J. D. Morrison, ed.). Academic Press, N.Y., 1984, p. 111. L.M. Jackmann and B.C. Lange, Tetrahedron 33, 2737 (1977). E.B. Dongala, D.L. Dull, C. Mioskowski and G. Solladie, Tetrahedron Lett. 4983 (1973). D.A. Evans and T.R. Taber, Tetrahedron Lett. 4695 (1980). D.A. Evans, J.V. Nelson, E. Vogel and T.R. Taber, J. Am. Chem. SOC.103, 3099 (1981). C.H. Heathcock, C.T. Buse, W.A. Kleschick, M.C. Pirrung, J.E. Sohn and J. Lampe, J. Org. Chem. 45, 1066 (1980). M. Guette, J. Capillon and J.-P. GuettC, Tetrahedron 29, 3659 (1973). T. Mukaiyama and N. Iwasawa, Chem. Lett. 753 (1984). J. Mulzer, P. delasalle, A. Chucholowski, U. Blascheck, G. Bruntrup, I. Jibril and G. Huttner, Tetrahedron 40, 2211 (1984). S. Masamune and W. Choy, Aldrichimica Acta 15, 47 (1982). H.B. Kagan and J.C. Fiaud, Top. Stereochem. 10, 175 (1978). C.H. Heathcock, Science 214, 395 (1981). C.H. Heathcock and C.T. White, J. Am. Chem. SOC.101, 7076 (1979). W. Ando and H. Tsumaki, Chem. Lett. 1409 (1983). D.H. Evans, J.V. Nelson and T.R. Taber, Top. Stereochem. 13, 1 (1981). R.W. Hoffmann, K. Ditrich and S. Froch, Tetrahedron 41, 5517 (1985). A. Bernardi, A.M. Capelli, A. Comotti, C. Gennari, M. Gardner, J. M. Goodman and I. Paterson, Tetrahedron 47, 3471 (1991). C. Gennari, S. Vieth, A. Comotti, A. Vulpetti, J. M. Goodman and I. Paterson, Tetrahedron 48, 4439 (1992). S. Yamago, D. Machii and E. Nakamura, J. Org. Chem. 56, 2098 (1991). J. Mulzer, G. Bruntrup, J. Finke and M. Zippel, J. Am. Chem. SOC.101, 7723 (1979). P.G.M. Wuts and M. A. Walters, J. Org. Chem. 49, 4573 (1984). J.E. Dubois and M. Dubois, Tetrahedron Lett. 4215 (1967). J.E. Dubois and P. Fellmann, Tetrahedron Lett. 1225 (1974). C. H. Heathcock and J. Lampe, J. Org. Chem. 48, 4330 (1 983).

3 38

References

[294] C.T. Buse and C. H. Heathcock, J. Am. Chem. SOC.99, 8109 (1977). [295] C.T. White and C.H. Heathcock, J. Org. Chem. 46, 191 (1981). [296] I. Mori, K. Ishihara and C.H. Heathcock, J. Org. Chem. 55, 1114 (1990). [297] D. Seebach, M. Ertas, R. Locher and W. B. Schweizer, Helv. Chim. Acta 68, 264 (1985). [298] P. Fellmann and J. E. Dubois, Tetrahedron 34, 1349 (1978). [299] E. A. Jeffery, A. Meisters and T. Mole, J. Organomet. Chem. 74, 365 (1 974). [300] H. 0. House, D. S. Crumrine, A. Y. Teranishi and H. D. Olmstead, J. Am. Chem. SOC.95, 3310 (1 973). [301] K.K. Heng and R.A. J. Smith, Tetrahedron 34, 425 (1979). [302] Y. Yamamoto and K. Maruyama, Tetrahedron Lett. 21, 4607 (1981). [303] M.T. Reetz and R. Peter, Tetrahedron Lett. 4691 (1981). [304] E. Nakamura and I. Kuwajima, Tetrahedron Lett. 3343 (1983). [305] W.W. Fenzl and R. Koster, Liebigs Ann. Chem. 1322 (1975). [306] S. Masamune, S. Mori, D. Van Horn and D. W. Brooke, Tetrahedron Lett. 1665 (1979). [307] T. Inoue and T. Mukaiyama, Bull. Chem. SOC.Jap. 53, 174 (1980). 13081 I. Kuwajima, M. Kato and A. Mori, Tetrahedron Lett. 4291 (1980). [309] R.W. Hoffmann and K. Ditrich, Tetrahedron Lett. 25, 1781 (1984). [310] C. Gennari, S. Cardani, L. Colombo and C. Scolastico, Tetrahedron Lett. 25, 2283 (1984). I3111 T. Mukaiyama, K. Banno and K. Narasaka, J. Am. Chem. SOC.96, 7503 (1974). [312] R. Noyori, K. Yokoyama, J. Sabata, I. Kuwajima, E. Nakamura and M. Shimizu, J. Am. Chem. SOC.99, 1265 (1977). [313] W.A. Kleschick, C.T. Buse and C.H. Heathcock, J. Am. Chem. SOC.99, 247 (1977). [314] I. Mori, K. Ishihara and C. H. Heathcock. J. Org. Chem. 55, 11 14 (1990). [315] R.W. Hoffmann, K. Ditrich and S. Froch, Liebigs Ann. Chem. 977 (1987). [316] Y. Yamamoto, H. Yatagai and K. Maruyama, J.C.S. Chem. Comm. 162 (1981). [317] S.S. Labadie and J.K. Stille, Tetrahedron 40, 2329 (1984). [318] S. Shenvi and J.K. Stille, Tetrahedron Lett. 627 (1982). [319] K. Kobayashi, M. Kawanisi, T. Hitomi and S. Nozima, Chem. Lett. 851 (1983). [320] T. Harada and T. Mukaiyama, Chem. Lett. 467 (1982). [321] T. Mukaiyama, R.W. Stevens and N. Iwasawa, Chem. Lett. 353 (1982). [322] T. Mukaiyama, T. Yura and N. Iwasawa, Chem. Lett. 809 (1985). [323] A.I. Meyers and P. J. Reider, J. Am. Chem. SOC.101, 2501 (1979). [324] C. H. Heathcock, M. C. Pirrung, S. H. Montgomery and J. Lampe, Tetrahedron 37, 4087 (1981). [325] M.C. Pirrung and C.H. Heathcock, J. Org. Chem. 45, 1727 (1980). [326] T. Harada and T. Mukaiyama, Chem. Lett. 161 (1982). [327] J. Canceill, J. J. Basselier and J. Jacques, Bull. Chim. France 1024 (1967). [328] J. Canceill and J. Jacques, Bull. SOC.Chim. France 2180 (1970). [329] A. Shanzer, L. Somekh and D. Butina, J. Org. Chem. 44, 3967 (1979). [330] D.A. Evans and L.R. McGhee, Tetrahedron Lett. 3975 (1980). [331] M. Hirama, D.S. Garvey, L.D.-L. Lu and S. Masamune, Tetrahedron Lett. 3937 (1979). [332] M. Hirama and S. Masamune, Tetrahedron Lett. 2229 (1979). [333] J. Mulzer, M. Zippel, G. Bruntrup, J. Segner and J. Finke, Liebigs Ann. Chem. 1108 (1980). [334] Y. Ito and S. Terashima, Tetrahedron Lett. 28, 6629 (1987). [335] P. J. Murphy, G. Procter and A.T. Russel, Tetrahedron Lett. 28, 2037 (1987). [336] N. Iwasawa, H. Huang and T. Mukaiyama, Chem. Lett. 1045 (1985). [337] Y. Tamaru, T. Harada, S. Nishi, M. Mizutani, T. Hoiki and 2. Yoshida, J. Am. Chem. SOC. 102, 7806 (1980). [338] C. Goasdoue, N. Goasdoue and M. Gaudemar, Tetrahedron Lett. 4001 (1983). [339] D.M. von Schriltz, E.M. Kaiser and C.R. Hauser, J. Org. Chem. 32, 2610 (1967). [340] E. Nakamura, K. Hashimoto and I. Kuwajima, Tetrahedron Lett. 2079 (1978). 13411 Y. Tamura, Y. Amino, Y. Furukawa, M. Kagotani and Z. Yoshida, J. Am. Chem. SOC.104, 4018.

References

339

I . Fleming and J. D. Kilburn, J. C. S. Chem. Comm. 305 (1 986). I. Fleming and A. K. Sarkar, J. C. S. Chem. Comm. 1199 (1986). Y. Kudo, M. Iwasawa, M. Kobayashi, Y. Senda and S. Mitsui, Tetrahedron Lett. 2125 (1972). Y. Tamuru, T. Hioki, S. Kawamura, H. Satomi and Z. Yoshida, J. Am. Chem. SOC.106, 3876 (1 984). [346] D.A. Evans, E. Vogel and J.V. Nelson, J. Am. Chem. SOC.101, 6120 (1979). 13471 H. Hamana, K. Sasakura and T. Sugasawa, Chem. Lett. 1729 (1984). [348] T. H. Chan, T. Aida, P. W.K. Lan, V. Gorys and D. H. Harpp, Tetrahedron Lett. 4029 (1979). [349] R. Noyori, I. Nishida, and J. Sakata, J. Am. Chem. SOC.103, 2106 (1981). [350] C.H. Heathcock and L.A. Flippin, J. Am. Chem. SOC.105, 1667 (1983). [351] S. Murata, M. Suzuki and R. Noyori, J. Am. Chem. SOC.102, 3248 (1980). [352] C. Gennari, M.G. Beretta, A. Bernardi, G. Moro, Tetrahedron 42, 893 (1986). [353] R.E. Ireland, P. Wipf and J.D. Armstrong, 111, J. Org. Chem. 56, 650 (1991). [354] C. Palazzi, L. Colombo and C. Gennari, Tetrahedron Lett. 26, 1735 (1985). [355] K. Homma, H. Takenoshita and T. Mukaiyama, Bull. Chem. SOC.Jap. 63, 1898 (1990). I3561 T. Mukaiyama, H. Akamatsu and J.S. Han, Chem. Lett. 889 (1990). [357] S. Shimada, K. Saigo, M. Abe, A. Sudo and M. Hasegawa, Chem. Lett. 1445 (1992). [358] N. Iwasawa and T. Mukaiyama, Chem. Lett. 1441 (1982). [359] R. W. Stevens and T. Mukaiyama, Chem. Lett. 1799 (1983). [360] D.A. Evans and L.R. McGee, J. Am. Chem. SOC.103, 2876 (1981). [361] E. Eichenauer, E. Friedrich, W. Lutz and D. Enders, Angew. Chem. 90, 219 (1978); Int. Ed. Engl. 17, 206 (1978). [362] M. Lucas and J.-P. Guette, J. Chem. Res. (S) 53 (1980). [363] R.W. Stevens and T. Mukaiyma, Chem. Lett. 595 (1983). [364] R. W. Stevens, N. Iwasawa and T. Mukaiyama, Chem. Lett. 1459 (1984). [365] 0. Piccolo, L. Filippini, L. Tinucci, E. Valoti and A. Citterlo, Helv. Chim. Acta 67, 739 (1984). [366] S. Kobayashi, H. Uchiro, Y. Fujishita, L. Shiina and T. Mukaiyama, J. Am. Chem. SOC.113, 4247 (1991). [367] E.R. Parmee, Y. Hong, 0. Tempkin and S. Masamune, Tetrahedron Lett. 33, 1729 (1992). [368] J. Corey, R. Imwinkelried, S. Pikul and Y.B. Xiang, J. Am. Chem. SOC.111, 5493 (1989). [369] S. Kiyooka, Y. Kaneko, M. Komura, H. Matsuo and M. Nakano, J. Org. Chem. 56, 2276 (1 991). [370] S. Kobayashi, Y. Fujishita and T. Mukaiyama, Chem. Lett. 2069 (1989). [371] N. Iwasawa, T. Yura and T. Mukaiyama, Tetrahedron 45, 1197 (1989). [372] S. Kobayashi and T. Mukaiyama, Chem. Lett. 297 (1989). [373] S. Kobayashi, T. Sano and T. Mukaiyama, Chem. Lett. 1319 (1989). [374] T. Mukaiyama, H. Uchiro, I. Shina and S. Kobayashi, Chem. Lett. 1019 (1990). [375] H. Yoda, T. Shirai, T. Kawasaki, T. Katagiri, K. Takabe, K. Kimata and K. Hosoya, Chem. Lett. 793 (1991). [376] T. Mukaiyama, H. Uchiro and S. Kobayashi, Chem. Lett. 1001 (1989). [377] T. Mukaiyama, H. Uchiro and S. Kobayashi, Chem. Lett. 1757 (1989). [378] T. Mukaiyama, M. Furuya, A. Ohtsubo and S. Kobayashi, Chem. Lett. 989 (1991). [379] T. Mukaiyama, H. Uchiro and S. Kobayashi, Chem. Lett. 1147 (1990). [380] S. Kobayashi, Y. Fujishita and T. Mukaiyama, Chem. Lett. 1455 (1990). [381] T. Mukaiyama, S. Kobayashi, H. Uchiro and I. Shiina, Chem. Lett. 129 (1990). [382] T. Mukaiyama, I. Shiina and S. Kobayashi, Chem. Lett. 1901 (1991). [383] S. Kobayashi, A. Ohtsubo and T. Mukaiyama, Chem. Lett 831 (1991). [384] A. Ando and T. Shioiri, Tetrahedron 45, 4969 (1989). [385] M. Muraoka, H. Kawasaki and K. Koga, Tetrahedron Lett. 29, 337 (1988). [386] J.A. Reid and E.E. Turner, J. Chem. SOC.3565 (1949); ibid 3694 (1950). [387] R.O. Duthaler, P. Herold, W. Lottenbach, K. Oertle and M. Riediker, Angew. Chem. 101, 490 (1989); Int. Ed. Engl. 28, 495 (1989). [342] [343] [344] [345]

340

References

[388] G. Bold, R.O. Duthaler and M. Riediker, Angew. Chem. 101, 491 (1989); Int. Ed. Engl. 28, 497 (1989). [389] R. Devant, U. Mahler and M. Braun, Chem. Ber. 121, 397 (1988). [390] J. E. Lynch, R. P. Volante, R. V. Wattley, L. Shinkai, Tetrahedron Lett. 28, 1385 (1987). [391] D.A. Evans, J. Bartroli and T.L. Shih, J. Am. Chem. SOC.103, 2127 (1981). [392] M. Nerz-Stormes and E.R. Thornton, J. Org. Chem. 56, 2489 (1991). [393] H. Danda, M.M. Hansen and C.H. Heathcock, J. Org. Chem. 55, 173 (1990). [394] A. Abdel-Magid, L. N. Pridgen, D. S. Egleston and I . Lantos, J. Am. Chem. Soc. 108, 4595 (1986). [395] L.N. Pridgen, A. Abdel-Magid and I. Lantos, Tetrahedron Lett. 30, 5539 (1989). [396] M.A. Walker and C.H. Heathcock, J. Org. Chem. 56, 5747 (1991). [397] Y. Nagao, S. Yamada, T. Kumagai, M. Ochiai and E. Fujita, J. C. S. Chem. Comm. 1418 (1 985). [398] K. H. Ahn, D. J. Yo0 and J. S. Kim, Tetrahedron Lett. 33, 6661 (1992). [399] W. Sankhavasi, M. Yamamoto, S. Kohmoto and K. Yamada, Bull. Chem. SOC.Jap. 64, 1425 (1991). [400] W. Oppolzer, I. Rodriguez, C. Starkemann and E. Walther, Tetrahedron Lett. 31, 5019 (1990). [401] K. H. Ahn, S. Lee and A. Lim, J. Org. Chem. 57, 5065 (1992). [402] K. Busch, U.M. Groth, W. Kuhnle and U. Schollkopf, Tetrahedron 48, 5607 (1992). [403] T. Katsuki and M. Yamaguchi, Tetrahedron Lett. 26, 5807 (1985). [404] D.A. Evans and A.E. Weber, J. Am. Chem. SOC.108, 6757 (1986). [405] D. A. Evans, E. B. Sjogren, A. E. Weber and R. E. Conn, Tetrahedron Lett. 28, 39 (1 987). [406] W. Oppolzer, J. Blagg, I. Rodriguez and E. Walther, J. Am. Chem. SOC.112, 2767 (1990). [407] W. Oppolzer, C. Starkemann, L. Rodriguez and G. Bernardinelli, Tetrahedron Lett. 47, 61 ( 1 99 I ). [408] W. Oppolzer and C. Starkemann, Tetrahedron Lett. 33, 2439 (1992). [409] S. Masamune, T. Sato, B. Kim and T. A. Wolmann, J. Am. Chem. SOC.108, 8279 (1986). [410] R.P. Short and S. Masamune, Tetrahedron Lett. 28, 2841 (1987). [411] M.T. Reetz, F. Kunisch and P. Heitmann, Tetrahedron Lett. 27, 4721 (1986). [412] M.R. Reetz, E. Rivadeneira and C. Niemeyer, Tetrahedron Lett. 31, 3863 (1990). [413] E. J. Corey and S.S. Kim, Tetrahedron Lett. 31, 3715 (1990). [414] E. J. Corey and S. S. Kim, J. Am. Chem. SOC.112, 4976 (1990). [415] E. J. Corey and S. Choi, Tetrahedron Lett. 47, 2857 (1991). [416] E. J. Corey, C.P. Decicco and R. C. Newbold, Tetrahedron Lett. 32, 5287 (1991). [417] I. Paterson, M.A. Lister and C.K. McClure, Tetrahedron Lett. 27, 4787 (1986). [418] C. Gennari, A. Bernardi, C. Scolastico and D. Potenza, Tetrahedron Lett. 26, 4129 (1985). [419] C. Gennari, L. Colombo, G. Bertolini and G. Schimperna, J. Org. Chem. 52, 2754 (1987). [420] C. Gennari, A. Bernardi, L. Colombo and C. Scolastico, J. Am. Chem. SOC.107, 5812 (1985). [421] C. Gennari, F. Molinari, P. Cozzi and A. Oliva, Tetrahedron Lett. 30, 5163 (1989). [422] G. Helrnchen, U. Leikauf, and I. Taufer-Knopfel, Angew. Chern. 874 (1985); Int. Ed. Engl. 24, 874 (1985). [423] R. 0. Duthaler, P. Herold, S. Wyler-Helfer and M. Riediker, Helv. Chim. Acta 73, 659 (1990). [424] C. Mioskowski and G. Solladie, J.C.S. Chem. Comrn. 162 (1977). [425] C. Mioskowski and G. Solladie, Tetrahedron 26, 277 (1980). [426] R. Anunnziata, M. Cinquini, F. Cozzi, F. Montanari and A. Restelli, J. C. S. Chem. Comm. 1138 (1983). [427] D. Seebach, V. Ehrig and M. Teschner, Liebigs Ann. Chem. 1357 (1976). [428] S. Masamune, W. Choy, F.A. J. Kerdesky and B. Imperiali, J. Am. Chem. SOC.103, 1566 (1 98 1). [429] W. Choy, P. Ma and S. Masarnune, Tetrahedron Lett. 3555 (1981). [430] C.H. Heathcock, M.C. Pirrung, J. Lampe, C.T. Buse and S.D. Young, J. Org. Chem. 46, 2290 (1981).

References

341

[431] C. H. Heathcock, M. C. Pirrung, C.T. Busse, J. R. Hagen, S.D. Young and J. E. Sohn, J. Am. Chem. SOC.101, 7077 (1979). [432] Siege1 and E. R. Thornton, J. Am. Chem. SOC.111, 5722 (1989). [433] C.H. Heathcock, S. D. Young, J.P. Hagen, M.C. Pirrung, C.T. White and D. VanDerveer, J. Org. Chem. 45, 3846 (1980). [434] N. Aktogu, H, Felkin and S.G. Davies, J.C.S. Chem. Comm. 1303 (1982). [435] S.G. Davies, I.M. Dordor and P. Warner, J.C. S. Chem. Comm. 957 (1984). [436] L.S. Liebeskind and M.E. Welker, Tetrahedron Lett. 25, 4341 (1984). [437] C.H. Heathcock and S. Arseniyadis, Tetrahedron Lett. 26, 6009 (1985). [438] I. Paterson, J.M. Goodman and M. Isaka, Tetrahedron Lett. 30, 7121 (1989). [439] A.I. Meyers and Y. Yamamoto, J. Am. Chem. SOC.103, 4278 (1981). [440] A.I. Meyers and Y. Yamamoto, Tetrahedron 40, 2309 (1984). [441] S. Masamune, S.A. Ah, D.L. Snitman and D.S. Garvey, Angew. Chem. 92, 573 (1980); Int. Ed. Engl. 19, 557 (1980). [442] M. T. Reetz and K. Kesseler, J. C. S. Chem. Comm. 1079 (1 984). [443] C. H. Heathcock, S. K. Davidsen, K. T. Hug and L. A. Flippin, J. Org. Chem. 51, 3027 (1986). [444] J.-I. Uenishi, H. Tomozane and M. Yamamoto, J.C.S. Chem. Comm 717 (1985). [445] F. Bigi, G. Casnati, G. Sartori, G. Araldi and G. Bocelli, Tetrahedron Lett. 30, 1121 (1989). [446] M.T. Reetz, B. Raguse, C. F. Marth, H.M. Hugel, T. Bach and D. N. A. Fox, Tetrahedron 48, 5731 (1992). [447] Y. Sakito, M. Asami and T. Mukaiyama, Chem. Lett. 455 (1980). [448] E. Nakamura, Y. Horiguchi, J. Shimada and I. Kuwajima, J. C. S. Chem. Comm. 796 (1 983). [449] U. Eder, G. Sauer and R. Wiechert, Angew. Chem. 83, 492 (1971); Int. Ed. Engl. 10, 496 (1 971). [450] Z.G. Hajos and D.R. Parrish, J. Org. Chem. 39, 1615 (1974). [451] J. Gutzwiller, P. Buchschacher and A. Furst, Synthesis 167 (1977). [452] P. Buchschacher, J. Casal, A. Furst and W. Meier, Helv. Chim. Acta 60, 2747 (1977). [453] Y. Tamai, Y. Mizutani, H. Hagiwara, H. Uda and N. Harada, Res. 148 (1985). [454] P. A. Grieco, N. Fukamiya and M. Miyashita, J. C. S. Chem. Comm. 573 (1976). [455] S. Danishefsky and P. Cain, J. Am. Chem. SOC.98, 4975 (1976). [456] I. Shimizu, Y. Naito and J. Tsuji, Tetrahedron Lett. 487 (1980). [457] T. Wakabayashi, K. Watanabe and Y. Kato, Synth. Comm. 7, 238 (1977). [458] S. Terashima, S. Sato and K. Koga, Tetrahedron Lett. 3469 (1979). [459] C. Agami and H. Senestre, J.C.S. Chem. Comm. 1385 (1984). [460] C. Agami, F. Meyner, C. Puchot, J. Guilhem and C. Pascard, Tetrahedron 40, 1031 (1984). [461] C. Agami, Bull. SOC.Chim. France 499 (1988). [462] Y. Ito, M. Sawamura, E. Shirakawa, K. Hajashizaki and T. Hayashi, Tetrahedron 44, 5253 (1988). [463] M. Sawamura, H. Hamashima and Y. Ito, J. Org. Chem. 55, 5935 (1990). [464] Y. Ito, M. Sawamura and T. Hayashi, J. Am. Chem. SOC.108, 6405 (1986). [465] Y. Ito, M. Sawamura, H. Hamashima, T. Emura and T. Hayashi, Tetrahedron Lett. 30, 4681 (f 989). [466] Y. Ito, M. Sawamura, E. Shirakawa, K. Hayashizaki and T. Hayashi, Tetrahedron Lett. 29, 235 (1988). [467] M. Sawamura and Y. Ito, Tetrahedron Lett. 31, 2723 (1990). [468] I. Matsuda, A. Ogiso, S. Sato and Y. Izumi, J. Am. Chem. SOC.111, 2333 (1989). [469] T. Hayashi, Y. Uozumi, A. Yamazaki, Tetrahedron Lett. 47, 2799 (1991). [470] Y. Ito, M. Sawamura and T. Hayashi, Tetrahedron Lett. 28, 6215 (1987). [471] G.E. Keck and D.A. Abbott, Tetrahedron Lett. 25, 1883 (1984). [472] A. Togni and S.D. Pastor, J. Org. Chem. 55, 1649 (1990). [473] M. Sawamura, Y. Ito and T. Hayashi, Tetrahedron Lett. 30, 2247 (1989). [474] Y. Ito, M. Sawamura, M. Kobayashi and T. Hayashi, Tetrahedron Lett. 29, 6321 (1988). [475] A. Togni and S. Pastor, Helv. Chim. Acta 72, 1038 (1989).

342

References

Chapter 6 [l] [2] [3] [4] [5] [6] [7] [8] 191 [lo] [Ill 1121 [13] [14] [15] [16] 1171 [18] [19] [20] [21] [22] [23] 1241 [25] 1261 [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43]

H. Wynberg and R. Helder, Tetrahedron Lett. 4057 (1975). K. Hermann and H. Wynberg, J. Org. Chem. 44, 2238 (1979). D. J. Cram and G.D.Y. Sogah, J.C.S. Chem. Comm. 625 (1981). M. Sawamura, H. Hamashima and Y. Ito, J. Am. Chem. SOC.114, 8295 (1992). T. Mukaiyama, Y. Hirato and T. Takeda, Chem. Lett. 461 (1978). D. Enders and B.M. Rendenbach, Chem. Ber. 120, 1223 (1987). D. A. Evans, M.T. Bilodeau, T.C. Sommers, J. Clardy, D. Cherry and Y. Kato, J. Org. Chem. 56, 5750 (1991). G. Castaldi, S. Cavicchioli, C. Giordano and F. Uggeri, Angew. Chem. 98, 273 (1986); Int. Ed. Engl. 25, 259 (1986). G. Stork, J.D. Winkler and N.A. Sacconamo, Tetrahedron Lett. 465 (1983). H. Kawano, Y. Ishii, M. Saburi and Y. Uchida, J.C.S. Chem. Comm. 87 (1988). T. Mukaiyama, Y. Sagawa and S. Kobayashi, Chem. Lett. 1821 (1986). C.H. Heathcock and D.E. Uehling, J. Org. Chem. 51, 2797 (1986). C.H. Heathcock, M.N. Norman and D.E. Uehling, J. Am. Chem. SOC.107, 2797 (1985). Y. Hashimoto and T. Mukaiyama, Chem. Lett. 1623 (1986). D.A. Oare and C.H. Heathcock, Tetrahedron Lett. 27, 6169 (1986). A. Bernardi, P. Dotti, G. Poli and C. Scolastico, Tetrahedron 48, 5597 (1992). D. A. Oare and C. H. Heathcock, J. Org. Chem. 55, 157 (1990). T. Mukaiyama, M. Tamura and S. Kobayashi, Chem. Lett. 1817 (1986). K. Kpegba, P. Metzner and R. Rakotonirina, Tetrahedron 45, 2041 (1989). M. Yamaguchi, M. Tsukamoto, S. Tanaka and I. Hirao, Tetrahedron Lett. 25, 5661 (1984). M. Yamaguchi, Chem. Lett. 377 (1990). M. Yamaguchi, K. Haseba, S. Tanaka and T. Minami, Tetrahedron Lett. 27, 959 (1986). D.A. Oare, M. A. Henderson, M.A. Sanner and C. H. Heathcock, J. Org. Chem. 55, 132 (1990). C.H. Heathcock, M.A. Henderson, D.A. Oare and M.A. Sanner, J. Org. Chem. 50, 3019 (1985). M. Yamaguchi, M. Hamada, S. Kawasaki and T. Minami, Chem. Lett. f085 (1986). D. Enders, K. Papadopoulos and B.E.M. Rendenbach, Tetrahedron Lett. 27, 3491 (1986). A. Bernardi, A.M. Capelli, C. Gennari and C. Scolastico, Tetrahedron Asym. 1, 21 (1990). W. Langer and D. Seebach, Helv. Chim. Acta 62, 1710 (1979). K. Soai, M. Okudu and M. Okamoto, Tetrahedron Lett. 47, 95 (1991). E. J. Corey, R. Naef and F. J. Hannon, J. Am. Chem. SOC.108, 71 14 (1986). K. Tanaka, J. Matsui, H. Suzuki and A. Watanabe, J.C.S. Perkin Trans. 1 1193 (1992). B.E. Rossiter and M. Eguchi, Tetrahedron Lett. 31, 965 (1990). R.K. Dieter and M. Tokles, J. Am. Chem. SOC.109, 2040 (1987). K. Yamamoto, M. Kanoh, N. Yamamoto and J. Tsuji, Tetrahedron Lett. 28, 6347 (1987). W. Oppolzer and T. Stevenson, Tetrahedron Lett. 27, 1139 (1986). W. Oppolzer and H. J. Ldher, Helv. Chim. Acta 64, 2808 (1981). W. Oppolzer, G. Poli, A. J. Kingma, C. Starkemann and G. Bernardinelli, Helv. Chim. Acta 70, 2201 (1987). W. Oppolzer and A. J. Kingma, Helv. Chim. Acta. 72, 1337 (1 990). W. Oppolzer, R. J. Mills, W. Pachinger and T. Stevenson, Helv. Chim. Acta 69, 1542 (1986). W. Oppolzer, A. J. Kingma and G. Poli, Tetrahedron 45, 479 (1989). R. Tamura, K. Watabe, N. Ono and Y. Yamamoto, J. Org. Chem. 57, 4895 (1992). W. Oppolzer, P. Dudfield, T. Stevenson and T. Godel, Helv. Chim. Acta 68, 212 (1985). G. Helmchen and G. Wegner, Tetrahedron Lett. 26, 6051 (1985).

References

343

D.M. Scott, A.T. McPhail and N.A. Porter, Tetrahedron Lett. 31, 1679 (1990). B. Giese, M. Zehnder, M. Roth and H.-G. Zeitz, J. Am. Chem. Soc. 112, 6741 (1990). S. Cardani, G. Poli, C. Scolastico and R. Villa, Tetrahedron 44, 5929 (1988). A. Bernardi, S. Cardani, G. Poli and C. Scolastico, J. Org. Chem. 51, 5041 (1986). J.P. Konopelski, K.S. Chu and G. R. Negrete, J. Org. Chem. 56, 1355 (1991). T. Mukaiyama, T. Takeda and M. Osaki, Chem. Lett. 1165 (1977). T. Mukaiyama, T. Takeda and K. Fujimoto, Bull. Chem. Soc. Japan 51, 3368 (1978). T. Mukaiyama and N. Iwasawa, Chem. Lett. 913 (1981). M. Asami and T. Mukaiyama, Chem. Lett. 569 (1979). K. Soai, A. Ookawa and Y. Nohara, Synth. Comm. 13, 27 (1983). J. Fujiwara, Y. Fukutani, M. Hasegawa, K. Maruoka and H. Yamamoto, J. Am. Chem. Soc. 106, 5004 (1984). [55] Y. Fukutani, K. Maruoka and H. Yamamoto, Tetrahedron Lett. 25, 591 1 (1984). [56] Y. Yamamoto and S. Nishii, J. Org. Chem. 53, 3597 (1988). [57] T. Ibuka, M. Tanaka, S. Nishii and Y. Yamamoto, J.C.S. Chem. Comm. 1596 (1987). [58] T. Ibuka, M. Tanaka, S. Nishii and Y. Yamamoto, J. Am. Chem. SOC. 111, 4864 (1989). [59] H.O. House and W.F. Fischer, Jr., J. Org. Chem. 33, 949 (1968). [60] M. Asaoka, K. Shima and H. Takei, J.C.S. Chem. Comm. 430 (1988). [61] E. J. Corey, F. J. Hannon and N. W. Boaz, Tetrahedron 45, 545 (1989). [62] A.B. Smith 111, N.K. Dunlap and G.A. Sulikowski, Tetrahedron Lett. 29, 439 (1988). [63] A.B. Smith 111, Tetrahedron Lett. 29, 443 (1988). [64] Y. Yamamoto, S. Nishii, and T. Ibuka, J.C.S. Chem. Comm. 1572 (1987). [65] Y. Yamamoto, S. Nishii and T. Ibuka, J. Am. Chem. Soc. 110, 617 (1988). [66] Y. Yamamoto, Y. Chounan, S. Nishii, T. Ibuka and H. Kitahara, J. Am. Chem. SOC. 114, 7652 (1992). [67] J. Leonard and G. Ryan, Tetrahedron Lett. 28, 2525 (1987). [68] S. Hanessian and K. Sumi, Synthesis 1083 (1991). [69] J. Crosby, Tetrahedron 47, 4789 (1991). (701 M. Larcheveque, G. Tamagnan and Y. Petit, J.C.S. Chem. Comm. 31 (1989). [71] M.T. Reetz and D. Rohring, Angew. Chem. 101, 1732 (1989); Int. Ed. Engl. 28, 1706 (1989). [72] M. Isobe, M. Kitamura and T. Goto, Tetrahedron Lett. 3465 (1979). [73] M. Isobe, M. Kitamura and T. Goto, Tetrahedron Lett. 4727 (1980). (741 M. Isobe, M. Kitamura and T. Goto, Tetrahedron Lett. 239 (1981). [75] C. Alcaraz, J.C. Carretero and E. Dominguez, Tetrahedron Lett. 47, 1385 (1991). [76] M. Isobe, Y. Ichikawa, Y. Funabashi, S. Mio and T. Goto, Tetrahedron 42, 2863 (1986). [77] G.H. Posner, in Asymmetric Synthesis, Vol. 2 (J.D. Morrison, ed.). Academic Press, New York, 1983, p. 225. [78] G.H. Posner, J.P. Mallamo and K. Miura, J. Am. Chem. SOC. 103, 2886 (1981). [79] G. Tsuchihashi, S. Mitamura, S. Inone and K. Ogura, Tetrahedron Lett. 323 (1973). [80] G.H. Posner and M. Hulce, Tetrahedron Lett. 25, 379 (1984). [81] G.H. Posner and L.L. Frye, Isr. J. Chem. 24, 8 (1984). [82] G.H. Posner, T.P. Kogan and M. Hulce, Tetrahedron Lett. 25, 383 (1984). [83] W.C. Still and I. Galynker, Tetrahedron 37, 3981 (1981). [84] S. Hashimoto, S. Yamada and K. Koga, J. Am. Chem. SOC. 98, 7450 (1976). [85] S. Hashimoto, N. Komeshima, S. Yamada and K. Koga, Tetrahedron Lett. 2907 (1977). [86] S. Hashimoto, S. Yamada and K. Koga, Chem. Pharm. Bull. 27, 771 (1979). [87] S. Hashimoto, H. Kogen, K. Tomioka and K. Koga, Tetrahedron Lett. 3009 (1979). [88] H. Kogen, K. Tomioka, S. Hashimoto and K. Koga, Tetrahedron 37, 3951 (1981). [89] K. Tomioka, F. Masumi, T. Yamashita and K. Koga, Tetrahedron Lett. 25, 333 (1984). [90] A. I. Meyers, R. K. Smith and C. E. Whitten, J. Org. Chem. 44, 2250 (1979). [91] A.I. Meyers and C.E. Whitten, J. Am. Chem. SOC.97, 6266 (1975). [92] D. J. Rawson and A. I. Meyers, J. Org. Chem. 56, 2292 (1991). [93] A.I. Meyers and M. Shipman, J. Org. Chem. 56, 7098 (1991).

[44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54]

344

References

J. Berlan, Y. Besace, G. Pourcelot and P. Cresson, J. Organomet. Chem. 256, 181 (1983). J. Berlan and Y. Besace, Tetrahedron 42, 4757 (1986). H. Felkin, G . Swierczewski and A. Tambute, Tetrahedron Lett. 707 (1969). M. Cherest, H. Felkin, C. Frajerman, C. Lion, G. Roussi and G. Swierczewski, Tetrahedron Lett. 875 (1966). [98] B. M. Trost, Acc. Chem. Res. 13, 385 (1980). [99] P. Pino and G. Consiglio, Pure Appl. Chem. 55, 1781 (1983). [loo] B. Bosnich and P.B. Mackenzie, Pure Appl. Chem. 54, 189 (1982). [IOI] B. Bosnich and M.D. Fryzuk, Top. Stereochem. 12, 119 (1980). [I021 B. M. Trost and P. E. Strege, J. Am. Chem. SOC.99, 1649 ( 1 977). [I031 P.B. MacKenzie, J. Whelan and B. Bosnich, J. Am. Chem. SOC.107, 2046 (1985). [I041 T. Hayashi, K. Kanehira, T. Hagihara and M. Kumada, J. Org. Chem. 53, 113 (1988). [lo51 T. Hayashi, Abstr. ESOC Conference, Freiburg, FRG, 1984. [I061 J. C. Fiaud, A. H. De Gournay, M. Larcheveque and H. B. Kagan, J. Organomet. Chem. 154, 175 (1978). [I071 J.-E. Backvall, R.E. Nordberg and D. Wilhelm, J. Am. Chem. SOC.107, 6892 (1985). [ 1081 G. Consiglio, 0. Piccolo, L. Roncetti and F. Morandini, Tetrahedron 42, 2043 ( 1 986). [I091 Y. Xu and B. Zhou, J. Org. Chem. 52, 974 (1987). [I 101 T. Takahashi, Y. Jinbo, K. Kitamura and J. Tsuji, Tetrahedron Lett. 25, 5921 (1984). [I 1 I] S. G. Davies, R. J. C. Easton, K. H. Sutton and J. C. Walker, J. C. S. Perkin Trans. 1 489 (1987). [I121 K. Fuji, F. Tanaka and M. Node, Tetrahedron Lett. 32, 7281 (1991). [I 131 G. Consiglio and P. Pino, Advances in Chem. Ser. 196, 371 (1982). [I 141 P. Pino, G. Consiglio, C. Botteghi and C. Salomon, Advances in Chem. Ser. 132, 295 (1974). [115] C. Botteghi, G. Consiglio and P. Pino, Chimia 26, 141 (1972). [I 161 T. Hayashi, M. Tanaka and I. Ogata, Tetrahedron Lett. 3925 (1978). [I171 T. Hayashi, M. Tanaka, Y. Ikeda and I. Ogata, Bull. Chem. SOC.Japan 52, 2605 (1979). [I 181 G. Parrinello, R. Deschenaux and J.K. Stille, J. Org. Chem. 51, 4189 (1986). [I191 G. Consiglio and S.C.A. Nefkens, Tetrahedron Asym. 1, 417 (1990). [I201 W.R. Jackson, P. Perlmutter and E. Tasdelen, Tetrahedron Lett. 31, 2461 (1990). [I211 H. Alper and N. Hamel, J. Am. Chem. SOC.112, 2803 (1990). [122] P. Duhamel, J.-Y. Valnot and J. J. Eddine, Tetrahedron Lett. 2863 (1982). [I231 P. Duhamel, J. J. Eddine and J.-J. Valnot, Tetrahedron Lett. 25, 2355 (1984). [I241 T. Yamashita, H. Mitsui, H. Watanabe and N. Nakamura, Bull. Chem. SOC.Japan 55, 961 (1982). [I251 M. Murakata, M. Nakajima and K. Koga, J.C.S. Chem. Comm. 1657 (1990). [I261 P.E. Sonnet and R.R. Heath, J. Org. Chem. 45, 3137 (1980). [I271 D.A. Evans and J.M. Takacs, Tetrahedron Lett. 4233 (1980). [I281 D.A. Evans, J. M. Takacs, L. R. McGee, M. D. Ennis, D. J. Mathre and J. Bartroli, Pure Appl. Chem. 53, 1109 (1981). [129] R.P. Alexander and I. Paterson, Tetrahedron Lett. 26, 5339 (1985). [I301 R.E. Ireland, R.H. Mueller and A.K. Willard, J. Am. Chem. SOC.98, 2868 (1976). [131] R. Schmierer, G. Grotemeier, G. Helmchen and A. Selim, Angew. Chem. 93, 207 (1981); Int. Ed. Engl. 20, 209 (1981). [132] G. Helmchen, A. Selim, D. Dorsch and I. Taufer, Tetrahedron Lett. 3213 (1983). [I331 E. Ade, G. Helmchen and G. Heiligenmann, Tetrahedron Lett. 1137 (1980). [I341 G. Helmchen and R. Schmierer, Tetrahedron Lett. 1235 (1983). [135] G. Helmchen and R. Wierzchowski, Angew. Chem. 96, 59 (1984); Int. Ed. Engl. 23, 60 (1984). [136] D. A. Evans, M. D. Ennis, T. Le, N. Mandel and G. Mandel, J. Am. Chem. SOC.106, 1154 (1 984). [I371 S. Najdi, D. Reichlin and M.J. Kurt, J. Org. Chem. 55, 6241 (1990). [I381 T.-H. Yan, V.-V. Chu, T.-C. Lin, C.-H. Wu and L.-H. Liu, Tetrahedron Lett. 32, 4959 (1991). [I391 W. Oppolzer, R. Moretti and S. Thomi, Tetrahedron Lett. 30, 5603 (1989). [94] [95] [96] [97]

References [I401 [I411 [I421 [I431 [I441 [I451 [I461 [I471 [I481 [I491 [ 1 501 [I511 [I521 [ I 531 [ 1541

[I551 [I561 [I571 [I581 [159] [160] [I611 [I621 [I631 [I641 [I651 [166] [I671 [I681 [I691 [I701 [I711 [I721 [I731 [174] [I751 [I761 [177] [I781 [I791 [180] [181]

[I821 [I831 [I841 [I851

345

Y. Nagao, W. M. Dai, M. Ochiai, S. Tsukagoshi and E. Fujita, J. Org. Chem. 55, 1148 (1990). G. Cardillo, M. Orena, M. Romero and S. Sandri, Tetrahedron 45, 1501 (1989). M. Enomoto, Y. Ito, T. Katsuki and M. Yamaguchi, Tetrahedron Lett. 26, 1343 (1985). K.-Y. Akiba, Y. Takasu and M. Wada, Tetrahedron Lett. 26, 2463 (1985). S. Ikegami, T. Hayama, T. Katsuki and M. Yamaguchi, Tetrahedron Lett. 27, 3403 (1986). W. Oppolzer, R. Moretti and S. Thomi, Tetrahedron Lett. 30, 6009 (1989). G. Lin, E. Hoegberg and T. Norin, Kexue Tongbao 29, 632 (1984). C.A. 101, 15374 (1984). M. Larcheveque, E. Ignatova and T. Cuvigny, Tetrahedron Lett. 3961 (1978). D. A . Evans, M.D. Ennis and D. J. Mathre, J. Am. Chem. SOC.104, 1737 (1982). S.G. Davies and C. L. Goodfelow, J.C.S. Perkin Trans. 1 192 (1989). Y. Watanabe and T. Endo, Tetrahedron Lett. 29, 321 (1 988). J.K. Whitesell and S. W. Felmann, J. Org. Chem. 42, 1663 (1977). G. J. Baird, S. G. Davies, R. H. Jones, K. Prout and P. Warner, J.C.S. Chem. Comm. 745 (1 984). P. J. Curtis and S. G. Davies, J. C. S. Chem. Comm. 747 (1984). S. L. Brown, S. G. Dagies, D. F. Foster, J. I. Seeman and P. Warner, Tetrahedron Lett. 27, 623 (1986). S.G. Davies and P. Warner, Tetrahedron Lett. 26, 4815 (1985). S. G. Davies, D. Middlemiss, A. Naylor and M. Wills, J. C. S. Chem. Comm. 797 (1990). S.P. Collingwood, S.G. Davies and S.C. Preston, Tetrahedron Lett. 31, 4097 (1990). Gy. Frater, U. Muller and W. Gunther, Tetrahedron Lett. 4221 (1981). D. Seebach, J. Zimmermann, U. Gysel, R. Ziegel and T.-K. Ha, J. Am. Chem. SOC.110,4763 (1988). D. Seebach, M. Boes, R. Naef and W.B. Schweizer, J. Am. Chem. SOC.105, 5390 (1983). D. Seebach and A. Fadel, Helv. Chim. Acta 68, 1243 (1985). T. Weber and D. Seebach, Helv. Chim. Acta 68, 155 (1985). T.M. Zudowsky, E. de Lara and S.G. Spanton, J. Org. Chem. 55, 5437 (1990). K.S. Chu, G. R. Negrete, J. P. Konopelski, F. J. Lakner, N,-T. Woo and M.M. Olmstead, J. Am. Chem. SOC.114, 1800 (1992). D. Seebach el al. (27 authors), Helv. Chim. Acta 75, 913 (1992). A.I. Meyers, Pure Appl. Chem. 51, 1255 (1979). A. I. Meyers and E.D. Mihelich, Angew. Chem. 88, 321 (1976); Int. Ed. Engl. 15, 270 (1976). A.I. Meyers, G. Knaus, K. Kamata and M.E. Ford, J. Am. Chem. SOC.98, 567 (1976). A. I. Meyers, A. Mazzu and C. E. Whitten, Heterocycles 6 , 971 (1977). S. Shibata, H. Matsushita, H. Kaneko, M. Noguchi, M. Saburi and S. Yoshikawa, Chem. Lett. 217 (1981). A.I. Meyers, E.S. Snyder and J.J.H. Ackerman, J. Am. Chem. SOC.100, 8186 (1978). M.A. Hoobler, D.E. Bergbreiter and M. Newcomb, J. Am. Chem. SOC.100, 8182 (1978). A. I. Meyers, Y. Yamamoto, E.D. Mihelich and R.A. Bell, J. Org. Chem. 45, 2792 (1980). A.I. Meyers and G. Knaus, J. Am. Chem. SOC.96, 6508 (1974). A.I. Meyers and K.T. Wahner, Tetrahedron Lett. 26, 2047 (1985). A.I. Meyers and B.A. Leiker, J. Org. Chem. 51, 1541 (1986). A. I. Meyers, R. H. Wallace, M. Hare and R. Garland, J. Org. Chem. 55, 3137 (1990). A.1. Meyers, B.A. Lefker, K. Th. Wanner and R.A. Aitken, J. Org. Chem. 56, 1936 (1991). U. Schollkopf, Pure Appl. Chem. 55, 1799 (1983). U. Schollkopf, Top. Curr. Chem. 109, 65 (1983). U. Schollkopf, Tetrahedron 39, 2085 (1983). U. Schollkopf, W. Hartwig and U. Groth, Angew. Chem. 91, 922 (2979); Int. Ed. Engl. 18, 863 (1979). U. Schollkopf, U. Groth and C. Deng, Angew. Chem. 93, 793 (1981); Int. Ed. Engl. 20, 798 (1 98 I). U. Schollkopf, U. Groth, K.-0. Westphalen and C. Deng, Synthesis 969 (1981). J. Nozulak and U. Schollkopf, Synthesis 866 (1982).

346

References

[186] U. Schollkopf and H. Neubauer, Synthesis 861 (1982). [187] U. Groth and U. Schollkopf, Synthesis 37 (1983). [I881 U. Schollkopf, U. Busse, R. Kilger and P. Lehr, Synthesis 271 (1984). [I891 U. Schollkopf, U. Groth and W. Hartwig, Liebigs Ann. Chem. 2407 (1981). [190] U. Groth and U. Schollkopf, Synthesis 673 (1983). [191] U. Groth, U. Schollkopf and Y. Chiang, Synthesis 864 (1982). [192] U. Schollkopf, U. Groth, M.-R. Gull and J. Nozulak, Liebigs Ann. Chem. 1133 (1983). [193] F. Sauriol-Lord and T.B. Grindley, J. Org. Chem. 46, 2831 (1981). 11941 F. Sauriol-Lord and T.B. Grindley, J. Org. Chem. 2833 (1981). [195] D. Seebach and D. Wasmuth, Helv. Chim. Acta 63, 197 (1980). [I961 Gy. Frater, Helv. Chim. Acta 62, 2825 (1979). [197] Gy. Frater, U. Muller and W. Gunther, Tetrahedron 40, 1269 (1984). [I981 Gy. Frater, Tetrahedron Lett. 425 (1981). [199] A.R. Chamberlin and M. Dezube, Tetrahedron Lett. 3055 (1982). [200] K. Tomioka, Y. Cho, F. Sat0 and K. Koga, Chem. Lett. 1621 (1981). [201] S. Takano, K. Chiba, M. Yonaga and K. Ogasawa, J.C.S. Chem. Comm. 616 (1980). [202] D. Seebach and J.D. Aebi, Tetrahedron Lett. 331 1 (1983). [203] H.M. Shieh and G.D. Prestwich, J. Org. Chem. 46, 4319 (1981). [204] I. Fleming and J.J. Lewis, J.C.S. Chem. Comm. 149 (1985). [205] I. Fleming, S. Gil, A. K. Sarkar and T. Schmidlin, J. C. S. Perkin Trans. 1 3363 (1 992). [206] I. Fleming, J.H.M. Hill, D. Parker and D. Waterson, J.C.S. Chem. Comm. 318 (1985). [207] G.J. McGarvey and J.M. Williams, J. Am. Chem. SOC. 107, 1435 (1985). [208] K. Narasaka, Y. Ukaji and K. Watanabe, Bull. Chem. SOC.Jap. 60, 1457 (1987). [209] Y. Ukaji and K. Narasaka, Bull. Chem. SOC.Jap. 61, 571 (1988). [210] E.L. Eliel, Tetrahedron 30, 1503 (1974). [211] E.L. Eliel, A.A. Hartmann and A.G. Abatjoglou, J. Am. Chem. SOC.96, 1807 (1974). [212] A.A. Hartmann and E.L. Eliel, J. Am. Chem. SOC.93, 2572 (1971). [213] E. L. Eliel, A. Abatjoglou and A. A. Hartmann, J. Am. Chem. SOC.94, 4786 (1 972). [214] J.J. Eisch and J.E. Galle, J. Org. Chem. 45, 4534 (1980). [215] T. Sato, T. Itoh and T. Fujisawa, Tetrahedron Lett. 28, 5677 (1987). [216] S. Hanessian, Y.L. Bennani and D. Delorme, Tetrahedron Lett. 31, 6461 (1990). [217] S. Hanessian and Y.L. Bennani, Tetrahedron Lett. 31, 6465 (1990). [218] D.P. Curran and J.-C. Chao, J. Am. Chem. SOC.109, 3036 (1987). [219] D. Hoppe, F. Hintze and P. Tebben, Angew. Chem. 102, 1457 (1990); Int. Ed. Engl. 29, 1422 (1 990). [220] S.T. Kerrick and P. Benk, J. Am. Chem. SOC.113, 9708 (1991). [221] M. Paetow, H. Ahrens and D. Hoppe, Tetrahedron Lett. 33, 5323 (1992). [222] H. Ahrens, M. Paetow and D. Hoppe, Tetrahedron Lett. 33, 5327 (1992). [223] J. Schwerdtfeger and D. Hoppe, Angew. Chem. 104, 1547 (1992); Int. Ed. Engl. 31, 1505 (1 992). [224] S. Yamada, K. Hiroi and K. Achiwa, Tetrahedron Lett. 4233 (1969). [225] S. Yamada and G. Otani, Tetrahedron Lett. 4237 (1969). [226] K. Hiroi, K. Achiwa and S. Yamada, Chem. Pharm. Bull. 20, 246 (1972). [227] G. Otani and S. Yamada, Chem. Pharm. Bull. 21, 2112 (1973). [228] G. Otani and S. Yamada, Chem. Pharm. Bull. 21, 2125 (1973). [229] T. Sone, K. Hiroi and S. Yamada, Chem. Pharm. Bull. 21, 2331 (1973). [230] T. Sone, S. Terashima and S. Yamada, Chem. Pharm. Bull. 24, 1288 (1976). I2311 C.C. Tseng, S. Terashima and S. Yamada, Chem. Pharm. Bull. 25, 29 (1977). [232] S. Yamada, M. Shibasaki and S. Terashima, Tetrahedron Lett. 381 (1973). [233] D. Mia-Jacheet and A. Horeau, Bull. SOC.Chim. France 4571 (1968). [234] S. J. Blarer and D. Seebach, Chem. Ber. 116, 2250 (1983). [235] R. H. Schlessinger and E. J. Iwanowicz, Tetrahedron Lett. 29, 1489 (1988). [236] J. Martens and S. Lubben, Tetrahedron 47, 1205 (1991).

References

347

[237] F. Felluga, P. Nitti, G. Pitacco and E. Valentin, Tetrahedron 45, 2099 (1989). [238] D. Seebach, A. K. Beck, J. Gollinski, J. N. Hay and T. Laube, Helv. Chim. Acta 68, 162 (1985). [239] A.I. Meyers, G.S. Pointdexter and Z. Brich, J. Org. Chem. 43, 892 (1978). I2401 A.I. Meyers, D.R. Williams and M. Druelinger, J. Am. Chem. SOC.98, 3032 (1976). [241] A. I . Meyers, D. R. Williams, G. W. Erickson, S. White and M. Druelinger, J. Am. Chem. SOC. 103, 3081 (1981). [242] J.K. Whitesell and M.A. Whitesell, J. Org. Chem. 42, 377 (1977). [243] S. Hashimoto and K. Koga, Tetrahedron Lett. 573 (1978). [244] S. Hashimoto and K. Koga, Chem. Pharm. Bull. 27, 2760 (1979). [245] K. Tomioka, K. Ando, T. Takemasa and K. Koga, J. Am. Chem. SOC.106, 2718 (1984). [246] K. Tomioka, K. Ando, T. Takemasa and K. Koga, Tetrahedron Lett. 5677 (1984). [247] A. I. Meyers, D.R. Williams, S. White and G. W. Erickson, J. Am. Chem. SOC.103, 3088 (1 981). [248] A. I. Meyers and D. R. Williams, J. Org. Chem. 43, 3245 (1 978). [249] S. Yamada, T. Oguri and T. Shioiri, J. C.S. Chem. Comm. 136 (1976). [250] J. A. Bajgrowicz, B. Cossec, C. Pigiere, R. Jacquier and P. Viallefont, Tetrahedron Lett. 3721 (1 983). [251] J.M. McIntosh, R.K. Leavitt, P. Mishra, K.C. Cassidy, J.E. Drake and R. Chadha, J. Org. Chem. 53, 1947 (1988). [252] M. Tabcheh, A. El Achqar, L. Pappalardo, M.-L. Roumestant and P. Viallefont, Tetrahedron 47, 4611 (1991). [253] D. Enders, in Asymmetric Synthesis, Vol. 3, (J. D. Morrison, ed.) Academic Press, New York, 1984, p. 275. 12541 D. Enders, Chemtech., 504 (1981). [255] D. Enders, in Curr. Dends in Org. Synth., (H. Nozaki, ed.). Pergamon Press, Oxford, 1983, p. 151. [256] D. Enders and H. Eichenauer, Angew. Chem. 88, 579 (1976); Int. Ed. Engl. 15, 549 (1976). [257] D. Enders and H. Eichenauer, Chem. Ber. 112, 2933 (1979). [258] D. Enders, H. Eichenauer and R. Pieter, Chem. Ber. 112, 3703 (1979). [259] P.M. Hardy, Synthesis 290 (1978). [260] K.G. Davenport, H. Eichenauer, D. Enders, M. Newcomb and D.E. Bergbreiter, J. Am. Chem. Soc. 101, 5654 (1979). [261] D. Enders and H. Eichenauer, Angew. Chem. 91, 425 (1979); Int. Ed. Engl. 18, 397 (1979). [262] D. Enders, H. Schubert and C. Nubling, Angew. Chem. 98, 11 18 (1986); Int. Ed. Engl. 25, 1109 (1986). [263] D.A. Claremon, P.K. Lumma and B.T. Phillips, J. Am. Chem. SOC.108, 8265 (1986). [264] D. Enders, H. Eichenauer, U. Baus, H. Schubert and K. A. M. Kremer, Tetrahedron 40, 1345 (1 984). [265] D. Enders and B.B. Lohray, Angew. Chem. 99, 359 (1987); Int. Ed. Engl. 26, 351 (1987). [266] D. Enders, W. Gatzweiler and U. Jegelka, Synthesis 1137 (1991). [267] S.I. Pennanen, Acta Chem. Scand. B 35, 555 (1981). [268] M.T. Reetz, R. Steinbach and K. Kesseler, Angew. Chem. 95, 872 (1982); Int. Ed. Engl. 21, 864 (1982). [269] M. Kolb and J. Barth, Liebigs Ann. Chem. 1068 (1983). [270] M. Kolb and J. Barth. Tetrahedron Lett. 2999 (1979). [271] A.I. Meyers, M. Boes and D.A. Dickman, Angew. Chem. 96, 448 (1984); Int. Ed. Engl. 23, 458 (1 984). [272] A.I. Meyers, L.M. Fuentes and Y. Kubota, Tetrahedron 40, 1361 (1984). [273] M.F. Loewe, M. Boes and A.I. Meyers, Tetrahedron Lett. 26, 3295 (1985). [274] L.A. Castonguay, J. W. Guiles, A. K. Rapp, and A. I. Meyers, J. Org. Chem. 57, 3819 (1992). [275] M. Kumada, Pure Appl. Chem. 52, 669 (1980). [276] D. F. Taber and R. E. RuckIe Jr., Tetrahedron Lett. 29, 3059 (1 985).

348

References

[277] G. Consiglio and C. Botteghi, Helv. Chem. Acta 56, 460 (1973). [278] T. Hayashi, M. Konishi, M. Fukushima, T. Mise, M. Kagotani, M. Tajika and M. Kumada, J. Am. Chem. Soc. 104, 180 (1982). [279] T. Hayashi, A. Yamamoto, M. Hojo and Y. Ito, J.C.S. Chem. Comm. 495 (1989). [280] T. Hayashi, M. Fukushima, M. Konishi and M. Kumada, Tetrahedron Lett. 79 (1980). [281] T. Hayashi, M. Konishi, M. Fukushima, K. Kanehira, T. Hioki and M. Kumada, J. Org. Chem. 48, 2195 (1983). [282] G. A. Cross and R. M. Kellogg, J. C. S. Chem. Comm. 1746 (1987). I2831 T. Hayashi, M. Konishi, H. Ito and M. Kumada, J. Am. Chem. Soc. 104, 4962 (1982). [284] F. Ozawa, A. Kubo and T. Hayashi, J. Am. Chem. Soc. 113, 1417 (1991). [285] T. Hayashi, K. Hayashizaki, T. Kioyi and Y. Ito, J. Am. Chem. Soc. 110, 8153 (1988). [286] T. Hayashi, T. Hagihara, Y. Katsuro and M. Kumada, Bull. Chem. Soc. Jap. 56, 363 (1983). [287] B. Bogdanovic, Angew. Chem. 85, 1013 (1973); Int. Ed. Engl. 12, 954 (1973). [288] B. Bogdanovic, B. Henc, B. Meister, H. Pauling and G. Wilke, Angew. Chem. 84, 1070 (1972); Int. Ed. Engl. 11, 1023 (1972). [289] G. Buono, G. Pfeiffer, A. Montreux and F. Petit, J. C. S. Chem. Comm. 937 (1 980). [290] P. Brun, A. Tenaglia and B. Waegell, Tetrahedron Lett. 385 (1983). [291] W.S. Johnson, R. Elliot and J.D. Elliot, J. Am. Chem. Soc. 105, 2904 (1983). [292] J.D. Elliott, W.M.F. Choi and W. S. Johnson, J. Org. Chem. 48, 2294 (1983). [293] V.M.F. Choi, J.D. Elliott and W.S. Johnson. Tetrahedron Lett. 25, 591 (1984). [294] W. S. Johnson, P. H. Crackett, J. D. Elliott, J. J. Jagodzinski, S. D. Lindell and S. Natarajan, Tetrahedron Lett. 25, 3951 (1984). [295] S.D. Lindell, J.D. Elliott and W.S. Johnson, Tetrahedron Lett. 25, 3947 (1984). [296] J.D. Elliot, J. Stelle and W.S. Johnson, Tetrahedron Lett. 26, 2535 (1985). [297] W.S. Johnson, C. Edington, J.D. Elliot and I. R. Silvermann, J. Am. Chem. Soc. 106, 7588 (1984). [298] J.F. Normant, A. Alexakis, A. Ghribi and P. Mangeney, Tetrahedron 45, 507 (1989). [299] S.E. Denmark, T.M. Willson and N.G. Almstead, J. Am. Chem. Soc. 111, 9258 (1989). [300] T. Harada, K. Sakamoto, Y. Ikemura and A. Oku, Tetrahedron Lett. 29, 3097 (1988). [301] D. Seebach, R. Imwinkelried and G. Stucky, Helv. Chim. Acta 70, 448 (1987). [302] J. M. Wilson and D. J. Cram, J. Am. Chem. Soc. 104, 881 (1982). [303] A.I. Meyers and K.A. Lutowski, J. Am. Chem. Soc. 104, 879 (1982). [304] K. Suzuki, E, Katayama and G. Tsuchihashi, Tetrahedron Lett. 4997 (1983). I3051 G. Tsuchihashi, K. Tomooka and K. Suzuki, Tetrahedron Lett. 25, 4253 (1984). [306] K. Suzuki, E. Katayama and G. Tsuchihasi, Tetrahedron Lett. 25, 1817 (1984). [307] K. Suzuki, M. Miyazawa, M. Shimazaki and G. Tsuchihashi, Tetrahedron 44, 4061 (1988). [308] Y. Honda, M. Sakai and G. Tsuchihashi, Chem. Lett. 1153 (1985). [309] Y. Honda, A. Ori and G. Tsuchihashi, Chem. Lett. 13 (1986). [310] D.F. Taber and K. Raman, J. Am. Chem. Soc. 105, 5935 (2983). [311] M. P. Doyle, A. van Oeveren, L. J. Westrum, M. N. Protopopova and T. W. Clayton, J. Am. Chem. Soc. 113, 8982 (1991).

Chapter 7 [l] R.B. Woodward and R. Hoffmann, Angew. Chem. 81, 797 (196); Int. Ed. Engl. 8, 781 (1969). [2] T. L. Gilchrist and R. C. Storr, Organic Reactions and Orbital Symmetry. University Press, Cambridge, 1972.

References

349

[3] R. Huisgen, Angew. Chem. 80, 329 (1968); Int. Ed. Engl. 7, 321 (1968). [4] H. Wurziger, Kontakte, 3 (1984). [5] IUPAC Nomenclature for Straightforward Transformation (Provisional). Pure Appl. Chem. 53, 305 ( I 98 1). [6] K.N. Houk, J. Am. Chem. SOC.95, 4092 (1973). [7] A. Korolev and V. Mur, Dokl. Akad. Nauk SSSR, 59, 251 (1948); C.A. 42, 6776 (1948). IS] H.M. Walborsky, L. Barash and T.C. Davis, Tetrahedron Lett. 19, 2333 (1963). [9] J. Sauer and J. Kredel, Tetrahedron Lett. 6359 (1966). [lo] E. J. Corey and H.E. Ensley, J. Am. Chem. SOC.97 6908 (1975). [ I I] H.E. Ensley, C.A. Parnell and E. J. Corey, J. Org. Chem. 43, 1610 (1978). [12] W. Oppolzer, M. Kurth, D. Reichlin, C. Chapuis, M. Mohnhaupt and F. Moffatt, Helv. Chim. Acta 64, 2802 (1981). [13] H.E. Ensley and R.V.C. Carr, Tetrahedron Lett. 513 (1977). [I41 W. Oppolzer, C. Chapuis and M. J. Kelly, Helv. Chim. Acta 66, 2358 (1983). [ I 51 W. Oppolzer, Curr. Trends in Org. Synth. Pergamon Press, Oxford, 1983, p. 131 . [I 61 J. D. Morrison and H. S. Mosher, Asymmetric Organic Reactions, Prentice-Hall. Englewood Cliffs, N. J., 1971; p. 256. [I71 T. Poll, G. Helmchen and B. Bauer, Tetrahedron Lett. 25, 2191 (1984). [I81 W. Oppolzer, C. Chapuis, D. Dupuis and M. Guo, Helv. Chim. Acta. 68, 2100 (1985). [I91 W. Oppolzer, C. Chapuis, M. D. Guo, D. Reichlin and T. Godel, Tetrahedron Lett. 4781 (1982). [20] W. Oppolzer, C. Chapuis and G. Bernardinelli, Tetrahedron Lett. 25, 5885 (1984). [21] D.A. Evans, K.T. Chapman and J. Bisaha, J. Am. Chem. SOC.110, 1238 (1988). [22] W. Oppolzer, M. Wils, M. J. Kelly, M. Signer and J. Blagg, Tetrahedron Lett. 31, 5015 (1990). [23] Y. Kawanami, T. Katsuki and M. Yamaguchi, Bull. Chem. SOC.Jap. 60, 4190 (1987). [24] E. J. Corey, X.-M. Cheng and K.A. Cimprich, Tetrahedron Lett. 32, 6839 (1991). [25] J. F. Buzinkai, D. M. Hrubowchak and F. X. Smith, Tetrahedron Lett. 26, 3095 (1 985). [26] T. Poll, A. F.A. Hady, R. Karge, G. Linz, J. Weetman and G. Helmchen, Tetrahedron Lett. 30, 5595 (1989). [27] H. Suzuki, K. Mochizuki, T. Hattori, N. Takahashi, 0. Tajima and T. Takiguchi, Bull. Chem. SOC.Jap. 61, 1999 (1988). [28] D. A. Evans, K. T. Chapman, D. T. Hung and A. T. Kawaguchi, Angew. Chem. 99, 1197 (1987); Int. Ed. Engl. 26, 1184 (1987). [29] H. Hartmann, A.F.A. Hady, K. Sartor, J. Weetman and G. Helmchen, Angew. Chem. 99, 1188 (1987); Int. Ed. Engl. 26, 1143 (1987). I301 K. Furuta, K. Iwanaga and H. Yamamoto, Tetrahedron Lett. 27, 4507 (1986). [31] D. Hebel and S. Rozen, J. Org. Chem. 53, 1125 (1988). [32] K. Tomioka, N, Hamada, T. Suenaga and K. Koga, J. C. S. Perkin Trans 1 426 (1 990). [33] T. Bauer, C. Chapuis, J. Kozak and J. Jurczak, Helv. Chim. Acta 72, 482 (1989). [34] J.K. Whitesell, J.N. Younathan, J. R. Hurst and M.A. Fox, J. Org. Chem. 50, 5499 (1985). [35] R.K. Boeckman, Jr., P.C. Naegely and S.D. Arthur, J. Org. Chem. 45, 752 (1980). [36] W. Oppolzer and C. Chapuis, Tetrahedron Lett. 4665 (1983). [37] W. Oppolzer, C. Chapuis, D. Dupuis and M. Guo, Helv. Chim. Acta 68, 2100 (1985). [38] W. Choy, L.A. Reed and S. Masamune, J. Org. Chem. 48, 1137 (1983). [39] S. Masamune, L.A. Reed, J.T. Davis and W. Choy, J. Org. Chem. 48, 4441 (1983). [40] J.K. Whitesell, A. Bhattacharya, D.A. Aguilar and K. Henke, J.C.S. Chem. Comm. 989 (1982). [41] M. J. Martinelli, J. Org. Chem. 55, 5065 (1990). [42] G. Helmchen and R. Schmierer, Angew. Chem. 93, 208 (1981); Int. Ed. Engl. 20, 205 (1981). [43] S. David, A. Lubineau and A. Thieffry, Tetrahedron, 34, 299 (1978). [44] P.A.T. W. Proskamp, R. C. Haltiwanger and R.C. Zwanenburg, Tetrahedron Lett. 2035 (1983). [45] V. Gouverneur and L. Ghosez, Tetrahedron Asym. 1, 363 (1990). [46] P. G. McDougal, J. G. Rico and D. VanDerveer, J. Org. Chem. 51, 4492 (1 986). [47] B.M. Trost, S.A. Godieski and J. P. Genet, J. Am. Chem. SOC.100, 3930 (1978).

350

References

[48] B. M. Trost, D. O’Krongly and J. L. Belletire, J. Am. Chem. SOC. 102, 7595 ( 1 980). [49] R. Rieger and E. Breitmaier, Synthesis 697 (1990). [50] M. M. Guseinov, I.M. Akhmedov and E.G. Mamedov, Azerb, Khim. Zh. 46 (1976); C.A. 85, 176925~( I 976). [ 5 11 S. Hashimoto, N. Komeshima and K. Koga, J. C. S. Chem. Comm. 437 (1979). [52] H. Takemura, N. Komeshima, I. Takahashi, S.-I. Hashimoto, N. Ikota, K. Tomioka and K. Koga, Tetrahedron Lett. 28, 5687 (1987). [53] P.N. Devine and T. Oh, J. Org. Chem. 57, 396 (1992). [54] K. Narasaka, H. Tanaka and F. Kanai, Bull Jap. 64, 387 (1991). [55] C. Chapuis and J. Jurczak, Helv. Chim. Acta 70, 436 (1987). [56] N. Iwasawa, J. Sugimori, Y. Kawase and K. Narasaka, Chem. Lett. 1947 (1989). [57] M. Terada, K. Mikami and T. Nakai, Tetrahedron Lett. 47, 935 (1991). I581 K. Maruoka, T. Itoh, T. Shirasaka and H. Yamamoto, J. Am. Chem. SOC.110, 310 (1988). [59] K. Maruoka and H. Yamamoto, J. Am. Chem. SOC.111, 789 (1989). [60] Q. Cao, T. Maruyama, M. Mouri and H. Yamamoto, J. Org. Chem. 57, 1951 (1992). [61] K. Furuta, S. Shimizu, Y. Miwa and H. Yamamoto, J. Org. Chem. 54, 1481 (1989). [62] K. Hattori and H. Yamamoto, J. Org. Chem. 57, 3264 (1992). [63] E.J. Corey and T.-P. Loh, J. Am. Chem. SOC.113, 8966 (1991). [64] E. J. Corey, T.-P. Loh, T.D. Roper, M.D. Azimiora and M.C. Noe, J. Am. Chem. SOC.114, 8290 (1992). [65] D. Craig, Quart. Rev. 16, 187 (1987). [66] L.-F. Tietze and G. Kiedrowski, Tetrahedron Lett. 219 (1981). [67] L.-F. Tietze and K.-H. Glusenkamp, Angew. Chem. 95, 901 (1983); Int. Ed. Engl. 22, 887 (1983). [68] T. Mukaiyama and N. Iwasawa, Chem. Lett. 29 (1981). [69] W. R. Roush, H. R. Gillis and A. I. KO, J. Am. Chem. SOC.104, 2269 (1982). [70] D.A. Evans, K.T. Chapman and J. Bisaha, Tetrahedron Lett. 25, 4071 (1984). [71] W. Oppolzer and D. Dupuis, Tetrahedron Lett. 26, 5437 (1985). [72] W. Oppolzer, Synthesis 793 (1978). [73] W. Oppolzer, Pure Appl. Chem. 53, 1181 (1981). [74] W. Oppolzer and D.A. Roberts, Helv. Chim. Acta 63, 1703 (1980). [75] C. Siege1 and E.R. Thorton, Tetrahedron Lett. 29, 5225 (1988). [76] J.A. Tucker, K.N. Houk and B.M. Trost, J. Am. Chem. SOC.112, 5465 (1990). [77] D. Poll, J.O. Metter and G. Helmchen, Angew. Chem. 97, 116 (1985); Int. Ed. Engl. 24, 112 (1985). [78] W. Oppolzer, I. Rodriguez, J. Blagg and G. Bernardinelli, Helv. Chim. Acta 72, 123 (1989). [79] S.D. Kahn and W. J. Hehre, J. Am. Chem. SOC.109, 663 (1987). [80] D.P. Curren, B. H. Kim, J. Daugherty and T.A. Heffner, Tetrahedron Lett. 29, 3555 (1988). [81] S. Kanemasa and K. Onimura, Tetrahedron 48, 8631 (1992). [82] S. Kanemasa and K. Onimura, Tetrahedron 48, 8645 (1992). [83] D. P. Curran, K.-S. Jeong, T.A. Heffner and J. Rebek, Jr., J. Am. Chem. SOC.111, 9238 (1989). [84] D.P. Curran and S.A. Gothe, Tetrahedron 44, 3945 (1988). [ 8 5 ] D. de Lange and B.L. Feringa, Tetrahedron Lett. 29, 5317 (1988). [86] J.P. Marino and R.F. de la Pradilla, Tetrahedron Lett. 26, 5381 (1985). [87] H. Kosugi, K. Tagami, A. Takahashi, H. Kanna and H. Uda, J.C.S. Perkin Trans. 1, 935 (1989). [88] C. Belzecki and I. Panfil, J. Org. Chem. 44, 1212 (1979). [89] K. Koizumi, H. Hirai and E. Yoshii, J. Org. Chem. 47, 4004 (1982). [90] A. Vasella and R. Voeffray, Helv. Chim. Acta 65, 1953 (1982). [91] A. Vasella and R. Voeffray, J.C.S. Chem. Comm. 97 (1981). [92] B. S. Green, A. T. Hagler, Y. Rabinson and M. Rejto, Isr. J. Chem. 15, 124 (1976/77). [93] P. Binger, A. Brinkmann, P. Roetke and B. Schafer, Liebigs Ann. Chem. 739 (1989). [94] L.M. Tolbert and M.B. Ali, J. Am. Chem. SOC.104, 1742 (1982).

References

35 1

[95] H. Koch, J. Runsink and H.-D. Scharf, Tetrahedron Lett. 3217 (1983). [96] H. Buschmann, H.-D. Scharf, N. Hoffmann and P. Esser, Angew. Chem. 103, 480 (1991); lnt. Ed. Engl. 30, 477 (1991). [97] H. Buschmann, H.-D. Scharf, N. Hoffmann, M. W. Plath and J. Runsink, J. Am. Chem. SOC. 111, 5367 (1989). 1981 I. Ojima and S. Inaba, Tetrahedron Lett. 2077 (1980). [99] I. Ojima and S. Inaba, Tetrahedron Lett. 2081 (1980). [loo] G.1. Georg and E. Akgon, Tetrahedron Lett. 31, 3267 (1990). [I011 K. Narasaka, Y. Hayashi, H. Shimadzu and S. Niihata, J. Am. Chem. SOC.114, 8869 (1992). [I021 H. Wynberg and E.G. J. Staring, J. Am. Chem. SOC.104, 166 (1982). [lo31 P.E.F. Ketelaar, E.G. J. Staring and W. Wynberg, Tetrahedron Lett. 26, 4665 (1985). [I041 H. Wynberg and E.G. J. Staring, J. Org. Chem. 50, 1977 (1985). [to51 L.-Y. Chen and L. Ghosez, Tetrahedron Lett. 31, 4467 (1990). [I061 A.I. Meyers and S.A. Fleming, J. Am. Chem. SOC.108, 306 (1986). [lo71 T. Aratani, Pure Appl. Chem. 57, 1839 (1985). [I081 H. Nozaki, H. Takaya, S. Moriuti and R. Noyori, Tetrahedron 24, 3655 (1968). [lo91 P. E. Krieger and J. A. Landgrebe, J. Org. Chem. 43, 4447 (1978). [I101 D. Holland, D.A. Laidler and D.J. Milner, J. Mol. Catal. 11, 119 (1981). [ill] T. Aratani, T. Yoneyoshi and T. Nagase, Tetrahedron Lett. 1707 (1975). [112] T. Aratani, Y. Yoneyoshi and T. Nagase, Tetrahedron Lett. 2599 (1977). [I131 T. Aratani, Y. Yoneyoshi and T. Nagase, Tetrahedron Lett. 685 (1982). [114] H. Fritschi, U. Leutenegger and A. Pfaltz, Angew. Chem. 98, 1028 (1986); Int. Ed. Engl. 25, 1005 (1986). [I 151 M. P. Doyle, R. J. Pieters, S. F. Martin, R. E. Austin, C. J. Oalman and P. Muller, J. Am. Chem. SOC.113, 1423 (1991). [116] M. P. Doyle, V. Bagheri, T. J. Wandless, N. K. Harn, D. A. Brinker, C. T. Eagle and K.-L. Loh, J. Am. Chem. SOC.112, 1906 (1990). [I171 H. Abdallah, R. GrCe and R. Carrie, Tetrahedron Lett. 503 (1982). [I 181 J. Vallgarda and U. Hacksell, Tetrahedron Lett. 32, 5625 (1991). [I191 G. Quinkert, U. Schwartz, H. Stark, W.-D. Weber, F. Adam, H. Baier, G. Frank and G. Durner, Liebigs Ann. Chem. 1999 (1982). [I201 G. Quinkert, H.-G. Schmalz, E. Walzer, S. Gross, G. Durner and J.W. Bats Angew. Chem. 98, 732 (1986); Int. Ed. Engl. 25, 732 (1986). [I211 M. J. De Vos and A. Krief, Tetrahedron Lett. 103 (1983). [I221 E.A. Mash, K.A. Nelson, Tetrahedron 43, 679 (1987). [I231 E.A. Mash, S.K. Math and J.B. Arterburn, J. Org. Chem. 54, 4951 (1989). [I241 E.A. Mash and D.S. Terok, J. Org. Chem. 54, 250 (1989). [125] E. A. Mash, S.B. Hemperly, K.A. Nelson, P. C. Heidt and S.V. Deusen, J. Org. Chem. 55, 2045 (1990). [I261 E.A. Mash, K.A. Nelson and P.C. Heidt, Tetrahedron Lett. 28, 1865 (1987). [I271 A. Mori, I. Arai and H. Yamamoto, Tetrahedron 42, 6447 (1986). [I281 T. Sugimura, T. Futagawa, M. Yoshikava and A. Tai, Tetrahedron Lett. 30, 3807 (1989). [I291 P. W. Ambler and S. G. Davies, Tetrahedron Lett. 29, 6979 (1988). [I301 A.I. Meyers, J.L. Romine and S.A. Fleming, J. Am. Chem. SOC.110, 7245 (1988). [I311 A. Misumi, K. Iwanaga, K. Furuta and H. Yamamoto, J. Am. Chem. SOC.107, 3343 (1985). [I321 H. Takahashi, M. Yoshioka, M. Ohno and S. Kobayashi, Tetrahedron 33, 2575 (1992). [I331 S. J. Rhoads, in Molecular Rearrangements Part 1, (P. de Mayo, ed.). Interscience, New York, 1963, p. 655. [134] W. von E. Doering and W.R. Roth, Angew. Chem. Int. Ed. Engl. 2, 115 (1963). [I351 R.K. Hill and N.W. Gilman, J.C.S. Chem. Comm. 619 (1967). [I361 R.K. Hill and A.G. Edwards, Tetrahedron Lett. 3239 (1964). [I371 G.B. Bennett, Synthesis 589 (1977). [I381 F.E. Ziegler, ACC.Chem. Res. 10, 227 (1977).

352

References

[I391 R.K. Hill, R. Soman and S. Sawada, J. Org. Chem. 37, 3737 (1972). [I401 K.-K. Chan, N. Cohen, J.P. De Noble, A.C. Specian, Jr. and G. Saucy, J. Org. Chem. 41, 3505 (1976). [I411 K. Mikami, K. Takahashi and T. Nakai, Tetrahedron Lett. 28, 5879 (1997). [I421 G. Saucy and N. Cohen, in New Synthetic Methodology and Biologically Active Substances, (Z. Yoshida, ed.). Elsevier, New York, 1981, p 155. [I431 N. Cohen, W.F. Eichel, R. J. Lopresti, C. Neukom and G. Saucy, J. Org. Chem. 41, 3512 (1976). [I441 K.-K. Chan, A. C. Specian Jr. and G. Saucy, J. Org. Chem. 43, 3435 (1978). [145] K.-K. Chan, N. Cohen, J.P. De Noble, A.C. Specian and G. Saucy, J. Org. Chem. 41, 3497 (1 976). [I461 R. E. Ireland, R. M. Mueller and A. K. Willard, J. Am. Chem. SOC.98, 2868 (1976). [I471 T. Sato, K. Tajima and T. Fujisawa, Tetrahedron Lett. 729 (1983). [I481 T. J. Could, M. Balestra, M. D. Wittman, J.A. Gary, L.T. Rossano and J. Kallmerten, J. Org. Chem. 52, 3889 (1987). [I491 R.E. Ireland and M.D. Varney, J. Am. Chem. Soc. 106, 3668 (1984). [I501 M. Nagatsuma, F. Shirai, N. Say0 and T. Nakai, Chem. LEtt. 1393 (1984). [151] T. Fujisawa, K. Tajima and T. Sato, Chern. Lett. 1669 (1984). [I521 T. Sato, H. Tsunekawa, H. Kohama and T. Fujisawa, Chem. Lett. 1553 (1986). [153] T. Sato, A. Tanaka, K. Tajima and T. Fujisawa, Chem. Lett. 1979 (1987). [154] M.J. Kurth and C,M. Yu, J. Org. Chern. 50, 1840 (1985). [I551 E. J. Corey and D.-H. Lee, J. Am. Chem. SOC.113, 4026 (1991). [I561 Y. Tamaru, Y. Furukawa, M. Mizutani, 0. Kitao and Z. Yoshia, J. Org. Chem. 48, 3631 (1 983). [157] Y. Tamaru, T. Harada, and Z. Yoshida, J. Am. Chem. SOC.100, 1923 (1978). [I581 P. Beslin and S. Perrio, Tetrahedron 47, 6275 (1991). [I591 M. J. Kurth and O.H.W. Decker, Tetrahedron Lett. 4535 (1983). [I601 M.J. Kurth and O.H.W. Decker, J. Org. Chem. 50, 5769 (1985). [I611 M.J. Kurth and O.H.W. Decker, J. Org. Chem. 56, 1377 (1991). [I621 M. J. Kurth and M. G. Brown, Synthesis 362 (1988). [I631 T. Tsunoda, 0. Sasaki and S. Ito, Tetrahedron Lett. 31, 727 (1990). [164] K. Maruoka, H. Banno and H. Yamamoto, J. Am. Chem. SOC.112, 7791 (1990). [I651 K. Mikami and T. Nakai, Synthesis 594 (1991). [I661 T. Nakai and K. Mikami, Chem. Rev. 86, 855 (2986). [I671 H. Priepke and R. Briickner, Chem. Ber. 123, 153 (1990). [I681 M.M. Midland and Y.C. Kwon, Tetrahedron Lett. 26, 5013 (1985). [I691 T. Nakai, K. Mikarni and S. Taya, J. Am. Chem. Soc. 103, 6492 (1981). [I701 K. Mikami, Y. Kimura, N. Kishi and T. Nakai, J. Org. Chem. 48, 279 (1983). [I711 K. Mikarni, K. Azuma and T. Nakai, Tetrahedron 40, 2303 (1984). [I721 K. Mikami, K. Fujimoto and T. Nakai, Tetrahedron Lett. 513 (1983). [I731 E. Nakai and T. Nakai, Tetrahedron Lett. 29, 5409 (1988). [I741 R. Hoffmann and R. Briickner, Angew. Chem. 104,646 (1992); Int. Ed. Engl. 31,647 (1992). [I751 K. Tomioka, T. Igarashi, M. Watanabe and T. Nakai, Tetrahedron Lett. 33, 5795 (1992). [I761 J.A. Marshall and T.M. Jenson, J. Org. Chem. 49, 1707 (1984). [I771 K. Hiroi and K. Nakazawa, Chem. Lett. 1077 (1980). [I781 M. Uchikawa, T. Hanamoto, T. Katsuki and M. Yamaguchi, Tetrahedron Lett. 27, 4577 (1986). [I791 0. Takahashi, K. Mikami and T. Nakai, Chem. Lett. 69 (1987). [180] M. Uchikawa, T. Katsuki and M. Yamaguchi, Tetrahedron Lett. 27, 4581 (1986). [I811 A. Ishikawa, H. Uchiyama, T. Katsuki and M. Yamaguchi, Tetrahedron Lett. 31, 2415 (1990). [I821 Y.-D. Wu, K.N. Houk and J.A. Marshall, J. Org. Chem. 55, 1421 (1990). [I831 R. Briickner, Chem. Ber. 122, 193 (1989). [I841 J.A. Marshall, E.D. Robinson and A. Zapata, J. Org. Chem. 54, 5854 (1989).

References [I851 [186] [I871 [I881

35 3

K.-K. Chan and G. Saucy, J. Org. Chem. 42, 3828 (1977). K. Mikami, K. Azuma and T. Nakai, Tetrahedron Lett. 24, 513 (1983). R. W. Hoffrnann, R. Gerlach and S. Goldmann, Tetrahedron Lett. 2599 (1978). P. Bickart, F. W. Carson, J. Jacobus, E. G. Muller and K. Mislow, J. Am. Chem. SOC.90, 4869 ( 1 968). [189] R.W. Hoffmann, G. Goldmann, R. Gerlach and N. Maak, Chem. Ber. 113, 845 (1980). [I901 K. Hiroi, P. Kitayama and S. Sato, J.C.S. Chem. Comm. 1470 (1983). 11911 W. Oppolzer, Angew. Chem. 101, 39 (1989); Int. Ed. Engl. 28, 38 (1989). 11921 R.K. Hill and M. Rabinovitz, J. Am. Chem. SOC.86, 965 (1964). [193] G. B. Gill and B. Wallace, J. C. S. Chem. Comm. 382 (1977). [I941 W.G. Dauben and T. Brookhart, J. Am. Chem. SOC.103, 237 (1981). [195] W.G. Dauben and T. Brookhart, J. Org. Chem. 47, 3921 (1982). [I961 J. V. Duncia, P.T. Lansbury, Jr., T. Miller and B. B. Snider, J. Am. Chem. SOC.104, 1930 (1 982). [I971 J.K. Whitesell, R.M. Lawrence and H.-H. Chen, J. Org. Chem. 51, 4779 (1986). [I981 J.K. Whitesell, H.-H. Chen and R.M. Lawrence, J. Org. Chem. 50, 4662 (1985). [199] K. Mikami, T.-P. Leh and T. Nakai, 'Tetrahedron Asym. 1, 13 (1990). [200] K. Mikami, M. Kaneko, T.-P. Loh, M. Terada and T. Nakai, Tetrahedron Lett. 31,3909 (1990). [201] K. Mikami, M. Shimizu and T. Nakai, J. Org. Chem. 56, 2952 (1991). [202] C. A. Townsend, T. Scholl and D. Arigoni, J. C. S. Chem. Comm. 921 (1 975). [203] W. Oppolzer, K.K. Mahanalabis and K. Battig, Helv. Chim. Acta 60, 2388 (1977). [204] W. Oppolzer, K. Battig and T. Hudlicky, Helv. Chim. Acta 62, 1493 (1979). [205] P.D. Kennewell, S.S. Matharu, J.B. Taylor and P.G. Sammes, J.C.S. Perkin 1 2563 (1982). [206] W. Oppolzer and C. Robbiani, Helv. Chim. Acta 63, 2010 (1980). 12071 W. Oppolzer, C. Robbiani and K. Battig, Helv. Chim. Acta 63, 2015 (1980). [208] W. Oppolzer, C. Robbiani and K. Battig, Tetrahedron 40, 1391 (1984). [209] K.L. Handoo, S.K. Handoo, K. Gadru and A. Kaul, Tetrahedron Lett. 26, 1765 (1985). [210] H. Felkin, L. D. Kwart, G. Swierczewski and J. D. Umpleby, J. C. S. Chem. Comm. 242 (1975). [21 I] W. Oppolzer and J.-M. Gaudin, Helv. Chim. Acta 70, 1474 (1987). [212] W. Oppolzer, T.H. Keller, D.L. Kuo and W. Pachinger, Tetrahedron Lett. 31, 1265 (1990). [213] W. Oppolzer, R. Pitteloud and H.F. Strauss, J. Am. Chem. SOC.104, 6476 (1982). [214] E.P. Lodge and C.H. Heathcock, J. Am. Chem. SOC.109, 2819 (1987). [215] K. Mikami, M. Terada and T. Nakai, J. Am. Chem. SOC.112, 3949 (1990). [216] K. Mikami, S. Narisawa, M. Shimizu and M. Terada, J. Am. Chem. SOC.114, 6566 (1992). 12171 S. Sakane, K. Maruoka and H. Yamamoto, Tetrahedron 42, 2203 (1986). [218] K. Maruoka, Y. Hoshino, T. Shirasaka and H. Yamamoto, Tetrahedron Lett. 29, 3967 (1988).

Chapter 8 [I] [2] [3] [4] [5]

D. J. Abbott, S. Colonna and J.M. Stirling, J. C. S. Perkin I 492 (1976). M. Furukawa, T. Okawara and Y. Terawaki, Chem. Pharm. Bull. 25, 1319 (1977). R.K. Hill and L.A. Renbaum, Tetrahedron 38, 1959 (1982). T. Wakabayashi, Y. Kato and K. Watanabe, Chem. Lett. 1283 (1976). N. Shida, T. Uyehara and Y. Yamamoto, J. Org. Chem. 57, 5049 (1992).

354

References

[6] H. Yoda, T. Shirai, T. Kawasaki, T. Katagiri, K. Takabe, K. Kimata and K. Hosoya, Chem. Lett. 793 (1991). [7] B. Lange, F. Bolhuis and B. L. Feringa, Tetrahedron 45, 6799 (1989). [8] S. G. Davies and 0. Ichihara, Tetrahedron Asym. 2, 183 (1 991). [9] M. Hirama, T. Iwakuma and S. Ito, J. C. S. Chem. Comm. 1523 (1987). [lo] M. Hirama, T. Shigemoto, Y. Yamazaki, S. Ito, J. Am. Chem. SOC.107, 1797 (1985). [ l l ] M. Hirama, T. Shigemoto and S. Ito, Tetrahedron Lett. 26, 4137 (1985). [12] M. Hirama, T. Shigemoto, Y. Yamazaki and S. Ito, Tetrahedron Lett. 26, 4133 (1985). [13] B.M. Trost and D.L. Van Vranken, Angew. Chem. 104, 194 (1992); Int. Ed. Engl. 31, 228 (1992). [14] T. Hayashi, A. Yamamoto and Y. Ito, Tetrahedron Lett. 29, 99 (1988). [15] T. Hayashi, A. Yamamoto, Y. Ito, E. Nishioka, H. Miura and K. Yanagi, J. Am. Chem. SOC. 111, 6301 (1989). [16] J.E. Backvall and E.E. Bjorkman, J. Org. Chem. 45, 2893 (1980). [I71 W. Oppolzer and 0. Tamura, Tetrahedron Lett. 31, 991 (1990). [18] W. Oppolzer, 0. Tamura, G. Sundarababu and M. Signer, J. Am. Chem. SOC.114, 5900 (1992). [19] D.A. Evans, T. C. Britton, R. L. Dorow and J. F. Dellaria, J. Am. Chem. SOC.108, 6395 (1986). [20] L.A. Trimble and J.C. Vederas, J. Am. Chem. SOC.108, 6397 (1986). [21] D. A. Evans, T. C. Britton, R. L. Dorow and J. F. Dellaria Jr., Tetrahedron 44, 5525 (1 988). [22] C. Gennari, L. Colombo and G. Bertolini, J. Am. Chem. SOC.108, 6394 (1986). [23] W. Oppolzer and R. Moretti, Helv. Chim. Acta 69, 1923 (1986). [24] B.M. Trost and E. Keinan, J. Am. Chem. SOC. 100, 7779 (1978). [25] H.C. Brown, A.M. Salunkhe and B. Singaram, J. Org. Chem. 56, 1170 (1991). [26] H.C. Brown, M.M. Midland and A.B. Levy, Tetrahedron 43, 4079 (1987). [27] M. Mikolajczyk, Pure Appl. Chem. 52, 959 (1980). [28] S.J. Abbott, S.R. Jones, S.A. Weinman and J.R. Knowles, J. Am. Chem. SOC.100, 2558 (1 978). [29] C. R. Hall and T. D. Inch, J. C. S. Perkin 1 1104 (1979). [30] T. Glowiak and W. Sawaka-Dobrowolska, Tetrahedron Lett. 42, 3965 (1977). [31] T. Sato, H. Ueda, K. Nakagawa and N. Bodor, J. Org. Chem. 48, 98 (1983). [32] H. C. Brown and G. Zweifel, J. Am. Chem. SOC.83, 486 (1961). [33] H.C. Brown, N.R. Ayyangar and G. Zweifel, J. Am. Chem. SOC.86, 1071 (1964). [34] H. C. Brown and N. M. Yoon, Isr., J. Chem. 15, 12 (1976177). [35] J.J. Partridge, N.K. Chadha and M.R. Uskokovic, J. Am. Chem. SOC.95, 7171 (1973). [36] G. Grethe, J. Sereno, T. H. Williams and M.R. Uskokovic, J. Org. Chem. 48, 5315 (1983). [37] D. J. Sandman, K. Mislow, W.P. Giddings, J. Dirlam and G.C. Hanson, J. Am. Chem. SOC. 90, 4877 (1968). [38] H. C. Brown, M. C. Desai and P.K. Jadhav, J. Org. Chem. 47, 5065 (1982). [39] H.C. Brown, P.K. Jadhav and A.K. Mandal, Tetrahedron 37, 3547 (1981). [40] A.K. Mandal and N.M. Yoon, J. Organomet. Chem. 156, 183 (1978). [41] A.K. Mandal, P.K. Jadhav and H.C. Brown, J. Org. Chem. 45, 3543 (1980). [42] H.C. Brown and P.K. Jadhav, J. Org. Chem. 46, 5047 (1981). [43] H. C. Brown, P.K. Jadhav and A.K. Mandal, J. Org. Chem. 47, 5074 (1982). [44] H. C. Brown, P. K. Jadhav and M. C. Desai, Tetrahedron 40, 1325 (1 984). [45] H.C. Brown, P.K. Jadhav and M.C. Desai, J. Am. Chem. SOC.104, 4303 (1982). [46] P.K. Jadhav and H.C. Brown, J. Org. Chem. 46, 2988 (1981). [47] H. C. Brown, J. V. N. V. Prasad and M. Zaidlewicz, J. Org. Chem. 53, 291 1 (1988). [48] S. Masamune, B. Kim, J.S. Petersen, T. Sato and S. J. Veenstra, J. Am. Chem. SOC.107, 4549 (1985). [49] W.C. Still and J. C. Barrish, J. Am. Chem. SOC.105, 2487 (1983). [50] T. Harada, Y. Matsuda, J. Uchimura and A. Oku, J.C.S. Chem. Comm. 1429 (1989). [51] D.H. Birtwistle, J.M. Brown and M. W. Foxton, Tetrahedron Lett. 27, 4367 (1986). [52] I. Fleming and N. J. Lawrence, Tetrahedron Lett. 29, 2077 (1988).

References

355

D.A. Evans, G. C. Fu and A. H. Hoveyda, J. Am. Chem. Soc. 110, 6917 (1988). D.A. Evans and G.C. Fu, J. Org. Chem. 55, 2280 (1980). T. Hayashi, Y. Matsumoto and Y. Ito, J. Am. Chem. Soc. 111, 3426 (1989). C. Zioudrou, I. Moustakali-Mavridis, P. Chrysochou and G. J. Karabatsos, Tetrahedron 34, 3181 (1978). [57] W.C. Still and K.R. Shaw, Tetrahedron Lett. 3725 (1981). [58] D. J. Morgens, Jr., Tetrahedron Lett. 3721 (1981). [59] W.C. Still and K.P. Darst, J. Am. Chem. Soc. 102, 7385 (1980). [60] K.N. Houk, N.G. Rondan, Y.-D. Wu, J.T. Metz and M. N. Paddon-Row, Tetrahedron 40, 2257 (1984). [61] J.-E. Backvall, P.G. Anderson and J.O. Vagberg, Tetrahedron Lett. 30, 137 (1989). [62] J.-E. Backvall, J.-E. Nystrorn and R.E. Nordberg, J. Am. Chem. SOC.107, 3676 (1985). [63] S.S. Jew, S. Terashima and K. Koga, Tetrahedron 35, 2337 (1979). [64] S.S. Jew, S. Terashima and K. Koga, Tetrahedron 35, 2345 (1979). [65] M. Hayashi, S. Terashima and K. Koga, Tetrahedron 37, 2797 (1981). [66] A.R. Chamberlin, M. Dezube and P. Dussault, Tetrahedron Lett. 22, 4611 (1981). [67] P.A. Bartlett, D.P. Richardson and J. Myerson, Tetrahedron 40, 2317 (1984). [68] Y. Tamaru, M. Mizutani, Y. Furukawa, S. Kuwamura, 2. Yoshida, K. Yanagi and M. Minobe, J. Am. Chem. Soc. 106, 1079 (1984). [69] Y. G. Kim and J. K. Cha, Tetrahedron Lett. 29, 201 1 (1 988). [70] A.R. Chamberlin and R.L. Mulholland, Jr., Tetrahedron 40, 2297 (1984). [71] J. Ichikawa, M. Asami and T. Mukaiyama, Chem. Lett. 949 (1984). [72] T. Mukaiyama, I. Tomioka and M. Shimizu, Chem. Lett. 49 (1984). [73] Y. Sucla, S. Yago, M. Shiro and T. Taguchi, Chem. Lett 389 (1992). [74] M. Ohshima and T. Mukaiyarna, Chem. Lett 377 (1987). (751 J. Gawronski, K. Gawronska and H. Wynberg, J.C.S. Chem. Cornm. 307 (1981). [76] H. Hiemstra and H. Wynberg, J. Am. Chem. SOC.103, 417 (1981). [77] H. Yamashita and T. Mukaiyama, Chem. Lett. 363 (1985). [78] K. Suzuki, A. Ikegawa and T. Mukaiyma, Bull. Chem. SOC.Jap. 5, 3277 (1982). (791 N. Kobayashi and K. Iwai, Macromolecules 13, 31 (1980). [80] N. Kobayashi and K. Iwai, Tetrahedron Lett. 2167 (1980). [81] G. Solladie, Synthesis 185 (1981). [82] C. Mioskowski and G. Solladie, Tetrahedron 36, 227 (1980). 1831 C. Mioskowski and G. Solladie, Tetrahedron Lett. 3341 (1975). [84] K. Hiroi and K. Makino, Chem. Lett. 617 (1986). [85] J.K. Whitesell and J.F. Carpenter, J. Am. Chem. SOC.109, 2839 (1987). [86] S.C. Benson and J.K. Snyder, Tetrahedron Lett. 32, 5885 (1991). (871 D. A. Evans, M.M. Fad, L. Colombo, J. J. Bisaha, J. Clady and P. Giaroni, J. Org. Chem. 57, 782 (1992). [88] E. Differding and R.W. Lang, Tetrahedron Lett. 29, 6087 (1988). [89] F.A. Davis and W. Han, Tetrahedron Lett. 33, 1153 (1992). [90] W. Oppolzer and P. Dudfield, Tetrahedron Lett. 26, 5037 (1985). [91] D.A. Evans, J.A. Ellmann and R.L. Dorov, Tetrahedron Lett. 28, 1123 (1987). [92] G. Castaldi, S. Cavicchioli, C. Giordano and F. Uggeri, J. Org. Chem. 52, 3018 (1987). [93] G. Castaldi and C. Giordano, Synthesis 1039 (1987). [94] C. Giordoni and L. Coppi, J. Org. Chem. 57, 2765 (1992). [95] L. Duhamel, A. Ravard, J.-C. Plaquevent and D. Davoust Tetrahedron Lett. 28, 5 5 17 ( 1 987). [96] P. Knochel, Angew. Chem. 104, 1486 (1992); Int. Ed. Engl. 31, 1459 (1992). [97] L. Duhamel and J.-C. Plaquevent, Tetrahedron Lett. 42, 2285 (1977). [98] L. Duhamel and J.-C. Plaquevent, Bull. SOC.Chim. France, 69 (1982). [99] L. Duhamel and J.-C. Plaquevent, Bull. SOC.Chim. France, 75 (1982). [loo] M. Asami and H. Kirihara, Chem. Lett. 389 (1987). [ l o l l M. Asami, Bull. Chem. SOC.Jap. 63, 1402 (1990). [53] [54] [55] [56]

356

References

[I021 M. Asami, Bull. Chem. SOC. Jap. 63, 721 (1990). [I031 J. Hine, W.-S. Li and J.P. Zeigler, J. Am. Chem. Soc. 102, 4403 (1980). [lo41 M. Yamaguchi, S. Yamamatsu, H. Oikawa, M. Saburi and S. Yoshikawa, Inorg. Chem. 20, 3179 (1981). [I051 T. Shiraiwa, S. Sakata, Y. Baba and H. Kurokawa, Chem. Lett. 233 (1990). [I061 S. R. Wilson and M. F. Price, J. Am. Chem. Soc. 104, 1124 (1982). [I071 S.R. Wilson and M.F. Price, Tetrahedron Lett. 569 (1983). [lo81 0. Piva, F. Henin, J. Muyart and J.-P. Pete, Tetrahedron Lett. 28, 4825 (1987). [I091 0. Piva and J.-P. Pete, Tetrahedron Lett. 31, 5157 (1990). [ I 101 R. Mortezaei, D. Awandi, F. Henin, J. Muzart and J.-P. Pete, J. Am. Chem. SOC.110, 4824 (1988). [ I 111 J.L. Charlton and V.C. Phani. Tetrahedron Lett. 33, 6073 (1992). [ I 12) Y. Naruse and H. Yainamoto, Tetrahedron 44, 6021 (1988). [I131 R. Shirai, M. Tanaka and K. Koga, J. Am. Chem. Soc. 108, 543 (1986). [114] H. Kumobayashi and S. Akutagawa, J. Am. Chem. Soc. 100, 3949 (1978). [ 1 1 5) H. J. Hansen, R. Schmid, M. Schmid, Eur. Pat. Appl. EP. 104.376; C. A. 101, 38124 (1 984). [I 161 H. J. Hansen, R. Schmid and M. Schmid, Eur. Pat. EP. Appl. 104.375; C.A. 101, 1 1 1 . I 19 ( I 984). [ I 171 K. Tani, T. Yamagata, S. Akutagawa, H. Kumobayashi, T. Taketomi, H. Takaya, A. Miyashita, R. Noyori and S. Otsuka, J. Am. Chem. Soc. 106, 5209 (1984). [ l 181 K. Tani, T. Yamagata, Y. Tatsuno, Y. Yamagata, K. Tomita, S. Akatugawa, H. Kumobayashi and S. Otsuka, Angew. Chem. 97, 232 (1985); Int. Ed. Engl. 24, 217 (1985). [I191 S. Otsuka and K. Tani, Synthesis 665 (1991). [I201 R. Schmid and H.-J. Hansen, Helv. Chim. Acta 73, 1258 (1990). [I211 R. Schmid, M. Cereghetti, B. Heiser, P. Schonholzer and H. J. Hansen, Helv. Chim. Acta 71, 897 (1988). [I221 S. Inoue, H. Takaya, K. Tani, S. Otsuka, T. Sato and R. Noyori, J. Am. Chem. Soc. 112,4897 (1990).

StereoselectiveSynthesis Mihily N6gradi Copyright 0VCH Verlagsgesellschaft mbH,1995

Subject Index

Acetal ring opening of 252f. Acetylide ion - addition of 150f. Acids - b-amino - by addition 220 - a,a-disubstituted - by alkylation 235f. - a-hydroxy - by enolate oxidation 134 - from dithianes 145 - from glyoxylate esters 149 - from oxazolines 149 - p-hydroxy - by aldol reaction 203 - from acetals 253 - from ketones 153, 180, 205 - from oxazolines 205 - y-hydroxy - by aldol reaction 204 - a-substituted - by hydrocarbalkoxylation 230 - from oxazolines 237f. - p-substituted - by addition 219ff. - from oxazolines 225 Acylaminocinnamic acids - hydrogenation of 56ff. N-Acylenamines - hydrogenation of 56ff. Acylation - of tin esters of diols 302 AD-mix 127 Addition - acetoxy, to 1,3-dienes 299 - anti-Cram 143 - diastereoselective 139 - homoallyl - by ally1 addition 157 - Lewis acid catalyzed 160

-

-

Michael 213ff allylic rearrangement in 217 - chiral catalysis of 214 - of thiols 303 of acetylide ion 150 of allylmetals - mechanism of 160ff. - stereochemistry of 157 ff. of allylsilanes 170f. of allylstannanes 173f. of allyltitanium compounds 167 f. of amines 294 of azodicarboxylate 295 of carbamates 294 of carbanions - to olefins 213ff. of chlorocarbene to olefins 21 of crotylboranes 162 of cyanide 30, 139, 146 of isonitriles 210f. of phosphoranes 275 of silanes to dienes 72 of silylcuprates 241 of silylketene acetals 208 of titanium alkyls 142f. of zinc alkyls 137 syn 258 to aldoximes 167 to alkoxyaldehydes, a- 206 to allylic alcohols 226 to aminals 145 to aminoketones 142 to azomethines, unsaturated 224 to bicyclic ketones 150 to C = N bonds 154ff. to cyclic ketones 150 to faces 9 - diastereotopic 22 to glyoxylate esters 175 to ketoesters 148 to metallated azomethines 218 to tricoordinate centers 9 ~

-

-

-

-

-

-

-

-

-

-

-

-

-

360

Subject Index

Alcohols - acetylenic - from allenes 165 - by hydrosilylation 72ff. - by Meerwein reduction 95 - by reduction 83ff., 91 ff., 1 IOff., 113, 117 - chiral primary, by reduction 93 - cyclic - by addition 150f. - by hydrosilylation 76 - by reduction 102, 108, 119ff. - hornoallylic 158 ff. - a-keto - by enolate oxidation 124, 134 - propargyl - by ketal cleavage 117 - by reduction 83f. sec-Alcohols - by acetal cleavage 117 - by addition to aldehydes 137ff., 147 - by catalytic reduction 28 f. - by hydroboration 26, 297ff. - by hydrogenation 66f., 17 - by hydrosilylation 71 ff. - by reduction 82ff., 90ff., 96ff., 102, 113ff. - with BINAL-H 84 - with Alpine-boranea 93 Aldehydes - a-hydroxy - from methyl glyoxylate 147 - by aldol reaction 207f. - reduction of 92 - p-substituted, by addition 225, 228 Aldol condensation 28, 208 fl" Aldol reaction 177 ff. - mechanism of 178ff. - stereochemistry of 178ff. - transition state of 178f. Aldoximes, addition to 167 Alkyl migration 254 Alk ylation - allylic 226 - of amide enolates 234 f - of amidines 248 - mechanism of 249 - of azornethines 244f. - of carbamates 243 - of enamines 244f. - of enolates 2320ff. - of hydrazones246

of oxazoline azaenolates 235 ff. of stabilized carbanions 240ff. Allylamines, rearragement of 308 Allylboronates, silylated 166 Allylic - alkylation 226 - intramolecular 227 - strain 231 Allylic alcohols - addition to 226 - from epoxides 306 - rearragement, in Michael addition 217 - substituted p- by addition 222 Allylic rearragement 160, 294 Allylsilanes, addition of 170f. Allylstannanes, addition of 173f. Allyltitanium compounds, addition of 167 Alpine borane 92f. Aluminum hydrides, chiral 81 Amides - by alkylation 231 - enolates of - chiral 234f. - in Michael addition 217 - a-hydroxy, by oxidation 134 - /I-hydroxy - by hydrosilylation 74 - by reduction 1 1 1 - a-substituted Amidines - addition to 145 - alkylation of 220, 247 - arylation of 220 Amines - by addition to C = N bonds 154f. - by allylmetal addition 170 - by hydrogenation 68 - by reduction 88 - from allylic acetate 295 - from boronic esters 296 - from dichloroboranes 296 - from hydrazones 155. 247 - from oxazolidines 146 - primary 146 -

-

-

sec-

- by hydrosilylation 73 by hydrogenation 79 - by reduction 88f., 92 Amino acids -

-

a-

by amide alkylation

235

Subject Index by hydrogenation 79f. from diketopiperazines 239 - from enamines 156 - from enolates 295 - from ketene acetals 295 - from ketones 155 - a,a-disubstituted, by alkylation 246 Aminoalcohols - by hydrogenation 66 - by reduction 1 1 1 f. Aminoalcohols - a- by addition 142 - from azomethines 155 by allylmetal addition - by reduction 112 - from azomethines 155 f. Aminoketones - a-, addition to 142 - /3-, reduction of 82 Aminosulfones, alkylation of 243 anti 6 Anticlinal 5 Antiperiplanar 4, 5 Antiperiplanarity effect 1 10, 142 Arylat ion - catalytic 251 - of aminals 220 Asymmetric - amplification 31, 138 - catalysis - in Diels-Alder reaction 264f. induction 32 f. - 1,2- 101, 142 - in acyclic systems 110 - in enone alkylation 222 - 1,3- 115, 146 - 1,4- 118, 147f., 153 - 1,5- 118, 153 - 1,6- 119, 148 - 1,7- 118 - by remote centers 118, 147f. - double 35, 162 - in allylmetal addition 170 - in Diels-Alder reaction 258f. - in epoxidation 132 - in halolactonization 302 - in hydroboration 298 - in hydrogenation 68 - in sulfide oxidation 133 - synthesis, Marckwald’s definition of 33 -

-

Ate complexes - Michael addition of 214 Azaenolates - alkylation of 245f. - of oxazolines, alkylation of 235ff. Azomethines - addition of cyanide to 154 - addition of sulfoxides to 156 - alkylation of 244f. - cycloaddition to 271 - metallated, addition to 218 - N-acyl, addition to 156 - unsaturated, addition to 224 BBNH 9, 89 - hydroboration with 298 Biaryls - by cross coupIing 250 - from oxazolines 254 - synthesis of 25 BINAL-H, - reduction with 83 - mechanism of 25 BINAP 48, 49, 71 - in hydroboration 299 - ruthenium catalyst with 67 BINOL (2,2’-dihydroxy-l,lf-biphenyl)264 - aluminum hydride complex of 83 - in allylic alkylation 228 - LiAIH, complex of 25 - Ti(1v) complex 31 Biphosphines 46 Bis-lactimethers, alkylation of 238f. Bond angles 4 Borabicyclo[3.3.1 jborane see BBNH Boronic acids, by hydroboration 298 BPPFA 58 BPPM 48, 59, 60, 66, 72 Bromination - of boron enolates 306 - of ketals 306 But-2-enes, addition to 21 C.I.P. convention 3 C = N bonds - hydrogenation of 79 - reduction of 88 CAMP 45, 48, 51 Carbamates, alkylation of 243 Carbenes - addition of 273 Carbonyl reductions, catalytic 50

361

362

Subject Index

Carboxylic acids - dianion of 194 - hydrogenation of 55 ff - a-hydroxy 144 - by reduction 118 - substituted, by hydrogenation 68f. Catalysis - chiral - in dihydroxylation 126f. - in alcoholysis 25 - in aldol condensation 209’ - in aldol reaction 198 - in amine addition 294 in carbene addition 273f. - in ene-reaction 292 - in isonitrile addition 210* - in zinc alkyl addition 138 - of decarboxylation 307 - of H-D exchange 307 - of Michael addition 214 - prediction of - stereoselectivity in 218 Catalysts for epoxidation 124 polymer supported 49 - rhodium-phosphine 50ff. ruthenium 49 - Wilkinson’s 45, 50 - zeolite linked 49 Chalcones, epoxidation of 124 Chelate model, of nucleophilic addition -

-

-

104

Chelates, rhodium diphosphine 51 ff. Chelation control 142, 146, 172 in nucleophilic addition 104 Chiral - auxiliaries in hydrogenation 78 - acetals, ring opening of 252f. - acrylates, Diels-Alder reaction of 258f. - additives 28 - aluminum hydrides 81 - amplification 138 - auxiliaries 17, 29 - recovery of 29 - base - in acylation of tin esters 302 - in enolization 307 diamines 136 dihyropyridines 95 - epoxidation catalysts 124 - iron complex 202, 276 -

-

-

lithium base in enolate alkylation 231 oxathianes 144 - oxaziridines 29 oxazolidines 40, 145, 148 - oxidants 123 - peracids 29 phosphate, synthesis of 296 - phosphine ligands 48 racemic 31 - sensitizers 25 - sulfoxides 152 Chirality 3 - deposition 220, 235 - transfer of 35f., 39 79, 220, 235, 238, 277ff., 288 - in alkyl migration 254 - two-dimensional 14 CHIRAPHOS 48, 52, 53, 60, 63 Chloroboranes 94 Chromium - ally1 176 - crotyl 176 cis 5 Claisen rearrangement 38 - chirality transfer in 39 Cobalamine - hydroganetion with 65 Cobalt catalysts, hydrogenation with 65 Complexes - binaphthyl 52 - dihydride-rhodium phosphine 50 - ferrocene 52 - iridium-phosphine 50 - rhenium 50 Compound, racemic 18 Configuration 5 52 - prediction of in hydrogenation - relative 5 - relative, in acyclic molecules 5 , 6 Conformation 4 Conformers 4 - reactivity of 16 Conglomerate 18 Connectedness 2, 7 Constitution 7 Control - kinetic 41 - product development 108, 109 - stereoelectronic 32 - steric approach 108 - thermodynamic 41 Cornforth’s rule 104 -

-

-

-

Subject Index Cram’s rule 142, 146, 166, 206 Cross coupling 249f. Cross-Aldol reaction 197 f. Crotylboranes, addition of 162 Curtin-Hammet principle 42, 103 Cyanide - addition 139, 146 - to azomethines 154 Cyclic model - of nucleophilic addition 103 C ycloadditions - [2+2] 270 - [3+2] 268 - Diels-Alder 257 ff. - to azomethines 273 Cyclobutanes by photodimerization 270 Cyclopropanation - with dihalomethanes 275 - with phosphoranes 275 Cyclopropanes - by carbene addition 273 Darvon alcohol 85 DBNE 138 Decarboxylation 307 Dehydropeptides, hydrogenation of 59 Deracemisation 15, 306 Dialkyl sulfides, oxidation of 133 Diamines - 1,2-, from azomethines 156 - chiral 136 Diastereomeric ratio 31 Diastereomers 2 - differentiation 27 - equilibration of 30 - formation of 22 - nomenclature for 5 - selectivity 12 Diastereoselectivit y - inducing of 31 - inversion of 32 - inversion temperature of 32 - substrate 12 Diastereotopic - faces 9 - addition to 22 - groups 7 Diastereotopos differentiation 27 Diazaborolidines 282 Diazoester insertion 254 DIBAL 116 - reduction with 91

363

/3-Dicarbonyl compounds, hydrogenation of 78 Dicarboxylic acids, hydrogenation of 64 Diels-Alder reaction 257 ff. - aza 263 - hetero 263 - intramolecular 266 - mechanism of 267 Diesters, by Michael addition 216f. Differentiation of molecules 1 Dihydropyridines, chiral 95 f. Dihydropyridones, by Diels-Alder reaction 265 2,2’-Dihydroxy-I ,l’-binaphthol see BINOL Di hydroxylation - diastereoselective 134 - with OsO, Diisobutylaluminum hydride see DIBAL Diisopinocampheylborane 90 - hydroboration with 27 - reduction with 35 Di ketones - by Michael addition 216 - 1,3-, hydrogenation of 67 - I$, by Michael addition 214 Diketopiperazines, alkylation of 238 Diols - 1,2- by addition 152 - by cis-hydroxylation 127 ff. - from aldehydes 154 - from diketones 143 - from hydroxyacids 144 - 1,3- by hydrogenation 67, 17 - by hydrosilylation 71 f., 74 - by reduction 112, 116f. - from acetals 253f. - from aldehydes 143, 149, 170 - from carbamates 243 - from hydroxyketones 146 - from ketols 146 - from nitroalkanols 149 - 1,4-, by reduction 116f. chiral, boronates of 164f. DlOP 47f., 51, 58, 60, 64, 72, 75f., 226, 229, 249 DIOXOP 48f. DIPAMP - rhodium complex of 46, 48, 53, 5Sf., 58, 60, 63, 70 ~

364

Subject Index

DIPAMP (cont.) - iridium complex of 60 Diphosphines, chiral, in allylic alkylation 227 Dipolar model, of nucleophilic addition 1,3-Dithianes, alkylation of 241 f. Double asymmetric induction 35

E 2, 5 E-2 isomerization 59, 196 Enamines, hydrogenation of 78 Enantioconvergent synthesis 36 Enantioface differentiation 27 Enantiorner differentiation 27 selectivity 13 Enantiorneric - excess 17 - purity 17 Enantiomers 2 Enantioselectivity - inducing of 27 - substrate 13 Enantiotopos differentiation 27 endo rule 258 Ene reaction 287 ff. - magnesium 292 - of N-sulfinylcarbamates 303 - retro 290 Enolates - Claisen rearrangement of 278 - E-Z isomerization of 196 - fluorination of 305 - generation of 184ff. - hydroxylation of 134 - inversion of configuration of 231 - of iron complexes 235 - oxidation of 29 - tin(ri) 235 Enolization - photo 307 - with chiral base 308 Enolsilanes 195 - Michael addition of 214 Enones - macrocyclic, addition to 224 - Michael addition to 214 Enzymatic transformations 14 Epimerization 13 Epoxidation - diastereoselective 130 ff. - enantioselective 123ff ~

~

-

104

kinetic resolution by 132 mechanism of 125f. of allylic alcohols 124 ff of chalcones 124 - of cyclic olefins 130 - of homoallylic alcohols 132 - of simple allylic alcohols 127 - Sharpless method of 125f. Epoxides - by oxidation 123ff., 130ff. - from ketones 154 Equilibria 13 erythro 5, 6 Esters - P-alkyl - by alkylation 231 ff., 240 - by reduction 89 - apdisubstituted, by alkylation 221 - hydroxy, by hydrosilylation 73 a-hydroxy - by reduction 96 from ester enolates 198 p-hydroxy - alkylation of dianions of 240 - by aldol reaction 203 - by hydrogenation 67, 77 179, 196, - from ester enolates 198 - p-substituted - by addition 220 - by Michael addition 217 Excess - diastereomeric 17 - enantiomeric 17 - values 17 ~

~

~

Faces - molecular 7 - diastereotopic 9 - enantiotopic 9 - addition to 10, 18, 19, 21 - homotopic 9 - addition to 18 - transformation of 18 Felkin-Anh model of nucleophilic addi tion 107 Fischli’s scheme 39 Fleming’s ruIe 240 Gauche 5 Geometrical parameters 4 Glyoxylate esters, addition to

175

Subject Index Grignard reagents - addition of 137 - chiral complexes of 136 - cross coupling 249f. Groups 7 - chiral auxiliary 17 - diastereotopic 7 - addition to 22 - enantiotopic 8 - addition to 19, 20 - heterotopic 8 - homotopic 8 - prochiral 10 Haloetherification 302 Halolactonization 300 f. Homoallylic alcohols - epoxidation of 132 - from aldehydes 158ff. Homotopic groups, transformation of 18 Hydrazino esters 295 Hydrazones - alkylation of 246 - metallated, in Michael addition 218 Hydride transfer 92 - from chiral donor 29 Hydroboration 26 f., 267 ff. - of dienes 299 Hydrocarbalkoxylation 229 Hydrocarbonylation 229 Hydroformylation 229 Hydrogen migration 306 f. Hydrogenation - chiral auxiliaries in 78 - diastereoselective 68 - effect of base addition in 55 - enantioselective 48 ff - heterogeneous catalytic 76 - mechanism of 60ff. - of P-dicarbonyl compounds 78 - of carboxylic acids, unsaturated 55ff. - of dehydropeptides 59 - of enamines 78 - of imines 66, 68 - of ketones 66, 79 - pressure effects in 54, 63, 64 - solvent effects in 53 - temperature effect in 55 H ydrosilylation - catalytic 70 - mechanism of 75 - of cyclic ketones 75

365

- of homoallyl alcohols 74 - of ketoesters 73 - of ketones 70 - Schiff bases 73 P-Hydroxy acids - by hydrogenation 66f., 69 - by hydrosilylation 73 - by reduction 118 Hydroxyaldehydes 145 P-Hydroxycarboxylic acids - by halolactonization 300 a Hydroxyesters by ene-reaction 31 a-Hydroxylation - of enolates 134 - cis, by OsO, 127ff.

Imines - hydrogenation of 66 Immonium salts, reduction of 88 Induction see Asymmetric induction Iridium, phosphine complexes of 60 Isoinversion - correlation 32 - principle 32, 271 - temperature 32 Isomerism 2 Isomerization - E / Z 28, 59 - of allylmetals 162, 166 - of enolates 231 Isomers - constitutional 2 - stereo 2 Isonitriles, addition of 210f. Isoxazolines - by cycloaddition 268 Ketene acetals 280 - bromination of 305 Ketenes, addition of alcohols to Ketoesters - addition to 148

33

- P

by alkylation 245 from diazoesters 255 - hydrosilylation of 73 Ketones - a-hydroxy, by enolate oxidation - alkylation of 30 - bicyclic, addition to 150 - cyclic, addition to 101, 15Of. - hydrogenation of 66, 79 -

124

366

Subject Zndex

Ketones (cont.) - hydroxy

- a from diketones 148 by enolate oxidation 124 - /I-, by aldol reaction 179ff., 200ff. - reduction of 83ff. - a-substituted - by addition 228, 241 - by alkylation 231 ff., 245 - by pinacol rearrangement 255 - from hydrazones 247f. - /I-substituted, by addition 222ff. Kinetic - control 24, 41 - resolution 13, 15, 235 - by epoxidation 132 -

I 21 /3-Lactams, by cycloaddition 270 &-Lactones 137 - by Claisen rearrangement 37 - by oxidation 129 - by cycloaddition 268 Light - circularly polarized 14, 15 - enantioselectivity induced by - polarized 6 like 21 Ik 21, 23

27

M 3 Mechanism 175 - linear, of allylmetal addition - of aldol reaction 178ff. - of allylamine rearragement 309 - of amidine alkylation 249 - of azaenolate alkylation 237 - of dihydroxylation 129 - of epoxidation 125f. - of hydrogenation 60ff. Meerwein-Ponndorf-Verley reduction 92 (-)-Menthol, synthesis of 309 Metal alkoxides, reduction with 95 Metal alkyls, reduction with 94 Michael addition - intramolecular of carbamates 294 - of thiols 303 Molecules, distinguishing of 1 Monoisopinocampheylborane 90, 298 NADPH models

95

Nitriles, a-hydroxy - by addition to aldehydes 139 - from ketals 253 Nitriloxides - cycloaddtion of 268 Nitrones cycloaddition to 269 Nucleophilic addition - to carbonyl compounds l00ff. - chelate model of 104 - cyclic model of 104 - dipolar model of 104 - Felkin-Anh model of 107 - to cyclic ketones 107 -

Olefins - addition of chlorocarbenes to 21 - codimerization of 251 - hydrogenation of 48 ff. - a-substituted, by cross coupling 250 Oppolzer’s sultam 202, 218, 275, 267, 305 Optical - purity 18 - rotation, sign of 6 Oxaborolidines 90 Oxathianes, chiral 144 Oxaziridine - decomposition of IS Oxazolidines - chiral 145 - cytcloaddition to 273 Oxazolines - chiral 40 - in Diels-Alder reaction 262 - from isonitriles 210 - unsaturated, addition to 224 Oxetanes - by photoaddition 33 Oxidation - of enolates 29 - of sulfides 126, 133 N-Oxides 15 Oximes - reduction of 89, 91

P 3 Paterno-Buchi reaction 270 Percamphoric acid 123 Pericyclic reactions 257 ff. Phase transfer catalysis 88 8-Phenylmenthol 148, 218, 259, 286, 291 Phosphate esters 40

Subject Index Phosphines, chiral 45 - in cross-coupling 249 - in olefin co-dimerization 251 Phosphonic acids, chiral, by alkylation 243 Photoaddition, to olefins 270, 273 Photochemical transformations 27 Photoisomerization of cyclooctenes 28 Phthalides 152 Pinacol cross coupling 119 Piperidines, by hydrogenation 65 PPFA 48, 51, 72 Prelog-Seebach notation 21, 22 Prelog’s rule 32, 102, 1 1 8, 147 Pressure effects, in hydrogenation 54, 63 pro-R 10 pro-S 10 Prochirality - descriptor 10 Product - development control 108, 109 - ratio 17 - selectivity 1 1 , 17 Projection, Fischer 5 PROPHOS 52, 53 Propionic acids, 3,3-disubstituted, preparation of 220 Protonation 306f. - of enamines 306 - of enolates 306 Pyrrolidines, by hydrogenation 65 o-Quinodimethanes, in Diels-Alder reaction 267

R 3 Racemic chiral auxiliary 3 1 Racemization 38 - during cleavage of chiral auxiliary - of diastereomers 39 Reagents - chiral 14 - optically impure 31 Rearragement - allylic 294 - aza-Claisen 282 - Beckmann 12 - Claisen 38, 278f. - Cope 36, 278 - of allylamines 308 - of epoxides 306 - of sulfinates 287 - of sulfoxides 286 - thio-Claisen 282

30

- Wittig 36, 283ff. Reduction - aldehydes, deuterated 92 - catalytic 45 ff - cyclic ketones 1 1 9ff. - diastereoselective l00ff. - dihydropyridines 95 - enantioselective 8 1 ff. - esters, unsaturated 56, 89 - immonium salts 88 - Meerwein-Ponndorf-Verley 92 - of aldehydes 92 - of alkynyl ketones 83, 91 - of a-aminoketones 111 - of cyclic ketones 32, 34 - of P-hydroxyketones 116 - of a-ketoesters 96 - of P-ketols 117 - of ketones 91 - of ketonitriles 94 - of oximes 91 - of sulfoxides 111 - of sulfoximides 111 - with metal alkoxides 95 - with metal alkyls 94 - with NaBH, mechanism of 109 Reformatsky reaction 200 Resolution, kinetic 13, 15 Ruthenium catalysts, in hydrogenation 55

s 3 Selectivity - diastereomer I2 - enantiomer 13 - kinetic control of 24 - product 11, 17 - substrate 11 - thermodynamic control of 24 Self immolative processes 29 Sequence rules 6, 10 Sharpless epoxidation 125 f. Sharpless oxidation 15 Sigmatropic - hydrogen shift 287, 307 - rearrangements - [2,3] 283ff. - [3,3] 278ff. Silylketene acetals, addition of 208 Simmons-Smith reaction 275 Solid phase, racemic 18 Solid solutions 18 Solvent effects in hydrogenation 53

361

368

Subject Index

Stereodifferentiation 27 Stereoelectronic control 31 Stereoisomers - characterization of 4 - formation of 18 Stereoselective reactions - kinetics of 41 - thermodynamics of 41 Stereoselectivity 17 - conditions for 23ff. - group theory of 32 - inducing of 27 - of nucleophilic addition 32 - theory of 32 - role of conformation in 42 Stereospecific reactions 17 Steric approach control 108 Steric hindrance 4, 27, 32 Sterochemical models 41 Structure, chemical 11 Substrate selectivity 11 Sulfinates, rearrangement of 287 Sulfones, from sulfinates 286 Sulfoxides . - by oxidation 124, 127, 133 - by Sharpless oxidation 127 - chiral 152 - from sulfinyl esters 303 - from sulfinamides 304 - from thioxazolidines 304 - rearrangement of 286 - unsaturated, addition to 224 Sulfylimines, by oxidation 133 Symmetry - intramolecular 7 - statistical 9 syn 5 , 6 - addition of dihydrogen 78 - dihydroxylation 126

Synclinal 4, 5 Synthesis, enantioconvergent

36

TADDOL 137, 264 Temperature - effect in hydrogenation 55 - of isoinversion 32 Thermodynamic control 24, 30, 41 Thioenolates in Michael addition 215 threo 5, 6 Titanium alkyls, addition of 142f. Topicity 7 Torsional angles 4 Torsional strain in addition 151 trans 5 Transamination of ketoacids 79f. Transition state - boat type 175 - chair type 160 - cyclic 36 - diastereomeric 23, 25 - in isonitrile addition 210 - in sigmatropic rearrangements 278 - in Wittig rearrangement 283 - of aldol reaction 181f. Two-dimensional model of stereoselectivity 24,34 u 22 ul 21, 23 unlike 21

Van der Waals radii 4 Woodward-Hoffmann rules 258, 307

z

2, 5 Zimmerman-Traxler model 160 Zinc alkyls, addition of 137f.

Reflections on Symmetry I n Chemistry....and Elsewhere E. Heilbronnerl J. D.DunitL Copublished with Helvetica Chimica Acta Publishers, Basel 1992. VI, 154 pages with 125 figurcs and 4 tables. Ilardcover. DM 58;. ISBN 3-527-28488-5

What do Schriidinger's wave functions, Escher's intricate tiling patterns and chemistry's latcst marvel, buckminster-fullcrene, have in common? This stunning book, written by two world-famous scientists, introduces the reader to one of nature's most fundamental, and ubiquitous, phenomena: Symmetry. Enormous in scope, this scientific tour-de-force covers topics as diverse as quantum mechanics, stereochemistry, Penrose tilings and Raphael's Madonna. The authors connect these apparently unrelated fields in a completely non-mathematical fashion, which should appeal to thc general and spccialist reader alike. Lucidly written and beautifully illustrated, this book provides an eminently readable discussion of the omnipresent concept of symmetry, its application in all branches of science, and its pcrception in nature, art and elsewhere.

$6 [@@"

International Edition in English

,

kL@

tr0"

To start your personal subscription please contact: r Ser 11 6 ublic Telephone +49 (6201) 60 64 27 Fax +49 (6201) 60 61 17

VCH 8 Wellington Court Cambridge CB1 1HZ UK Fax (0223)31 33 21 (for customers in the UK)

VCH 220 East 23rd Street NewYork NY 10010-4606 USA Fax (212) 481 0897 (for customers in the USA and Canada)

VCH Hardstrasse 10 Postfach CH-4020 Basel Switzerland Fax (061) 2 71 06 18 (for customers in Switzerland)