New Methods for Chiral Cyanohydrin Synthesis

Erica Wingstrand

Doctoral Thesis

Stockholm 2009

Akademisk avhandling som med tillstånd av Kungl Tekniska Högskolan i Stockholm framlägges till offentlig granskning för avläggande av filosofie doktorsexamen i kemi med inriktning mot organisk kemi fredagen den 8 maj kl 10.00 i sal F3, KTH, Lindstedtsvägen 26, Stockholm. Avhandlingen försvaras på engelska. Opponent är Professor Benjamin List, Max-Planck-Institut für Kohlenforschung, Tyskland.

ISBN 978-91-7415-263-0 ISSN 1654-1081 TRITA-CHE-Report 2009:12

© Erica Wingstrand, 2009 E-PRINT AB, Stockholm Erica Wingstrand, 2009: ”New Methods for Chiral Cyanohydrin Synthesis”, Organic Chemistry, KTH Chemical Science and Engineering, Royal Institute of Technology, SE-100 44 Stockholm, Sweden.

Abstract

This thesis deals with method development in asymmetric catalysis and specifically syntheses of enantioenriched O-functionalized cyanohydrins.

The first part describes the development of a method for the synthesis of O-alkoxycarbonylated and O-acylated cyanohydrins. Ethyl cyanoformate and acyl cyanides were added to aldehydes in a reaction catalyzed by a chiral dimeric Ti-salen complex together with a tertiary amine. High yields and enantioselectivities were in most cases obtained. Mechanistic studies were performed and a reaction mechanism was proposed.

The second part describes a method in which the undesired minor enantiomer in a Lewis acid–Lewis base-catalyzed acylcyanation is continuously recycled into prochiral starting material. Close to enantiopure O-acylated cyanohydrins were obtained in high yields.

The third part deals with asymmetric acylcyanations of ketones. Acetyl cyanide was found to add to α-ketoesters in a reaction catalyzed by a chiral Lewis base. Yields up to 77% and 82% ee were obtained.

The final part describes an enzymatic method for high-throughput analysis of O-acylated cyanohydrins. The enantiomeric excess and conversion were determined for products obtained from a number of aromatic and aliphatic aldehydes.

Keywords: acyl cyanides, asymmetric catalysis, biocatalysis, cyanide, cyanohydrins, dual activation, enzymes, high-throughput screening, Lewis acid, Lewis base, metal catalysis, minor enantiomer recycling, salen, titanium.

Abbreviations

Abbreviations and acronyms used in agreement with the ACS standards1 are not listed here.

BINOL 1,1-bi-2-naphthol (DHQD)2AQN hydroquinidine (anthraquinone-1,4-diyl) diether DIPEA N,N-diisopropylethylamine E electrophile ee enantiomeric excess EMDee enzymatic method for determining ee LA Lewis acid LB Lewis base M metal MER minor enantiomer recycling n.d. not determined PyBOX pyridine bis(oxazoline) r.d.s. rate determining step TMSCN trimethylsilyl cyanide

List of Publications

This thesis is based on the following papers, referred to in the text by their Roman numerals I-VI:

I. Dual Lewis Acid–Lewis Base Activation in Enantioselective Cyanation of Aldehydes Using Acetyl Cyanide and Cyanoformate as Cyanide Sources Stina Lundgren, Erica Wingstrand, Maël Penhoat, and Christina Moberg J. Am. Chem. Soc. 2005, 127, 11592-11593

II. Lewis Acid–Lewis Base-Catalysed Enantioselective Addition of α- Ketonitriles to Aldehydes Stina Lundgren, Erica Wingstrand, and Christina Moberg Adv. Synth. Catal. 2007, 349, 364-372

III. O-Acylated Cyanohydrins: Versatile Intermediates and Products Erica Wingstrand, Fei Li, Stina Lundgren, Maël Penhoat, and Christina Moberg Chim. Oggi, 2007, 25 (Suppl.), 14-15

IV. Minor Enantiomer Recycling – Metal Catalyst and Biocatalyst Working in Concert Erica Wingstrand, Linda Fransson, Anna Laurell, Karl Hult, and Christina Moberg Manuscript

V. Chiral Lewis Base-Catalyzed Enantioselective Acetylcyanation of α-Ketoesters Fei Li, Erica Wingstrand, Khalid Widyan and Christina Moberg Eur. J. Org. Chem. provisionally accepted

VI. High-Throughput Synthesis and Analysis of Acylated Cyanohydrins Anders Hamberg, Stina Lundgren, Erica Wingstrand, Christina Moberg and Karl Hult Chem. Eur. J. 2007, 13, 4334-4341

Table of Contents

Abstract Abbreviations List of publications

1. Introduction ...... 1 1.1. Aim of the Thesis ...... 2 1.2. Chiral Cyanohydrins ...... 2 1.2.1. Cyanide Sources ...... 3 2. Enantioselective Cyanation of Aldehydes Using Acyl Cyanides and Cyanoformate as Cyanide Sources ...... 5 2.1. Introduction ...... 5 2.1.1. Catalysts for Asymmetric Cyanation of Aldehydes ...... 5 2.1.2. Salen-Based Catalysts ...... 7 2.1.3. Dual Lewis Acid–Lewis Base Activation ...... 8 2.1.4. Aim of the Study ...... 9 2.2. Synthesis of O-Alkoxycarbonyl functionalized Cyanohydrins ...... 9 2.3. Synthesis of O-Acylated Cyanohydrins ...... 12 2.4. Synthesis of a Herbicide Intermediate ...... 16 2.5. Mechanistic Aspects ...... 17 2.6. Conclusions ...... 22 3. Minor Enantiomer Recycling in Enantioselective Cyanohydrin Synthesis ...... 23 3.1. Introduction ...... 23 3.1.1. Methods for Transforming a Racemate into a Single Enantiomer ...... 23 3.1.2. Aim of the Study ...... 25 3.2. Minor Enantiomer Recycling (MER) ...... 26 3.2.1. Development of the MER System ...... 26 3.2.2. Use of MER in O-Acylated Cyanohydrin Synthesis ...... 28 3.2.3. Modifications and Improvements ...... 29 3.3. Conclusions and Outlook ...... 30 4. Enantioselective Addition of Acyl Cyanides to Ketones ...... 31 4.1. Introduction ...... 31 4.1.1. Catalysts for Asymmetric Cyanation of Ketones ...... 31 4.1.2. Aim of the Study ...... 33 4.2. Chiral Lewis Base-Catalyzed Addition of Acetyl Cyanide to α-Ketoesters .... 33 4.2.1. Mechanistic Aspects ...... 35 4.3. Acylcyanations of Methyl Ketones ...... 37 4.4. Conclusions ...... 38 5. Enzymatic Method for High-Throughput Analysis of O-Acylated Cyanohydrins ...... 39 5.1. Introduction ...... 39 5.1.1. High-Throughput Methods for Determination of Enantiomeric Excess ...... 39 5.1.2. Aim of the Study ...... 40 5.2. Determination of Enantiomeric Excess Using an Enzymatic Method with a pH Indicator ...... 40 5.3. Determination of Enantiomeric Excess and Conversion Using an Enzymatic Method with NADH ...... 42 5.4. Conclusions ...... 44 6. Summary and Concluding Remarks ...... 45 7. Acknowledgements ...... 47 8. References ...... 49

1. Introduction

The demand for chiral compounds in enantiopure form has increased during the past decades. The pharmaceutical industry has been the main driving force, but also industries producing flavorings, agricultural chemicals, fragrances, fine chemicals, and materials.2 In nature, chiral compounds are synthesized in highly enantioenriched form in reactions often run in a catalytic mode, wherein material consumption, waste and energy are minimized.3 For organic chemists, nature therefore has been and still is a source of great inspiration.

Asymmetric catalysis is an excellent tool for producing chiral compounds synthetically. Only a small amount of a chiral catalyst is needed to promote the formation of a large amount chiral product in enantioenriched form, in the same way as in nature.4 Numerous asymmetric methods using chiral metal catalysts, nature’s own enzymes or mutants thereof, catalytic antibodies, and organocatalysts have been developed.

Several asymmetric catalytic methods are employed in industry and more will be utilized, but other techniques for preparation of enantioenriched compounds are still used more. The unwillingness from industry to change to new catalytic methods is due to many factors; first of all to high expenses related to introducing new techniques, but sometimes also due to inadequate catalysts. Functional group tolerance, enantioselectivity, availability, price, scaling, toxicity, stability, and activity are parameters which at times can be problematic. These aspects need to be considered before a catalyst reaches industrial applications.5 Therefore, new and improved methods are still desired to meet the needs from industry and to simplify the synthesis of highly enantioenriched compounds in general.

1.1. Aim of the Thesis The aim of this thesis was to develop new tools and methods in asymmetric catalysis with applications in chiral cyanohydrin synthesis. The goal was to investigate new types of cyanide sources, develop efficient catalytic systems, and fast analytical methods. To increase the knowledge of the reactions, their mechanisms were to be examined.

1.2. Chiral Cyanohydrins The addition of cyanide to carbonyl compounds to form cyanohydrins6 (Scheme 1) was first reported by Winkler as early as 1832.7 Winkler synthesized mandelonitrile (2-hydroxyphenylacetonitrile) by adding hydrogen cyanide, which had been discovered by Scheele in 1782,8 to benzaldehyde. Only five years later the reverse reaction, although less synthetically useful, was found to be catalyzed by an oxynitrilase enzyme, which was identified by Wöhler in almonds.9 Seventy years later, Lapworth demonstrated that the actual nucleophile in cyanohydrin synthesis is the cyanide ion and that the reaction rate can be increased by the addition of base.10 Another 70 years later, Ching and Kallen showed that the reaction is reversible, with the position of the equilibrium being determined by a combination of electronic and steric factors.11

Scheme 1. Cyanohydrin formation.

O HO CN + HCN R R' R R'

Chiral cyanohydrins can be found in nature in the form of glycosides6b-d,12 and they can act as antifeedants in more than 2500 plants, fungi and insects. The first asymmetric synthesis of a cyanohydrin, and one of the first asymmetric routes known in organic synthesis, was reported in 1908 by Rosenthaler who in an enzyme-catalyzed reaction used emulsin (extract from almonds) as the source of the enzymatic catalyst.13 Emulsin contains both glycosidase and oxynitrilase and its natural reaction is to break down the glycoside amygdalin to mandelonitrile, which is further degraded to benzaldehyde and hydrogen cyanide (Scheme 2). By controlling the reaction conditions Rosenthaler managed to instead run the reaction in reverse and add hydrogen cyanide to benzaldehyde producing enantioenriched mandelonitrile.

Scheme 2. The breakdown of amygdalin occurring in almonds.

OH O H CN HO 2 HO OH OH HCN O Ph OH OH O O O HO OH HO O OH glycosidase Ph CN oxynitrilase Ph H OH HO

Several synthetic applications of chiral cyanohydrins are known, as their hydroxy and nitrile groups can be transformed into a large number of functional groups under conditions that preserve high optical purity.6 Primary14 and secondary15 β-hydroxyamines, α-hydroxyacids,16 2,3-substituted piperidines,17 α-aminonitriles,18 and α-fluronitriles19 are only a few examples, many of which are used in further stereoselective reactions (Scheme 3). Chiral cyanohydrins are not only useful synthons, but are also found in different O- protected forms in several commonly used pyrethroid insecticides.20

Scheme 3. Example of functional groups that can be synthesized from cyanohydrins.

OH OH F H N R CN R R1 R CN

NH OH 2 OH R CN R OH O n

N R1 Cbz

1.2.1. Cyanide Sources Despite the many applications of chiral cyanohydrins and their long history, it was not until the past decades that they have become intensively studied. This might be due to the difficulties in handling and the high toxicity of hydrogen cyanide, which has been the most frequently used cyanide source. In 1973 trimethylsilyl cyanide (TMSCN) was added for the first time to aldehydes and ketones to give the corresponding TMS-protected cyanohydrins21 and since then TMSCN has replaced hydrogen cyanide as the cyanide source of choice in research labs.6f TMSCN has many advantages in comparison with hydrogen cyanide since it is easy to handle and the resulting TMS ether prevents racemization of the cyanohydrin (Scheme 1). Despite these advantages, some serious disadvantages still remain with TMSCN such as its high toxicity, its flammability, and its high price. Therefore several alternative cyanide sources have been explored,22 such as metal cyanide salts,23 acetone cyanohydrin,24

3 cyanoformates25 and cyanophosphonates,26 but there was still need for new, cheap alternatives (Figure 1).

HO CN O O HCN TMS CN MCN P RO CN RO CN OR Figure 1. Cyanide sources used in cyanohydrin synthesis.

2. Enantioselective Cyanation of Aldehydes Using Acyl Cyanides and Cyanoformate as Cyanide Sources

(Papers I-III)

2.1. Introduction The asymmetric addition of cyanide to a prochiral aldehyde is a powerful reaction for carbon-carbon bond formation. The process opens up several opportunities for further chemical transformations. It is, therefore, an interesting challenge for the chemist when it comes to creating efficient, cheap, high-yielding methods that will give the desired enantiopure cyanohydrin. Many excellent procedures have already been invented, but there is still room for improvements, especially when it comes to finding alternative cyanide sources and designing more general catalysts, which will give high enantioselectivity and yield for a broad range of substrates.

2.1.1. Catalysts for Asymmetric Cyanation of Aldehydes Today chiral Lewis acidic Ti(IV)-Schiff base complexes are the most studied catalysts for asymmetric cyanohydrin synthesis. This was not the case 15-20 years ago when instead cyclic dipeptides, such as 1 (Figure 2), were the most promising catalysts, followed by enzymes and transition metal complexes.6e The change started in 1986 when Reetz reported the first asymmetric TMSCN addition to an aldehyde catalyzed by an optically active Lewis acidic boron catalyst 2.27 Despite the low yield (45-55% after 140 h) and the low enantioselectivity (12-16% ee), numerous researchers extended the field by inventing new chiral Lewis acid-catalyzed systems.

In industry mainly enzymes, or more specifically oxynitrilases, are used as catalysts. Both (R)- and (S)-oxynitrilases have been isolated from different sources, although the R-form is far more common in nature. By cloning the enzymes and over-expressing them in suitable organisms sufficient amounts can be obtained and both forms are now commercially available. The asymmetric hydrocyanation of a broad range of aldehydes has been performed using oxynitrilases, usually giving high enantioselectivities and high yields.28

N O HN NH t-Bu S H Ph Ph N HN Ph B N N H H O Cl O NPr2 1 2 3 Figure 2. Examples of non‐metal catalysts used in asymmetric cyanohydrin synthesis.

Organocatalysts are becoming increasingly used in cyanohydrin synthesis. Inoue made early progress when using chiral dipeptide 1 in the hydrocyanation of benzaldehyde.29 Since then little work using organocatalysts was reported until this century when several catalysts were developed. Feng and co-workers described the use of a proline-based N,N’-dioxide in the cyanosilylation of aromatic aldehydes30 and the corresponding cyanoformylation using a chiral quaternary ammonium salt together with triethylamine.31 Nájera and co- workers also used quaternary ammonium salts together with triethylamine in a similar kind of cyanoformylation reaction.32 Both groups obtained good enantioselectivities with aromatic aldehydes (up to 72 and 88% ee, respectively). Excellent enantioselectivity (96% ee) was reported by Jacobsen in the cyanosilylation of benzaldehyde using only 0.05 mol % of the hydrogen- bonding thiourea derived catalyst 3 (Figure 2).33

A huge number of chiral Lewis acidic metal complexes have been investigated in asymmetric cyanohydrin synthesis.34 The most common class of chiral ligands used are Schiff bases, but amides, PyBOXes, BINOL, BINOL derivatives and other types of alcohols have also been frequently studied. More than 50% of the catalysts have a titanium metal center, although aluminum, vanadium and lanthanides are also common.35 Due to the large number of examples available, only a few will be discussed here (Figure 3).

O TG2 O O Ph2P

OH OH t-Bu HO N OH O O O NH HN O N HO N N OH Ph Ph Ph P Ph Ph 2 O 4 5 6 7 Figure 3. Examples of chiral ligands used in asymmetric cyanohydrin synthesis.

Oguni reported a chiral C1-symmetric Schiff base-titanium complex (4, Figure 3) for the cyanosilylation of aldehydes with varying enantioselectivies (20- 96% ee) and yields (54-76%).36 The cyanosilylation of aldehydes was also investigated by Uang using amide ligand 5 together with Ti(OiPr)4; excellent enantioselectivies (93-99% ee) were observed for both aromatic and aliphatic aldehydes.37 PyBOX ligands together with lanthanides have previously been investigated in our group in the silylcyantion of aldehydes. A polymer- supported version, 6, together with YbCl3 gave the TMS-protected mandelonitrile in 30 minutes in 88% yield and 80% ee and the catalyst could be recycled four times without diminishing the enantioselectivity.38 Shibasaki reported a bifunctional BINOL-derived catalyst 7, which together with AlCl3, catalyzed the TMSCN addition to aldehydes in excellent yields and enantioselectivites, although the reaction required at least 36 hours for completion.39

2.1.2. Salen‐Based Catalysts Metal complexes of Schiff bases derived from salicylaldehydes and N,N’- substituted ethylenediamines, commonly known as salen ligands, have been employed in a wide range of different reactions such as epoxidation, epoxide ring-opening, Diels-Alder, conjugate addition, and imine addition.40 In cyanations of aldehydes salen derivatives complexed to titanium,41 vanadium,42 aluminum,43 and manganese44 have been used as catalysts. The first two examples in cyanohydrin synthesis were reported about the same time 41 by Jiang, and Belokon’ and North in 1996. Both groups used Ti(OiPr)4 and TMSCN, obtaining similar results (up to 87% ee and 90% yield) even though Jiang used a salen ligand derived from 1,2-diphenylethylenediamine and Belokon’ and North used a salen ligand obtained from 1,2- diaminocyclohexane. Belokon’ and North later discovered that the dimeric titanium complex of a similar salen ligand (8) was far more reactive and selective in the silylcyanation reaction (Scheme 4) than the previous complex. Only 0.1 mol % of the catalyst was now needed to obtain full conversion for a range of aldehydes after one hour at room temperature with up to 92% ee (<5 min, 86% ee for benzaldehyde) whereas 20 mol % of the catalyst and -80 °C were earlier needed to obtain high enantiomeric excess in 24 hours.45 By lowering the temperature, even higher enantioselectivity was obtained in the reaction using 8 although longer reaction time was required (96% ee, 3 days, - 50 °C).

Scheme 4. The highly efficient silylcyanation system developed by Belokon’ and North.

O OTMS 0.1 mol % 8 H + TMSCN CN CH2Cl2 rt, 5 min 86% ee tBu tBu

tBu tBu N O O N O Ti Ti O N O O N tBu tBu

tBu 8 tBu

Several groups have followed in the footsteps of Belokon’ and North and developed other salen-based catalysts for use in the silylcyantion of aldehydes. Belokon’ and North themselves have constructed modified systems, but none of the new designs have been as efficient as complex 8.46

2.1.3. Dual Lewis Acid–Lewis Base Activation In many asymmetric reactions either the nucleophile or the electrophile is activated by a chiral catalyst in order for the reaction to proceed smoothly and to give the enantiomerically enriched product. In nature it is common for both the nucleophile and the electrophile to be activated during the reaction. To make use of comparable dual activating systems has become increasingly common in otherwise sluggish reactions.47

Two different approaches can be used for dual Lewis acid–Lewis base activation. The nucleophile and the electrophile can either be activated by one bifunctional catalyst containing both a Lewis acidic and a Lewis basic site (A, Figure 4) or two separate catalysts can be used (B).35 Both methods have their advantages. In B there is a large risk for reaction of the Lewis base with the Lewis acid and thereby quenching of the desired reaction. In A the backbone of the catalyst can be constructed in such a manner that the Lewis acidic and the Lewis basic parts cannot come in close contact, thereby diminishing the risk of quenching. System B is easier to modify than A since either or both the Lewis acid and/or the Lewis base can be modified without the need of re-synthesizing the complete catalytic system.

Nu E Nu E

LB LA LB LA Catalyst Additive Catalyst (A)(B) Figure 4. Dual Lewis acid–Lewis base activation by (A) a bifunctional catalyst or (B) two separate catalysts.

Corey was the first to apply dual activation in an asymmetric cyanohydrin synthesis.48 The Lewis acidic magnesium complex 9 was used to activate the aldehyde and a Lewis basic bisoxazoline 10 (Figure 5) to activate hydrogen cyanide, which was formed in situ from TMSCN and moisture. Shibasaki later applied the bifunctional BINOL-derived catalyst 7 (Figure 3) in the same reaction.39 A similar bifunctional BINOL-derived catalyst 11 (Figure 5), where the Lewis basic phosphane oxide groups in 7 had been exchanged for diethylamino substituents, was reported by Nájera and Saá.49 The silylcyanation of aldehydes was also recently investigated by Feng who used a pyridine N-oxide as the Lewis basic moiety in the bifunctional catalyst 12.50

O CN Et N 2 Ph + O O O O N HN O - Cl Al O Ph N N N N O Mg Ph Ph Ph Ph + - Cl Et2N N O 910 11 12 Figure 5. Catalysts and ligands applied in dual Lewis acid–Lewis base activation.

2.1.4. Aim of the Study The aim of this study was to investigate ethyl cyanoformate and acyl cyanides as alternative cyanide sources to TMSCN in the asymmetric synthesis of cyanohydrins from aldehydes. We also wanted to study the effect of using Lewis acid–Lewis base activation in the reaction and investigate the mechanism.

2.2. Synthesis of O‐Alkoxycarbonyl functionalized Cyanohydrins The addition of TMSCN to aldehydes catalyzed by a chiral titanium bispyridylamide complex had been studied previously in our group.51 It was observed that the enantiomeric excess increased with time to a maximum of 70%, probably due to the formation of a more selective catalyst. By first employing the catalyst for the silylcyanation of valeraldehyde, a “sacrificial”

9 aldehyde, before TMSCN and the desired aldehyde were added, only a small variation of enantiomeric excess with time was observed and higher enantioselectivity was obtained (79%).52 Soon after the experiments had been performed in our laboratory Belokon’ and North reported similar findings in cyanation reactions with Ti-salen complexes.53

Scheme 5. Addition of ethyl cyanoformate to benzaldehyde catalyzed by 16‐Ti(OiPr)4.

O Ph Ph O O O O O O Ti(OiPr)4, 16 NH HN DMAP H + NC O CN CH2Cl2,rt N N 13a 1415a 16 70% conversion 58% ee

At the same time we became interested in studying alternatives to TMSCN as cyanide sources. Cyanoformates had been used in the cyanation of aldehydes catalyzed by inorganic bases,25a Lewis bases54 and by BINOL-55 and salen- based catalysts56, and we decided to investigate ethyl cyanoformate as cyanide source in the chiral titanium bispyridylamide catalyzed reaction (Scheme 5). The reaction proceeded smoothly in the presence of 4 mol % 16-Ti(OiPr)4 and 2 mol % DMAP (Figure 6) to give 70% conversion and 58% ee of the product at room temperature. The presence of Lewis base in the reaction turned out to be crucial as no conversion was otherwise observed.

N H H N N N N N N N H H DM AP DABCO DBU sparteine

H H

H H HO N HO N H H MeO

N N cinchonidine quinine Figure 6. Lewis bases used in the dual Lewis base‐Lewis acid activation.

Encouraged by this result of dual activation we chose to investigate the effect of adding a Lewis base in the addition of ethyl cyanoformate to aldehydes catalyzed by the dimeric Ti-salen complex 8 (Scheme 4). The reaction in the absence of Lewis base had previously been studied by Belokon’ and North and

10 an enantiomeric excess of 95% was observed after 18 hours at -42 °C (Table 1, entry 1).56 They could decrease the catalyst loading to 1 mol % and still obtain full conversion after 19 hours, although lower enantiomeric excess was achieved (83%, entry 2).

We first studied the effect of the amount of added Lewis base (Table 1, entries 3-6). A significant rate enhancement was observed in the presence of DMAP (2-8 h, compared to 18 h in the absence of Lewis base) without any major loss in enantioselectivity (86-95% ee, compared to 95% ee). Lowering the amount ethyl cyanoformate from 2 to 1.2 equivalents required a longer reaction time (4 vs. 8 h entries 3 and 7), but did not result in any loss of enantioselectivity.

Other Lewis bases (Figure 6) were then investigated (entries 8-16). Using 10 mol % Et3N as Lewis base and 5 mol % 8 resulted in 97% conversion within 3 hours and 92% ee (entry 9). Use of ent-8 instead of 8 together with the three chiral bases (entries 11-16) resulted in negligible changes in selectivity, but in slower reactions in the cases of cinchonidine and quinine.

Table 1. Cyanation of benzaldehyde with 14 catalyzed by 8 and Lewis bases at ‐40 °C.

O O O O O 8, Lewis base H + NC O CH2Cl2,-40°C CN 13a 14 15a entry % 8 Lewis base (%) equiv of 14 time (h) % conva % eeb 1c 5 2 18 100 95 (S) 2c 1 2 19 100 83 (S) 3 5 DMAP (10) 2 4 99 93 (S) 4 5 DMAP (15) 2 3 96 91 (S) 5 5 DMAP (20) 2 2 93 86 (S) 6 1 DMAP (2) 2 8 78 95 (S) 7 5 DMAP (10) 1.2 8 95 94 (S) 8 5 DABCO (10) 1.2 7 90 90 (S)

9 5 Et3N (10) 1.2 3 97 92 (S) 10 5 DIPEA (10) 1.2 3 96 89 (S) 11 5 sparteine (10) 1.2 3 98 78 (S) 12 5 cinchonidine (10) 1.2 4 98 94 (S) 13 5 quinine (10) 1.2 4 93 93 (S) 14 5d sparteine (10) 1.2 3 98 79 (R) 15 5d cinchonidine (10) 1.2 7 93 94 (R) 16 5d quinine (10) 1.2 7 97 94 (R) a Determined by GC/MS. b Determined by chiral GC. c Run at -42 °C (Ref. 56). dent-8 was used.

A variety of both aromatic and aliphatic aldehydes were then studied in the presence of Et3N. The O-alkoxycarbonylated cyanohydrins were in all cases readily formed within 4-7 hours in high isolated yields (Table 2). The product from 13f was obtained with somewhat lower enantioselectivity (73%, entry 6) than the other cyanohydrins, which were formed in good to excellent enantiomeric excesses (89-94%).

Table 2. Cyanation of aldehydes with 1.2 equiv 14 catalyzed by 8 and Et3N at ‐40 °C.

O O O 8 (5 mol %) Et3N(10mol%) O O + R H NC O CH2Cl2,-40°C R CN 13 14 15 entry 13 (R) time (h) % yielda % eeb 1 a (Ph) 4 95 92 (S)

2 b (4-CH3-C6H4) 6 88 94 (S) 3 c (4-CH3O-C6H4) 6 79 94 (S) 4 d (4-Cl-C6H4) 4 90 93 (S) 5 e ((E)-PhCH=CH) 7 97 93 (S)

6 f ((CH3)3C) 5 81 73 (S) 7 g (CH3(CH2)4) 5 83 89 (S) a Isolated yield. b Determined by chiral GC.

2.3. Synthesis of O‐Acylated Cyanohydrins We considered acyl cyanides to be attractive alternatives to TMSCN as cyanide source in asymmetric cyanations of aldehydes since acylated cyanohydrins are important synthons which can undergo further transformations without loss in enantiomeric purity.6 Hanefeld, for example, used O-acylated cyanohydrins for the synthesis of N-acylated β-aminoalcohols.57 Acylated cyanohydrins are also themselves interesting since compounds containing cyanoester functions are used as potent insecticides.20

Acyl cyanides had not been used in the enantioselective addition to aldehydes, although a few non-selective processes yielding racemic acylated cyanohydrins had been developed. Already in 1949 benzoyl cyanide, which is more reactive than acetyl cyanide, was added to aromatic aldehydes in the presence of a catalytic amount of potassium hydroxide.58 The same reaction has later been performed with potassium carbonate in aqueous acetonitrile25a and with DMSO together with 4Å molecular sieves.59 DABCO60 and tributyltin cyanide61 have also been shown to catalyze acetyl cyanide addition to several different aldehydes.

Enantioenriched O-acylated cyanohydrins have been synthesized by a few two-step methods. Potassium or sodium cyanide together with acetic anhydride have been used in an one-pot procedure, which required a large excess of the reagents.23b-c,62 Enzymatic one-pot procedures have been used where racemic cyanohydrins are acylated by a lipase leading to enantioenriched cyanohydrin esters via dynamic kinetic resolution.63

We decided to investigate the commercially available acetyl cyanide as cyanide source in the addition to benzaldehyde together with the dimeric Ti- salen catalyst 8 (Table 3). Running the reaction with 2 equivalents of acetyl cyanide (17a) and 5 mol % of 8 at -40 °C resulted to our disappointment in no reaction after 24 hours (entry 1). Raising the temperature to 25 °C gave slow product formation and low enantioselectivity (entry 2).

Table 3. Cyanation of benzaldehyde with 17a catalyzed by 8 and Lewis bases.

O O O 8 (5 mol %) O Lewis base H + CN CH2Cl2 CN 13a 17a 18a entry Lewis base (%) equiv of 17a T (°C) time (h) % conva % eeb 1 2 -40 24 0 n.d. 2 2 25 24 30 53 (S) 3 DMAP (10) 2 -40 6 57 94 (S) 4 DMAP (15) 2 -40 6 67 91 (S) 5 DMAP (20) 2 -40 6 76 91 (S) 6 DMAP (10) 2 -10 6 78 89 (S) 7 DMAP (10) 2 25 4 97 67 (S) 8 DMAP (10) 1 -10 6 63 90 (S) 9 DABCO (10) 2 -40 9 67 92 (S)

10 Et3N (10) 2 -40 8 96 94 (S) 11 DIPEA (10) 2 -40 8 97 81 (S) 12 DBU (10) 2 -40 2 99 20 (S) 13 diethylamine (10) 2 -40 8 41 81 (S) 14 sparteine (10) 2 -40 8 93 65 (S) 15 cinchonidine (10) 2 -40 9 78 96 (S) 16 quinine (10) 2 -40 9 80 92 (S) 17c sparteine (10) 2 -40 8 96 67 (R) 18c cinchonidine (10) 2 -40 9 75 92 (R) 19c quinine (10) 2 -40 9 73 95 (R) a Determined by GC/MS. b Determined by chiral GC. cent-8 was used.

To test if the same dual activation protocol could be applied using acetyl cyanide (17a) as observed when employing ethyl cyanoformate (14), a range of different Lewis bases (Figure 6) were examined in the reaction (Table 3, entries 3-19). Satisfactorily high conversion and enantioselectivity were now obtained and the best results, 96% conversion and 94% ee, were achieved with 10 mol % Et3N and 5 mol % 8 within 8 h at -40 °C. Replacing Et3N by DBU resulted in a faster reaction, but at the expense of the enantioselectivity (entry 12). Three chiral bases were used together with both 8 and ent-8, but only minor differences in reactivity and selectivity were observed (entries 14-19).

The addition of acetyl cyanide to benzaldehyde was also attempted in the presence of only Lewis base (Table 4) to enable comparison of the reactivity with that in the reaction catalyzed by Lewis acid 8 and a Lewis base. After 8 hours at -40 °C no or only minor conversion was observed and the chiral bases resulted in product with no or low enantiomeric excess. These results confirmed that the reaction needs dual Lewis acid–Lewis base activation to proceed smoothly with high enantioselectivity.

Table 4. Cyanation of benzaldehyde with 2 equiv 17a catalyzed by Lewis bases at ‐40 °C.

O O O Lewis base (10 mol%) O H + CN CH 2Cl2,-40°C CN 13a 17a 18a entry Lewis base time (h) % conva % eeb 1 DMAP 8 < 1 2 DABCO 8 < 1

3 Et3N 8 13 4 DIPEA 8 16 5 sparteine 8 23 0 6 cinchonidine 8 9 40 (S) 7 quinine 8 2 15 (S) a Determined by GC/MS. b Determined by chiral GC.

The utility of the system was tested by investigating both aromatic and aliphatic aldehydes in the acetylcyanation in the presence of Et3N and 8 (Table 5). The desired products were with a few exceptions obtained within 6-12 hours in high isolated yields and enantiomeric excesses. The two aliphatic aldehydes (entries 10-11) provided the corresponding products in high yields after only 6 hours, although the highly branched 18j was obtained in somewhat lower enantiomeric excess (76%). 3-Phenoxybenzaldehyde 13f (entry 6) required prolonged reaction time (48 h) and 2-pyridinecarboxaldehyde (entry

8) gave only 20% ee, but all other aromatic aldehydes gave high enantiomeric excesses and yields.

Table 5. Cyanation of aldehydes with 2 equiv 17a catalyzed by 8 and Et3N at ‐40 °C.

O O O 8 (5 mol %) Et3N(10mol%) O + R H CN CH2Cl2,-40°C R CN 13 17a 18a-k entry 13 (R) product time (h) % yielda % eeb 1 a (Ph) 18a 10 89 94 (S)

2 b (4-CH3-C6H4) 18b 10 90 96 (S) 3 c (4-CH3O-C6H4) 18c 12 72 94 (S) 4 d (4-Cl-C6H4) 18d 8 89 95 (S) 5 e (2-furyl) 18e 12 93 89 (R)

6 f (3-PhO-C6H4) 18f 48 84 85 (S) 7 g (3-pyridyl) 18g 12 91 86 (n.d.) 8 h (2-pyridyl) 18h 12 87 20 (n.d.) 9 i ((E)-PhCH=CH) 18i 12 64 93 (S)

10 j ((CH3)3C) 18j 6 84 76 (S) 11 k (CH3(CH2)4) 18k 6 89 90 (S) a Isolated yield. b Determined by chiral GC or chiral HPLC.

Six acyl cyanides were then synthesized (Scheme 6) by reacting acyl bromides with dry copper(I) cyanide without solvent (for low-boiling acyl cyanides 17b- 17f)64 or by reacting acyl chlorides with copper(I) cyanide in acetonitrile (for less volatile acyl cyanides such as 17g).65

Scheme 6. Synthesis of acyl cyanides 17b‐g from acyl halides and CuCN.

O O +CuCN R X R CN 17 b R=CH2CH3, X = Br, 74% c R=(CH2)2CH3,X=Br,71% d R=(CH2)3CH3,X= Br,75% e R=CH(CH3)2,X=Br,79% f R=CH(CH3)3,X=Br,68% g R=(E)-CH=CHPh, X = Cl, 81%

The synthesized acyl cyanides and the commercially available benzoyl cyanide (17h) were added to benzaldehyde and the results compared to that of the reaction run with commercial acetyl cyanide (Table 6). Acyl cyanides 17b-e (entries 2-5) resulted in the desired products in high isolated yields (85-90%) and with high enantioselectivities (92-93% ee) within 10 hours. Acyl cyanide 17f (entry 6) exhibited lower reactivity and the reaction had to be carried out at

15 room temperature, while 17g (entry 7) instead showed higher reactivity and provided full conversion in only 4 hours at -40 °C with excellent enantioselectivity (94% ee).

Table 6. Cyanation of aldehydes with 2 equiv 17 catalyzed by 8 and Et3N at ‐40 °C.

O O O 8 (5mol%) Et3N(10mol%) O R + H R CN CH Cl ,-40°C 2 2 CN 13a 17 18a,l-r entry 17 (R) product time (h) % yielda % eeb

1 a (CH3) 18a 10 89 94 (S) 2 b (CH2CH3) 18l 10 89 93 (S) 3 c ((CH2)2CH3) 18m 10 90 92 (S) 4 d ((CH2)3CH3) 18n 10 85 93 (S) 5 e (CH(CH3)2) 18o 10 86 92 (S) c 6 f (C(CH3)3) 18p 10 81 79 (S) 7 g ((E)-CH=CHPh) 18q 4 89 94 (S) 8 h (Ph) 18r 26 76 75 (S) a Isolated yield. b Determined by chiral GC or chiral HPLC. c The reaction was carried out at rt.

Baeza et al. shortly after us reported the benzoylcyanation of aldehydes catalyzed by the bifunctional catalyst 11 in up to 68% ee at room temperature.66 With catalyst 11, acetyl cyanide did not promote the reaction. List later reported the first asymmetric acyl cyanation of imines with a chiral thiourea-derived catalyst, which was also applied in a catalytic three- component acyl-Strecker reaction.67

2.4. Synthesis of a Herbicide Intermediate Of all pesticides currently used about 25% are chiral and for economical reasons many are used as racemates or, if they contain more than one stereocenter, even as mixtures of diastereomers.68 However, it is known that the two enantiomers of the compounds differ significantly in aquatic toxicity and degradation rate. As mentioned earlier, several of the chiral insecticides contain cyanoester functions (see examples in Figure 7).20

Cl O O O Cl O O CN O CN Cl Fenvalerate Cypermethrin Figure 7. Examples of chiral cyanoester‐containing insecticides.

Cyanoesters, such as 19,69 are also used as synthons in the synthesis of insecticides as in a route to the widely used herbicide glufosinate 20 (Scheme 7).70 Cyanoester 19 itself has also been evaluated as an insecticide in racemic form.71

We wanted to illustrate that enantioenriched insecticides can be prepared in a straight-forward reaction by applying the dual Lewis acid–Lewis base catalytic system in the synthesis of cyanoester 19 from acrolein (21) and acetyl cyanide (17a) (Scheme 7).

Scheme 7. Synthesis of cyanoester 19, an intermediate for glufosinate (20).

O O O O O 8 (5 mol %) O + Et3N(10mol%) P H CN OH CH2Cl2 HO CN NH 21 17a-40 to -25 °C 19 20 2 79%, 73% ee

The reaction was first run according to the procedure used for other substrates, using 2 equivalents of 17a, which resulted in only 40% conversion, probably due to polymerization of 21 during the reaction. By instead using an excess of acrolein (21), cyanoester 19 was isolated in 79% yield after 22 hours. Racemization of product 19 during purification resulted in a decrease in enantiomeric excess from 64 to 54%. Traces of Et3N present during distillation were found to cause the racemization and washing the crude product with weak acid before further purification solved the problem of enantiomeric excess decrease. By adding the acetyl cyanide slowly to the reaction mixture over 6 hours at -40 °C higher enantiomeric excess was obtained (73% ee).

2.5. Mechanistic Aspects The mechanism of the silylcyanation of aldehydes catalyzed by 8 (Schemes 8 and 9) has been comprehensively studied by Belokon’ and North.23c,53,72 They found that complex 8 is in a concentration- and temperature-dependent equilibrium with complex 21 in dichloromethane and chloroform (Scheme 8).

Scheme 8. The concentration‐ and temperature‐dependent equilibrium between dimer 8 and monomer 21.

t Bu tBu tBu

tBu tBu tBu N O O N O N O Ti Ti 2 O Ti N O O O N O N tBu t Bu tBu

tBu tBu tBu 8 21

While running the addition of potassium cyanide to aldehydes catalyzed by 8 in the presence of different anhydrides, North and Belokon’ observed an increase of enantiomeric excess with time. Their explanation was that complex 8 is slowly converted into catalytically active species (Scheme 9), and during the early stage of the reaction, less enantioselective catalysis occurs due to other species.53

Scheme 9. Proposed mechanism for the silylcyanation of aldehydes catalyzed by 8.

- TMS 2CN O (salen)Ti Ti(salen) [(salen)TiO]2 2 x (salen)Ti=O O TMSCN 821 TMS PhCHO

Ph O OTMS CN (salen)Ti O Ti(salen) + Ti(salen) OTMS CN

H - (salen)Ti CN (salen)Ti O CN PhCHO O O Ph (salen)Ti CN (salen)Ti CN

OTMS r.d.s. Ph CN H (salen)Ti O Ph CN O TMSCN (salen)Ti CN

We observed a similar increase of enantiomeric excess in our acylcyantion using 8 (Figure 8, graph a). Stirring complex 8 for 3 hours at -40 °C before introducing aldehyde, Lewis base and acyl cyanide (graph b) resulted in product formation with a constant enantiomeric excess throughout the reaction. When instead 8 was stirred for 3 hours at room temperature prior to cooling to -40 °C followed by immediate addition of aldehyde, Lewis base, and acyl cyanide (graph c), the enantiomeric excess again increased with time from a low value in the initial part of the reaction.

100

85 a

ee (%) 80 b c 75

65 0 50 100 150 200 250 300 350 400 450 500 time (min) Figure 8. Addition of 17a to benzaldehyde under a) normal conditions, b) Ti‐salen complex 8 kept at ‐40 °C for 3 h before addition of Lewis base, aldehyde and acyl cyanide, c) Ti‐salen kept at rt for 3 h before cooling down to ‐40 °C and addition of Lewis base, aldehyde and acyl cyanide.

In order to understand this effect (Figure 8) 1H NMR spectra of 8 were recorded at intervals at -40 °C. The dimer 8 to monomer 21 ratio (Scheme 8) was 80:20 at room temperature. Spectra recorded immediately after cooling showed a dimer to monomer ratio of 90:10. This ratio increased with time to a maximum of 95:5 after 1.5 hours and was then constant. By heating the sample back to room temperature the original 80:20 ratio was again observed. It therefore seems likely that the dimeric complex 8 is needed to obtain high enantiomeric excess in the acylcyanation reaction.

Acetyl cyanide has been shown to dimerize in the presence of base (Scheme 10) by initial attack of base on the carbonyl group followed by acylation of the alcoholate.73 Based on this knowledge the mechanism for the non-selective acylcyanation of aldehydes catalyzed by DABCO60 or DBU74 has been suggested to start with attack of the Lewis base on the carbonyl group of the acyl cyanide followed by formation of a cyanide ion and a potent acylating

19 agent. The cyanide ion and the acylating agent then react with the aldehyde to form the observed product.

Scheme 10. The base‐catalyzed dimerization of acetyl cyanide.

O O 2 base CN CN CN

Baeza et al. suggested an alternative mechanism for the benzoylcyanation of aldehydes catalyzed by 11. They proposed that the reaction is initiated by cyanide ions obtained by deprotonation of hydrogen cyanide present as an impurity in the acyl cyanide.66

We could not detect any triethylammonium ions by recording the 1H NMR spectrum of a mixture of Et3N and hydrogen cyanide in CD2Cl2. This is an indication, but not a strict proof that our acylcyanation reaction does not resemble the suggested mechanism by Baeza et al., since deprotonation is highly inefficient in dichloromethane and a minute amount of cyanide ions may be enough to initiate the reaction. To further investigate if hydrogen cyanide is part of the mechanism, two equivalents of 13C-labelled hydrogen cyanide were bubbled through the reaction solution containing 8 before addition of the reactants. A mass spectrum recorded after 5% conversion showed no 13C incorporation in the product, which indicates that free cyanide ions are not present in the solution since scrambling between the 13C-labelled hydrogen cyanide and free cyanide should be observed in that case.

Scheme 11. Proposed mechanism by Belokon’ and North for the cyanation of aldehydes in presence of acetic anhydride.

H - (salen)Ti OAc (salen)Ti O OAc Ac2O PhCHO [(salen)TiO]2 Ph O TMSCN O 8 (salen)Ti OAc (salen)Ti CN 22

OAc r.d.s. H Ph CN (salen)Ti O Ph CN O Ac O 2 (salen)Ti OAc

Treatment of complex 8 with acetic anhydride was shown by Belokon’ and North to form the bimetallic bis-acetate 22.23c From this observation they proposed the mechanism shown in Scheme 11 for the formation of O-acylated cyanohydrins using potassium cyanide or TMSCN as cyanide source and an

20 anhydride as acylating reagent. By mixing acetyl cyanide (17a) and 8 we found that the 13C NMR signal for the carbonyl carbon atom in 17a was shifted which indicates that 17a, or acetate, is coordinated to titanium.

Based on the observations by Belokon’ and North and our own experimental data we suggest the mechanism shown in Scheme 12 for the acylcyantion of aldehydes catalyzed by 8 and a Lewis base. The Lewis base first attacks the carbonyl group of the acyl cyanide (17a) with formation of a cyanide ion complexed to an acylating agent (23). Catalyst 8 and 23 form complex 24 that after coordination to aldehyde 13a forms complex 25 with acetate as a counter ion. After formation of the titanium-bound cyanohydrin and re-coordination of acetate to form 26, acylation of the cyanohydrin will generate product 18a and regenerate complex 8. In this proposed mechanism the cyanide and the acyl group within the product are assumed to originate from the same acyl cyanide molecule, which could explain the lack of incorporation of 13C from free hydrogen cyanide. The Lewis base is not part in the stereochemistry- determining step, which probably is the reason why only minor changes in enantioselectivity were observed between reactions using the chiral Lewis bases together with either 8 or ent-8.

Scheme 12. Mechanism suggestion for dual Lewis acid–Lewis base‐catalyzed acylcyanation of aldehydes.

O 17a CN O

OAc NE t CN [(salen)TiO] 3 2 23 Ph CN 8 Et3N 18a

H (salen)Ti OAc (salen)Ti O Ph CN O 24 O (salen)Ti CN (salen)Ti OAc 26 O Ph H H 13a (salen)Ti O AcO O Ph (salen)Ti CN 25

2.6. Conclusions We have developed a new efficient method for the synthesis of enantioenriched O-alkoxycarbonylated and O-acylated cyanohydrins from aldehydes by Lewis acid–Lewis base activation. Yields up to 97% and enantiomeric excesses up to 96% were obtained. A mechanism for this direct 100% atom economical reaction has been suggested and experimental data support the proposed addition of cyanide to aldehyde within the dimeric Ti-salen complex.

3. Minor Enantiomer Recycling in Enantioselective Cyanohydrin Synthesis

(Paper IV)

3.1. Introduction To create chiral enantiopure compounds in high yields, preferably at a low cost, is a challenging but important task. Enormous progress in asymmetric synthesis using biocatalysis, metal catalysis, and organocatalysis has produced an array of efficient methods for the preparation of chiral enantioenriched substances on both industrial and laboratory scale.40,75 Many of these reactions proceed with high enantioselectivity, but additional purification to get rid of any remaining unwanted enantiomer is still often needed to meet the tough demands regarding high enantiomeric excess required for many applications. To avoid additional purification there is a need for new efficient asymmetric methods that produce close to enantiopure compounds in high yields.

3.1.1. Methods for Transforming a Racemate into a Single Enantiomer Separation of enantiomers via diastereomeric derivatives and kinetic resolution are the most common ways to prepare enantioenriched compounds in industry (Scheme 13).28b In the first method the racemate is separated into its enantiomers by the formation of pairs of diastereomers with an enantiopure resolving agent followed by separation, usually by crystallization of one of the diastereomers, and removal of the resolving agent. In kinetic resolution the enantiomers are separated by making use of their different reactivity with a chiral enantiopure molecule or catalyst. The major drawback of both methods is that a maximum of 50% yield can be obtained. Techniques wherein a racemate can be transformed into a single enantiomer in 100% yield are desired.

Scheme 13. Schematic drawings of resolution via diastereomers (left) and kinetic resolution.

(+)-S A* cat* + (+)-SA* + (-)-SA* (+)-S P fast (-)-S Up to 50% separation

cat* (-)-S ent-P (+)-SA* (-)-SA* slo w

-A* -A* (+)-S (-)-S

Several dynamic methods have been developed in order to overcome the disadvantage of a maximum of 50% yield. One common technique is dynamic kinetic resolution (DKR) wherein a racemization of the starting compound is combined with a resolution, a process which can result in a theoretical yield of 76 100% (Scheme 14). If the racemization is fast (krac>kA) the enantiomeric excess of the product is determined only by the selectivity (kA/kB) of the resolution step. Dynamic kinetic asymmetric transformation (DYKAT) is a process related to DKR, although DYKAT proceeds through equilibration of diastereomeric rather than enantiomeric intermediates (Scheme 14).77

Scheme 14. Dynamic kinetic resolution (left) and dynamic kinetic asymmetric transformation ((+)‐Scat*, (‐)‐Scat* = diastereomeric substrate‐catalyst complexes; Scat* = intermediate complex).

cat* k(+)-S kP (+)-S P (+)-S (+)-Scat* P kA Up to 100% fast k( +)-Scat * krac racemization Scat* cat* k (-)-S ent-P (-) Sc at* kB k(-)-S k en t -P slow (-)-S (-)Scat* ent-P

At first sight, crystallization-induced asymmetric transformation (CIAT) seems to closely resemble DKR (Scheme 15).78 This is not exactly the case since in CIAT the equilibrium is driven by the crystallization, usually via seeding or salt-formation (non-covalent bonds), while in DKR the driving force is the differences in rates for formation of a new covalent bond.

Scheme 15. Crystallization induced asymmetric transformation.

(+)-S

racemization

crystallization (-)-S (-)-S (or as salt) in solution crystalline

In order to obtain quantitative yield of enantiopure product, an alternative to racemization of the substrate enantiomers is to convert both into the same product isomer. In a so called enantioconvergent process the substrate enantiomers must react with inversion and retention of configuration, respectively, at the pre-existing stereocenter (Scheme 16).77

Scheme 16. Enantioconvergent process.

k( +)-S (+)-S P (retention)

k(-)-S (inversion) (-)-S

Cyclic deracemization is an elegant way of turning a racemate into an enantiopure compound (Scheme 17). Since it is a deracemization there is no “product”, but simply an optical enrichment of the substrate.76a Turner and co- workers have reported several examples of cyclic deracemizations of chiral amines via an imine intermediate using an enzyme and a stoichiometric amount of an achiral reagent (Scheme 17).79

Scheme 17. Cyclic deracemization exemplified by the work of Turner and co‐workers.79

NH 2 enzyme

NH3:BH 3

NH2

All methods mentioned above are based on racemic starting materials and one enantioselective step in order to achieve enantiopure products in quantitative yield. Asymmetric reactions starting from prochiral substrates and achiral reagents still have to rely on excellent chiral catalyst selectivity to reach the high yields and selectivities observed from use of different resolution techniques, unless a subsequent purification step is performed.

3.1.2. Aim of the Study The aim of this study was to develop a recycling process for the undesired minor enantiomer in our Lewis acid–Lewis base-catalyzed acylcyanation of aldehydes. The goal was to increase enantioselectivity without decreasing the yield.

3.2. Minor Enantiomer Recycling (MER)

3.2.1. Development of the MER System We wanted to design a system where the minor enantiomer in the dual Lewis acid–Lewis base-catalyzed acylcyanation would selectively be recycled back to the starting aldehyde, which could then re-react (Scheme 18). By using two selective catalysts, the selectivity of the overall process can be shown to be equal to the product of the enantioselectivities of the individual catalysts. The overall selectivity can, thus, be higher than the selectivity of one single catalyzed step. A one-pot cyclic process in which both the forward and the backward reactions could work in concert was desired in order to obtain the product as a single enantiomer in high yield.

Scheme 18. Proposed design of a MER system for O‐acylated cyanohydrins.

chiral induction

O O O O O O H + + CN CN CN

minor major

ee correction

For the re-formation of aldehyde, a catalyst capable of selectively hydrolyzing one of the product enantiomers was needed. We believed Candida antarctica lipase B (CALB) would be a good choice since it is known to hydrolyze a variety of O-acylated cyanohydrins with high selectivity and efficiency in kinetic resolutions.80

Our first attempts to create a MER system consisted of adding immobilized CALB and water to a reaction mixture of acetyl cyanide, benzaldehyde, ent-8, and Et3N in dichloromethane (Scheme 19). The acylcyanation was supposed to run in the organic phase while the water phase was needed for the enzymatic hydrolysis. Not surprisingly, neither the forward Lewis acid–Lewis base- catalyzed reaction nor the backward enzyme-catalyzed reaction turned out to work in this biphasic system. The reason turned out to be that Et3N dissolved in the water phase, which stopped the forward reaction and that the immobilized enzyme layered on-top of the water phase and therefore could not come in close contact with its substrate.

Scheme 19. First failed attempt to create a MER system.

O O

O O O O Et3N, ent-8 H + CN + CN CN CH2Cl2,rt 13a 17a 18a

O OH 2

ALB, H CN C

27a

Solvent, Lewis base, temperature, and pH were among the parameters needed to be modified. The solvent should not be miscible with water, or otherwise the forward reaction does not work, and the solvent was to be of lower density than water so that the immobilized CALB could form a layer in between the two solvent layers. The Lewis base had to show higher solubility in the organic phase than the aqueous phase and it should be efficient in the dual activation protocol. The temperature had to be adjusted so both the forward and the backward reactions run smoothly. Finally, the pH needed to be regulated and then kept constant for optimal enzyme performance and to prevent spontaneous hydrolysis of the product.

After extensive optimization a functioning MER system was obtained (Scheme 20). The two-phase system used consisted of toluene and a 1M phosphate buffer of pH 8. To achieve rapid forward reaction DBU was used as Lewis base and 40 °C was chosen as the optimal temperature in order to obtain efficient hydrolysis by the enzyme.

The system worked perfectly in terms of enantioselectivity (>99% ee) for the test-substrate benzaldehyde, but yields were poor (<50%). Decomposition of acetyl cyanide into acetic acid in the water phase was the cause of the low yields. By adding acetyl cyanide continuously over several hours the problem was solved and yields exceeding 80% were obtained.

Scheme 20. The functioning two‐phase one‐pot MER system.

O O CN HCN Ph H O ent-8,DBU PhMe O

Ph CN 40 °C major O OH O Ph CN minor Ph CN CALB 1M buffer pH8 O OH

3.2.2. Use of MER in O‐Acylated Cyanohydrin Synthesis Five O-acylated cyanohydrins were synthesized using the MER method (Scheme 21).81 For benzaldehyde (13a), which was used for optimization of the method, a yield of 84% (GC) and an enantiomeric excess of 99.6% were obtained after 14.5 hours when acetyl cyanide (17a) was added over 11 hours. Somewhat higher enantioselectivity (>99.7% ee) was obtained by letting the reaction mixture stir for a longer time (up to 25.5 h). Product 18a was also isolated from a large scale reaction mixture (10 times larger scale than previously), to which 17a was added over 18 hours followed by 3 hours additional stirring, in 86% yield and 99.3% ee.

Exchanging 17a for butanoyl cyanide (17c) produced the butanoyl ester 18m in 83% yield and 95% ee after 25 hours (Scheme 21). Addition of 17a to 13c over 22 hours resulted in 18c with excellent selectivity (99.2% ee), although in lower yield (67%). Addition of 17a over a longer period of time would probably result in higher yield. From the acetylcyanation of the 4- chlorosubstituted benzaldehyde (13d) 18d was obtained in decent yield (74%) and excellent enantioselectivity (98% ee). Best results for the addition of 17a to furfural (13e) were obtained by adding the acyl cyanide over a shorter period of time (5.5 h). Product 18e was then obtained in 81% yield and 97% ee. Longer reaction time resulted in lower yield due to enzymatic hydrolysis of both enantiomers of the product.

Scheme 21. Synthesis of O‐acylated cyanohydrins using the MER method.81

O O 17a added O over 11 h

H CN 13a 18a 14.5 h, 84%, 99.6% ee 25.5 h, 80%, > 99.7% ee O O 17c added O over 20.5 h H CN 13a 18m 25 h,83%, 95% ee O O 17a added O over 22 h H3CO H3CO H CN 13c 18c 25.3 h, 67%, 99.2% ee O 17a added O O over 10.75 h Cl Cl H CN 13d 18d 23. 5 h, 74%, 98% ee O O 17a added O over 5.5 h O H O CN 13e 18e 7.25 h, 81%, 97% ee

The MER method could also be run with the Lewis base as the sole catalyst for the acetylcyanation. Racemic product is then initially formed and the enantioselectivity is determined only by the enzyme. For the reaction with benzaldehyde 98% ee and 77% yield were obtained without ent-8. More acetyl cyanide is also consumed in reactions performed without chiral Lewis acid. A reaction carried out with the chiral Lewis acid–Lewis base-catalyzed reaction resulting in 60% ee and the enzymatic hydrolysis proceeding with 100% selectivity would for example need 4 cycles and consume 1.25 equivalents acetyl cyanide to produce the product in 99.6% enantiomeric execess and 100% yield. Using only achiral Lewis base would lead to the same results only after 9 cycles and consumption of 1.99 equivalents of acetyl cyanide

3.2.3. Modifications and Improvements In the five cases presented in Scheme 21 higher enantioselectivites were obtained using the MER method than using the simple dual activation protocol (compare Scheme 21 with Table 5, entries 1, 3-5 and Table 6, entry 3). The yields were, however, in the same range or lower in the reactions using the MER method. Therefore, there was still room for modifications of the MER

29 system, which could lead to improved yields, but also to improved enantioselectivity in the Lewis acid–Lewis base-catalyzed step since at 40 °C using 20 mol % DBU the enantiomeric excess was below 10% for the addition of 17a to 13a.

To increase the selectivity in the Lewis acid–Lewis base-catalyzed step lower amounts of DBU were tested in the acetylcyanation of benzaldehyde. Using 10 mol % DBU at 40 °C increased the enantiomeric excess to 24%. Running the same reaction at room temperature increased the enantiomeric excess somewhat to 36%. Another way of running the MER system was also examined. The Lewis acid–Lewis base-catalyzed reaction was first run for 9 hours at room temperature in presence of 1.1 equivalents of 17a with 5 mol % each of ent-8 and DMAP before addition of CALB and buffer. Further continuous addition of 17a (up to a total of 3 equivalents) for 16.5 hours at 40 °C followed by stirring of the reaction mixture for an additional 5 hours resulted in a final yield of 92% of product 18a with 99% ee. An even higher yield (97%) was obtained, still with >99% ee, by adding more 17a (up to 4.5 equiv).

3.3. Conclusions and Outlook A minor enantiomer recycling (MER) system has been developed in which the minor product enantiomer from a chiral Lewis acid–Lewis base-catalyzed reaction is continuously re-transferred to starting material by an enzyme. In this recycling one-pot procedure chiral O-acylated cyanohydrins were synthesized from prochiral aldehydes and acyl cyanides in up to 97% yield and >99% ee.

Improvements of the MER system can however still be made when it comes to shortening reaction times, lowering the amount of added acyl cyanide, and increasing the selectivity in the Lewis acid–Lewis base-catalyzed step. More substrates would be of interest to investigate and to develop a system in which the other product enantiomer is obtained. Application of this concept to other catalytic processes is also of high interest.

4. Enantioselective Addition of Acyl Cyanides to Ketones

(Paper V)

4.1. Introduction Quaternary stereocenters are found in wide range of important pharmaceutical compounds and in many natural products. Therefore access to efficient enantioselective methods for the synthesis of compounds with such stereocenters is important.82

Ketones are more sterically hindered around the carbonyl group than aldehydes and the carbonyl group is also less electrophilic (unless the ketone has strong electron withdrawing groups). Therefore cyanation of ketones is generally more difficult than cyanation of aldehydes.6f-g,35 Successful examples of asymmetric cyanation of ketones using diverse catalysts have however been reported, but in most cases long reaction times are needed, the substrate scope is often narrow and only a few cyanide sources have been employed. From the developed systems it is difficult to draw general conclusions about which kind of catalytic system is most efficient. A few examples will be discussed here.

4.1.1. Catalysts for Asymmetric Cyanation of Ketones As with aldehydes, enzymes were used first as catalysts for enantioselective hydrocyanation of ketones. Both (R)- and (S)-oxynitrilases have been used with enantiomeric excesses up to 99% and yields up to 90% for a variety of methyl ketones.6c-e,28c,83 However, for ethyl alkyl ketones, yields were considerably lower (not exceeding 33%) which indicates that these substrates are on the limit of what the enzyme can tolerate.84

In 1997 Choi reported the first asymmetric chemically catalyzed cyanation of a ketone and as in most enantioselective ketone cyanations, excluding biocatalyzed reations, TMSCN was used as the cyanide source.85 Catalyst 28 (1 mol %, Figure 9) and high pressure (0.8 GPa) were used by the Choi group to cyanosilylate acetophenone in 93% yield and 60% ee in 18 hours. The first ambient-temperature system was reported by Belokon’ and North, in which 8 (Scheme 4) catalyzed the silylcyanation of unhindered simple methyl and ethyl aromatic ketones with enantiomeric excesses of 32-77% after 96 hours.86

O Ph H P O O Ph O N O O O N N Ti O H HO O O O OH O

28 29 30 O NHCPh3 HO

NH2 H COONa

N O B TfO H 31 32

Figure 9. Catalysts and ligands used for asymmetric cyanations of ketones.

Ligand 29 (Figure 9), developed by Shibasaki and co-workers, together with Ti(OiPr)4 or Gd(OiPr)3 turned out to be the first more general catalytic systems that could tolerated a broad range of ketones in the silylcyanation reaction. Up to 97% ee and 97% yield were obtained with either of the two metals, although considerably longer reaction times were needed in reactions using titanium (36-96 h) than gadolinium (1-36 h).87 Hoveyda and Snapper reported the silylcyanation of both aromatic and aliphatic ketones catalyzed by peptide- derived ligand 30 and Al(OiPr)3 in up to 98% yield after 48 hours and with 95% ee.88 A chiral oxazaborolidinium ion (31), developed by Corey and co- workers, enabled silylcyanation of methyl ketones with high enantioselectivities (85-96% ee) and yields (73-95%) after 2-10 days at room temperature when diphenylmethylphosphine oxide was used as co-reactant.89 Feng and co-workers used a chiral amino acid salt (32) as catalyst in the silylcyanation of ketones. More than 90% ee and high yields (75-96%) were reported for methyl aromatic ketones.90

Organocatalysts have also been successfully employed in the cyanation of ketones. Deng and co-workers added cyanoformates to unconjugated ketones, using natural or modified cinchona alkaloids as catalysts.25b In most examples good yields (55-99%) and high enantiomeric excesses (81-97%) were obtained within 2-7 days. Modified cinchona alkaloids catalyzed the cyanosilylation of acetal ketones as well, in excellent yields (92-99%) and enantioselectivities (90-98% ee).91 Jacobsen and co-workers used the thiourea derived catalyst 3 (Figure 2) in cyanosilylation of a wide range of ketones. Excellent enantiomeric excesses (86-98%) and high yields (81-97%) were obtained within 12-48 hours.33

4.1.2. Aim of the Study The goal of this study was to expand the field of asymmetric ketone cyanations by developing a catalytic system which could be used for asymmetric addition of acyl cyanides to ketones.

4.2. Chiral Lewis Base‐Catalyzed Addition of Acetyl Cyanide to α‐Ketoesters After the successful results using Lewis acid–Lewis base activation in acylcyanations of aldehydes, we decided to investigate ketones as substrates. No examples of direct asymmetric cyanoester formation had been reported, but a few examples producing racemic product had been described.74,92

Methyl benzoylformate (33a, Scheme 22) was first used as substrate in the acylcyanation under the same reaction conditions as those used for aldehydes (5 mol % 8, 10 mol % Et3N, CH2Cl2). At room temperature 50% conversion to a 84:16 mixture of racemic O-acetylated cyanohydrin 34a and non-protected cyanohydrin 35a was obtained after 5 hours. Running the reaction at lower temperature (-40 °C) still did not lead to enantioenriched product. By adding 17a over 30 minutes to the reaction mixture at room temperature 97% conversion was obtained with racemic 34a as the sole product. It was found that the reaction was also catalyzed by Et3N alone; at room temperature 30% conversion of a 67:33 mixture of 34a and 35a was obtained after 16 hours.

Scheme 22. Acetylcyanation of α‐ketoester 33a catalyzed by Lewis bases.

O Lewis base MeOCO CN HO CN OMe O (Lewis acid) OMe OMe + + CH Cl O CN 2 2 O O 33a17a 34a 35a

We then decided to investigate if selectivity could be induced by using a chiral Lewis base (Figure 10) as the sole catalyst in the reaction (Table 7).

H H N N H H O O HO N HO N H H O O O O MeO

N N N N cinchonidine quinine (DHQD)2AQN Figure 10. Chiral Lewis bases used as catalysts for cyanation of α‐ketoesters.

To our delight 34a was obtained in 66% ee and 69% conversion (99:1 ratio between 34a and 35a) was observed after 6 hours at -40 °C using cinchonidine (10 mol %) as catalyst (entry 1, Table 7). Addition of 10 mol % methanol led to increased conversion (77%), still with the same product ratio and enantiomeric excess (entry 2). Even higher conversion (91%) was observed when 17a was added over 3 hours although in somewhat lower 34a:35a ratio (93:7, entry 3). Changing Lewis base to quinine or (DHQD)2AQN resulted in decreased selectivities and conversions as compared with cinchonidine (entries 4-7). Product 34a was isolated in 77% yield and 66% ee.

Table 7. Cyanation of ketoester 33a with 2 equiv 17a catalyzed by Lewis bases at ‐40 °C.

O Lewis base Me OCO CN HO CN O OMe (MeOH) OMe OMe + + CH Cl O CN 2 2 O O -40 °C 33a17a 34a 35a entry Lewis base (%) time (h) % conva 34a:35a % eeb 34a 1 cinchonidine (10) 6 69 99:1 66 2 c cinchonidine (10) 6 77 99:1 66 3 c cinchonidine (10) 6d 91 93:7 66 4 quinine (10) 48 7 71:29 42 5 c quinine (10) 21 21 63:37 34

6 (DHQD)2AQN (5) 6 41 96:4 44 c 7 (DHQD)2AQN (5) 6 53 99:1 34 a Determined by 1H NMR. b Determined by chiral HPLC. cReaction run in presence of 10 mol % MeOH. d 17a added over 3 h.

Ethyl benzoylformate (33b, Scheme 23) reacted almost equally well as 33a under optimized conditions with cinchonidine (10 mol %) and MeOH (10 mol %) at -40 °C. The O-acylated product 34b was obtained in 67% isolated yield and 64% ee. Racemic 34b had previously been used for the preparation of β-lactones acting on the central nervous system93 and racemic naphtyl analogues had been prepared by radical reaction.94

Scheme 23. Acetylcyanation of α‐ketoester 33a catalyzed by cinchonidine under optimized conditions.

O cinchonidine (10 mol %) MeO CO CN O OEt MeOH (10 mol% OEt + O CN CH 2Cl2 O -40°C 33b17a 34b added 67%, 64% ee over 3 h

Ethyl pyruvate (33c) proved to be far more reactive than the two aromatic α-ketoesters. After 4 hours at -40 °C using cinchonidine and MeOH close to full conversion was observed, although the selectivity was moderate (41% ee, entry 1, Table 8). Running the reaction at -78 °C resulted in slightly higher enantioselectivity (47%), but lower conversion (60% after 3 hours, entry 2). In a reaction run without methanol 53% conversion and 53% ee were observed. Replacement of cinchonidine with (DHQD)2AQN (5 mol %) resulted in higher enantioselectivties (entries 4-6), in contrast to when aromatic α-ketoesters were used. Running the reaction at -40 °C with or without methanol (entries 4 and 5, respectively) resulted in full conversion after 3 hours and 64% ee in presence of alcohol and in 54% conversion and somewhat higher selectivity (69% ee) without alcohol. The best results, 99% conversion and 68% ee, were observed in presence of methanol at -78 °C after 6 hours (entry 6). Even higher selectivity (82% ee) was observed in the reaction of the more sterically hindered tert-butylester 33d (entry 7).

Table 8. Cyanation of ketoesters 33c and 33d with 2 equiv 17a catalyzed by Lewis bases.

O Me OCO CN O Lewis base OR' MeOH (10 mol %) OR' R + R O CN CH2Cl2 O 33c R=Me,R'=Et 17a 34c 33d R= Et, R' = tBu entry ketoester Lewis base (%) time (h) temp (°C) % conva % eeb 1 33c cinchonidine (10) 4 -40 98 41 2 33c cinchonidine (10) 1 -78 63 47 3 -78 60 47 3 c 33c cinchonidine (10) 3 -78 53 53

4 33c (DHQD)2AQN (5) 1 -40 89 64 3 -40 99 64 c 5 33c (DHQD)2AQN (5) 1 -40 54 69 3 -40 58 69

6 33c (DHQD)2AQN (5) 1 -78 71 68 3 -78 96 69 6 -78 99 68 7 33d cinchonidine (10) 3 -78 99 82 a Determined by 1H NMR. b Determined by chiral GC. cReaction run without MeOH.

4.2.1. Mechanistic Aspects From our results it can be concluded that the Lewis base is responsible for chirality induction. During the reactions it was observed that the ratio between the O-acylated product 34a and the non-protected cyanohydrin 35a increased with reaction time while the conversion remained constant. From this observation it is assumed that 35a is an intermediate in the formation of

35 product 34a. To elucidate if the addition of cyanide to the prochiral ketone or the acylation of the initially formed cyanohydrin is the enantiodetermining step in the reaction, the enantiomeric excess of 35a was determined from a mixture of 34a and 35a (35a was propionated to 36 to enable separation of the enantiomers, Scheme 24). While the enantiomeric excess of 34a was 53%, 36 was found to be racemic.

Most probably the reaction proceeds by a non-selective cyanation followed by a dynamic kinetic resolution to form the enantioenriched O-acyalted cyanohydrin. Deng and co-workers had previously proposed a similar mechanism in the cyanocarbonation of ketones.25b

Scheme 24. Acylation of 35a in presence of 34a.

O O

MeOCO CN HO CN O MeOCO CN EtOCO CN OMe OMe OMe OMe + Sc(OTf)3 + O O O O 34a 35a 34a 36

Based on these observations we propose the mechanism shown in Scheme 25. Acetyl cyanide (17a) reacts with methanol to liberate hydrogen cyanide, which is supposed to initiate the reaction. The chiral Lewis base (LB*) and 17a form 37 and cyanide. Racemic 35, which is in equilibrium with 33, is then acylated by 37 resulting in a dynamic kinetic resolution and formation of enantioenriched O-acylated cyanohydrin 34.

Scheme 25. Suggested mechanism for asymmetric acetylcyanation of α‐ketoesters.

O O MeOH HCN + CN OMe 17a

O -O CN NC O- OR OR OR + O O O 33 35

O O - + LB* CN + CN LB* 17a 37

enantioenriched 34

4.3. Acylcyanations of Methyl Ketones An obvious continuation of the studies on acylcyanations of α-ketoesters and aldehydes was to investigate asymmetric acylcyanation of methyl ketones. For the acetylcyanation in Scheme 26, we tested the dual activation protocol, the conditions used for α-ketoesters, a range of Lewis bases derived from tertiary amines as catalysts, and thiourea derived catalysts similar to those used by Jacobsen33 and List.67 Until today we have still not succeeded to carry out this reaction by any of these methods or catalysts.

Scheme 26. Failed acetylcyanation of methyl ketones.

O NC OCOMe

chiral catalyst 38a and/or 39a O Lewis base or + or O CN NC OCOMe

38b 39b

However, benzoylcyanations of 38a and b were possible under very specific conditions (Table 9). The reaction conditions are similar to the ones reported by Shi and co-workers in benzoylcyanation of cyclohexanone catalyzed by 30 mol % DBU.74 When a 1:1 or higher ratio between ketone and benzoyl cyanide (17h) was used in presence of 8 the reaction proceeded (entries 1-3, 5). The use of only Lewis base as catalyst resulted in low or no conversion (entries 4 and 6). An excess of 17a, relative to ketone, led to a dramatic decrease in reactivity (<11% conv, 48 h, entries 7-8). When chiral catalyst 8 was used product 39 was obtained as a racemate.

Table 9. Benzoylcyanation of ketones catalyzed by DBU and 8.

O O DBU (10 mol %) 8 (5 mol %) NC OCOPh + R Ph CN CH 2Cl2 R 38 17h rt 39 entry 38 (R) equiv 38 equiv. 17h time (h) % conva 1 a (Ph) 1 1 4 27

2 b (CH3(CH2)5) 1 1 4 73 3 a (Ph) 2 1 4 60 4b a (Ph) 2 1 4 25

5 b (CH3(CH2)5) 2 1 4 99 b 6 b (CH3(CH2)5) 2 1 4 0 7 a (Ph) 1 2 48 11 8b a (Ph) 1 2 48 7 a Determined by GC-MS. b 8 was not used.

We decided to test Shi’s reaction conditions as well. With benzoyl cyanide (17h) the reaction worked as expected, but exchange of 17h for acetyl cyanide (17a) resulted in no conversion.

4.4. Conclusions Asymmetric acetylcyanation of α-ketoesters has been performed using chiral Lewis bases as catalysts. Yields up to 77% and enantiomeric excesses of 82% were obtained. The reaction proceeds via dynamic kinetic resolution of the initially formed racemic cyanohydrin during the acetylation step, which is different from the mechanism for acylcyanation of aldehydes. Acetylcyanation of methyl ketones turned out to be considerably more difficult and no conversion was observed. However, benzoylcyanation of methyl ketones proceeded well when DBU together with 8 were used as catalysts and when the ketone was in excess.

5. Enzymatic Method for High‐Throughput Analysis of O‐Acylated Cyanohydrins

(Paper VI)

5.1. Introduction High-throughput methods are becoming standard techniques for the discovery, development, and optimization of new compounds, reactions, and reaction conditions.95 Synthesis of catalyst libraries and ligand optimization are a few examples in which combinatorial methods have been employed.96 When high- throughput techniques are applied in asymmetric catalysis, the screening of enantioselectivity is problematic since the analytical techniques using conventional chiral columns in GC or HPLC applications are too time- consuming for sequential high-throughput purposes.95 To solve this problem several new techniques have been developed, but none of these methods is general for broad ranges of substrates.97

5.1.1. High‐Throughput Methods for Determination of Enantiomeric Excess Reetz and co-workers pioneered the field of high-throughput methods for enantiomeric excess determination and they have developed several useful techniques. In 1998 one of the first high-throughput methods, which used IR thermography, was developed by Reetz et al.98 Several other techniques based on circular dichroism,99 mass spectrometry,100 fluorescence,101 capillary electrophoresis,102 color tests based on liquid crystals,103 and enantioselective indicator-displacement assays104 have since then been developed by the Reetz group and by others.

High-throughput methods using biocatalysis have been developed as well. Taran et al. used an antibody for fast screening of enantioselectivity in the synthesis of chiral α-hydroxy acids.105 Enzymatic methods for the determination of enantiomeric excess (EMDee), in which two enantiomers selectively are converted into chemically different species, thereby making it possible to measure the enantiomeric excess, have been employed in a few cases. Reetz screened the selectivity of lipases by measuring the absorption of p-nitrophenolate anion released upon hydrolysis of enantiopure R- and S- substrate.106 p-Nitrophenolate could also be used as a pH indicator, which made it possible to measure the amount acetic acid formed in a lipase- catalyzed hydrolysis of a scalemic mixture of allylic acetates.107

The enantiomeric excess of secondary alcohols was measured by Seto and co- workers in an alcohol dehydrogenase catalyzed selective oxidation of one of the enantiomers with NADP+. The formation of NADPH was followed by UV spectroscopy at 340 nm and from the rate of oxidation the enantioselectivity could be calculated.108 The same group also used alcohol dehydrogenase for determination of chiral sulfoxides by measuring the inhibition of ethanol oxidation catalyzed by the enzyme. The two enantiomers of the sulfoxides inhibited the enzyme with different rates.109 The formation of NADH or NADPH was measured by Witholt and co-workers who used two dehydrogenases for selective oxidation of enantiomeric alcohols. Thereby the concentration of each enantiomer could be determined and the enantiomeric excess and yield be calculated.110 Berkowitz used two dehydrogenases with different enantiomeric preferences for determining enantioselectivity and relative rates in a kinetic resolution.111

5.1.2. Aim of the Study The aim of this study was to develop an alternative method to GC and HPLC for determination of conversion and enantiomeric excess of O-acylated cyanohydrins. The goal was to obtain an efficient method in which several samples could be analyzed in a short period of time with high accuracy.

5.2. Determination of Enantiomeric Excess Using an Enzymatic Method with a pH Indicator We decided to use two enzymes together with a pH-indicator for determination of the enantiomeric excess of O-acylated cyanohydrins (Scheme 27). In the first step the selective enzyme Candida antarctica lipase B (CALB), which in this case hydrolyzed the S enantiomer, was used. In the second step the unselective enzyme pig liver esterase (PLE), which hydrolyzed the remaining R enantiomer, was added. The amount of acetic acid produced in each step was determined by titration with yellow p-nitrophenolate (yellow) to colorless p-nitrophenol; this indicator had previously been used in a similar way by Seto and co-workers.107 The enantiomeric excess was then calculated from recording the color change at 405 nm over the two steps.

Scheme 27. Determination of enantiomeric excess after protonation of p‐nitrophenolate (p‐NPO‐) at 405 nm.

O CALB OH a) + AcOH AcO- R CN R CN p-NPO- p-NPOH O

O PLE OH b) + AcOH AcO- R CN R CN p-NPO- p-NPOH

The correlation between GC determination and the enzymatic method for determination of enantiomeric excess is shown in Figure 11. As seen, the enzymatic method worked well, although slightly lower values were observed than with GC. When the analysis was performed on microtiter plates instead of in cuvettes the results were, however, difficult to reproduce, probably due to change of pH in the weak buffer. This change of pH may be caused by contamination by carbon dioxide from the atmosphere.

0 -75 -50 -25 0 25 50 75

EMDee (%) -25

-50

-75 GCee (%)

Figure 11. Enzymatic determination of enantiomeric excess as a function of the corresponding GC values. Positive values correspond to an excess of the R enantiomer and n to an excess of the S enantiomer. The dashed line corresponds to GCee = EMDee.

5.3. Determination of Enantiomeric Excess and Conversion Using an Enzymatic Method with NADH We have in collaboration with professor Karl Hult and his group designed a new method (Scheme 28) for determination of both conversion and enantiomeric excess of O-acetylated mandelonitrile (18a).112 The problems related to detection of protons in the enzymatic method using a pH indicator were solved by using this new method.

The method is outline in Scheme 28. Horse liver alcohol dehydrogenase (HLADH) and NADH are added to the reaction mixture containg O-acylated cyanohydrin (a, Scheme 28). The enzyme catalyzes the reduction of any remaining aldehyde with NADH, which will be oxidized to NAD+. In the second step CALB, which hydrolyzes the S enantiomer selectively, is added. The formed unprotected cyanohydrin is in equilibrium with aldehyde, which is reduced in accordance with the unreacted aldehyde. In the final step (c) unselective PLE is added and the remaining enantiomer is hydrolyzed followed by reduction of the formed aldehyde with NADH. Since NADH absorbs at 340 nm while NAD+ does not, the consumption of NADH in each step can be observed as a decrease in absorbance. From the change in absorbance in the three steps conversion and enantiomeric excess can be determined.

Scheme 28. Determination of enantiomeric excess and conversion after NADH reduction at 340 nm.

O OH a) HLADH R H R H NA DH, H + NAD+ H + O

O OH O OH b) CALB HLA DH H R CN R CN R H R NADH, H+ NA D+ H + O

O OH O c) PLE HLADH OH H R CN R CN R H R NADH, H+ NAD+ H

Eighteen structurally different O-acylated cyanohydrins were screened (Figure 11). For ten of the substrates both conversion and enantiomeric excess could be accurately determined (18a-e, 18g, 18i, 18l-m, and 18o), as illustrated in Figure 13 by comparison of values obtained from the enzymatic method with values obtained by GC for 18b-d.

O O O O O

CN CN CN CN CN O 18a 18b MeO 18c Cl 18d 18e #¤ #¤ #¤ #¤ #¤ O O O O O

O O O O O PhO CN CN CN CN CN N 18f N 18g 18h 18i 18j ¤ #¤ ¤ #¤ O O O O

O O O O

CN CN CN CN 18k 18l 18m 18n #¤ #¤ O O O O

O O O Ph O Ph

CN CN CN CN

18o 18p 18q 18r #¤ Figure 12. Different O‐acylated cyanohydrins analyzed with the EMDee using NADH. Enantiomeric excess was accurately determined for compounds labeled with # and conversion was accurately determined for compounds labeled with ¤.

For substrates 18f and 18h, the enantiomeric excess could not be determined. The method provided data which indicated a high excess of the S enantiomer. However, analysis by GC showed that these data were incorrect. One explanation could be racemization during the analysis. Both conversion and enantiomeric excess were difficult to determine for substrates 18j, 18k, 18n and 18p-r. For 18j, 18k, and 18n the rates of hydrolysis were very slow with both CALB and PLE, which made the analysis troublesome. For 18p-r no hydrolysis was observed with CALB, most probably due to the acyl groups being too large to fit in the pocket of the active site.

a) 100 b) 100 18b 18b 18c 50 75 18c 18d 18d

0 50 EMDee -100 -50 0 50 100 25 -50

Enzymatic conversion (%) conversion Enzymatic 0 -100 0 255075100 GCee GC conversion (%)

Figure 13. Comparison between the enzymatic method using NADH and GC for determination of enantiomeric excess (a) and conversion (b). Positive ee values correspond to an excess of the R enantiomer and negative values to an excess of the S enantiomer. The dashed line corresponds to GC values = EMD values.

Crude reaction samples of cyanohydrins, for which the EMDee worked well in cuvettes, were also analyzed on microtiter plates. The results were adequate and could be used for initial screening of reaction conditions.

5.4. Conclusions An enzymatic method for determination of conversion and enantiomeric excess of O-acylated cyanohydrins has been developed. Compounds obtained from aromatic aldehydes and acyl cyanides containing up to four carbon atoms could be accurately analyzed. The method can be used for high-throughput initial screening of reaction conditions.

6. Summary and Concluding Remarks

This thesis deals with the development of new methods for synthesis and analysis of O-functionalized cyanohydrins.

A method in which a chiral Lewis acidic Ti-salen complex and a Lewis base catalyzed the formation of O-alkoxycarbonylated and O-acylated cyanohydrins from aldehydes was developed. By employing this dual activation yields up to 97% and enantiomeric excesses up to 96% were obtained. When either the Lewis acid or the Lewis base was excluded low or no conversion was observed. A mechanism, in which cyanide was added to aldehyde within the dimeric Ti-salen complex, was proposed.

Close to enantiopure O-acylated cyanohydrins in up to 97% yield were obtained when the dual activation system together with CALB were employed in a biphasic system. In this minor enantiomer recycling method the enzyme hydrolyzed the minor product enantiomer, aldehyde was re-formed, and reacted again in the Lewis acid–Lewis base-catalyzed reaction. Application of this minor enantiomer recycling concept to other reactions would be appealing.

Acylcyanations of ketones were also studied. A method for asymmetric acetylcyanation of α-ketoesters catalyzed by chiral Lewis bases was developed and yields up to 77% and 82% ee were obtained. The reaction was found to proceed by a dynamic kinetic resolution, which is different from Lewis acid– Lewis base catalyzed reactions with aldehydes. Acylcyanations of methyl ketones were problematic. Reaction leading to racemic product was observed only with benzoyl cyanide as cyanide source. A general efficient catalytic system for asymmetric acylcyanations of ketones still remains to be developed.

An enzymatic method for determination of conversion and enantiomeric excess was described and was utilized for high-throughput screening. The method could be used for a number of O-acylated cyanohydrins derived from aromatic and aliphatic aldehydes, but is very specific for this class of compounds. However, modifications of the method are expected to lead to wider scope.

7. Acknowledgements

I would like to express my gratitude to:

Christina Moberg. Thank you for accepting me as PhD student, for your guidance the past five years, and for sharing your passion for chemistry.

The Swedish Foundation for Strategic Research (SSF) and KTH for funding.

The Aulin-Erdtman Foundation, the Knut and Alice Wallenberg Foundation, and the Foundation in Memory of Lars Gunnar Sillén for making it possible to present this work at several conferences.

My co-authors Christina Moberg, Karl Hult, Stina Lundgren, Maël Penhoat, Anders Hamberg, Fei Li, Khalid Widyan, Anna Laurell, and Linda Fransson for fruitful collaborations.

Torbjörn Norin, Ester Fjellander, Anna Laurell, and Brinton Seashore-Ludlow for valuable comments on this thesis.

Lena Skowron, Henry Challis, Ann Ekqvist, and Ulla Jacobsson for providing help with practical matters.

All past and present co-workers at the organic chemistry department and especially the Ki-group for creating a nice atmosphere, for good discussions, and for all the help in the lab. Special thanks to Stina Lundgren for a great, rewarding collaboration with the dual activation project and to Anna Laurell for taking the MER project to a higher level.

My friends for all great times. Mom, Dad, and my brother Dag for encouragement, support, and for always being around when I need you. Liselotte for your love, support and for making me smile.

8. References

1 http://pubs.acs.org/paragonplus/submission/joceah/joceah_abbreviations.pdf 2 Halpern, J.; Trost, B. M. Proc. Natl. Acad. Sci. USA 2004, 101, 5437. 3 Thayer, A. M. Chem. Eng. News 2007, 85, 11-19. 4 Lin, G.-Q.; Li, Y.-M.; Chan, A. S. C. Priciple and Applications of Asymmetric Synthesis; Wiley; New York, 2001; pp 52-53. 5 Blaser, H.-U.; Pugin, B.; Spindler, F. J. Mol. Cat. A: Chem. 2005, 231, 1-20. 6 For reviews of cyanohydrin synthesis, see: (a) Mowry, D. T. Chem. Rev. 1948, 42, 189-283. (b) North, M. Synlett 1993, 807-820. (c) Effenberger, F. Angew. Chem. Int. Ed,. Engl. 1994, 33, 1555-1564. (d) Gregory, R. J. H. Chem. Rev. 1999, 99, 3649-3682. (e) North, M. Tetrahedron: Asymmetry 2003, 14, 147-176. (f) Brunel, J.-M.; Holmes, I. P. Angew. Chem. Int. Ed. 2004, 43, 2752-2778. (g) North, M.; Usanov, D. L.; Young, C. Chem. Rev. 2008, 108, 5146-5226. 7 Winkler, F. W. Liebigs Ann. Chem. 1832, 4, 242 (as referred to in ref. 2a). 8 Scheele, K. W. Opuscula 1782, 2, 48 (as referred to in ref. 2a). 9 Wöhler, F. Liebigs. J. Ann. Chim. 1837, 22, 1 (as referred to in ref. 2g). 10 Lapworth, A. J. Chem. Soc. 1903, 995-1005. 11 Ching, W.-M.; Kallen, R. G. J. Am. Chem. Soc. 1978, 100, 6119-6124. 12 Zagrobelny, M.; Bak, S.; Rasmussen, A. V.; Jørgensen, B.; Naumann, C. M.; Møller, B. L. Phytochemistry 2004, 65, 293-306. 13 Rosenthaler, L. Biochem. Z. 1908, 14, 238-253. 14 Brown, R. F. C.; Donohue, A. C.; Jackson, W. R.; McCarthy, T. D. Tetrahedron 1994, 50, 13739-13752. 15 Effenberg, F.; Eichhorn, J. Tetrahedron: Asymmetry 1997, 8, 469-476. 16 Schwindt, M. A.; Belmont, D. T.; Carlson, M.; Franklin, L. C.; Hendrickson, V. S.; Karrick, G. L.; Poe, R. W.; Sobieray, D. M.; Van De Vusse, J. J. Org. Chem. 1996, 61, 9564-9568. 17 Monterde, M. I.; Brieva, R.; Gotor, V. Tetrahedron: Asymmetry 2001, 21, 525-528. 18 Gaucher, A.; Ollivier, J.; Salaün, J. Synlett 1991, 151-153. 19 Stelzer, U.; Effenberger, F. Tetrahedron: Asymmetry 1993, 4, 161-164. 20 (a) Shan, G.; Hammock, B. D. Anal. Biochem. 2001, 299, 54-62. (b) Beckmann, M.; Haack, K.-J. Chem. Unserer Zeit, 2003, 37, 88-97. (c) Wheelock, C. E.; Wheelock, Å. M. ; Zhang, R.; Stok, J. E.; Morisseau, C.; Le Valley, S. E.; Green, C. E.; Hammock, B. D. Anal. Biochem. 2003, 315, 208-222. (d) Huang, H.; Stok, J. E.; Stoutamire, D. W.; Gee, S. J.; Hammock, B. D. Chem. Res. Toxicol. 2005, 18, 516-527. 21 (a) Lidy, W.; Sundermeyer, W. Chem. Ber. 1973, 106, 587-593. (b) Evans, D. A.; Truesdale, L. K.; Carroll, G. L. J. Chem. Soc., Chem. Commun. 1973, 55-56. 22 A few of many examples are listed in references 23-26 and more examples can be found in the reviews listed in reference 6.

23 (a) Smitskamp-Wilms, E.; Brussee, J.; van der Gen, A.; Van Scharrenburg, G. J. M.; Sloothaak, J. B. Recl. Trav. Chim. Pays-Bas 1991, 110, 209-215. (b) Belokon’, Y. N.; Gutnov, A. V.; Moskalenko, M. A.; Yashkina, L. V.; Lesovoy, D. E.; Ikonnikov, N. S.; Larichev, V.; North, M. Chem. Commun, 2002, 244-245. (c) Belokon, Y. N.; Carta, P.; Gutnov, A. V.; Maleev, V.; Moskalenko, M. A.; Yashkina, L. V.; Ikonnikov, N. S.; Voskoboev, N. V.; Khrustalev, V. N.; North, M. Helv. Chim. Acta, 2002, 85, 3301- 3312. 24 (a) Kobayashi, Y.; Hayashi, H.; Miyaji, K.; Inoue, S. Chem. Lett. 1986, 931-934. (b) Mori, A.; Kinoshita, K.; Osaka, M.; Inoue, S. Chem. Lett, 1990, 1171-1172. (c) Ognyanov, V. I.; Datcheva, V. K.; Kyler, K. S. J. Am. Chem. Soc. 1991, 113, 6992- 6996. 25 (a) Okimoto, M.; Chiba, M. Synthesis, 1996, 1188-1190. (b) Tian, S.-K.; Deng, L. J. Am. Chem. Soc. 2001, 123, 6195-6196. 26 (a) Yoneda, R.; Harusawa, S.; Kurihara, T. J. Org. Chem, 1991, 56, 1827-1832. (b) Baeza, A.; Casas, J.; Nájera, C.; Sansano, J. M.; Saá, J. M. Angew. Chem. Int. Ed. 2003, 42, 3143-3146. 27 Reetz, M. T.; Kunisch, F.; Heitman, P. Tetrahedron Lett. 1986, 27, 4721-4724. 28 (a) Gröger, H. Adv. Synth. Catal. 2001, 343, 547-558. (b) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.; Stürmer, R.; Zelinski, T. Angew. Chem. Int. Ed. 2004, 43, 788-824. (c) García-Urdiales, E.; Alfonso, I.; Gotor, V. Chem. Rev., 2005, 105, 313-354. 29 Oku, J.-I.; Inoue, S. J. Chem. Soc., Chem. Commun. 1981, 229-230. 30 Wen, Y.; Huang, X.; Huang, J.; Xiong, Y, Qin, B.; Feng, X. Synlett 2005, 2445-2448. 31 Peng, D.; Zhou, H.; Liu, X.; Wang, L.; Feng, X. Synlett 2007, 2448-2450. 32 Chinchilla, R.; Nájera, C.; Ortega, F. J. Terahedron: Asymmetry 2008, 19, 265-268. 33 Fuerst, D. E.; Jacobsen, E. N. J. Am. Chem. Soc. 2005, 127, 8964-8965. 34 For a recent excellent review, see reference 6g. 35 Khan, N. H.; Kureshy, R. I.; Abdi, S. H. R.; Agrawal, S.; Jasra, R. V. Coord. Chem. Rev. 2008, 252, 593-623. 36 Hayashi, M.; Miyamoto, Y.; Inoue, T.; Oguni, N. J. Org. Chem. 1993, 58, 1515- 1522. 37 Chang, C.-W.; Yang, C.-T.; Hwang, C.-D.; Uang, B.-J. Chem. Commun. 2002, 54-55. 38 Lundgren, S.; Lutsenko, S.; Jönsson, C, Moberg, C. Org. Lett. 2003, 5, 3663-3665. 39 Hamashima, Y.; Sawada, D.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 2641-2642. 40 Yoon, T. P.; Jacobsen, E. N. Science 2003, 299, 1691-1693. 41 For the first examples see: (a) Pan, W.; Feng, X.; Gong, L.; Hu, W.; Li, Z.; Mi, A.; Jiang, Y. Synlett 1996, 337-338. (b) Belokon’, Y.; Ikonnikov, N.; Moscalenko, M.; North, M.; Orlova, S.; Tararov, V.; Yashkina, L. Tetrahedron: Asymmetry 1996, 7, 851- 855. 42 For the first example see: Belokon’, Y. N.; North, M.; Parsons, T. Org. Lett. 2000, 2, 1617-1619. 43 For the first example see: Kim, S. S.; Song, D. H. Eur. J. Org. Chem. 2005, 1777- 1780.

44 For the first example see: Kim, S. S.; Lee, S. H. Synth. Commun. 2005, 35, 751-759. 45 Belokon’, Y. N.; Caveda-Cepas, S.; Green, B.; Ikonnikov, N. S.; Khrustalev, V. N.; Larichev, V. S.; Moscalenko, M. A.; North, M.; Orizu, C.; Tararov, V. I.; Tasinazzo, M.; Timofeeva, G. I.; Yashkina, L. V. J. Am. Chem. Soc. 1999, 121, 3968-3973. 46 See Ref. 6g and references therein. 47 (a) Steinhagen, H.; Helmchen, G. Angew. Int. Ed. Engl. 1996, 35, 2339-2342. (b) Ma, J.-A.; Cahard, D. Angew. Chem. Int. Ed. 2004, 43, 4566-4583. 48 Corey, E. J.; Wang, Z. Tetrahedron Letters, 1993, 34, 4001-4004. 49 Casas, J.; Nájera, C.; Sansano, J. M.; Saá, J. M. Org. Lett. 2002, 4, 2589-2592. 50 Zeng, B.; Zhou, X.; Liu, X.; Feng, X. Tetrahedron, 2007, 63, 5129-5136. 51 Belda, O.; Duquesne, S.; Fischer, A.; Moberg, C. J. Organometal. Chem. 2004, 689, 3750-3755. 52 Wingstrand, E.; Lundgren, S.; Penhoat, M.; Moberg, C. Pure Appl. Chem. 2006, 78, 409-414. 53 Belokon’, Y. N.; Blacker, A. J.; Carta, P.; Clutterbuck, L. A.; North, M. Tetrahedron 2004, 60, 10433-10447. 54 Deardoff, D. R.; Taniguchi, C. M.; Tafti, S. A.; Kim, H. Y.; Choi, S. Y.; Downey, K. J.; Nguyen, T. V. J. Org. Chem. 2001, 66, 7191-7194. 55 (a) Tian, J.; Yamagiwa, N.; Matsunaga, S.; Shibasaki, M. Angew. Chem. Int. Ed. 2002, 41, 3636-3638. (b) Casas, J.; Baeza, A.; Sansano, J. M.; Nájera, C.; Saá, J. M. Tetrahedron: Asymmetry 2003, 14, 197-200. (c) Tian, J.; Yamagiwa, N.; Matsunaga, S.; Shibasaki, M. Org. Lett. 2003, 5, 3021-3024. (d) Yamagiwa, N.; Tian, J.; Matsunaga, S.; Shibasaki, M. J. Am. Chem. Soc. 2005, 127, 3413-3422. 56 Belokon’, Y. N.; Blacker, A. J.; Clutterbuck, L. A.; North, M. Org. Lett. 2003, 5, 4505-4507. 57 Veum, L.; Pereira, S. R. M.; van der Waal, J. C.; Hanefeld, U. Eur. J. Org. Chem. 2006, 1664-1671. 58 Marvel, C. S.; Brace, N. O.; Miller, F. A.; Johnson, A. R. J. Am. Chem. Soc. 1949, 71, 34-36. 59 Watahiki, T.; Ohba, S.; Oriyama, T. Org. Lett. 2003, 5, 2679-2681. 60 Hoffman, H. M. R.; Ismail, Z. M.; Hollweg, R.; Zein, A. R. Bull. Chem. Soc. Jpn. 1990, 63, 1807-1810. 61 Scholl, M.; Lim, C.-K.; Fu, G. C. J. Org. Chem. 1995, 60, 6229-6231. 62 Khan, N. H.; Agrawal, S.; Kureshy, R. I.; Abdi, S. H. R.; Mayani, V. J.; Jasra, R. V. Eur. J. Org. Chem. 2006, 3175-3180. 63 (a) Veum, L.; Kanerva, L. T.; Halling, P. J.; Maschmeyer, T.; Hanefeld, U. Adv. Synth. Catal. 2005, 347, 1015-1021. (b) Veum, L.; Hanefeld, U. Synlett 2005, 2382- 2384. (c) Sakai, T.; Wang, K.; Ema, T. Tetrahedron 2008, 64, 2178-2183. 64 Procedures partly taken from: Roth, P.; Hädener, A.; Tamm, C. Helv. Chim. Acta 1990, 73, 476-482. 65 Normant, J. F.; Peichucki, C. Bull. Soc. Chim. Fr. 1972, 2402-2403. 66 Baeza, A.; Nájera, C.; Sansano, J. M.; Saá, J. M. Tetrahedron: Asymmetry 2005, 16, 2385-2389.

67 (a) Pan, S. C.; Zhou, J.; List, B. Synlett 2006, 3275-3276. (b) Pan, S. C.; Zhou, J.; List, B. Angew. Chem. Int. Ed. 2007, 46, 612-614. (c) Pan, S. C.; List, B. Synlett 2007, 318-320. (d) Pan, S. C.; List, B. Org. Lett. 2007, 9, 1149-1151. (e) Pan, S. C.; List, B. Chem. Asian J. 2008, 3, 430-437. 68 (a) Liu, W.; Gan, J.; Schlenk, D.; Jury, W. A. Proc. Natl. Acad. Sci. USA 2005, 102, 701-706. (b) Qin, S.; Budd, R.; Bondarenko, S.; Liu, W., Gan, J. J. Agric. Food Chem. 2006, 54, 5040-5045. 69 Haak, K.-J. Chem. Unserer Zeit 2003, 37, 128-138. 70 Glufosinate was sold by Bayer Crop Science for a value of 237 million euro worldwide in 2007 (information from www.bayercropscience.com). 71 (a) Peterson, C. J.; Tsao, R.; Eggler, A. L.; Coats, J. R. Molecules 2000, 5, 648-654. (b) Park, D.-S.; Coats, J. R. J. Pestic. Sci. 2005, 30, 99-102. 72 (a) Belokon’, Y. N.; Green, B.; Ikonnikov, N. S.; Larichev, V. S.; Lokshin, B. V.; Moscalenko, M. A.; North, M.; Orizu, C.; Peregudov, A. S.; Timofeeva, G. I. Eur. J. Org. Chem. 2000, 2655-2661. (b) Belokon, Y. N.; Clegg, W.; Harrington, R. W.; Young, C.; North, M. Tetrahedron 2007, 63, 5287-5299. 73 Tate, B. E.; Bartlett, P. D. J. Am. Chem. Soc. 1956, 78, 5575-5580. 74 Zhang, W.; Shi, M. Org. Biomol. Chem. 2006, 4, 1671-1674. 75 (a) Patel, R. N. Coord. Chem. Rev. 2008, 252, 659-701. (b) Catalytic Asymmetric Synthesis (Ed.: Ojima, I.), Wiley, New York, 2000. (c) Comprehensive Asymmetric Catalysis (Eds.: Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H.) Springer, Berlin, 1999. (d) Dalko, P. I., Moisan, L. Angew. Chem. Int. Ed. 2004, 43, 5138-5175. (e) Hawkins, J. M.; Watson, T. J. N. Angew. Chem. Int. Ed. 2004, 43, 3224-3228. 76 (a) Asymmetric Organic Synthesis with Enzymes (Eds.: Gotor, V.; Alfonso. I.; García- Urdiales, E.) Wiley-VCH, Weinheim, 2008. (b) Pellissier, H. Tetrahedron 2008, 64, 1563-1601. (c) Pàmies, O.; Bäckvall, J.-E. Chem. Rev. 2003, 103, 3247-3262. 77 Steinreiber, J.; Faber, K.; Griengl, H. Chem. Eur. J. 2008, 14, 8060-8072. 78 (a) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Chemistry; Wiley; New York, 1994; pp 316-317. (b) Anderson, N. G. Org. Process. Res. Dev. 2005, 9, 800-813. 79 (a) Beard, T. M.; Turner, N. J. Chem. Commun. 2002, 246-247. (b) Alexandre, F.-R.; Pantaleone, D. P.; Taylor, P. P.; Fotheringham, I. G.; Ager, D. J.; Turner, N. J. Tetrahedron Letters 2002, 43, 707-710. (c) Carr, R.; Alexeeva, M.; Enright, A.; Eve, T. S. C.; Dawson, M. J.; Turner, N. J. Angew. Chem. Int. Ed. 2003, 42, 4807-4810. (d) Dunsmore, C. J.; Carr, R.; Fleming, T.; Turner, N. J. J. Am. Chem. Soc. 2006, 128, 2224-2225. 80 (a) Hanefeld, U.; Li, Y.; Sheldon, R. A.; Maschmeyer, T. Synlett 2000, 1775-1776. (b) Hanefeld, U.; Straathof, A. J. J.; Heijnen, J. J. J. Mol. Catal. B. Enzym. 2001, 11, 213-218. (c) Veum, L.; Kuster, M.; Telalovic, S.; Hanefeld, U.; Maschmeyer, T. Eur. J. Org. Chem. 2002, 1516-1522. 81 The reactions were carried out with 13 (1 equiv), 17 (3 equiv), ent-8 (5 mol %), DBU (20 mol %), and immobilized CALB (10 mg) in toluene/phosphate buffer pH 8 at 40 °C. Yields were determined by GC with n-undecane as internal standard. Enantiomeric excess was determined with chiral GC. 82 (a) Corey, E. J.; Guzman-Perez, A. Angew. Chem. Int. Ed. 1998, 37, 388-401. (b) Christoffers, J.; Mann, A. Angew. Chem. Int. Ed. 2001, 40, 4591-4597. (c) Christoffers, J.; Baro, A. Adv. Synth. Catal. 2005, 347, 1473-1482.

83 Effenberger, F.; Hörsch, B.; Weingart, F.; Ziegler, T.; Kühner, S. Tetrahedron Letters 1991, 32, 2605-2608. 84 Effenberger, F.; Heid, S. Tetrahedron: Asymmetry 1995, 6, 2945-2952. 85 Choi, M. C. K.; Chan, S. S.; Matsumoto, K. Tetrahedron Letters 1997, 38, 6669- 6672. 86 (a) Belokon, Y. N.; Green, B.; Ikonnikov, N. S.; North, M.; Tararov, V. I. Tetrahedron Letters 1999, 40, 8147-8150. (b) Belokon, Y. N.; Green, B.; Ikonnikov, N. S.; North, M.; Parsons, T.; Tararov, V. I. Tetrahedron 2001, 57, 771-779. 87 (a) Hamashima, Y.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2000, 122, 7412- 7413. (b) Yabu, K.; Masumoto, S.; Yamasaki, S.; Hamashima, Y.; Kanai, M.; Du, W.; Curran, D. P.; Shibasaki, M. J. Am. Chem. Soc. 2001, 123, 9908-9909. 88 Deng, H.; Isler, M. P.; Snapper, M. L.; Hoveyda, A. H. Angew. Chem. Int. Ed. 2002, 41, 1009-1012. 89 Ryu, D. H.; Corey, E. J. J. Am. Chem. Soc. 2005, 127, 5384-5387. 90 Liu, X.; Qin, B.; Zhou, X.; He, B.; Feng, X. J. Am. Chem. Soc. 2005, 127, 12224- 12225. 91 Tian, S.-K.; Hong, R.; Deng, L. J. Am. Chem. Soc. 2003, 125, 9900-9901. 92 Baeza, A.; Nájera, C.; de Gracia Retamosa, M.; Sansano, J. M. Synthesis 2005, 2787- 2797. 93 Testa, E.; Fontanella, L.; Cristiani, G. F.; Mariani, L. Ann. 1961, 639, 166-180. 94 Citterio, A.; Fancelli, D.; Santi, R.; Pagani, A.; Bonsignore, S. Gazz. Chim. Ital. 1988, 118, 405-408. 95 Maier, W. F.; Stöwe, K.; Sieg, S. Angew. Chem. Int. Ed. 2007, 46, 6016-6067. 96 (a) Shimizu, K. D.; Snapper, M. L.; Hoveyda, A. H. Chem. Eur. J. 1998, 4, 1885- 1889. (b) Reetz, M. T. Angew. Chem. Int. Ed. 2001, 40, 284-310. (c) Reetz, M. T. Angew. Chem. Int. Ed. 2002, 41, 1335-1338. (d) Gennari, C.; Piarulli, U. Chem. Rev. 2003, 103, 3071-3100. 97 Tsukamoto, M.; Kagan, H. B. Adv. Synth. Catal. 2002, 344, 453-463. 98 Reetz, M. T.; Becker, M. H.; Kühling, K. M.; Holzwarth, A. Angew. Chem. Int. Ed. 1998, 37, 2647-2650. 99 (a) Ding, K.; Ishii, A.; Mikami, K. Angew. Chem. Int. Ed. 1999, 38, 497-501. (b) Yamauchi, M.; Mawatari, K.; Hibara, A.; Tokeshi, M.; Kitamori, T.; Anal. Chem. 2006, 78, 2646-2650. (c) Nieto, S.; Lynch, V. M.; Anslyn, E. V.; Kim, H.; Chin, J. J. Am. Chem. Soc. 2008, 130, 9232-9233. 100 (a) Guo, J.; Wu, J.; Siuzdak, G.; Finn, M. G. Angew. Chem. Int. Ed. 1999, 38, 1755- 1758. (b) Reetz, M. T.; Becker, M. H.; Klein, H.-W.; Stöckigt, D. Angew. Chem. Int. Ed. 1999, 38, 1758-1761. 101 (a) Copeland, G. T.; Miller, S. J. J. Am. Chem. Soc. 1999, 121, 4306-4307. (b) Klein, G.; reymond, J.-L. Helv. Chim. Acta 1999, 82, 400-407. (c) Korbel, G. A.; Lalic, G.; Shair, M. D. J. Am. Chem. Soc. 2001, 123, 361-362. (d) Matsushita, M.; Yoshida, K.; Yamamoto, N.; Wirsching, P.; Lerner, R. A.; Janda, K. D. Angew. Chem. Int. Ed. 2003, 43, 5984-5987.

102 (a) Reetz, M. T.; Kühling, K. M.; Deege, A.; Hinrichs, H.; Belder, D. Angew. Chem. Int. Ed. 2000, 39, 3891-3893. (b) Ludwig, M.; Kohler, F.; Belder, D. Electrophoresis 2003, 24, 3233-3238. 103 (a) van Delden, R. A.; Feringa, B. L. Angew. Chem. Int. Ed. 2001, 40, 3198-3200. (b) Eelkema, R.; van Delden, R. A.; Feringa, B. L. Angew. Chem. Int. Ed. 2004, 43, 5013-5016. 104 (a) Zhu, L.; Anslyn, E. V. J. Am. Chem. Soc. 2004, 126, 3676-3677. (b) Leung, D.; Anslyn, E. V. J. Am. Chem. Soc. 2008, 130, 12328-12333. 105 Taran, F.; Gauchet, C.; Mohar, B.; Meunier, S.; Valleix, A.; Renard, P. Y.; Créminon, C.; Grassi, J.; Wagner, A.; Mioskowski, C. Angew. Chem. Int. Ed. 2002, 41, 124-127. 106 Reetz, M. T.; Zonta, A.; Schimossek, K.; Liebeton, K.; Jaeger, K.-E. Angew. Chem. Int. Ed. 1997, 36, 2830-2832. 107 Onaran, M. B.; Seto, C. T. J. Org. Chem. 2003, 68, 8136-8141. 108 Abato, P.; Seto, C. T. J. Am. Chem. Soc. 2001, 123, 9206-9207. 109 Sprout, C. M.; Seto, C. T. Org. Lett. 2005, 7, 5099-5102. 110 Li, Z.; Bütikofer, L.; Witholt, B. Angew. Chem. Int. Ed. 2004, 43, 1698-1702. 111 Dey, S.; Karukurichi, K. R.; Shen, W.; Berkowitz, D. B. J. Am. Chem. Soc. 2005, 127, 8610-8611. 112 Hamberg, A.; Lundgren, S.; Penhoat, M.; Moberg, C.; Hult, K. J. Am. Chem. Soc. 2006, 128, 2234-2235.