Chapter One
CATALYTIC ENANTIOSELECTIVE ALDOL ADDITION REACTIONS
Erick M. Carreira, Alec Fettes, and Christiane Marti
Laboratory of Organic Chemistry, Swiss Federal Institute of Technology (ETH-Z), HCI H337 ETH-Hnggerberg, CH 8093 Zrich, Switzerland
CONTENTS
Page Introduction 2 Background 6 Mechanism and Stereochemistry 8 General Aspects 8 Catalyst Turnover 11 Scope and Limitations 15 Acetate, Methyl Ketone, and Unsubstituted Dienolates 15 Enoxysilanes and Stannanes 15 Direct Aldol Addition Reactions of Unsubstituted Systems 28 Propionates and Substituted Enolates 30 Enoxysilanes and Stannanes 30 Direct Aldol Addition Reactions of Substituted Systems 48 Enoxysilanes of Acetoacetate and Furan 52 Domino Reactions Involving Aldol Additions 58 Comparison With Other Methods 61 Experimental Conditions 70 Experimental Procedures 71 Phenyl (2R,3S)-2-(Benzyloxy)-5-(tert-butyldimethylsiloxy)-3-hydroxypentanoate 71 Phenyl (2S,3S)-3-Hydroxy-2-methyl-3-(4-methoxyphenyl)propanoate 72 (R)-1-Hydroxy-1-phenyl-3-heptanone 73 Methyl (R)-3-Hydroxy-8-phenyloct-4-ynoate 73 (R)-S-tert-Butyl 4-Benzyloxy-3-hydroxybutanethioate 74 (S)-4-Hydroxy-4-phenyl-2-butanone 75 (R)-4,4-Dimethyl-1-hydroxy-1-phenyl-3-pentanone 75 (S)-3-Hydroxy-4-methyl-1-(3-nitrophenyl)-1-pentanone 76 (R)-3-Cyclohexyl-3-hydroxy-1-phenylpropan-1-one 77 (R)-4-Hydroxy-5-methylhexan-2-one 77 S-tert-Butyl (3S)-3-Hydroxy-3-methoxycarbonylbutanethioate 77 (2S,1'R)-2-(Hydroxyphenylmethyl)cyclohexanone 78 (4S,5R)-4-(Methoxycarbonyl)-5-phenyl-2-oxazoline 79 (2R,3S)-2,3-Dihydroxy-2,3-O-isopropylidene-1-(2-methoxyphenyl)-5-phenyl-1-pentanone 80 (3S,4S)-4-Cyclohexyl-3,4-dihydroxybutan-2-one 80 (R)-6-(2-Hydroxy-4-phenylbut-3-enyl)-2,2-dimethyl-[1,3]-dioxin-4-one 81 (2S,3S)-Methyl 4-Benzyloxy-3-hydroxy-2-methylbutanoate 82 Tabular Survey 82 Table 1A. Silver-Catalyzed Mukaiyama-Type Aldol Additions 84 Table 1B. Boron-Catalyzed Mukaiyama-Type Aldol Additions 88 Table 1C. Copper-Catalyzed Mukaiyama-Type Aldol Additions 106 Table 1D. Tin-Catalyzed Mukaiyama-Type Aldol Additions 117 Table 1E. Titanium-Catalyzed Mukaiyama-Type Aldol Additions 124 Table 1F. Other Metal-Catalyzed Mukaiyama-Type Aldol Additions 139 Table 1G. Non-Metal-Catalyzed Mukaiyama-Type Aldol Additions 150 Table 2A. Gold-Catalyzed Aldol Additions of Unmodified Nucleophiles 164 Table 2B. Non-Gold Metal-Catalyzed Aldol Additions of Unmodified Nucleophiles 175 Table 2C. Non-Metal-Catalyzed Aldol Additions of Unmodified Nucleophiles 189 Table 3. Aldol Tandem Reactions 204 References 208
INTRODUCTION
The aldol addition reaction has become a strategically important, reliable transformation that is widely employed in the asymmetric synthesis of complex molecules (Eq. 1). It can be counted upon not only to provide access to polyketide fragments with their characteristic 1,3-oxygenation pattern, but also to numerous other classes of compounds, such as oxo-heterocycles, [alpha]- and [beta]-amino acids, and nucleosides. Indeed, its numerous applications in synthesis attest to the versatile role of this reaction in the pantheon of organic transformations.
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 1)
Two general approaches have emerged for the asymmetric aldol reaction: (1) diastereoselective additions, wherein stoichiometric quantities of a covalently bound, chiral controlling element shepherds the stereochemical course of the reaction, and, alternatively, (2) enantioselective methods, wherein a chiral catalyst functions as the stereochemical controlling element, preferably in substoichiometric amounts.
Given the intensive research activities in this area by numerous groups over the last three decades, diastereoselective methods remain dominant in practice to date with respect to the frequency of use. This is hardly surprising as such methods have enjoyed a long period of intense scrutiny and possess a high degree of predictability and reliability. By comparison, useful catalytic, enantioselective methods are fairly recent arrivals on the scene. Nonetheless, explosive developments in the field are beginning to provide the practitioner with additional useful tools for complex molecule assembly via catalytic asymmetric aldol transformations. In this respect, it is important to note that enantioselective asymmetric, catalytic aldol addition methods can furnish ketide fragments such as polyacetate or unsubstituted skipped polyol fragments that have not otherwise been conveniently accessible through more traditional diastereoselective approaches.
Enantioselective aldol methods use small-molecule (often termed chemical catalysts) as well as macromolecule catalysis (commonly referred to as biological catalysts, such as enzymes or antibodies). The origins of the differentiation between chemical or biological is likely historical. It is debatable whether such a categorization is at all reasonable, because the only obvious differentiation between catalysts of either group is molecular weight or, correspondingly, size, properties that vary continually. Moreover, the advances in mutagenesis techniques, library synthesis, and high-throughput screening methods for the generation of non-natural catalysts in both classes serve to further blur the distinction between bio- and abiological catalysis. It is certainly the case that members of either group share many more features in common than an artificial designation would suggest. Thus examples of catalytic processes in both types can be found that are metal-mediated as well as metal-free, operate in organic or aqueous media, and exhibit both high and low efficiencies.
The early work with isocyanoacetates and ferrocenyl bis-phosphine 1 established that catalytic enantioselective aldol additions could be carried out with in-situ generation of an enolate (Eq. 2), and the commonly held distinction that only
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] Eq. (2)
biocatalytic aldol additions would be tolerant of nucleophilic substrate partners possessing polar unprotected groups (e.g., hydroxyls) is no longer tenable (Eq. 3). The two approaches (biological and chemical) are complementary in scope. Certain classes of reactions are at present best effected with macromolecular catalysts, whereas others are best effected with small-molecule catalysts such as 2-9 (Eqs. 3-13). Given the immense breadth of the field, a compilation of catalytic methods for asymmetric aldol reactions is best when it is focused. Consequently, the coverage in this chapter is limited to catalytic enantioselective aldol addition methods using small-molecule catalysts, wherein the products in general
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 3)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 4)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 5)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 6)
retain [beta]-hydroxycarbonyl functionality. For historical reasons there are three exceptions to such boundary conditions on the topical discussion, namely the aldol addition reactions leading to oxazoline products (Eq. 2), aldehyde ene addition reactions of enol silanes, and domino processes (cf Eq. 5). In the aldol reactions that form the basis of this compilation, catalytic turnover is a necessary requirement, with reactions in which the turnover number is unity excluded, as these are considered best classified with processes that rely on the use of chiral auxiliaries or stoichiometric controlling groups or additives.
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 7)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 8)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 9)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 10)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 11)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 12)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 13)
Background
The use of catalysts to channel the stereochemical outcome of the aldol addition reaction process is feasible because the aldol addition reaction is overall an atom-transfer reaction (Fig. 1). In the reactions involving enolate additions to aldehydes, an O-Si or C-Sn bond in the starting material is exchanged for another O-Si or O-Sn bond in the product with concomitant trade of the C=O bond in the electrophile and C=C in the enolate component for a C-C bond and a new C=O in the adduct. In aldol additions involving direct addition of enolizable carbonyls, the C-H and C=O bonds in the educts are exchanged for C-C and O-H bonds in the adduct.
Although the larger number of catalytic enantioselective aldol processes involve the use of enoxysilanes, the direct addition of enolizable ketones and esters to aldehydes and ketones in aldol additions have been documented as well. The classic in this respect is rooted in the proline-catalyzed Robinson annulation reaction (Eq. 14)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 14)
for the preparation of the Wieland-Miescher ketone and the direct addition of isocyanoacetates to aldehydes to give isoxazolines catalyzed by gold complexes prepared from 1 (Eq. 2). More recently, impressive advances have been made in catalytic, enantioselective aldol reactions involving the direct addition of enolizable carbonyls to aldehydes (Eqs. 15-18), mediated by chiral metal phenoxides such as 10 or alkoxides (11) as well as by amines, such as proline or derivatives like 11.
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 15)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 16)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 17)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Eq. 18)
MECHANISM AND STEREOCHEMISTRY
General Aspects. The diastereoselective aldol addition reactions that are best understood and provide adducts with a high degree of stereochemical control and predictability are those proceeding from a metal enolate through closed, Zimmerman-Traxler-like transition states, such as those derived from alkali or alkaline earth metals or boron. By contrast, catalytic enantioselective counterparts lack thorough mechanistic understanding. Earlier analyses suggesting open transition states (Fig. 2, A-C), despite their historical importance and utility, are likely oversimplifications, because at the very least they tend to ignore the fate of the silyl group in the course of C=O addition and C-C bond formation. Given the inherent high energy of a trialkylsilyl cation, its formation in the course of the aldol process is unlikely. In this respect, it is interesting to note that more recent models (D-H) take into account the silicon in the transition state. Indeed, focusing on the fate of the silyl group can lead to useful new catalytic processes.
Under typical conditions, most enoxysilanes are unreactive toward aldehydes, notable exceptions being silyl enolates derived from amides, enoxy-silacyclobutanes, and trihalosilyl enolates. The typical lack of reactivity for the parent trialkylsilyl enolates is critical, as it ensures that the uncatalyzed background aldol additions are precluded. The enoxytrihalosilanes represent an interesting exception. Thus, although their reactions with aldehydes are fast even at low temperature, the Lewis base catalyzed processes are even faster.
In the simplest analysis, the enoxysilanes and aldehydes can be induced to react by either of two fundamentally different mechanisms: electrophilic activation of the C=O component or nucleophilic activation of the enolate or enol surrogate (Fig. 3). In reality, it is likely that mechanistic pathways form a continuum of possibilities and possess some component of both activation modes. Recent reports involving the use of enolizable carbonyl substrates that undergo in situ conversion to the corresponding enolate or enamine provide new mechanistic options for this transformation. Moreover, these add to the class of traditional closed cyclic transition-state arrangements found in the aldol addition reactions such as those of alkali, alkaline earth, and boron enolates.
Close inspection of the few cases that have been studied in some detail reveals that the mechanistic possibilities are as numerous as the catalysts and conditions that have been described for this venerable reaction. Given the complexity of the reaction, there are a number of important issues that can be examined in the process: substrate binding and activation (C=O electrophile, enolate nucleophile, or both), C=O face differentiation, as well as catalyst turnover.
(Continues...)
Excerpted from Organic Reactions, Volume 67 Copyright © 2006 by Organic Reactions, Inc.. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
