手性相转移催化

Asymmetric Synthesis DOI: 10.1002/anie.200601737 Recent Advances in Asymmetric Phase-Transfer Catalysis Takashi Ooi and Keiji Maruoka* Angewandte Chemie Keywords: asymmetric synthesis · chiral onium salts · crown compounds · enantioselectivity · phase-transfercatalysis K. Maruoka and T. OoiReviews 4222 www.angewandte.org � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2007, 46, 4222 – 4266 1. Introduction In 1971, Starks introduced the term “phase-transfer catalysis” to explain the critical role of tetraalkylammonium or phosphonium salts (Q+X�) in the reactions between two substances located in different immiscible phases.[1] For example, the displacement reaction of 1-chlorooctane with aqueous sodium cyanide is accelerated many thousandfold by the addition of hexadecyltributylphosphonium bromide (1) as a phase-transfer catalyst (Scheme 1). Key to this tremendous enhancement in reactivity is the generation of a quaternary phosphonium cyanide, which makes the cyanide anion soluble in organic solvents and sufficiently nucleophilic. The high rate of displacement is mainly due to two of the three character- istic features of the pairing cation (Q+): high lipophilicity and the large ionic radius. Although it was not the first observation of the catalytic activity of quaternary onium salts,[2] the foundations of phase- transfer catalysis were laid by Starks together with Makosza and Br2ndstr3m in the mid to late 1960s. Since then, the chemical community has witnessed an exponential growth of phase-transfer catalysis as a practical methodology for organic synthesis. The advantages of this method are its simple experimental procedures, mild reaction conditions, inexpensive and environmentally benign reagents and sol- vents, and the possibility of conducting large-scale prepara- tions.[3] Nowadays, it appears to be the most important [*] Dr. T. Ooi,[+] Prof. K. Maruoka Department of Chemistry Graduate School of Science Kyoto University Sakyo, Kyoto 606–8502 (Japan) Fax: (+81)75-753-4041 E-mail: maruoka@kuchem.kyoto-u.ac.jp [+] Current address: Department of Applied Chemistry Graduate School of Engineering Nagoya University Chikusa, Nagoya 464-8603 (Japan) The use of chiral nonracemic onium salts and crown ethers as effective phase-transfer catalysts have been studied intensively primarily for enantioselective carbon–carbon or carbon– heteroatom bond-forming reactions under mild biphasic conditions. An essential issue for optimal asymmetric catalysis is the rational design of catalysts for targeted reaction, which allows generation of a well-defined chiral ion pair that reacts with electrophiles in a highly efficient and stereoselective manner. This concept, together with the synthetic versatility of phase-transfer catalysis, provides a reliable and general strategy for the practical asymmetric synthesis of highly valuable organic compounds. From the Contents 1. Introduction 4223 2. General Mechanism of Asymmetric Phase-Transfer Catalysis 4224 3. Alkylation 4225 4. Michael Addition 4251 5. Aldol and Related Reactions 4257 6. Darzens Reaction 4257 7. Neber Rearrangement 4258 8. Horner–Wadsworth–Emmons Reaction 4259 9. Cyclopropanation 4259 10. Epoxidation 4259 11. Aziridination 4261 12. Oxidation 4262 13. Reduction 4262 14. Fluorination 4263 15. Sulfenylation 4263 16. Cyanation 4263 17. Conclusions 4263 Scheme 1. Tetraalkylonium salts as phase-transfer catalysts. Phase-Transfer Catalysis Angewandte Chemie 4223Angew. Chem. Int. Ed. 2007, 46, 4222 – 4266 � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim synthetic method used in various fields of organic chemistry, and has also found widespread industrial applications. On the other hand, the development of asymmetric phase- transfer catalysis based on the use of structurally well-defined chiral, nonracemic catalysts has progressed rather slowly, despite its potential to create a new area of asymmetric catalysis by taking full advantage of structurally and stereo- chemically modifiable tetraalkylonium ions (Q+). However, recent efforts toward this direction have resulted in notable achievements, thus making it feasible to perform various bond-formation reactions under the mild conditions used in phase-transfer catalysis. This Review aims to illustrate the evolution of this active research field. Since several excellent reviews on this topic have been published,[4] the main focus will be on recent progress. It is the goal of this Review to provide a better understanding of the current situation and future perspectives of asymmetric phase-transfer catalysis. 2. General Mechanism of Asymmetric Phase- Transfer Catalysis Two representative reaction systems can be considered for phase-transfer-catalyzed bond formations using chiral cata- lysts. One involves the functionalization of active methylene or methine groups, typically under basic conditions. These reactions generally follow an interfacial mechanism.[5] Most of the successful asymmetric transformations under phase-trans- fer conditions belong to this category. The alkylation of an active methylene group, specifically the glycinate Schiff base 2,[4g,j,10a] is selected to illustrate the crucial parameters and key problems in such reactions. As depicted in Figure 1, the first step of the alkylation is the interfacial deprotonation of the a- proton of 2 with base (MOH) to give the corresponding metal enolate 3, which stays at the interface of the two layers. Subsequent ion-exchange of the anion with the catalyst (Q*+X�) generates a lipophilic chiral onium enolate 4. This step results in the enolate going deep into the organic phase, where it reacts with an alkyl halide to afford the optically active monoalkylation product 5 with concomitant regener- ation of the catalyst.[4f,10c,d] This type of reaction is only successful if the chiral onium cation (Q*+) can lead to the generation of highly reactive chiral onium enolate 4 through sufficiently fast ion-exchange and effective shielding of one of the two enantiotopic faces of the enolate anion. The former minimizes the intervention of the direct alkylation of metal enolate to give racemic 5, and the latter rigorously controls the absolute stereochemistry. An additional important issue to be considered is the effect of the strongly basic conditions, which could primarily cause decomposition of the catalyst, although hydrolysis of the substrate (ester and imine moi- eties), product racemization, and dialkylation could also be problematic. Such undesirable processes associated with the starting materials and products could be prevented by appropriate choice of protecting groups. In fact, the tert- butyl ester of 2 resists saponification, and the benzophenone imine moiety is essential not only for facilitating the initial deprotonation but also for leaving the remaining a-proton of 5 intact. In general, the type of phase-transfer system (liquid– liquid or solid–liquid) and other reaction variables (base, solvent, temperature, substrate concentration, and stirring rate) can be tuned to optimize the reactions. Another, relatively less-studied system is the nucleophilic addition of an organic or inorganic anion lacking a prochiral center to prochiral electrophiles. In these reactions an extraction mechanism is operative.[1] The anion is used as an aqueous solution or solid of its inorganic salt, and it is transferred into the organic phase as a chiral ion pair by ion- exchange with the catalyst. It then most commonly attacks a prochiral electrophile, and a new stereogenic center is created. The asymmetric epoxidation of a,b-unsaturated Takashi Ooi received his PhD (1994) from Nagoya University with Professor Hisashi Yamamoto, and was a postdoctoral fellow in the group of Professor Julius Rebek, Jr. at MIT (1994–1995). He was appointed as an assistant professor at Hokkaido University in 1995 and promoted to a lecturer in 1998. He moved to Kyoto University as an asso- ciate professor in 2001, and became a full professor of Nagoya University in 2006. He has been awarded the Chugai Award in Synthetic Organic Chemistry, Japan (1997), the Japan Chemical Society Award for Young Chemist (1999), and the Thieme Journal Award (2006). Keiji Maruoka received his PhD (1980) from the University of Hawaii with Prof. Hisashi Yamamoto. He then became an assistant professor at Nagoya University and an asso- ciate professor in 1990. He moved to Hok- kaido University as a full professor (1995), and since 2000 has been a professor at Kyoto University. His research interests focus on organic synthesis with bidentate Lewis acids and designer chiral organocatalysts. His awards include the Ichimura Prize for Science (2001), the Japan Synthetic Organic Chemistry Award (2003), Nagoya Silver Medal (2004), and the GSC award (2006). Figure 1. General mechanism for the asymmeric alkylation of active methylene compounds, with a glycine Schiff base used as an example. K. Maruoka and T. OoiReviews 4224 www.angewandte.org � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2007, 46, 4222 – 4266 ketones using an aqueous solution of sodium hypochlorite represents a typical example (Figure 2). The chiral onium hypochlorite (Q*+OCl�) is responsible for the enantiofacial discrimination of the prochiral enone 6.[141,142] The pH value of the reaction is nearly neutral, so the possibility of side reactions is considerably reduced. However, an even more precise catalyst design seems to be required because the chiral cation (Q*+) should recognize the enantiotopic faces of the electrophilic reacting partner. 3. Alkylation 3.1. Pioneering Studies The enantioselective alkylation of active methylene com- pounds occupies the central position in the field of asym- metric phase-transfer catalysis, and its development was triggered by the pioneering study by a Merck research group in 1984.[6] Dolling and co-workers utilized the cincho- nine-derived quaternary ammonium salt 8a as the catalyst for the methylation of phenylindanone derivative 9a under liquid–liquid phase-transfer conditions (toluene/50% aq NaOH solution) and succeeded in obtaining the correspond- ing alkylated product 10a in excellent yield and high enantiomeric excess (Scheme 2). The authors made system- atic studies of this reaction, and proposed the tight ion pair intermediate 11, formed through hydrogen bonding as well as electrostatic and p–p stacking interactions, to account for the result. The effectiveness of the catalysis was also demon- strated in the reaction of a-propyl analogue 9b with 1,3- dichloro-2-butene.[7] Diederich and Ducry synthesized a series of diastereo- meric chiral quaternary ammonium bromides 12a–d which incorporated the quinuclidinemethanol fragment of cinchona alkaloids and a 1,1’-binaphthyl moiety. The ability of these compounds to function as phase-transfer catalysts in the asymmetric allylation of 9a under similar conditions were evaluated. These studies revealed that, without any optimi- zation, 12a was superior to the other three diastereomeric catalysts 12b–d, although both the chemical yield and enantiomeric excess of 10c were unsatisfactory (Scheme 2).[8] This phase-transfer-catalyzed alkylation strategy was successfully applied to the asymmetric cyanomethylation of oxindole 13 by the use of catalyst 8b with a 3,4-dichlorophe- nylmethyl group appended on the nitrogen atom. This reaction allowed a simple and stereoselective synthesis of (�)-esermethole (15), a precursor to the clinically useful anticholinesterase agent (�)-physostigmine (Scheme 3).[9] 3.2. Asymmetric Synthesis of a-Amino Acids and their Derivatives 3.2.1. Monoalkylation of Schiff Bases Derived from Glycine In 1989, five years after the pioneering work by the Merck research group, this type of catalyst was successfully utilized for the asymmetric synthesis of a-amino acids by OEDonnell et al. , who used glycinate Schiff base 2 as a key substrate.[10] Figure 2. General mechanism for the nucleophilic addition of anions to prochiral electrophiles, with the asymmetric epoxidation of a,b- unsaturated ketones used as an example. Scheme 2. Asymmetric phase-transfer-catalyzed alkylation of indanone derivatives. Scheme 3. Asymmetric cyanomethylation of oxindole 13 as a step in the synthesis of (�)-esermethole (15). Phase-Transfer Catalysis Angewandte Chemie 4225Angew. Chem. Int. Ed. 2007, 46, 4222 – 4266 � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org The asymmetric alkylation of 2 proceeded smoothly under mild phase-transfer conditions, with N-(benzyl)cinchoninium chloride (8c) as a catalyst, to give the alkylation product (R)- 16 in good yield and moderate enantioselectivity (Scheme 4). By simply switching to the cinchonidine-derived catalyst 17a, the product could be obtained with the opposite absolute configuration (S) but with a similar degree of enantioselec- tivity. Further optimization with the hydroxy-protected catalyst 17b (second-generation catalyst) enhanced the enantioselectivity to 81% ee.[10c,11] A single recrystallization and subsequent deprotection of 16 afforded essentially optically pure a-amino acids. An important aspect of this reaction is the selective formation of the monoalkylated product 16, without con- comitant production of the undesired dialkylated product, provided the Schiff base of benzophenone is employed as the starting material.[12] This effect results from the much lower acidity of the remaining a-proton of 16 (compared to that of 2). This reduced acidity is also crucial for securing the configurational stability of the newly created a-stereogenic center under the reaction conditions. In fact, exposure of optically pure Schiff base (S)-18 to typical alkylation con- ditions without alkyl halide did not cause racemization regardless of the addition of phase-transfer catalyst (Bu4NBr; Scheme 5). Interestingly, however, a similar prod- uct racemization experiment in the presence of 17a showed the formation of 35% of (R)-18 in two hours, and then no further racemization. Moreover, no racemization was detected if an alkyl halide such as benzyl bromide was present during the reaction with 17a. These results suggested that the racemization of (S)-18 was controlled by the organic soluble ammonium alkoxide, and its in situ benzylation generated the ammonium bromide 17c, a possible active catalyst in the asymmetric phase-transfer-catalyzed alkylation of 2.[11] Although asymmetric phase-transfer alkylation of the glycinate Schiff base 2 can be achieved by using chiral phase- transfer catalysts derived from the relatively inexpensive, commercially available cinchona alkaloids, research in this area was slow. However, a new class of cinchona alkaloid derived catalysts bearing an N-anthracenylmethyl group (third-generation catalyst) developed by two independent research groups have opened up a new era of asymmetric phase-transfer catalysis. In 1997, Lygo et al. developed the N- anthracenylmethylammonium salts 8d and 17d, and applied them to the asymmetric phase-transfer alkylation of 2 to synthesize a-amino acids with much higher enantioselectivity (Scheme 6).[13] Scheme 4. Asymmetric synthesis of a-amino acids from glycine derivative 2 by phase-transfer catalysis. Scheme 5. Racemization experiments on (S)-18. Scheme 6. The third-generation catalysts developed by Lygo et al. K. Maruoka and T. OoiReviews 4226 www.angewandte.org � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2007, 46, 4222 – 4266 At the same time Corey et al. prepared O-allyl-N- anthracenylmethyl cinchonidinium salt 17e. By using solid cesium hydroxide monohydrate (CsOH·H2O) at very low temperature, they achieved a high asymmetric induction in the enantioselective alkylation of 2 (Scheme 7). The catalyst was characterized by X-ray analysis of O-allyl-N-anthrace- nylmethylcinchonidium p-nitrophenoxide.[14] Recently, Lygo et al. demonstrated that this type of chiral quaternary ammonium salt can be generated in situ and directly used as a catalyst for the subsequent asymmetric alkylation of 2. For example, treatment of dihydrocinchonine (19) with 9-bromomethylanthracene in toluene at 60–75 8C for 5 h followed by the addition of 2, benzyl bromide, and aqueous KOH at room temperature and continuous stirring for 18 h afforded the desired protected phenylalanine 18 with 93% ee (Scheme 8).[15] The observed enantioselectivity was comparable to that obtained with pre-prepared catalyst. This approach is likely to be useful for the identification of optimal catalyst structures for a given asymmetric transformation. In 1999 we prepared the structurally rigid, chiral spi- roammonium salts of type 20, derived from commercially available (S)- or (R)-1,1’-bi-2-naphthol, as a new C2-symmet- ric phase-transfer catalyst and successfully applied them to the highly efficient, catalytic enantioselective alkylation of 2 under mild phase-transfer conditions.[16] The key finding was a significant effect of an aromatic substituent at the 3,3’-position of one binaphthyl subunit of the catalyst (Ar) on the enantiofacial discrimination. (S,S)- 20e proved to be the catalyst of choice for the preparation of a variety of essentially enantiopure a-amino acids by this transformation (Table 1). In general, 1 mol% of 20e is sufficient for the smooth alkylation, and the catalyst loading can be reduced to 0.2 mol% without loss of enantiomeric excess (entry 6). The use of aqueous cesium hydroxide (CsOH) as a basic phase at lower reaction temperature is recommended for the reaction with simple alkyl halides such as ethyl iodide (entry 7). Since both enantiomers of the catalyst of type 20 can be readily assembled in exactly the same manner starting from either (S)- or (R)-1,1’-bi-2-naphthol, a wide variety of natural and unnatural a-amino acids can be synthesized in an enantiomerically pure form by the phase-transfer-catalyzed alkylation of 2. The salient feature of 20e is its ability to catalyze the asymmetric alkylation of glycine methyl ester and ethyl ester derivatives 21 and 22 with excellent enantioselectivities. Since methyl and ethyl esters are certainly more susceptible toward nucleophilic additions than tert-butyl esters, the synthetic Scheme 7. The third-generation catalysts developed by Corey et al. Scheme 8. In situ generation of chiral phase-transfer catalysts. Table 1: Effect of aromatic substituents (Ar) and general applicability of 20e for the phase-transfer-catalyzed alkylation of 2. Entry Catalyst RX Yield [%] ee [%] 1 20a PhCH2Br 73 79 (R) 2 20b PhCH2Br 81 89 (R) 3 20c PhCH2Br 95 96 (R) 4 20d PhCH2Br 91 98 (R) 5 20e PhCH2Br 90 99 (R) 6[a] 20e PhCH2Br 72 99 (R) 7[b] 20e EtI 89 98 (R) 8 20e 80 99 (R) 9 20e 98 99 (R) 10 20e 86 98 (R) [a] With 0.2 mol% of (S,S)-20e. [b] With saturated CsOH at �15 8C. Phase-Transfer Catalysis Angewandte Chemie 4227Angew. Chem. Int. Ed. 2007, 46, 4222 – 4266 � 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org advantage of this process is clear, as highlighted by the facile transformation of the alkylation products (Scheme 9).[17] A similar electronic effect of fluoroaromatic substituents was utilized by Jew, Park, and co-workers for the develop- ment of efficient catalysts derived from cinchona alkaloids. Evaluation of the effect of electron-withdrawing groups on the benzylic group of dihydrocinchonidinium salt 26 revealed that an ortho-fluoro substituent on the aromatic ring led to dramatic enhancement of the enantioselectivity. Catalyst 26e with a 2’,3’,4’-trifluorobenzyl group showed the highest selectivity in the transformation of a variety of alkyl halides (Scheme 10).[18] It has been proposed that a hydrogen-bonding interaction between the oxygen atom at C9 and the fluorine atom at C2’ in 26b might rigidify its conformation, thus leading to high enantioselectivity. Recent st