Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan i Stockholm

framlägges till offentlig granskning för avläggande av doktorsexamen i kemi fredagen

den 2 oktober kl 13.00 i sal F3, KTH, Lindstedtsvägen 26, Stockholm. Avhandlingen

försvaras på engelska. Fakultetsopponent är Professor Frank Glorius, Westfälische

Wilhelms-Universität Münster.

Enantioenriched Cyanohydrins and

Acetoxyphosphonates – Synthesis and

Applications

Robin Hertzberg

Doctoral Thesis

Stockholm 2015

ISBN 978-91-7595-627-5

ISSN 1654-1081

TRITA-CHE-Report 2015:31

© Robin Hertzberg, 2015

Universitetsservice US AB, Stockholm

Robin Hertzberg, 2015: ”Enantioenriched Cyanohydrins and

Acetoxyphosphonates – Synthesis and Applications”, KTH Chemical Science

and Engineering, Royal Institute of Technology, SE-100 44 Stockholm,

Sweden.

Abstract

In this thesis, the synthesis of enantioenriched compounds using novel

methodologies that employ metal- and biocatalysis is described.

In the first part, the synthesis of enantioenriched cyanohydrins, which are

highly versatile synthetic intermediates, is described. A minor enantiomer

recycling methodology, which uses a catalytic system consisting of a titanium

salen dimer and a lipase, was highly successful in yielding the desired

products, often in essentially enantiopure form. Alternatively, when the minor

enantiomer recycling method gave unsatisfactory results, the same two

catalysts were used in a sequential two-step process. The minor enantiomer

recycling procedure was used to synthesize three different β-adrenergic

antagonists with very high enantiomeric excesses via the corresponding O-

acetylated cyanohydrins. With the same cyclic process, O-(α-bromoacyl)

cyanohydrins were synthesized and subsequently transformed to

aminofuranones via an intramolecular Blaise reaction. In addition, substitution

of the bromide in the O-(α-bromoacyl) cyanohydrins with different nitrogen

nucleophiles followed by reduction gave N-substituted β-amino alcohols. This

reaction sequence was applied to the synthesis of the β3-adrenergic receptor

agonist solabegron. Finally, the O-(α-bromoacyl) cyanohydrins were subjected

to a palladium catalyzed cross-coupling with a range of boronic acids. This

reaction proceeded with high yields, and was performed with enantiopure

substrates with no or only minor racemization of the resulting products.

In the second part, the first asymmetric direct addition of acylphosphonates to

aldehydes is described. This transformation is catalyzed by a tridentate Schiff

base aluminum(III) Lewis acidic complex, a Lewis base, and a Brønstedt base.

Several aromatic and aliphatic acetoxyphosphonates were isolated, in most

cases in high yields. Unfortunately, the enantioselectivity in the reaction was

only moderate. Therefore, an investigation to develop a minor enantiomer

recycling system for the synthesis of acetoxyphosphonates was initiated, but a

working cyclic process could not be found in this work.

Keywords: asymmetric synthesis, cyanohydrins, acetoxyphosphonates,

biocatalyst, titanium catalyst, minor enantiomer recycling, cross-coupling,

amino alcohols

Abbreviations

Ac Acetyl

ATHase Artificial transfer hydrogenase

aq. Aqueous

BINOL 1,1'-Bi-2-naphthol

CALA Candida antarctica lipase A

CALB Candida antarctica lipase B

CRL Candida rugosa lipase

dba Dibenzylideneacetone

ee Enantiomeric excess

GC Gas chromatography

GC-MS Gas chromatography–mass spectrometry

DABCO 1,4-Diazabicyclo[2.2.2]octane

DBU 1,8-Diazabicyclo[5.4.0]undec-7-ene

DCM Dichloromethane

DIPEA N,N-Diisopropylethylamine

DMAP 4-(Dimethylamino)pyridine

DKR Dynamic kinetic resolution

DPEPhos (Oxydi-2,1-phenylene)bis(diphenylphosphine)

DYKAT Dynamic kinetic asymmetric transformation

equiv Equivalents

GABA γ-Aminobutyric acid

h Hours

HIV Human immunodeficiency virus

HNL Hydroxynitrile Lyase

HOMO Highest occupied molecular orbital

HPLC High performance liquid chromatography

IBS Irritable bowel syndrome

KR Kinetic resolution

LUMO Lowest unoccupied molecular orbital

MAO-N Monoamine oxidase

MER Minor enantiomer recycling

MTBE tert-Butyl methyl ether

NHC N-Heterocyclic carbene

NMR Nuclear magnetic resonance

Nu Nucleophile

OAB Overactive bladder

PPL Lipase from porcine pancreas

PSL Pseudomonas cepacia lipase

r.t. Room temperature

temp Temperature

THF Tetrahydrofuran

Tf Trifluoromethanesulfonyl

TMS Trimethylsilyl

Ts p-Toluenesulfonyl

WHO World health organization

List of Publications

This thesis is based on the following papers, referred to in the text by Roman

numerals I-VI:

I. Enantioenriched ω-Bromocyanohydrin Derivatives. Improved

Selectivity by Combination of Two Chiral Catalysts Robin Hertzberg, Khalid Widyan, Berenice Heid, and Christina Moberg

Tetrahedron 2012, 68, 7680–7684

II. One-Step Preparation of O-(α-Bromoacyl) Cyanohydrins by Minor

Enantiomer Recycling: Synthesis of 4-Amino-2(5H)-furanones Robin Hertzberg, and Christina Moberg

J. Org. Chem. 2013, 78, 9174−9180

III. Minor Enantiomer Recycling: Application to Enantioselective

Syntheses of Beta Blockers Ye-Qian Wen, Robin Hertzberg, Inanllely Gonzalez, and Christina

Moberg

Chem. - Eur. J. 2014, 20, 3806–3812

IV. Enantioselective Acylphosphonylation - Dual Lewis Acid-Lewis

Base Activation of Aldehyde and Acylphosphonate Ye-Qian Wen

†, Robin Hertzberg

†, and Christina Moberg

J. Org. Chem. 2014, 79, 6172−6178 †

Authors contributed equally

V. Synthesis of the β3-Adrenergic Receptor Agonist Solabegron and

Analogous N-(2-Ethylamino)-β-amino Alcohols from O-Acylated

Cyanohydrins − Expanding the Scope of Minor Enantiomer

Recycling Robin Hertzberg, Guillermo Monreal Santiago, and Christina Moberg

J. Org. Chem. 2015, 80, 2937−2941

VI. Palladium-Catalyzed C(sp3)-C(sp

2) Cross-Couplings Between O-(α-

Bromoacyl) Cyanohydrins and Boronic Acids Robin Hertzberg and Christina Moberg

Manuscript

Paper not included in this thesis:

VII. Dual Lewis Acid-Lewis Base Catalyzed Acylcyanation of Aldehydes.

A Mechanistic Study Anna Laurell Nash, Robin Hertzberg, Ye-Qian Wen, Björn Dahlgren,

Tore Brinck, and Christina Moberg

Manuscript

Table of Contents

Abstract

Abbreviations

List of Publications

1. Introduction ..................................................................................... 1 1.1. Strategies for Obtaining Enantiopure Compounds ................................ 1

1.1.1. Kinetic Resolution ..................................................................................... 2 1.1.2. Sequential Processes ............................................................................... 4 1.1.3. Dynamic Kinetic Resolution ...................................................................... 4 1.1.4. Minor Enantiomer Recycling ..................................................................... 6

1.2. Aim of this Thesis .................................................................................. 9 2. Synthesis and Applications of O-Acylated Cyanohydrins. ............ 11

2.1. Introduction .......................................................................................... 11 2.1.1. The Importance of Cyanohydrins .............................................................11 2.1.2. Strategies for the Preparation of Enantioenriched Cyanohydrins ..............13

2.1.2.1. Biocatalytic Methods for the Preparation of Cyanohydrins ................... 13 2.1.2.2. Cooperative Catalytic Systems for the Preparation of Cyanohydrins ... 15 2.1.2.3. Minor Enantiomer Recycling for the Preparation of Cyanohydrins ....... 20

2.2. Synthesis of Enantioenriched ω-Bromocyanohydrin Derivatives ......... 22 2.2.1. Introduction ..............................................................................................22 2.2.2. Minor Enantiomer Recycling ....................................................................22 2.2.3. Sequential Process ..................................................................................26 2.2.4. Conclusions .............................................................................................27

2.3. Application of Minor Enantiomer Recycling in the Synthesis of β-

Adrenergic Antagonists ................................................................................... 27 2.3.1. Introduction ..............................................................................................27 2.3.2. Synthesis of (S)-Propranolol ....................................................................28 2.3.3. Synthesis of (R)-Dichloroisoproterenol .....................................................31 2.3.4. Synthesis of (R)-Pronethalol ....................................................................31 2.3.5. Conclusions .............................................................................................32

2.4. Synthesis and Applications of O-(α-Bromoacyl) Cyanohydrins ............ 32 2.4.1. Introduction ..............................................................................................32 2.4.2. Synthesis of O-(α-Bromoacyl) Cyanohydrins and Subsequent

Transformation to Aminofuranones ............................................................................33 2.4.2.1. Minor Enantiomer Recycling ............................................................... 35 2.4.2.2. Cyclizations to Aminofuranones .......................................................... 39

2.4.3. Functionalization of O-(α-Bromoacyl) Cyanohydrins ................................40 2.4.3.1. Synthesis of Solabegron ..................................................................... 40 2.4.3.2. C(sp

3)-C(sp

2) Cross-Coupling Between O-(α-Bromoacyl)

Cyanohydrins and Boronic Acids........................................................................... 45 2.4.4. Conclusions .............................................................................................52

3. Enantioselective Acylphosphonylation.......................................... 53 3.1. Introduction .......................................................................................... 53 3.2. Dual Activation in the Synthesis of Acetoxyphosphonates .................. 54 3.3. Minor Enantiomer Recycling ................................................................ 62 3.4. Conclusions ......................................................................................... 64

4. Concluding Remarks .................................................................... 65

Acknowledgements

Appendix

References

1

1. Introduction

Chiral molecules are essential for life on earth. Most of these molecules consist

of single enantiomers. The amino acids that build up all natural enzymes,

which catalyze chemical reactions taking place in the human body, consist

solely of L-amino acids. Organic compounds which consist of one single

stereoisomer have therefore an increasingly important role in the

pharmaceutical and agricultural industries. A large part of the pharmaceutical

drugs sold today contain one or more stereocenters, but several of these are

sold as mixtures of stereoisomers. Since the drug targets most often are

enzymes, the effect of different stereoisomers of the drug can differ to a large

extent.1 Stereoisomers of one drug might not only have different activity, but

can also have different modes of action; one stereoisomer might even be toxic

while the other gives the desired effect.2,3

The β-adrenergic antagonist propranolol (1) is an example of a chiral drug in

which the stereochemistry is vital for its biological activity (Figure 1).

Propranolol is used to treat numerous disorders and is listed in the WHO

model list of essential medicines.4 Despite the fact that the (S)-enantiomer of

this compound is over 100 times more active than the (R)-enantiomer,

propranolol is sold as a racemate, i.e. an equal mixture of both enantiomers.3,5

Another example is the pyrethroid insecticide esfenvalerate (2), which is the

(S,S)-isomer of fenvalerate and is responsible for the majority of the

insecticidal activity (Figure 1).6

Figure 1. (S)-Propranolol and esfenvalerate

1.1. Strategies for Obtaining Enantiopure Compounds

The transformation of prochiral starting material to enantioenriched product

with the aid of a chiral reagent can result in high enantiomeric excess (ee)

(Figure 2), but if the selectivity is not sufficient, separation of the unwanted

stereoisomer is necessary. The separation can be accomplished by

chromatographic techniques or by crystallization. The obvious disadvantage

with these separations is that the minor stereoisomer is taken out of the system,

thus unavoidably resulting in a reduced total yield.

2

Figure 2. Transformation of a prochiral compound (A) to enantioenriched product

(PR)

Systems that employ more than one catalyst are in many cases favorable over

single catalytic systems; fewer synthetic steps might be necessary, which

results in lower amounts of produced waste.7,8

In addition, cooperative

catalytic systems can result in higher enantiopurity of the desired products

since a reinforcing effect is obtained when two enantioselective catalysts work

in concert.9

1.1.1. Kinetic Resolution

In a kinetic resolution (KR) (Figure 3), two enantiomers of a substrate (SR and

SS) undergo a transformation with different rates under the influence of a chiral

catalyst. In the ideal case the selectivity is very high for one of the starting

enantiomers and very low for the other enantiomer (kR >> kS), such that at 50%

conversion only one of the enantiomers has been transformed to product (PR).10

Figure 3. Kinetic resolution (KR)

The stereoselectivity factor s is defined as the ratio between the rate constants

for the formation of the enantiomeric products (kR/kS). The stereoselectivity

factor can be calculated from the formulas below by using either the ee of the

substrate or the ee of the product. The stereoselectivity factor s is used for

chemocatalyzed reactions while the biochemical stereoselectivity factor, also

known as the “enantiomeric ratio”, E is often used for biocatalyzed reactions.

The equations are identical for both s and E (c is the conversion):10-13

Substrate: 𝑠=ln[(1-c)(1-ee(S))]

ln[(1-c)(1+ee(S))] Product: s=

ln[1-c(1+ee(P))]

ln[1-c(1-ee(P))]

As long as the stereoselectivity factor is >1 and the reaction is first order or

pseudo-first order in the substrate, any enantiomeric purity of the substrate can

be obtained by driving the reaction to the appropriate conversion. Thus, if the

selectivity of the KR is low, the yield of the recovered substrate will also be

low at high enantiomeric purity. The product ee is however limited to the

selectivity of the catalyst and has its maximum at the initial part of the

resolution; as the conversion increases the ee of the product decreases.9

3

The use of biocatalysts for kinetic resolution can offer advantages over the use

of chemocatalysts, such as high enantioselectivity at ambient temperature. In

addition, directed evolution gives access to enzymes with broad substrate

tolerability.14

Enzymatic KR is used in several industrial processes,15,16

for

example in the synthesis of the GABA analogue pregablin (5), which was

developed for treatment of central nervous system disorders. Several enzymes

were tested for the hydrolysis of the diester rac-3 and the lipase Lipolase was

selected as it showed an excellent E value (>200). The resolution was possible

to perform at a ton scale and gave the desired product (S)-4 with >98% ee at

45-50% conversion (Scheme 1). The unreacted enantiomer (R)-3 could be

racemized in a separate step in the presence of a strong base, thus allowing a

second KR and thereby resulting in a higher total yield.17

Scheme 1. Kinetic resolution in the synthesis of pregablin

Although large focus has been on enzymatic kinetic resolution, there have also

been several reports on the use of other catalytic systems.18

Fu and coworkers

have used planar-chiral DMAP derivatives as catalysts for the resolution of

alcohols and amines, often with high selectivity.19

Catalyst 6 was used in a

kinetic resolution to obtain an intermediate for the synthesis of epothilone A.

The undesired enantiomer (S,S)-7 was selectively acetylated to compound

(S,S)-8, leaving (R,R)-7 with 98% ee in 47% yield, corresponding to a

stereoselectivity factor of 107 (Scheme 2).20

Scheme 2. Kinetic resolution using a planar-chiral DMAP derivative

4

1.1.2. Sequential Processes

By combining an asymmetric transformation of a prochiral substrate with a

KR, higher yield of enantiopure product compared to only a KR is possible.

When a scalemic mixture of product enantiomers is obtained after an

enantioselective reaction, the minor enantiomer is selectively transferred to a

component which can be separated from the desired product (Figure 4).21,22

The two transformations can either be performed by the same catalyst or by

two different catalysts.9

Figure 4. Sequential process

Spilling and co-workers reported the enantioselective synthesis of α-

hydroxyphosphonates using a metal-catalyzed reaction between an aldehyde

and dimethylphosphite (Scheme 3). Either the minor or the major enantiomer

of the acetylated hydroxyphosphonate was then hydrolyzed by lipases in order

to increase the ee of the desired product.23

Scheme 3. Sequential process in the synthesis of α-hydroxyphosphonates

1.1.3. Dynamic Kinetic Resolution

By combining a kinetic resolution with a simultaneous secondary process

where the substrate is continuously racemized, a theoretical yield of 100% is

possible to obtain. As the faster-reacting and slower-reacting enantiomers are

present in equal amounts, and the resolution step is practically irreversible, the

ee of the product is constant throughout the process (Figure 5).11,12

The

selectivity in a dynamic kinetic resolution (DKR) is limited to the selectivity of

the resolution catalyst. Ideally the rate of racemization should be higher than

that of the resolution (kRac > kR > kS), and the stereoselectivity factor should be

higher than 20. Alternatively, the rate of the resolution is higher than that of

the racemization (kR > kRac >> kS), but in order to have a high enantioselectivity

the resolution needs to have very high stereoselectivity (s > 200).24

5

Figure 5. Dynamic kinetic resolution (DKR)

Dynamic kinetic resolutions which employ combinations of biocatalysts and

metal catalysts have been used in the synthesis of a wide range of compounds

with high selectivity using substrates such as secondary alcohols and amines.24-

26 Bäckvall and co-workers applied DKR in the synthesis of (S)-propranolol

(1) from the corresponding β-azido alcohol 9 using Candida antarctica lipase

B (CALB or Novozyme-435) and the hydrogen transfer catalyst 11 (Scheme

4). The azido acetate 10 was obtained in 71% yield and 86% ee.

Recrystallization of the final compound resulted in almost enantiomerically

pure product.27

Scheme 4. Dynamic kinetic resolution in the synthesis of (S)-propranolol

Dynamic kinetic asymmetric transformation (DYKAT) involves a de-

racemization where the substrate equilibrates through diastereomeric

intermediates (Figure 6). Several types of DYKATs have been developed and

used in different synthetic transformations. Because the substrate isomers (SR

and SS) are converted through diastereomeric complexes (SRCat and SSCat)

they will be present in unequal amounts. If the equilibrium leading to the

major stereoisomer of the product (kREq) is faster than the equilibrium leading

to the minor stereoisomer (kSEq), the total selectivity will be higher than that of

the corresponding KR. If there is instead a mismatch case where the

equilibrium leading to the major isomer is the slowest, the selectivity will be

lower.11

6

Figure 6. Dynamic kinetic asymmetric transformation (DYKAT) type I

1.1.4. Minor Enantiomer Recycling

In the methods described so far, the minor unwanted stereoisomer of the

product is either separated from the desired stereoisomer, which will result in a

reduced yield, or it will remain in the mixture and deplete the enantiopurity of

the product. If the minor unwanted stereoisomer (PS) instead is selectively

transformed back to the prochiral substrate (A), higher enantioselectivity and a

theoretical quantitative yield of the product are possible (Figure 7). In addition,

if enantioselective catalysts are used for both the product-forming step and the

recycling step, a reinforcing effect is obtained and the overall

enantioselectivity of the system will be higher than that of the individual steps.

Figure 7. Minor enantiomer recycling (MER). Ps is the minor enantiomer which is

selectively transformed back to prochiral substrate A. Transformation of sacrificial reagent X to low energy product Y provides the thermodynamic driving force for the

system

In this type of processes, the maximum ee is given by (E1E2-1)(E1E2+1), where

E1 and E2 are the stereoselectivity factors of the separate steps. In contrast, the

maximum ee in a DKR is dependent only on one enantioselective catalyst and

is thus given by (E-1)/(E+1).9 For a cyclic process to be feasible, the system

needs to be kept out of equilibrium and coupled to a second exergonic

reaction, which feeds the process with chemical energy and drives the system

by mass balance. This secondary process can be the transformation of a

7

sacrificial reagent to a byproduct with lower chemical potential (X → Y in

Figure 7). Without this coupled reaction, a cyclic process would not be

possible since the system would violate the principle of microscopic

reversibility, which states that the forward and reverse reaction must proceed

via the same transition state.9,28,29

Deracemizations of racemic alcohols and amines have in several cases been

accomplished by concurrent oxidation and reduction sequences. In these

systems an enzyme catalyzes the selective oxidation of one of the enantiomers

to give the prochiral ketone or imine. A second enzyme then reduces the

ketone or imine selectively to the other enantiomer (Figure 8). Whole-cells,

which sometimes contain an unknown catalytic composition, have most often

been used for the concurrent oxidation and reduction.30

Kroutil and co-workers

utilized purified enzymes in this methodology and could deracemize several

secondary alcohols to obtain enantiopure alcohols in essentially quantitative

yields. Both enantiomers of the alcohols were available by selecting the

appropriate enzyme system.31,32

A deracemization of 3H-indolines and

tetrahydroquinolines using chemocatalyzed enantioselective reduction and

oxidation was recently reported by Toste and co-workers. The

enantioselectivity of the oxidation step in this system was however very low,

but since the enantioselectivity for the reduction was high, the desired products

were obtained in high ee.33

Similarly, several procedures using an

enantioselective oxidation catalyst together with a non-selective stoichiometric

reduction reagent have been reported.34

The selectivity in these systems is,

however, dependent on only one chiral catalyst.9,11,12

Figure 8. Concurrent enantioselective oxidation and reduction

Hollmann, Turner, Ward, and co-workers recently developed a double

stereoselective cascade system employing an artificial transfer hydrogenase

(ATHase) together with a monoamine oxidase (MAO-N).35

In this system the

artificial ATHase reduces an imine (12) to an amine (13) with high

enantioselectivity to form the (R)-enantiomer. The MAO-N then selectively

oxidizes the minor (S)-enantiomer of 13 back to the starting imine (12), which

can react again. A third enzyme, a catalase, was necessary to destroy the H2O2

formed as it was thought to poison the ATHase (Scheme 5). It was observed

that the ee of the formed product (13) increased during the course of the

reaction at the same time as the conversion increased and reached a maximum,

a characteristic of a minor enantiomer recycling process. The product (R)-13

8

was obtained in enantiopure form and with >99% conversion. Identical results

were obtained when starting the reaction from the imine or the racemic amine.

Scheme 5. Synthesis of enantioenriched amine (R)-13 using an oxidation/reduction

cascade with two enantioselective catalysts

In our group a minor enantiomer recycling protocol was developed for the

synthesis of O-acylated cyanohydrins. In this system, a chiral Lewis acid

catalyzes the reaction between a prochiral aldehyde and an acyl cyanide to

form a scalemic mixture of enantiomeric O-acylated cyanohydrins. The minor

product enantiomer is then selectively hydrolyzed by a lipase to the

unprotected cyanohydrin and a carboxylic acid (Scheme 6). In the two-phase

system used, consisting of toluene and aqueous buffer, the carboxylic acid is

irreversibly deprotonated to a carboxylate ion. The transformation of acyl

cyanide to carboxylate ion is an exergonic process that provides the necessary

driving force of the system. Under the reaction conditions, the unprotected

cyanohydrin releases HCN and forms the starting aldehyde, which then reacts

again in the forward reaction.36,37

Enantiopure product can always be obtained

by allowing the lipase to hydrolyze any remaining traces of minor product

enantiomer when all of the acyl cyanide has reacted.

Scheme 6. Minor enantiomer recycling to produce O-acylated cyanohydrins

The minor enantiomer recycling methodology can result in higher selectivity

and higher yields than other methods and is therefore an attractive area of

research. Expanding the methodology used for the synthesis of O-acylated

9

cyanohydrins to other types of substrates and reactions should be possible if

appropriate conditions are found. By replacing the cyanide with other

nucleophiles, a range of different compound classes should be accessible by

this methodology. One possible type of compounds is acetoxyphosphonates

which have structures analogous to those of acylated cyanohydrins (Figure 9).

Figure 9. Targets for minor enantiomer recycling

The major challenge in achieving a successful recycling system is to find

conditions by which two enantioselective catalysts are able to function

together and reinforce each other. In addition, a sacrificial process which

provides a thermodynamic driving force of the system is required.29

1.2. Aim of this Thesis

We wanted to continue to explore the asymmetric synthesis of cyanohydrins by

using combinations of chemical catalysts and biocatalysts in either the minor

enantiomer recycling methodology or in a sequential process. We were

especially interested in synthesizing cyanohydrins which are easily

transformed to compounds with synthetic and biologic importance. In addition,

focus is on the development of new synthetic transformations of the obtained

cyanohydrins to products which otherwise can be tedious to access.

The asymmetric synthesis of acetoxyphosphonates via the direct addition of

acylphosphonates to aldehydes has previously not been described. We wanted

to explore this reaction with different enantioselective catalysts and also

investigate if the transformation could be achieved by employing the minor

enantiomer recycling methodology.

10

11

2. Synthesis and Applications of O-Acylated Cyanohydrins.

(Papers I–III and V–VI)

2.1. Introduction

2.1.1. The Importance of Cyanohydrins

Enantioenriched cyanohydrins constitute a highly important class of

compounds as they can be transformed into a wide range of valuable products.

Therefore, several different strategies for the asymmetric synthesis of

cyanohydrins have been developed.38-44

Scheme 7. Equilibrium between unprotected cyanohydrins and aldehydes

Unprotected cyanohydrins readily undergo racemization through elimination

of HCN, which results in formation of the corresponding prochiral carbonyl

compound (Scheme 7). O-Acylated cyanohydrins are not only much more

stable towards racemization and decomposition in comparison to the

unprotected cyanohydrins, they are also synthetically important.45,46

O-

Acylated cyanohydrins have found wide application in agriculture as many

pyrethroid insecticides have this type of structure (Figure 10).47,48

Many

pyrethroids are used as a mixture of stereoisomers even though it is known that

they vary in activity.6

Figure 10. Examples of O-acylated cyanohydrin insecticides

In addition, O-acylated cyanohydrins can be directly transformed to N-acylated

β-amino alcohols by reduction of the nitrile group accompanied by

intramolecular acyl transfer to the formed amino group.49

With this approach

the acyl group will be part of the final product and at the same time protect the

cyanohydrin from racemization and decomposition. Alternatively, protected or

unprotected cyanohydrins can be reduced with a hydride reagent to the free

amino alcohol, which subsequently is either alkylated or acylated (Scheme

8).43

This strategy is less attractive with regard to atom economy.

12

Scheme 8. Reduction of cyanohydrins to N-substituted β-amino alcohols

The β-amino alcohol and β-amido alcohol structural motifs are present in many

compounds with interesting biological properties (Figure 11). Examples

include the α1-adrenoreceptor agonist midodrine (16),50

the β3-adrenergic

receptor agonists solabegron (17)51,52

and mirabegron (19),53

and the β1-

adrenergic receptor agonist denopamine (18).54

Other examples are the

naturally occurring compounds (R)-tembamide (20) and (R)-aegelin (21),

which have both been used in traditional Indian medicine. In addition, the

previously mentioned β-adrenergic antagonist (S)-propranolol (1) is another

important example. Solabegron (17) has shown promising results for the

treatment of both overactive bladder (OAB) and irritable bowel syndrome

(IBS) and is currently in clinical trials.55-57

Figure 11. Biologically active compounds containing the β-amino alcohol and β-

amido alcohol structural motifs

13

2.1.2. Strategies for the Preparation of Enantioenriched Cyanohydrins

The most extensively used sources of cyanide in cyanations of carbonyl

compounds are HCN and TMSCN. HCN is extremely toxic and volatile, and

addition to aldehydes or ketones results in the unstable parent cyanohydrin. It

is therefore advantageous to instead use TMSCN as source of cyanide due to

its high reactivity and the instant protection as the silyl ether, which prevents

decomposition and racemization. However, TMSCN is also highly toxic and

volatile, and it is relatively expensive compared to that of other cyanide

sources.39,41

Cyanoformic esters have also been used to a relatively large extent

as source of cyanide while acyl cyanides have been more scarcely used. Acyl

cyanides have the advantage of producing highly stable O-acylated

cyanohydrins as products. Therefore, a method for the direct esterification of

O-TMS-cyanohydrins to give O-acylated cyanohydrins has been developed.58

2.1.2.1. Biocatalytic Methods for the Preparation of Cyanohydrins

Enzymes that catalyze the release of HCN from cyanohydrins exist naturally in

for example almonds as a protection against herbivores and microbial attack.

Hydroxynitrile Lyases (HNL’s) have been used for the highly enantioselective

synthesis of cyanohydrins by employing both aldehydes and ketones together

with an excess of HCN (Scheme 9).59-63

Scheme 9. HNL-catalyzed addition of HCN to aldehydes

Many compounds obtained by the use of these enzymes have been used in the

synthesis of pharmaceuticals, agrochemicals, and other biologically active

compounds.62

Hydroxynitrile Lyase from Hevea brasiliensis was used in the

kilogram synthesis of a β-amino alcohol (24) from furfural (22), resulting in

essentially enantiopure product (>99% ee) (Scheme 10).64

Scheme 10. Kilogram synthesis of a β-amino alcohol involving a HNL-catalyzed

HCN-addition to furfural

Candida antarctica lipase B (CALB) has been used in the kinetic resolution of

several aromatic O-acetylated cyanohydrins. In the transesterification of the

racemic compounds, the lipase is highly selectivity (E = 35–360) for the (S)-

enantiomers (Scheme 11).59,65-67

The high selectivity of CALB towards

cyanohydrins was utilized in our group for a high throughput ee analysis of O-

acetylated cyanohydrins.68,69

The enantiomeric preference of the lipase can be

predicted according to the steric properties of the substrate; these predictions

are known as Kazlauskas’ rules.70

In the active site, the (S)-enantiomer of the

14

substrate resides with the cyanide group in a pocket which has a very limited

size, and for the (R)-enantiomer to be hydrolyzed, the large aryl group needs to

fit in this pocket.71

Scheme 11. Kinetic resolution of O-acetylated cyanohydrins with CALB. The

enantiopreference of the enzyme is also shown; L represents the large substituent and M represents the medium substituent

CALB is a robust enzyme that can be used between pH 3.5 and 9.5 and the

immobilized variant can be used at temperatures up to 80 °C. CALB is also

stable in non-polar solvents as well as in more polar solvents.72

Lipases belong

to the hydrolases and follow the serine hydrolase mechanism (Scheme 12).

Scheme 12. The serine hydrolase mechanism for the hydrolysis of an ester. The

catalytic amino acids in CALB are indicated

In the active site, three amino acids are involved in the catalytic triad; serine

acts as a nucleophile and is activated by a histidine together with an aspartic or

a glutamic acid through hydrogen bonding. The reaction sequence is initiated

15

by attack of serine on the ester substrate (I) and the tetrahedral intermediate

formed is stabilized by the oxyanion hole which consists of amino acids with

hydrogen bond donor groups (II). Elimination of the alcohol product results in

an enzyme-substrate complex (III). Water subsequently attacks this complex

and again results in a tetrahedral intermediate that is stabilized by the oxyanion

hole (IV). Elimination of the acid product allows the enzyme to react again. In

the mechanism the water molecule can be replaced with another nucleophile

such as an alcohol. Depending on the reaction conditions, a lipase can work in

both directions and can thus either cleave or form an ester bond.73,74

Unprotected cyanohydrins are in equilibrium with the corresponding aldehydes

and HCN under basic conditions and thus readily racemize. By combining

enzymatic enantioselective acylation with in situ racemization, several

effective dynamic kinetic resolutions (DKR) have been developed.59

Initial

reports utilized anion-exchange resins or quinidines for the racemization and

either lipases from Pseudomonas sp. or Pseudomonas cepacia for the

transacetylation.75-78

Kanerva and co-workers expanded the scope of the

system, by employing lipase from Pseudomonas cepacia or Candida

antarctica lipase A (CALA), to more synthetically useful cyanohydrins.79-81

Hanefeld and co-workers could obtain the O-acetylated cyanohydrin (S)-28 in

97% yield and with 98% ee after four days reaction when using immobilized

CALB for the transacetylation step (Scheme 13).82,83

A disadvantage of the

method was that aliphatic substrates resulted in lower enantioselectivity than

aromatic substrates; this was believed to be due to slow racemization and/or

base-catalyzed non-selective acylation.84

Scheme 13. Dynamic kinetic resolution in the synthesis of O-acetylated

cyanohydrins using CALB

2.1.2.2. Cooperative Catalytic Systems for the Preparation of Cyanohydrins

Simultaneous activation of both the electrophile and the nucleophile occurs in

most enzymatic transformations, but can also improve the reactivity and the

enantioselectivity of chemically catalyzed organic reactions.85

Activation by a

combination of a Lewis acid and a Lewis base lowers the energy of the LUMO

of the electrophile and at the same time raises the energy of the HOMO of the

nucleophile, and thus increases the reactivity of the reagents.86

Cooperative

catalytic activation of the nucleophilic cyanide group and the electrophilic

carbonyl group has been extensively used in the asymmetric synthesis of

cyanohydrins.41,86-88

One of the earliest examples, which was reported by

16

Shibasaki and co-workers, employed the bifunctional BINOL derived catalyst

29 in the cyanosilylation of aldehydes with very good results (Scheme 14). The

aluminum center acted as a Lewis acid, activating the aldehyde, while the

phosphine oxide acted as a Lewis base and activated TMSCN.88,89

Scheme 14. Cyanosilylations catalyzed by bifunctional catalyst 29

The titanium salen dimeric complex 30 was introduced for the asymmetric

cyanosilylation of aldehydes and ketones by Belokon, North, and co-workers.

This catalyst is obtained by treating the monomeric titanium chloride-complex

with an aqueous buffer, which leads to elimination of HCl and formation of the

dimeric catalyst. High ee and 100% conversion could be obtained in the

cyanosilylation of several different aldehydes with only 0.1 mol % of catalyst

(R,R)-30 (Scheme 15).90

Reactions with ketones were slower and resulted in

products with lower enantiopurity.91

Scheme 15. Cyanosilylations catalyzed by (R,R)-30

It was also possible to use catalyst (R,R)-30 in the addition of cyanoformates to

aldehydes, which resulted in O-protected cyanohydrins. However, in this case

it was necessary to use 5 mol % of the catalyst at -40 °C in order to have

practical reaction times.92

It was later discovered that the reaction could be

accelerated, and the catalyst loading reduced, by using KCN as a co-catalyst.93

Catalyst (R,R)-30 together with KCN and different carboxylic acid anhydrides

was also used to synthesize O-acylated cyanohydrins in a one-pot protocol

(Scheme 16). Good to high enantiomeric excesses (60–93%) were observed

when employing acetic anhydride as acylating agent. The system was however

dependent on which acylating agent used; benzoic anhydride, for example,

resulted in much lower enantioselectivity.94-96

17

Scheme 16. Synthesis of O-acylated cyanohydrins using carboxylic acid anhydrides

and KCN

From mechanistic and kinetic studies it was concluded that the dimeric

complex 30 is the active species and that it simultaneously activates both the

aldehyde and the cyanide component in the transition state (Scheme 17a).97

The dimeric complex 30 was shown to be in equilibrium with the monomeric

compound 30a in solution (Scheme 17b). Low temperatures and high

concentration of the catalyst favor the dimeric complex. The equilibrium is

also solvent dependent; NMR in deuterated benzene shows only the dimeric

form while in deuterated chloroform and dichloromethane a mixture of

complexes is observed.98

Scheme 17. a) Transition state for the attack of cyanide to the carbonyl carbon with

catalyst 30 b) Equilibrium between the monomer 30a and dimer 30

Ding and co-workers expected that the equilibrium between the monomeric

and dimeric species would influence the activity of the catalyst 30. The two

monomeric units were therefore covalently linked by different spacers in order

to maximize the cooperative catalysis (Figure 12).99

Catalyst 31 showed a

dramatic increase in activity compared to 30 and the silyl-protected

18

cyanohydrin from benzaldehyde was obtained in almost quantitative yield and

with 96% ee after just five minutes at room temperature. Catalyst loadings

could be reduced to as low as 0.0005 mol % when longer reaction times were

used, and still with very high enantioselectivity. The catalyst could also be

used to synthesize O-acetylated cyanohydrins with high enantiopurity by

utilizing NaCN and acetic anhydride. One drawback of catalyst 31 is that the

synthesis of the required ligand is rather complicated. In contrast, the ligand

used in catalyst 30 is commercially available.100

Figure 12. Linked catalyst 31

North and co-workers coupled their previously developed synthesis of O-

acylated cyanohydrins to a lipase-catalyzed kinetic resolution (Scheme 18).101

The minor enantiomer from the product-forming reaction, which employed

(S,S)-30, KCN, and acetic anhydride, was subsequently hydrolyzed by CALB

to the unprotected cyanohydrin which could be separated from the desired

product. Initially, the DCM used in the first step was evaporated before the

enzymatic resolution, but it was later discovered that the enzyme was able to

retain its function in a mixed solvent system consisting of DCM and MTBE.

With this procedure the enantiopurity of the O-acetylated cyanohydrins

obtained in the first step was in several cases improved. However, a sequential

process such as this one involves the destruction of the minor enantiomer, and

the yield is therefore unavoidably decreased.

Scheme 18. Sequential process in the synthesis of O-acetylated cyanohydrins

19

The non-selective direct addition of acyl cyanides to carbonyl compounds was

early discovered to be catalyzed by base,102

and since then several variations of

this reaction have been developed.45,103-108

For example, N-heterocyclic

carbenes (NHCs) have been used as catalysts in the synthesis of racemic O-

acylated cyanohydrins.109

The advantage of using acyl cyanides is that both the

cyanide group and the acyl group originate from the same reagent, thereby

providing a transformation with 100% atom economy. The first asymmetric

direct addition of acyl cyanides to aldehydes was developed in our group and

employs a Lewis acid – Lewis base dual activation system (Scheme 19).110-112

The reaction is catalyzed by the titanium salen dimer 30, but in contrast to the

addition of TMSCN to aldehydes, a Lewis base is also necessary for the

reaction to proceed at low temperature. The role of the Lewis base is to

activate the acyl cyanide by adding to the carbonyl carbon thus releasing a

cyanide ion and forming a reactive acylating agent. The aldehyde is activated

by the titanium complex 30, probably in a fashion similar to that in the reaction

with TMSCN, thus allowing an enantioselective nucleophilic attack by the

cyanide ion. The resulting intermediate is then acylated by the acyl ammonium

ion. This methodology was successful for several different aldehydes and acyl

cyanides and gave the desired products in high yields and with high ee,

although some substrates gave products with lower enantioselectivity and/or

lower yield. Cyanoformates could be used in place of acyl cyanides with the

same catalytic system, also with good results.

Scheme 19. Addition of acyl cyanides to aldehydes catalyzed by 30 and Et3N

Recent mechanistic studies in our group included a Hammett correlation study

that showed that different steps are rate-determining depending on whether

electron-rich or electron-poor aldehydes are used. With the electron-rich

aldehydes the attack of cyanide on the carbonyl is the rate limiting step,

whereas with the electron-poor aldehydes the acylation step instead becomes

the slower step. It was also shown by using different combinations of

chiral/achiral base and enantiopure/racemic titanium complex 30 that acylation

20

occurs while the cyanohydrin is bound to titanium and that the unprotected

cyanohydrin most likely is not an intermediate in the reaction.113,114

Another catalytic system, consisting of a TiIV

complex of a BINOLAM-ligand

(32), has been used in the synthesis of O-aroyl cyanohydrins from aldehydes

and aroyl cyanides (Scheme 20).115,116

Several aldehydes gave the desired

products in high yields, however with moderate enantiopurity. Only aroyl

cyanides were used in the reaction as no product was formed when using

acetyl cyanide. The catalyst ligand contained a diethylaminomethyl group

which, in contrast to what is the role of the triethylamine in our dual activation

system, was proposed to act as Brønsted base which deprotonates trace

amounts of HCN present in commercial benzoyl cyanide. An aluminum

complex of ligand 32 was also used in the addition of methyl

cyanoformate117,118

and diethyl cyanophosphonate119,120

to aldehydes.

Scheme 20 Acylcyanation of aldehydes catalyzed by Ti(OiPr)4 complexed with

BINOLAM-ligand 32

2.1.2.3. Minor Enantiomer Recycling for the Preparation of Cyanohydrins

A minor enantiomer recycling procedure for the highly enantioselective

synthesis of O-acylated cyanohydrins was developed in our group (Scheme

21).37

In this system, the forward reaction between aldehyde and acyl cyanide

is catalyzed by the titanium complex (S,S)-30, while the kinetic resolution of

the resulting scalemic mixture is catalyzed by CALB. In order for the two

reactions to occur simultaneously, a mixture of a buffer phase and an organic

phase with lower density phase was used. The buffer phase is required to

ensure that the carboxylic acid formed in the hydrolysis step is deprotonated to

carboxylate ions. The buffer is also important in order to maintain the enzyme

activity which decreases at low pH.72

The acyl cyanide is added slowly into the

organic phase throughout the reaction in order to avoid hydrolysis in the

aqueous phase. Initially a tertiary amine was used as a co-catalyst, in analogy

to the dual activation procedure, but it was later discovered that the presence of

a Lewis base was not necessary in the two-phase system. Several different

aldehydes were used in the system and produced (R)-O-acylated cyanohydrins

with very high enantiomeric purity. The opposite product enantiomer was

obtained by using the enantiomeric titanium complex (R,R)-30 together with

Candida rugosa lipase (CRL).37

In addition, by using either the minor

enantiomer recycling system or a dynamic kinetic resolution, both enantiomers

21

of the O-acetylated cyanohydrin originating from (E)-2-butenal could be

obtained with the same biocatalyst.121

Scheme 21. Minor enantiomer recycling in the synthesis of O-acylated

cyanohydrins

Kinetic modelling and experimental studies of the system clearly demonstrated

that a reinforcing effect is obtained when combining two chiral catalysts.

Modelling also showed that the final ee of the product is the same regardless of

the ratio between the two catalysts, as long as the total selectivity is the same.

Interestingly, the yield is dependent on this ratio, thus showing that the yield

can be improved by varying the catalyst composition.36

Recently the reaction scope of minor enantiomer recycling was expanded to

the use of alkyl cyanocarbonates as source of cyanide.28

Titanium catalyst

(S,S)-30 was used for the product-forming reaction, and a lipase from porcine

pancreas (PPL) selectively hydrolyzed the minor product enantiomer. In this

study, the importance of having a thermodynamic driving force coupled to the

recycling was clearly demonstrated. Bovine carbonic anhydrase catalyzed the

decomposition of carbonic acid, formed in the enzymatic hydrolysis, to CO2.

In addition, a nitrogen stream was bubbled through the reaction mixture to

assure an efficient release of CO2. The required exergonic process was thus

provided by the release of CO2 from the system. When the process was

performed without either carbonic anhydrase or a nitrogen stream, the ee of the

product dramatically decreased, thus showing that when the CO2 is not

released from the system the cyclic process is halted.

22

2.2. Synthesis of Enantioenriched ω-Bromocyanohydrin Derivatives

2.2.1. Introduction

In order to demonstrate the synthetic importance of the methods developed in

our group, we wanted to synthesize the enantioenriched O-acetylated

cyanohydrins derived from 4-bromobutanal (33a) and 5-bromopentanal (33b).

The unprotected cyanohydrins have previously been synthesized by the

hydroxynitrile lyase-catalyzed HCN-addition to aldehydes 33a-b and gave the

unprotected cyanohydrins with 90–94% ee.122,123

It has been shown that the

unprotected cyanohydrins of these aldehydes can be used to synthesize 2-

cyanotetrahydrofuran (35a) and 2-cyanotetrahydropyran (35b) (Scheme 22).122

Chiral tetrahydrofurans are found in several natural products and are thus of

synthetic interest.124,125

The ω-bromocyanohydrins have also been used in the

synthesis of enantioenriched 2-substituted piperidines,126

azacycloalkan-3-

ols,127

and 2,3-disubstituted piperidines.128

Scheme 22. Retrosynthesis of 2-cyanotetrahydrofuran (35a) and 2-

cyanotetrahydropyran (35b)

2.2.2. Minor Enantiomer Recycling

We expected that the use of the minor enantiomer recycling protocol would be

beneficial for the synthesis of the O-acetylated cyanohydrins 34a and 34b from

4-bromobutanal (33a) and 5-bromopentanal (33b). We therefore initiated our

study by investigating the two individual steps of the cyclic process separately.

The forward reaction between the aldehydes 33a or 33b and acetyl cyanide,

catalyzed by titanium complex (S,S)-30, was performed in a two-phase system

consisting of toluene and pH 8 buffer (Scheme 23). To replicate the cyclic

process, three equivalents of acetyl cyanide (36) were slowly added to the

organic phase while the yield and ee were monitored by GC. As expected the

ee of the formed products remained constant throughout the reaction, although

at room temperature the reaction was only moderately selective to form the

(R)-enantiomers; 42% ee for 34a and 22% ee for 34b. During NMR analysis of

the crude reaction using 33a, we could observe the formation of the cyclized

byproduct (R)-35a. This compound was determined to have an ee of 40%,

which is almost identical to that of (R)-34a. We therefore believe that the two

products are formed either from a titanium-cyanohydrin complex or from the

free cyanohydrin.

23

Scheme 23. Forward reaction to synthesize (R)-34a and (R)-34b in the two-phase

system

We next tested some different lipases in the kinetic resolution of racemic O-

acetylated cyanohydrins 34a and 34b in the two-phase system (Scheme 24).

We followed the hydrolysis at different temperatures and pH:s by monitoring

the composition of enantiomers over time by chiral GC. CALB was highly

selective for the (S)-enantiomers of both substrates, although the hydrolysis

was slower for 34b (Figure 14) than for 34a (Figure 13). The enzyme

maintained the high selectivity when the pH was decreased to 6 at room

temperature, but when the temperature was increased to 40 °C the

enatioselectivity in the hydrolysis of 34a decreased. In contrast, hydrolysis by

Candida antarctica lipase A (CALA) was much slower and had lower

selectivity. As observed previously,37,66,129

Candida rugosa lipase (CRL)

showed a selectivity opposite of CALB for both substrates, but the rate of

hydrolysis was much slower than with CALB.

Scheme 24. Kinetic resolution of racemic O-acetylated cyanohydrins 34a and 34b

24

Figure 13. Kinetic resolution of 34a with CALB at r.t. with pH 8 buffer

Figure 14. Kinetic resolution of 34b with CALB at r.t. with pH 8 buffer

When the enzyme for the reverse step had been selected, we next continued

with the investigation of the cyclic process. Unfortunately, with aldehyde 33a

we detected large amounts of the cyclized product 35a when performing the

reaction at pH 8. This resulted in low yields of the desired product (Table 1,

entries 1 and 2), but very high ee (enantiopure product at 40 °C). Higher

temperature usually results in a faster enzymatic hydrolysis, but in the present

case also the formation of 35a was accelerated. We assumed that the

cyclization was mediated by base, and therefore the reaction was also

performed at pH 7 and 6. Indeed, lower pH gave lower amounts of byproduct;

at pH 6, 70% yield of 34a was obtained, but with only 82% ee (Table 1, entry

5). We could however obtain almost enantiopure product in 62% yield by

continued stirring of the reaction mixture after the addition of acetyl cyanide

was finished, thus hydrolyzing the remaining (S)-enantiomer (reaction profile

can be seen in Figure 15). Under these conditions cyclized product 35a was

formed in 17% yield and with 33% ee. This product had opposite absolute

configuration to that formed in the forward reaction, which indicates that the

0

10

20

30

40

50

60

0 5 10 15 20 25

%

Time (h)

(R)-38a (%)

(S)-38a (%)

0

10

20

30

40

50

60

0 5 10 15 20 25

%

Time (h)

(R)-38b (%)

(S)-38b (%)

(R)-34a (%)

(S)-34a (%)

(R)-34b (%)

(S)-34b (%)

25

cyclic product can form in the forward reaction as well as in the resolution

step.

With aldehyde 33b byproduct formation was not a problem and the desired

product was formed in high yield, although with moderate ee (Table 1, entry

7). The low selectivity can probably be explained by slow enzymatic

hydrolysis. Performing the reaction at 40 °C gave a somewhat higher ee, and

leaving the mixture to stir after complete addition of acetyl cyanide (36) gave

enantiopure (R)-34b in 67% yield (Table 1, entry 8).

Table 1. Minor enantiomer recycling with aldehydes 33a and 33b

entry aldehyde pH temp

(°C)

yielda of 34

(%)

ee of 34

(%)

1 33a 8 r.t. 46 82

2 33a 8 40 26 100

3 33a 7 r.t. 57 87b

4 33a 7 40 21 99

5 33a 6 r.t. 70 82c

6 33a 6 40 42 90d

7 33b 7 r.t. 84 61

8 33b 7 40 78 86e

aDetermined by GC. bContinued stirring for 13 h gave 50% of (R)-34a with 100% ee; cContinued stirring for 15 h gave 62% of (R)-34a with 99% ee, 17% (S)-35a with 33%

ee, and 10% 37a; dContinued stirring for 15 h gave 34% of (R)-34a with 100% ee, 28%

racemic 35a and 3% of 37a; eContinued stirring for 15 h gave 67% of (R)-34b with

>99% ee.

26

Figure 15. Reaction profile for the minor enantiomer recycling in the synthesis of

(R)-34a (Table 1, entry 5). Addition of acetyl cyanide (36) was stopped after 25 hours

2.2.3. Sequential Process

The minor enantiomer recycling process was compared to the dual activation

methodology, employing titanium catalyst (S,S)-30 and triethylamine in DCM

at -20 °C, that was developed in our group. Under these conditions the desired

products (R)-34a and (R)-34b were obtained in almost quantitative yields

(determined by GC), but with merely 81 and 80% ee, respectively. By

removing the DCM and dissolving the residue in toluene and stirring with

CALB and pH 8 buffer, essentially enantiopure products were obtained in high

yields (Scheme 25). The opposite enantiomer ((S)-34a) was obtained in

quantitative yield and 78% ee using catalyst (R,R)-30. Treatment of this

compound with CRL gave after 125 hours (S)-34a in 87% yield (determined

by GC) and 91% ee.

Scheme 25. Sequential process in the synthesis of (R)-34a and (R)-34b, and

subsequent deprotection and cyclization to give (R)-35a and (R)-35b

0

20

40

60

80

100

0 10 20 30 40

%

Time (h)

ee (%)

yield (%)

27

As the sequential procedure gave an overall better result than the minor

enantiomer recycling process it was used on a preparative scale (2.4 mmol)

and resulted in that 71% of 34a and 61% of 34b could be isolated in almost

enantiopure form. Finally (R)-34a and (R)-34b were transformed to the cyclic

compounds by initial hydrolysis to the unprotected cyanohydrins (R)-37a,b

using p-TsOH in EtOH. Cyclization using AgClO4122

gave compounds (R)-35a

and (R)-35b with no or very little racemization in moderate yields (Scheme

25).

2.2.4. Conclusions

Essentially enantiopure O-acetylated ω-bromocyanohydrins were obtained in

high yields using a sequential procedure. Minor enantiomer recycling could

also be used for the synthesis of these compounds but was less efficient due to

byproduct formation and low rate of enzymatic hydrolysis.

The highly enantioenriched cyanohydrins were transformed to (R)-2-

cyanotetrahydrofuran and (R)-2-cyanotetrahydropyran.

2.3. Application of Minor Enantiomer Recycling in the Synthesis of β-Adrenergic Antagonists

2.3.1. Introduction

We wanted to further expand the usefulness of the minor enantiomer recycling

methodology and at the same time show that the method can be used to access

compounds which are otherwise not readily available. As cyanohydrins easily

can be reduced to β-amino alcohols, we envisioned that the β-adrenergic

antagonists (S)-propranolol (1),130

(R)-dichloroisoproterenol (38),131

and (R)-

pronethalol (39)130

would all be available with a common strategy employing

enantioenriched cyanohydrins obtained via minor enantiomer recycling

(Scheme 26).

Scheme 26 Retrosynthesis to obtain enantioenriched β-adrenergic antagonists via

minor enantiomer recycling

28

Several different strategies have been used to synthesize (S)-propranolol (1),

but there are no reports which include the enantioselective addition of cyanide

to the prochiral aldehyde to obtain the corresponding cyanohydrin. (S)-

Propranolol was however synthesized by kinetic resolution of the racemic

unprotected cyanohydrin by Pseudomonas cepacia lipase. The cyanohydrin

was obtained with 96% ee at 56% conversion.132

The (R)-enantiomer of the

acylated cyanohydrin was obtained in 68% yield and merely 78% ee in a

dynamic kinetic resolution.77

The (R)-O-acetylated cyanohydrin from 3,4-dichlorobenzaldehyde, which is

the precursor to 38, has previously been synthesized in 40% yield and 93% ee

by enzymatic kinetic resolution of the racemic O-acetylated cyanohydrin.133

Addition of TMSCN to the same aldehyde gave the silyl-protected

cyanohydrin with 78% ee.134

Several successful syntheses of the cyanohydrin which is the precursor to 39

have been reported using different cyanide sources.41

2.3.2. Synthesis of (S)-Propranolol

We started by investigating the selectivity of the dual activation system in the

reaction between aldehyde 40 and acetyl cyanide (36) using catalyst (S,S)-30

and Et3N at -30 °C. The acylated cyanohydrin (S)-41 could be isolated in 69%

yield, however with only 15% ee. Using TMSCN as source of cyanide together

with catalyst (S,S)-3090

gave a slightly higher ee (Scheme 27). Preliminary

studies performed in our laboratory indicated that low selectivity is obtained in

cyanations of aldehydes containing a heteroatom in the α-position when using

catalyst 30.135

The aldehyde 40 has also been reported to produce racemic

cyanohydrin when subjected to (R)-hydroxynitrilase-catalyzed HCN-

addition.136

Scheme 27 Addition of acetyl cyanide or TMSCN to aldehyde 40 using catalyst

(S,S)-30

29

As the direct methods attempted gave low enantioselectivity, it was thought

that the minor enantiomer recycling procedure would be an advantageous

method. The racemic acetylated cyanohydrin 41 was therefore treated with

different enzymes in order to identify the optimal biocatalyst for the recycling

step. CALB was the superior enzyme and showed high selectivity for the (R)-

enantiomer (Scheme 28, Figure 16). Pseudomonas cepacia lipase (PSL) was

selective but the rate of hydrolysis was lower; both CALA and CRL were

unselective and slow.

Scheme 28. Kinetic resolution of 41 by CALB

Figure 16. Kinetic resolution of 41 with CALB at r.t. with pH 8 buffer

The forward reaction in the two-phase system, which was catalyzed by (S,S)-

30, was unselective and gave racemic product. The low selectivity was not

surprising as the dual activation methodology at low temperature with the

same catalyst also gave low ee. We still believed that it was possible to achieve

a successful recycling system as the enzymatic hydrolysis step was highly

selective.

We started by employing our standard amounts of catalysts at room

temperature, but this unfortunately gave both low yield and low ee (Table 2,

entry 1). We could improve the results somewhat by raising the temperature to

40 °C (entry 2). Addition of acetyl cyanide (36) at a lower rate allowed the

system to reach steady state and gave almost enantiopure product, although in

low yield (entries 3 and 4). Increasing the amount of titanium catalyst (S,S)-30

gave an increased yield, although with a slightly lower ee (entries 5 and 6).

0

10

20

30

40

50

60

0 5 10 15 20 25 30

%

Time (h)

Series1

Series2

(S)-41 (%)

(R)-41 (%)

30

Table 2. Optimization of the minor enantiomer recycling using aldehyde 40

entry Temp

(°C)

Addition

time (h)

(S,S)-30

(mol %)

CALB

(mg)

Yield

(%)a ee

(%)a

1 r.t. 25 5 20 44 41

2 40 25 5 20 50 77

3 40 50 5 20 35 98

4 40 50 5 40 11 100

5 40 50 7.5 20 34 95

6 40 50 10 20 61 96 aDetermined by GC using undecane as internal standard.

The yields in the cyclic process were low to moderate in all cases and we

therefore suspected that the starting aldehyde was not stable under the reaction

conditions. Stirring the aldehyde together with CALB and (S,S)-30 in the two-

phase system indeed substantially lowered the amount of aldehyde found in the

organic phase. The effect of varying the stirring rate of the reaction mixture

was also investigated; the yields did not appear to be affected, but higher ee:s

were observed with a faster stir rate.

The conditions in entry 6 in Table 2 were used on a preparative scale. The

desired product (S)-41 was then isolated in 59% yield and with 98% ee. This

result is an improvement in the synthesis of this cyanohydrin compared to the

kinetic resolution of the unprotected cyanohydrin, which gave 96% ee at 56%

conversion,132

and to the dynamic kinetic resolution, which gave 68% yield

and 78% ee.77

Reduction of the nitrile and ester in (S)-41 with LiAlH4 to the

amino alcohol (S)-43, followed by reductive amination with acetone and

NaBH4, afforded (S)-propranolol (1) in 56% yield over two steps with very

high ee (Scheme 29).137

31

Scheme 29. Synthesis of (S)-propranolol

2.3.3. Synthesis of (R)-Dichloroisoproterenol

When subjecting 3,4-dichlorobenzaldehyde (44) to the minor enantiomer

recycling system, we observed that the ee of the acylated cyanohydrin (R)-45

initially increased over time, as expected, but after approximately 30 hours

addition of acetyl cyanide (36) the ee started to decrease. We believe that this

decrease in ee might be explained by inhibition of CALB by acetyl cyanide.

Inhibition would result in slower hydrolysis of the minor enantiomer and a

shift in the balance between the less selective forward reaction and the reverse

resolution. By successively decreasing the rate of addition of acetyl cyanide

during the course of reaction the decrease in ee was no longer observed.

However, when performing the reaction on a preparative scale we used an

alternative approach which was less time-consuming. The acetyl cyanide was

added over eight hours which gave the product with 70% ee. Addition of an

additional batch of CALB to that already present in the reaction mixture and

hydrolysis of the remaining minor enantiomer gave (R)-45 in 54% isolated

yield and with 96% ee (Scheme 30).

It was not possible to utilize LiAlH4 for the reduction of (R)-45 as this gave the

final product 38 in merely 58% ee. To avoid racemization, the ester was

hydrolyzed to give the unprotected cyanohydrin, and subsequent reduction of

the nitrile under milder conditions with BH3·THF gave (R)-

dichloroisoproterenol (38) without any depletion of the ee.

Scheme 30. Synthesis of (R)-dichloroisoproterenol

2.3.4. Synthesis of (R)-Pronethalol

In the minor enantiomer recycling with 2-naphtylaldehyde (46) acetyl cyanide

(36) was added to the reaction mixture over 50 hours at 40 °C. After this

32

addition the ee was 91%, but by allowing the mixture to stir for an additional 5

hours and 40 minutes the ee increased to 98% and (R)-47 could be isolated in

82% yield (Scheme 31). This again showed an advantage of minor enantiomer

recycling; very high ee can be obtained by allowing the enzyme to perform a

kinetic resolution of the scalemic mixture. (R)-Pronethalol (39) was obtained

with 96% ee by using the LiAlH4 reduction followed by reductive amination.

Scheme 31 Synthesis of (R)-pronethalol

2.3.5. Conclusions

The β-adrenergic antagonists (S)-propranolol, (R)-dichloroisoproterenol, and

(R)-pronethalol were all obtained with very high enantiomeric purity using the

minor enantiomer recycling procedure. In the synthesis of the cyanohydrin

which is the precursor of (S)-propranolol, other methods showed inferior

enantioselectivity, which seems to be due to the presence of the heteroatom in

the aldehyde. When employing 3,4-dichlorobenzaldehyde in the cyclic

procedure, problems with enzyme inhibition were encountered but high ee was

obtained by allowing the enzyme to hydrolyze the remaining minor enantiomer

at the end of the reaction.

2.4. Synthesis and Applications of O-(α-Bromoacyl) Cyanohydrins

2.4.1. Introduction

Cyanohydrins functionalized with an O-(α-bromoacyl) group can undergo

several different transformations and are thus precursors to a range of

important compounds (Scheme 32). Intramolecular cyclization of O-(α-

bromoacyl) cyanohydrins results in aminofuranones and hydroxyfuranones,

which can undergo further transformations (routes a and b, Scheme 32). The

acyl chain in the O-(α-bromoacyl) cyanohydrins can be varied by either

substituting the bromo group with different nucleophiles or by performing a

cross-coupling at the α-position resulting in a new C-C bond (routes c and d,

Scheme 32). Reductions of the obtained products provide biologically active

N-substituted β-amino alcohols.

Enantioenriched O-(α-bromoacyl) cyanohydrins would be available by

employing α-bromoacyl cyanides in the minor enantiomer recycling. This

33

highly divergent approach would have a very high atom economy and would

thus allow further functionalization of the obtained products after the

enantioselective step.

Scheme 32. Possible transformations of O-(α-bromoacyl) cyanohydrins; a:

Cyclization to aminofuranones. b: Cyclization to hydroxyfuranones. c: Nucleophilic

substitution of the bromo-group. d: Cross-coupling at the α-position

2.4.2. Synthesis of O-(α-Bromoacyl) Cyanohydrins and Subsequent Transformation to Aminofuranones

The 2(5H)-furanone structural motif can be found in a range of compounds

possessing biological properties (Figure 17).138-141

Additionally, both

aminofuranones and hydroxyfuranones serve as synthetic intermediates.142,143

Examples are found in the syntheses of the antibiotic virginiamycin M2144

and

the pheromone (+)-eldanolide.145

In addition, reduction of the double bond in

the hydroxyfuranones has been used to access several interesting

compounds.146,147

Figure 17. The 2(5H)-furanone structural motif

We planned to access enantioenriched aminofuranones from O-(α-bromoacyl)

cyanohydrins obtained in the minor enantiomer recycling process. α-

Halocarbonyl compounds form zinc enolates upon treatment with Zn0, and the

addition of these zinc enolates to nitriles is known as the Blaise reaction.148

Zinc enolates of O-(α-bromoacyl) cyanohydrins react in an intramolecular

fashion and produces either aminofuranones or hydroxyfuranones, depending

34

on conditions for work-up (Scheme 32, routes a and b).149

Under strongly

acidic conditions the intermediate imine is hydrolyzed to the hydroxyfuranone,

whereas work-up with NH4Cl instead gives the aminofuranone (Scheme

33).149-151

Scheme 33. Intramolecular Blaise reaction in the synthesis of aminofuranones

Other syntheses of these types of compounds include for example the base-

mediated cyclization of O-acylated cyanohydrins.152,153

However, this is

limited to cyanohydrins obtained from ketones since deprotonation of the α-

proton would otherwise occur.

In a recent example the importance of the O-(α-bromoacyl) cyanohydrins as a

synthetic intermediate was demonstrated by Fürstner and co-workers.147

In the

synthesis of nominal gobienine A, a BINOL-derived ligand was used in the

asymmetric TMSCN addition to an aliphatic aldehyde (48) which afforded the

corresponding cyanohydrin with 89% ee. The cyanohydrin was subsequently

acylated with 2-bromopropionyl bromide in a separate step to give the O-(α-

bromoacyl) cyanohydrin 49. Intramolecular cyclization followed by acidic

work-up resulted in the desired hydroxyfuranone 50. Interestingly, the authors

could also show that by transforming the hydroxyl group to a triflate group, it

was possible to perform a palladium-catalyzed methoxycarbonylation at this

position (Scheme 34).

Scheme 34. Synthesis of nominal gobienine A which include an intramolecular

Blaise reaction

35

2.4.2.1. Minor Enantiomer Recycling

The acyl cyanides (53a,b), required for the synthesis of O-(α-bromoacyl)

cyanohydrins via the minor enantiomer recycling process, were initially

synthesized according to a literature procedure from TMSCN and the

corresponding acyl bromide.154

We showed that it was possible to replace the

expensive TMSCN with CuCN, although purification was sometimes difficult

and resulted in low yields (Scheme 35).

Scheme 35. Synthesis of acyl cyanides 53a and 53b using either TMSCN or CuCN

Initially, we attempted the synthesis of the O-(α-bromoacyl) cyanohydrins

using our dual activation catalytic system consisting of titanium catalyst (S,S)-

30 together with triethylamine. The reaction between acyl cyanide 53a and

benzaldehyde (25) gave only low conversion and low amounts of the desired

O-acylated cyanohydrin 54a (Scheme 36). This is probably due to

incompatibility between the Lewis base and the acyl cyanide.

Scheme 36. Synthesis of 54a using the dual activation system

Since the two-phase system does not require the presence of a Lewis base in

order for the reaction to proceed, it was believed to be more compatible with

the acyl cyanides compared to the dual activation system. As the acyl cyanide

53a contains a stereocenter, and is used as a racemate, four stereoisomers of

the product 54 are formed in the forward step. Remarkably, the biocatalyst

(CALB) was highly selective in hydrolyzing the two stereoisomers with (S)-

configuration at the C-O stereocenter ((S,R)-54a and (S,S)-54a), whereas no

particular selectivity was observed towards the C-Br stereocenter (Figure 18).

In the subsequent cyclization the stereocenter on the acyl group would be lost

and we were therefore satisfied with the observed selectivity.

36

Figure 18. Kinetic resolution of a scalemic mixture of 54a. CALB hydrolyzes (S,R)-

54a and (S,S)-54a selectively.

The minor enantiomer recycling procedure with acyl cyanide 53a and various

aldehydes resulted in high yields and high isomeric ratios (between

(R,R)+(R,S) and (S,R)+(S,S)) of the desired products 54a-g (Scheme 37).

Conditions were optimized for each substrate in terms of addition rate of acyl

cyanide, temperature, and catalyst amount. The amount of titanium catalyst

(S,S)-30 was increased from 5 to 10 mol % in the reaction of 4-

methoxybenzaldehyde since otherwise low yield was obtained. As was seen in

the modelling studies previously discussed, alteration of catalyst composition

can improve the yield of the reaction.36

In most cases the acyl cyanide 53a

needed to be added over 50 hours, since otherwise the system did not reach

steady state ee.

0

10

20

30

40

50

0 5 10 15 20

%

Time (h)

(S,R)-54a

(S,S)-54a

(R,R)-54a

(R,S)-54a

37

Scheme 37. Minor enantiomer recycling in the synthesis of O-(α-bromoacyl)

cyanohydrins 54a-g. Ratios between isomers (R,R)+(R,S) and (S,R)+(S,S) are shown

The product from (E)-butenal (54f) gave only moderate yield as a result of the

presence of relatively high amounts of unprotected cyanohydrin 55f, but the

desired product could still be isolated with high isomeric ratio. When

subjecting pentanal to the catalytic cycle a decrease in isomeric ratio of 54g

was observed after the usual initial increase. We previously observed this

phenomenon with another substrate and it was then attributed to enzyme

inhibition by the acyl cyanide (see section 2.3.3). We therefore performed

CALB-catalyzed kinetic resolutions of 54g with and without two equivalents

of acyl cyanide 53a present in order to investigate if any enzyme inhibition

38

could be observed. From these experiments we saw that the hydrolysis was

slower when the acyl cyanide was present. A successive decrease in addition

rate of 53a helped to avoid the effect of the inhibition (a reaction profile is

seen in Figure 19). It was, however, more convenient to add all three

equivalents of 53a over eight hours, which gave an isomeric ratio of 80:20

(between (R,R)+(R,S) and (S,R)+(S,S)), after which the enzyme was allowed to

hydrolyze the remaining minor isomers. With this procedure compound 54g

was obtained in 65% yield with 99:1 isomeric ratio.

Figure 19. Reaction profile for the synthesis of 54g. a:53a added with a rate of 0.06

equiv/h. b:53a added with a rate of 0.036 equiv/h. c: 53a added with a rate of 0.015 equiv/h. A total of 3 equiv added

Acyl cyanide 53b could also be used in the catalytic reaction together with

benzaldehyde (25) and compound (R)-56a was obtained in good yield with

very high ee (Scheme 38). Initial problems with solubility of 53b in toluene

were solved by adding the acyl cyanide dissolved in cyclopentyl methyl ether.

Scheme 38. Minor enantiomer recycling in the synthesis of (R)-56a

0

0,5

1

1,5

2

2,5

0

20

40

60

80

100

0 20 40 60 80 100 120

de

(%)

Time (h)

Diastereomeric

excess

Relative yield

b

a

b

c

c

a

Relative

yield

39

2.4.2.2. Cyclizations to Aminofuranones

The products (54) obtained from the minor enantiomer recycling were cyclized

using the zinc-mediated Blaise reaction. The reaction mixtures were quenched

with NH4Cl which resulted in the formation of aminofuranones (R)-58. The

reactions proceeded in high to very high yields with no or little racemization

(Scheme 39).

Scheme 39. Intramolecular Blaise reaction in the synthesis of aminofuranones

The Blaise cyclization was less straightforward with the O-acylated

cyanohydrin 56a. Under the same conditions as previously, formation of large

amounts of dehalogenated compound 28 and only small amounts of the desired

product 59 was observed (Scheme 40). Since the formation of the

dehalogenated byproduct 28 was observed, it was clear that the oxidative

addition of zinc took place in the reaction. However, only minor amounts of

the enolate reacted with the nitrile to give the cyclic product 59.

Scheme 40. Intramolecular Blaise reaction using compound 56a

Several reaction parameters, such as reaction time and solvent, were varied

without improved results. Substantially less of 28 was observed when using

magnesium instead of zinc, but the yield was still very low and other

decomposition products were observed. As there are several examples of

40

primary α-bromo esters reacting in intermolecular Blaise reactions,148

it is

possible that the cyclization with the secondary O-(α-bromoacyl) cyanohydrins

54 is accelerated by steric factors and this is why the primary compound 56a is

less reactive. In addition, the protonated aminofuranone, obtained through acid

mediated lactonization, is known to be highly unstable.150

This instability

might be one contributing factor for the low yields observed.

2.4.3. Functionalization of O-(α-Bromoacyl) Cyanohydrins

O-Acylated cyanohydrins are used as insecticides47,48

and are easily

transformed to N-substituted β-amino alcohols.49,155

Therefore, it is attractive

to have access to simple methods for the functionalization of the acyl chain in

these compounds. The disadvantage with other enantioselective syntheses of

acylated cyanohydrins is that the required acyl cyanides or acylating agents

might not be readily accessible or have unacceptable enantioselectivity.96,112

In

addition, using already functionalized acyl cyanides in the minor enantiomer

recycling could be troublesome as CALB has shown to be sensitive to

sterically demanding groups at the acyl part.69

A highly divergent strategy is

instead to use α-bromoacyl cyanides which give products that have the

possibility to react with different nucleophiles or to take part in cross-coupling

reactions after the enantioselective step (Scheme 32, routes c and d). In this

way, a variety of enantioenriched O-acetylated cyanohydrins are made

available from one common intermediate and only one single asymmetric

transformation is required.

2.4.3.1. Synthesis of Solabegron

To demonstrate the synthetic potential of O-(α-bromoacyl) cyanohydrins we

decided to synthesize the β3-adrenergic receptor agonist solabegron (17).51,52

We envisioned the synthesis to start with the preparation of a O-(α-bromoacyl)

cyanohydrin via minor enantiomer recycling, followed by substitution with an

aniline derivative and reduction of the nitrile to obtain the N-substituted β-

amino alcohol (Scheme 41). This would be the first synthesis of solabegron

that proceeded through an enantioenriched cyanohydrin. In previous syntheses

the chiral center originated either from (R)-3-chlorostyrene oxide51,156

or (R)-3-

chloromandelic acid,52,157,158

both of which are commercially available.

Scheme 41 Retrosynthetic analysis of solabegron

41

To access the desired final compound, the bromine in the acylated cyanohydrin

needed to be substituted by an aniline derivative. When treating enantiopure

O-(α-bromoacyl) cyanohydrin (R)-56a with aniline, together with DIPEA as

base, depletion of product ee was observed. Fortunately we discovered that

aniline, which is less basic, could act as a base in the reaction. Thus, by the use

of an excess of anilines containing both electron-donating and electron-

withdrawing groups the products (R)-60a-c could be isolated in high yields

without any racemization (Scheme 42).

Scheme 42. Substitutions with aniline derivatives

Also when the secondary amines dibenzylamine and N-benzylmethylamine

were used the substituted products were obtained with good yields without any

racemization (72% yield 98% ee and 71% yield >99% ee respectively). The

use of the aliphatic dietylamine resulted in product with a decreased ee. The

primary amine propylamine did not give the desired substitution product.

Instead, the amine attacked the ester to give the unprotected cyanohydrin

which decomposed to aldehyde, which in turn reacted with a second equivalent

of amine to give imine 61. Attack on this species by HCN gave compound 62

(Scheme 43).

Scheme 43. Substitution using propylamine

The next step in the synthesis of the desired N-substituted β-amino alcohols

was the reduction of the nitrile group followed by intramolecular acyl transfer

to the formed amine. Under the previously optimized conditions for this

transformation, nickel on alumina was used as catalyst and the reaction was

performed in dioxane at 120 °C under 20 bars of hydrogen.49

Hanefeld and co-

workers applied these conditions to enantioenriched O-acetylated (S)-

mandelonitrile (28) and obtained the reduced product in 45% yield, but they

42

observed a partial racemization of the product (from 95 to 75% ee). In our

studies, we used the more readily available Raney nickel which was reported

to give similar yields as nickel on alumina. With this catalyst it was possible to

perform the reaction at a substantially lower temperature of 80 °C under 20

bars of hydrogen, and we could reduce substrates (R)-60a-b to amido alcohols

(R)-63a-b with practically no racemization, although with low yields (Scheme

44). The low yields can probably be attributed to cleavage of the benzylic C-O

bond. Reduction of 60c was also possible, but we were not able to isolate the

pure product.

Scheme 44. Catalytic hydrogenation to amido alcohols (R)-63a-b

Direct reduction of the O-(α-bromoacyl) cyanohydrins 54 or 56 would be a

more divergent strategy as only one reduction would be required. We were

unfortunately not able to obtain the desired product using the catalytic

hydrogenation. After several attempts, only reduction of the bromide was

observed (Scheme 45). Several hydride reduction agents were also attempted

for this transformation, but we could not obtain the desired products.

Scheme 45. Failed reductions of compound 54 and 56

The synthesis of solabegron was initiated with the minor enantiomer recycling

using 3-chlorobenzaldehyde (64) and acyl cyanide 53b. Unfortunately, we

observed once again a decrease in ee during addition of the acyl cyanide (see

sections 2.3.3 and 2.4.1.2). The enzymatic kinetic resolution of rac-56b

(Scheme 46) in the presence of two equivalents of acyl cyanide 53b was

extremely slow compared to the resolution without acyl cyanide added (Figure

20 and Figure 21), which indicated that CALB was inhibited by 53b.

43

Scheme 46. Kinetic resolution of rac-56b using CALB with or without two

equivalents of 53b

Figure 20. Kinetic resolution of rac-56b using CALB

Figure 21. Kinetic resolution of rac-56b using CALB with two equivalents of 53b

present

By increasing the loading of CALB used in the cyclic process, an ee of 86%

was obtained when all acyl cyanide had been added. Subsequent hydrolysis of

the remaining minor enantiomer by CALB gave product (R)-56b in 69% yield

with 98% ee (Scheme 47).

0

10

20

30

40

50

60

0 5 10 15 20 25

%

Time (h)

(R)-60b (%)

(S)-60b (%)

0

10

20

30

40

50

60

0 5 10 15 20 25

%

Time (h)

(R)-60b (%)

(S)-60b (%)

(R)-56b (%)

(S)-56b (%)

(R)-56b (%)

(S)-56b (%)

44

Scheme 47. Minor enantiomer recycling in the synthesis of compound (R)-56b

Aniline derivative 69 could be synthesized in a one-pot two-step procedure in

which Pd/C catalyzed both the initial Suzuki coupling between 66 and 67 and

the subsequent reduction of the nitro group in 68 (Scheme 48).159

The

substitution employing (R)-56b, obtained in the minor enantiomer step,

proceeded in high yield to give (R)-70 with practically no racemization. Three

equivalents of the aniline derivative 69 were required in order to obtain

satisfactory yields; fortunately it was possible to recover most of the excess

amine during the purification. In the catalytic hydrogenation we could use our

conditions to obtain the reduced product (R)-71 with only a slight decrease in

enantiomeric purity, although with low yield. Next, the amide group was to be

reduced to a secondary amine without simultaneous reduction of the ester

group. This transformation was achieved without depletion of the ee by

employing two equivalents of borane dimethylsulfide complex;160

reduction of

the ester was observed if more of the reducing agent was used. Finally,

solabegron (17) was obtained by hydrolysis of the ester group in (R)-72 under

basic conditions, followed by protonation to give the carboxylic acid.52

45

Scheme 48. Synthesis of solabegron

2.4.3.2. C(sp3)-C(sp

2) Cross-Coupling Between O-(α-Bromoacyl)

Cyanohydrins and Boronic Acids

α-Halocarbonyl compounds have been extensively used in various cross-

couplings. Hartwig and co-workers generated zinc enolates from α-

bromocarbonyl compounds which successfully coupled with aryl bromides

under palladium catalysis (Scheme 49).161-163

This method successfully

employed a range of different aryl bromides in the coupling with esters and

amides to give the desired products in high isolated yields.

46

Scheme 49. Hartwig’s cross-coupling which employed zinc enolates

We expected the O-(α-bromoacyl) cyanohydrins 56 to be suitable substrates

for this transformation since they readily form the corresponding zinc enolates,

and only to a low extent react with the nitrile group. Unfortunately, the zinc

enolate derived from compound 56a did not yield the desired product in the

reaction with bromobenzene, but instead reduced compound 28 was the major

product (Scheme 50).

Scheme 50 Attempts employing compound 56a in cross-coupling via the zinc

enolate

Fu and co-workers developed an asymmetric Hiyama cross-coupling between

racemic secondary α-bromo esters and aryl silanes.164

When employing O-(α-

bromoacyl) cyanohydrin 54a in this system we could not detect the desired

product (Scheme 51).

Scheme 51. Attempts to utilize Fu’s asymmetric Hiyama cross-coupling with

compound 54a

The cross-coupling between an α-halocarbonyl compound and an organoboron

derivative was first reported by Suzuki and co-workers, and employed a

system consisting of Pd(PPh)4 together with Tl2CO3.165

Later Gooβen further

developed this transformation and showed that several boronic acids could be

coupled with simple α-bromo esters and amides in good yields.166,167

Several

modifications of reaction conditions and possible substrates that can be used

have been described.168-175

Therefore, this type of cross-coupling appeared to

be a promising alternative for the O-(α-bromoacyl) cyanohydrins.

47

The O-(α-bromoacyl) cyanohydrins contain an ester bond that is labile towards

hydrolysis and an acidic hydrogen, and therefore a Suzuki-type cross-coupling

with these substrates requires mild reaction conditions. Several natural

products were recently synthesized which included α-arylation of esters with

an adjacent stereocenter,176,177

but so far relatively simple substrates lacking

racemization-prone stereocenters have been employed in these types of cross-

couplings. This reaction is usually performed in THF or in toluene with

K2CO3, Cs2CO3, K3PO4, or KF as base and either a palladium or nickel

catalyst. We decided to use KF as a base as we assumed that metal carbonates

and phosphates would be too basic for our substrates. We started with the

cross-coupling between O-(α-bromoacyl) cyanohydrin 56a and phenylboronic

acid employing KF, Pd(OAc)2, and P(o-Tol)3 in THF.166

Unfortunately, we

detected only starting material and degradation to aldehyde and unprotected

cyanohydrin, and heating to 60 °C increased the level of degradation. In

contrast, replacing THF with toluene resulted in full conversion of 56a within

30 minutes, and only minor amounts of the protodehalogenated compound 28

could be detected (<10 %) (Scheme 52). It was essential to heat the reaction

mixture; at room temperature the reaction required approximately two days to

reach full conversion. Without phosphine ligand present, practically no

reaction occurred.

Scheme 52. Cross-coupling between 56a and phenylboronic acid

We investigated the substrate scope of the reaction using enantiopure O-(α-

bromoacyl) cyanohydrin (R)-56a in order to see if any racemization occurred

under the above conditions. Aromatic boronic acids substituted in 2-, 3-, and 4-

position with both electron-withdrawing and electron-donating groups yielded

the desired products (R)-73a-i with no or minor loss of the initial ee and with

good to high yields (Scheme 53). The O-(α-bromoacyl) cyanohydrin 56b,

obtained in the synthesis of solabegron (see section 2.4.2.1), could also be used

in the cross-coupling with phenylboronic acid with good results.

The majority of the reported cross-couplings between α-halocarbonyl

compounds and organoboron derivatives utilize several hours of reaction time.

In contrast, most of our substrates reached full conversion within 30–90

minutes. However, 3,4-dimethoxyphenylboronic acid reacted slowly and in the

initial part of the reaction the mixture turned into a thick suspension, but after

24 hours at 60 °C the desired product (R)-73g could be isolated in acceptable

yield (66%). Boronic acids containing electron-withdrawing groups reacted in

some cases sluggishly under our conditions. For example, 4-

48

ethoxycarbonylphenylboronic acid gave very low conversion to product (R)-

73h even after 20 hours at 60 °C. When we replaced Pd(OAc)2 with

Pd2(dba)3·CHCl3 we could achieve full conversion of (R)-56a after 23 hours at

60 °C and isolate (R)-73h in high yield. 4-Cyanophenylboronic acid was even

more unreactive and no product formation could be detected with Pd(OAc)2 as

a catalyst, but with Pd2(dba)3·CHCl3 we could after 24 hours at 100 °C observe

a ratio of 1:1.1 between starting material and product (according to 1H NMR).

In the cross-coupling between α-bromoesters and boronic acids, Gooβen

observed that when using Pd(OAc)2 a larger amount of biaryl product was

obtained compared to when Pd2(dba)3 was used. This was suggested to be

because the boronic acid initially reduces PdII to Pd

0;166

a reason for the slow

reaction with certain boronic acids in our system might be that the reduction of

Pd(OAc)2 is inefficient.

Scheme 53. Substrate scope for the cross-coupling reaction

Unfortunately, we were not able to obtain any product when using our

conditions with the O-(α-bromoacyl) cyanohydrin 54a which contains a

secondary bromine. Other solvents and ligands were screened but the desired

49

product could not be detected. In these reaction mixtures we could often

observe complex mixtures containing starting material and protodehalogenated

product 74 (Scheme 54). It was not possible to achieve this transformation

using a Ni(PPh3)4-catalyzed procedure that was reported to work well for

secondary α-bromoesters.171

Scheme 54. Attempted cross-couplings between compound 54a and phenylboronic

acid

A proposed mechanism for the cross-coupling is shown in Scheme 55. The

reaction is initiated by oxidative addition of Pd0 to the O-(α-bromoacyl)

cyanohydrin I, generating complex IIA, which is in equilibrium with oxygen-

bound palladium enolate IIB. Transmetallation between IIA and the boronic

acid generates compound IIIA which is in equilibrium with IIIB. Reductive

elimination from IIIA results in the desired product IV and regenerates Pd0.

Scheme 55. Proposed mechanism for the cross-coupling between O-(α-bromoacyl)

cyanohydrins and boronic acids

The reason that the O-(α-bromoacyl) cyanohydrin 54a did not react with

boronic acids in the cross-coupling might be explained by this mechanism. The

intermediate palladium enolates II and III are probably favored in the oxygen-

50

bound forms IIB and IIIB when R' ≠ H, possibly because of steric

interactions. Transmetallation followed by reductive elimination gives

protodehalogenated product V and homocoupled product VI. This hypothesis

is based on the fact that α-halocarbonyl compounds previously have been used

to promote the palladium-catalyzed homocoupling of boronic acids, and

substitution at the α-position then favored the homocoupling over the cross-

coupling.178

Since the cross-couplings between 56 and most boronic acids tested proceeded

very rapidly, we wanted to investigate if this high reactivity was general for

this type of reactions or if the structure of 56 contributed to the high rate. In

order to investigate if the presence of the nitrile group accelerated the reaction

we prepared compound 75, which contains a methyl group in place of a nitrile,

and subjected this compound to our conditions in the cross-coupling with

phenylboronic acid. In addition, since 75 is more sterically hindered then 56,

we subjected ethyl bromoacetate (77) to the same conditions (Scheme 56). All

reactions were carried out at four times lower concentration than usual (25 mM

substrate) to be able to follow the reaction over a convenient period of time.

We monitored the concentrations of substrates and products in the cross-

coupling reaction by 1H NMR using an internal standard.

Scheme 56. Variation of the structure of the α-halocarbonyl compounds in the

cross-coupling reaction

51

Figure 22 Concentrations of substrates and products in the cross-coupling reaction.

The concentrations were calculated from 1H NMR using 1-methoxynaphthalene as internal standard

From the graph, which shows the concentrations of substrates and products

over time, it is obvious that the reaction rate using 56a is significantly higher

than that using 75 or 77 (Figure 22). In addition, large amounts of 75 and 77

are remaining even after several hours of reaction. Presumably the reaction is

so slow at this concentration that other side reactions prevail and either the

boronic is consumed or the catalyst is destroyed. At the concentration used in

the preparative reactions (0.1 M), both 75 and 77 reacted more rapidly and

reached much higher conversions. These results indicate that the nitrile group

in 56a has an effect on the rate of the reaction. The rate acceleration might be

explained by coordination of the nitrile to the palladium atom, which possibly

facilitates the oxidative addition step, or assures that in the enolate

intermediates palladium is bound to carbon and not to oxygen (IIA and IIIA in

Scheme 55). The effect is probably not an inductive effect since the nitrile is

far away from the reacting center.

Finally we wanted to demonstrate that the obtained products could be

transformed to the important N-substituted β-amino alcohols. Compounds 73a

and 73i were therefore subjected to the previously used conditions for

hydrogenation and the products 79a and 79i were isolated with only minor

depletion of ee, although in moderate yields (Scheme 57).

0

5

10

15

20

25

30

35

0 50 100 150 200

Co

ncen

tra

tio

n

(mM

)

Time (min)

[S.M.]

Cyano

[Prod]

Cyano

[S.M.]10

41

[Prod]

1041

[S.M.]

1042

[Prod]

1042

56a

73a

75

76

77

78

52

Scheme 57. Catalytic hydrogenations of compounds 73a and 73i

2.4.4. Conclusions

The minor enantiomer recycling methodology was used to synthesize a variety

of O-(α-bromoacyl) cyanohydrins in high yields and with very high isomeric

ratios. When employing α-bromoacyl cyanides with a secondary bromine,

CALB was highly selective in hydrolyzing two out of the four stereoisomers.

Problems with enzyme inhibition were encountered for some substrates but the

effects could be avoided by using higher enzyme loading and allowing the

enzyme to hydrolyze the remaining minor enantiomer after the addition of acyl

cyanide was finished. The O-(α-bromoacyl) cyanohydrins were not readily

available with the dual activation system. This demonstrates that the minor

enantiomer recycling methodology can be used to access compounds which

are not otherwise easily accessible.

The O-(α-bromoacyl) cyanohydrins proved to be highly useful as synthetic

intermediates. Cyclization of these compounds, using an intramolecular Blaise

reaction, gave a range of aminofuranones in high yields and with very high ee.

The bromide in the synthesized cyanohydrins was efficiently substituted by

different amines. Subsequent reduction and intramolecular acyl transfer gave

access to N-substituted β-amino alcohols. The usefulness of this methodology

was demonstrated with the enantioselctive synthesis of the β3-adrenergic

receptor agonist solabegron.

In addition, the O-(α-bromoacyl) cyanohydrins could participate in a

palladium-catalyzed cross-coupling with a range of boronic acids to give

products in high yields, practically without any racemization. Certain boronic

acids reacted very sluggishly; replacement of the source of palladium

increased the reactivity.

53

3. Enantioselective Acylphosphonylation.

(Paper IV)

3.1. Introduction

Derivatives of α-hydroxy phosphonic acids show biological activity for a wide

range of targets. The absolute configuration of the compounds is important for

their biological activity, and several strategies for the asymmetric synthesis of

α-hydroxy phosphonic acid derivatives have therefore been developed.179,180

The Pudovik reaction,181

which is the base-catalyzed hydrophosphonylation of

aldehydes by dialkylphosphites, has been widely used in the synthesis of both

racemic and enantioenriched hydroxyphosphonates (Scheme 58).179

Scheme 58. The Pudovik reaction

The syntheses of acetoxyphosphonates have received much less attention

despite the interesting biological properties which are shown by these types of

compounds.182,183

Racemic acetoxyphosphonates have for example been

synthesized by the in situ acetylation of hydroxyphosphonates obtained from

the MgO- or K2CO3-catalyzed addition of diethylphosphite to aldehydes.184,185

Enzymatic kinetic resolutions have been used to produce both enantioenriched

hydroxyphosphonates and acetoxyphosphonates.186

For example,

Hammerschmidt and co-workers resolved racemic acetoxyphosphonates

derived from benzaldehyde using the lipase from Rhizopus oryzae (F-AP

15).187

The enantioselectivity was claimed to be very high (E > 100), although

another group later determined the enatioselectivity factor to be merely 32 for

the same kinetic resolution.188

CALB proved to be highly enantioselective in

the acetylation of hydroxyphosphonates derived from short-chain aliphatic

aldehydes.189

Since the phosphonic group will be the larger substituent

according to Kazlauskas’ rules, the size of the group derived from the aldehyde

is highly limited (Scheme 59).186

Thus, phenyl and propyl groups were too

large to fit in the medium-size pocket.

Scheme 59. Kinetic resolution of hydroxyphosphonates by CALB

54

The high selectivity of CALB for these types of substrates was utilized in the

dynamic kinetic resolution (DKR) of racemic hydroxyphosphonates.190

Essentially enantiopure acetoxyphosphonates were obtained in high yields, but

the reaction was only investigated for substrates obtained from short aliphatic

aldehydes (Scheme 60).

Scheme 60. Dynamic kinetic resolution of hydroxyphosphonates

As previously discussed (section 1.1.2), Spilling and co-workers used a

sequential process which encompassed a lipase-catalyzed hydrolysis of the

minor enantiomer of the acetoxyphosphonates obtained from asymmetric

addition of dimethylphosphite to aldehydes.23

An alternative strategy to access acetoxyphosphonates, analogous to that used

for acylcyanations of aldehydes, would be to add acylphosphonates to

aldehydes (Scheme 61). This would be a transformation with perfect atom

economy and would avoid the use of auxiliary acylating agents.

Scheme 61. Analogy between acylcyanation and acylphosphonylation

This approach has previously been used in the samarium-catalyzed synthesis

of racemic acetoxyphosphonates, but the transformation was limited to

aromatic acylphosphonates.191

The asymmetric synthesis of

acetoxyphosphonates via addition of acylphosphonates to aldehydes had not

previously been reported.

3.2. Dual Activation in the Synthesis of Acetoxyphosphonates

In the initial attempts we used the phosphite 80192

in the reaction with

aldehydes as this type of compounds are known to form acetoxyphosphonates

even without the presence of catalyst (Scheme 62).193

However, the reaction

proved to be sluggish and was not accelerated by Lewis acid catalysis using

titanium complex 30.

55

Scheme 62. Attempts of using phosphite 80 in the synthesis of

acetoxyphosphonates

Therefore we continued the investigation with acylphosphonates in the

reaction with aldehydes catalyzed by a combination of Lewis acid and Lewis

base. When treated with nucleophiles, acylphosphonates behave as activated

carboxylic acid derivatives and eliminate a dialkylphosphoryl group.194

Primary and secondary amines are known to react with acylphosphonates to

form amides and phosphites.195

In addition, DBU has been shown to catalyze

the acylation of alcohols with acylphosphonates; aliphatic acylphosphonates

required one equivalent of DBU in order to reach acceptable yields.196

For the reaction between an aldehyde and an acylphosphonate to be initiated,

the carbon-phosphorus bond in the acylphosphonate needs to be cleaved by an

amine, resulting in a phosphite and an acylated amine. Activation of the

aldehyde by the Lewis acid would allow the phosphite to add to the carbonyl

carbon. Acylation of the resulting phosphonate would give the desired

acetoxyphosphonate and regenerate the catalyst.

Scheme 63. Acylphosphonylation of benzaldehyde

In the optimization study we employed acylphosphonate 81 together with

benzaldehyde (25) (Scheme 63). Using Et3N, DABCO, or DMAP without

Lewis acid, the formation of a tetrahedral intermediate between the base and

the acylphosphonate was observed,197

but no desired product 82a. With DBU,

which is more basic, formation of the enolate of acetylphosphonate was also

observed.198

Only trace amounts of 82a were detected when using the system

consisting of titanium catalyst 30 and a Lewis base, a catalytic system which

was highly successful in the acylcyanation of aldehydes.110,112

The N-acylated form of 1H-1,2,3-benzotriazole is known to be an efficient

acylating agent.199

The unprotected hydroxyphosphonate 83 was indeed

acylated with full conversion when treated with this acylating agent in the

presence of DBU; in the absence of base the reaction was very slow (pKa of

1H-benzotriazole is ca 8200

). Employing a catalyst system consisting of

titanium catalyst (S,S)-30, DBU, and 1H-benzotriazole (85) gave only

56

formation of byproduct 84.201

Slow addition of 81 helped to suppress the

formation of this byproduct and the desired product 82a was formed, although

with significant amounts of 84. Replacement of (S,S)-30 with the aluminum

complex 86202

as Lewis acid gave only trace amounts of product. In contrast,

the aluminum complex 87a derived from a tridentate L-valinol-based Schiff

base resulted in a highly improved result. This catalyst has previously shown

to give highly enantioenriched products in high yields in the

hydrophosphonylation of aldehydes.203

High to very high yields of

acetoxyphosphonate 82a were obtained by optimizing the amounts of 1H-

benzotriazole and DBU (Table 3). It was obvious from this optimization study

that the presence of 1H-benzotriazole was essential for the reaction as

otherwise low yields were obtained even when 50 mol % DBU was used

(entries 1–2). Higher loading of 1H-benzotriazole gave higher yield but with

slightly lower enantioselectivity (entry 5). In contrast, a larger amount of DBU

resulted in reduced yield, possibly because of enolate formation (entries 6–7).

The best enantioselectivity was observed when 20 mol % of 1H-benzotriazole

was used (entry 4). Replacing DBU with DMAP, DABCO, or DIPEA resulted

in lower yields (entries 8–10). Chiral bases had only minor effects on the

enantioselectivity; reactions in the presence of cinchonidine and cinchonine

gave ee:s of 32 and 28%, respectively.

Table 3. a Optimization of the amounts of 1H-benzotriazole (85) and DBU

entry 85

(mol %)

base

(mol %) 82a:83

(mol ratio)

yield

(%)b ee (%)c

1 0 DBU (15) 1:0.13 37 28 (R)

2 0 DBU (50) 1:0.16 38 18 (R)

3 10 DBU (15) 1:0.13 53 28 (R)

4 20 DBU (15) 1:0.05 79 38 (R)

5 50 DBU (15) 1:0 100 30 (R)

6 20 DBU (30) 1:0 36 32 (R)

7 20 DBU (50) 1:0 32 26(R)

8 20 DMAP (15) 1:0.04 63 34 (R)

9 20 DABCO (15) 1:0.09 61 38 (R)

10 20 DIPEA (15) 1:1.7 16 8 (S) aReaction conditions: benzaldehyde (0.25 mmol), 87a (10 mol %), toluene (1 mL),

room temperature; 81 (3 equiv), diluted to 200 µL with toluene, was added over 20 h

using a syringe pump. bDetermined by 1H NMR with 1-methoxynaphthalene as internal

standard. cDetermined by HPLC using chiral IA column.

57

In order to improve the enantioselectivity, we decided to vary the structure of

the Schiff base ligand (Table 4). Variation of the steric properties of the amino

alcohol part (entries 1–3 and 5) or in the 3-position of the aromatic ring

(entries 4–6) did not lead to any improved result. In addition, changing the

electronic properties of the ligand had no positive effect on the

enantioselectivity (entries 7–8). We therefore continued our investigation with

ligand 87a.

Table 4.a Catalyst optimization

entry catalyst

(10 mol %) 82a:83

(mol ratio)

yield

(%)b ee (%)c

1 87a 1:0.05 79 38 (R)

2 87b 1:0.04 48 2 (R)

3 87c 1:0.25 35 28 (R)

4 87d 1:0.31 38 12 (R)

5 87e 1:0.51 23 4 (S)

6 87f 1:0.28 13 10 (S)

7 87g 1:0 61 20 (R)

8 87h 1:0 46 32 (R) aReaction conditions: benzaldehyde (0.25 mmol), 1H-benzotriazole (20 mol %), DBU

(15 mol %) in toluene (1 mL), room temperature, 81 (3 equiv), diluted to 200 µL with

toluene, was added over 20 h using a syringe pump. bDetermined by 1H NMR with 1-

methoxynaphthalene as internal standard. cDetermined by HPLC using chiral IA

column.

Fewer equivalents of acylphosphonate 81 could be used when the addition rate

was reduced; a quantitative yield was obtained with merely 1.5 equiv, although

under these conditions the enantioselectivity was lower.

58

The enantioselectivity could be improved by performing the reaction at a lower

temperature. At 0 °C the ee of the product was 52% (Table 5), but reducing the

temperature to -10 °C gave lower yield without improvement of the

enantioselectivity. Higher loading of 1H-benzotriazole (50 mol %) was

required at 0 °C in order to maintain a high yield. In addition, catalyst 87a

proved to be insoluble in toluene below room temperature, and DCM had to be

used as solvent. We could observe that the byproduct 84 was formed in lower

amount in DCM than in toluene; instantaneous addition of 81 to the reaction

mixture at room temperature gave 35% yield after 47 hours, compared to the

reaction in toluene which gave merely 5% yield.

Table 5. aAcylphosphonylations at reduced temperature

entry temp

(°C) 85

(mol %)

DBU

(mol %) 82a:83

(mol ratio)

yield

(%)b

ee

(%)c

1 0 20 15 1:07 60 56

2 0 50 20 1:0 100 52 aReaction conditions: benzaldehyde (0.25 mmol) and 87a (10 mol %) in DCM (1 mL).

81 (3 equiv), diluted to 300 µL with DCM, was added over 45 h using a syringe pump. bDetermined by 1H NMR with 1-methoxynaphthalene as internal standard. cDetermined

by HPLC using chiral IA column.

The optimized reaction conditions were applied to aromatic aldehydes

substituted in the 2-, 3-, and 4-positions with both electron-withdrawing and

electron-donating groups, and high isolated yields of the desired products were

obtained in most cases, although with moderate enantioselectivities (Scheme

64). Propanal proved to be less reactive and the reaction had to be performed at

room temperature in order to reach acceptable yields. The

acetoxyphosphonates we obtained were shown to have R absolute

configuration, which is opposite to that obtained in the

hydroxyphosphonylation using the same ligand.203

This indicates that the

structure of the catalyst-substrate complex in the enantioselective step in our

case differs from that in the hydroxyphosphonylation.

59

Scheme 64. Substrate scope of the acylphosphonylation

We also attempted to use some other acylphosphonates in our catalytic system

(Scheme 65). Replacing the acetyl group in 81 with a benzoyl group, as in

compounds 88, gave the desired product in very low yield. Using the more

sterically demanding 89 gave the corresponding acetoxyphosphonate in good

yield, but with very low ee. Use of acylphosphonate 90204

resulted in a

product in low yield without any improvement of the enantioselectivity.

Scheme 65. Other acylphosphonates used in the acylphosphonylation of

benzaldehyde

In our proposed mechanism we believe that the reaction is initiated by the

deprotonation of 1H-benzotriazole (85) by DBU, resulting in an increase in its

60

nucleophilicity.205,206

The deprotonated benzotriazole then attacks

acylphosphonate 81, resulting in phosphite 91 and the acylated benzotriazole

(92) (Scheme 66). The formed phosphite is probably coordinated to the

aluminum center, as was proposed in the hydrophosphonylation with the same

catalyst.207

Phosphite 91 attacks the Lewis acid-activated aldehyde, resulting in

hydroxyphosphonate 93, which subsequently is acylated by 92 to obtain final

product 82. Phosphite 91 can alternatively attack an unreacted

acylphosphonate 81 to give byproduct 84. The cleavage of the

acylphosphonate by benzotriazole is supported by the presence of acetylated

benzotriazole 92 according to NMR of the crude reaction mixture.

Scheme 66. Proposed mechanism for the acylphosphonylation

In order to determine whether the acyl- and the phosphite groups in the product

are derived from the same molecule of acylphosphonate, we performed an

experiment which employed the two acylphosphonates 90 and 94.204

With one

equivalent of benzaldehyde (25) and 0.5 equivalents each of the

acylphosphonates we detected the compounds 82a and 95 by GC-MS (Scheme

67). Therefore we concluded that the acylation does not have to take place

within the same complex as the P-C bond formation.

Scheme 67. Acylphosphonylation with acylphosphonates 90 and 94

Feng and co-workers observed a strong positive nonlinear effect with catalyst

87a in the hydrophosphonylation of aldehydes, which indicated the

involvement of a dimeric species.203

Calculations showed that the reaction with

61

dimeric catalyst was thermodynamically and kinetically more favorable than

the reaction with monomeric catalyst.207

In addition, the same ligand in

complex with Et2AlBr was shown by X-ray crystallography to have a dimeric

structure.208

To gain insight into the nature of the catalyst in our system we investigated the

correlation between ligand ee (of catalyst 87a) and product (82a) ee. In

contrast to what Feng and co-workers observed,203

our reaction displayed a

weak negative nonlinear effect (Figure 23). A positive nonlinear effect might

be an indication that a heterochiral dimer is formed which is less reactive and

also more stable than the homochiral dimer. In this way, some of the racemic

catalyst will be removed from the catalytic cycle and more enantioenriched

catalyst will take part in the reaction. A negative nonlinear effect might

indicate that the heterochiral dimer is the more reactive species. However, it

cannot be definitely concluded whether our system involves a dimeric species

or not; a negative nonlinear effect might arise for other reasons such as the

competition with a non-catalyzed background reaction.209

Figure 23. Correlation between product ee and ligand ee

0

10

20

30

40

50

60

0 20 40 60 80 100

ee product

(%)

ee ligand

(%)

62

3.3. Minor Enantiomer Recycling

We anticipated that acetoxyphosphonates with higher enantiomeric purity

would be accessible by a minor enantiomer recycling procedure analogous to

that applied in the synthesis of acylated cyanohydrins (Scheme 68).

Scheme 68. Envisioned synthesis of acetoxyphosphonates via minor enantiomer

recycling

For a minor enantiomer recycling process to be successful, the forward

reaction needs to take place in the two-phase system; the aqueous phase is

required for the formation of carboxylate ions, which supply the

thermodynamic driving force for the system. Additionally, a catalyst which

hydrolyzes the minor enantiomer obtained in the forward reaction is essential.

Finally, the hydroxyphosphonate formed in the reverse step needs to

spontaneously eliminate a dialkylphosphite and give back the starting

aldehyde.

Initially an experiment was performed to investigate if the

hydroxyphosphonate formation is reversible. The hydroxyphosphonate 83 was

stirred together with one equivalent of p-tolualdehyde (96) and one equivalent

of DBU in THF; formation of benzaldehyde (25) was indeed observed, thus

demonstrating that the reaction is reversible (Scheme 69).

Scheme 69. Reversibility experiment

Unfortunately the forward reaction in the two-phase system resulted in low

yields of the desired acetoxyphosphonates. However, we discovered that

hydroxyphosphonate 83 was almost completely acetylated in the two-phase

system when in the presence of acetylated benzotriazole 92210

and DBU.

Therefore we attempted the reaction between diethyl phosphite (98) and

benzaldehyde (25) in the presence of acetylated benzotriazole 92 catalyzed by

the aluminum catalyst 87a and DBU. This resulted in low yield of the product

63

together with hydroxyphosphonate 83 (Scheme 70). Using acylphosphonate 81

in place of diethyl phosphite (98) in the two-phase system gave no improved

yield.

Scheme 70. Forward reaction in the two-phase system

For the reverse step we investigated several different enzymes for the kinetic

resolution of acetoxyphosphonates. CALB has been reported to catalyze the

acetylation of racemic hydroxyphosphonates derived from short-chain

aldehydes with high selectivity.189

In the kinetic resolution of racemic

acetoxyphosphonate 99190

using CALB or lipase from Pseudomonas cepacia,

we discovered that the substrate seemed to partly dissolve in the aqueous phase

since both enantiomers of 99 were rapidly consumed. No enantioselective

hydrolysis was observed by CALB when the organic phase was toluene, but

with MTBE as solvent we could observe some selective hydrolysis, although

still large amounts of 99 dissolved in the aqueous phase (Scheme 71). Lipase

from Pseudomonas cepacia was also slightly enantioselective in toluene.

Scheme 71. Kinetic resolution of rac-99

The kinetic resolution of aromatic acetoxyphosphonates was investigated with

several different enzymes without success. For example, the enantioselective

hydrolysis of racemic 101 or 82a with Rhizopus oryzae (F-AP 15)187

was not

successful in a two-phase system consisting of toluene and 2 M pH 7

phosphate buffer (Scheme 72).

Scheme 72. Kinetic resolution of racemic 101 and 82a

64

3.4. Conclusions

A one-step, direct addition of acylphosphonates to aldehydes in the synthesis

of enantioenriched acetoxyphosphonates was developed. The system utilized a

combination of Lewis base, Lewis acid, and Brønstedt base in a cooperative

catalytic activation. A range of aldehydes was used and gave products in most

cases in high yields. The enantioselectivity of the reaction was however only

moderate. Several derivatives of a Schiff-based ligand were synthesized and

assessed in the catalytic reaction, but without any improvement of the

enantioselectivity.

An investigation aimed at synthesizing acetoxyphosphonates via a minor

enantiomer recycling procedure was also initiated. Several difficulties were

however encountered. The forward reaction in the two-phase system was

inefficient and only low yields of the product were obtained. In addition, we

did not succeed in finding an enzyme to catalyze the reverse reaction under our

conditions.

65

4. Concluding Remarks

In this thesis, asymmetric syntheses of enantioenriched cyanohydrins and

acetoxyphosphonates have been investigated. Problems that arise when truly

enantiopure products are desired have been highlighted, and large focus has

been on the development and application of the minor enantiomer recycling

methodology. In the recycling procedure, a titanium salen dimer catalyzes the

reaction between an aldehyde and an acylcyanide to give an O-acylated

cyanohydrin, and a lipase selectively hydrolyzes the unwanted stereoisomer to

the unprotected cyanohydrin, which is in equilibrium with the aldehyde. In

several cases the enantioselectivity of the cyclic procedure was superior to that

of other available methods. The benefit of using minor enantiomer recycling

was obvious in the synthesis of the cyanohydrin that is a precursor to (S)-

propranolol. In cases where the minor enantiomer recycling process failed to

give product with sufficient enantiopurity, the remaining unwanted

stereoisomer could be subjected to a kinetic resolution with the lipase in the

same reaction mixture.

When the minor enantiomer recycling gave unsatisfactory results the same

enantioselective catalysts were used in a sequential process. The O-acylated

cyanohydrins were first obtained using a combination of the titanium salen

dimer and a Lewis base at low temperature, and subsequent enzymatic kinetic

resolution yielded practically enantiopure products.

It was demonstrated that the various cyanohydrins produced were highly

useful as synthetic intermediates. O-(α-Bromoacyl) cyanohydrins proved to be

especially versatile and could be further transformed to several biologically

active compounds. For example, a novel synthesis of the highly

enantioenriched β3-adrenergic receptor agonist solabegron was developed.

In addition, the first enantioselective acylphosphonylation of aldehydes was

developed. In analogy to the synthesis of O-acylated cyanohydrins, a

combination of a chiral Lewis acid and a Lewis base catalyzed the addition of

acylphosphonates to aldehydes. The system developed gave access to a variety

of enantioenriched acetoxyphosphonates that were obtained in high yields. The

enantioselectivity of this reaction was however only moderate. More work

concerning catalyst design and investigation of the reaction mechanism might

be necessary to improve the results. We did not succeed to find conditions

where we could use the minor enantiomer recycling in the synthesis of the

acetoxyphosphonates; the main problem was finding an efficient enzyme to

catalyze the reverse reaction. Further optimization of reaction conditions or

enzyme engineering might facilitate this step.

66

In conclusion, the work presented in this thesis provides chemists with new

tools for the asymmetric synthesis of several important types of compounds.

Since the biological activity of stereoisomers often differs, the development of

efficient syntheses of enantiopure compounds will reduce costs and

environmental impact in the manufacturing of pharmaceuticals and

agrochemicals. The continuing efforts to develop synthetic methodologies that

produce enantiopure compounds with high yields are therefore of vital interest.

67

Acknowledgements

Firstly I want to thank my supervisor, Professor Christina Moberg, for

accepting me as a Ph.D. student, always having time for discussions, and

teaching me how to conduct research. Your enthusiasm for chemistry is highly

inspiring.

I also want to thank my co-supervisor, Dr. Krister Zetterberg, for always

having useful advices during group meetings.

My appreciation also goes to Vetenskapsrådet (VR) for funding, and the Aulin

Erdtman foundation for helping me to present my work at conferences.

Thanks to Dr. Peter Dinér, Dr. Anna Laurell Nash, and Dr. Laure Theveau for

reading my thesis and giving me valuable comments.

Thanks to my co-authors: Professor Christina Moberg, Dr. Ye-Qian Wen, Dr.

Khalid Widyan, Berenice Heid, Inanllely Gonzalez, Dr. Anna Laurell Nash,

Björn Dahlgren, Professor Tore Brinck, and Guillermo Monreal Santiago.

Thanks to Professor Karl Hult and Dr. Linda Fransson for many valuable

discussions.

Many thanks to the talented students who have contributed to my research:

Berenice, Inanllely, Robin, Alexis, Tina, Lorraine, and Guillermo.

Thanks to Anna for being a great colleague and friend!

Thanks to all past and present members in the Ki-group and to everybody at

the department of organic chemistry. Special thanks to: Brinton, Juho, Anna,

Jakob, Tessie, Piret, Fredrik, Björn, Lei H, Lei W, Yan, Laure, Brian, Erik, Ye-

Qian, Hui, Youcai, Ende, Khalid, Johan, Rosalba, Antanas, and Quentin.

Thanks to Dr. Zoltán Szabó for help with NMR studies, and thanks to Dr. Ulla

Jacobsson, Henry Challis, and Lena Skowron for help with everything else!

Thanks to Professor Ingemar Kvarnström, Dr. Veronica Sandgren, and Dr.

Marcus Bäck at LiU for introducing me to research in organic chemistry.

Of course I also want to thank all of my friends for all the good times during

the years!

Till sist vill jag tacka Åsa, mina föräldrar och min bror med familj för det stöd

och den uppmuntran ni alltid gett mig.

68

Appendix

The following is a description of my contributions to Publications I to VI, as

requested by KTH:

Paper I: I contributed to the formulation of the research problem, performed

the majority of the experimental work, and wrote part of the manuscript. I

supervised the undergraduate student Berenice Heid.

Paper II: I contributed to the formulation of the research problem and shared

the writing of the manuscript. I performed the experimental work assisted by

undergraduate students Inanllely Gonzalez, Alexis Bordet, and Robin

Lafficher.

Paper III: I contributed to the formulation of the research problem, performed

some of the experimental work, and wrote part of the manuscript. I supervised

the undergraduate student Inanllely Gonzalez.

Paper IV: I contributed to the formulation of the research problem, shared the

experimental work, and wrote part of the manuscript.

Paper V: I contributed to the formulation of the research problem, shared the

experimental work, and wrote the majority of the manuscript. I supervised the

undergraduate student Guillermo Monreal Santiago.

Paper VI: I contributed to the formulation of the research problem, performed

the experimental work, and wrote the majority of the manuscript.

69

References

1 Cossy, J. R. In Comprehensive Chirality; Carreira, E. M., Yamamoto, H., Eds.; Elsevier:

Amsterdam, 2012; Vol. 1, p 1-7.

2 Islam, M.; Mahdi, J.; Bowen, I. Drug Saf. 1997, 17, 149-165.

3 Nguyen, L. A.; He, H.; Pham-Huy, C. Int. J. Biomed. Sci. 2006, 2, 85-100.

4 19th WHO Model List of Essential Medicines, April 2015.

http://www.who.int/medicines/publications/essentialmedicines/en/

5 Barrett, A. M.; Cullum, V. A. Br. J. Pharmacol. 1968, 34, 43-55.

6 Pérez-Fernández, V.; García, M. Á.; Marina, M. L. J. Chromatogr. A 2010, 1217, 968-989.

7 Denard, C. A.; Hartwig, J. F.; Zhao, H. ACS Catal. 2013, 3, 2856-2864.

8 Köhler, V.; Turner, N. J. Chem. Commun. 2015, 51, 450-464.

9 Fransson, L.; Moberg, C. ChemCatChem 2010, 2, 1523-1532.

10 Kagan, H. B.; Fiaud, J. C. In Top. Stereochem.; Eliel, E. L., Wilen, S. H., Eds.; John Wiley

& Sons, Inc.: Hoboken, NJ, USA, 1988, p 249-330.

11 Steinreiber, J.; Faber, K.; Griengl, H. Chem. – Eur. J. 2008, 14, 8060-8072.

12 Faber, K. Chem. – Eur. J. 2001, 7, 5004-5010.

13 Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104, 7294-

7299.

14 Timms, N.; Daniels, A. D.; Berry, A.; Nelson, A. In Comprehensive Chirality; Carreira, E.

M., Yamamoto, H., Eds.; Elsevier: Amsterdam, 2012; Vol. 7, p 21-45.

15 Patel, R. N. In Comprehensive Chirality; Carreira, E. M., Yamamoto, H., Eds.; Elsevier:

Amsterdam, 2012; Vol. 9, p 288-317.

16 Muñoz Solano, D.; Hoyos, P.; Hernáiz, M. J.; Alcántara, A. R.; Sánchez-Montero, J. M.

Bioresour. Technol. 2012, 115, 196-207.

17 Martinez, C. A.; Hu, S.; Dumond, Y.; Tao, J.; Kelleher, P.; Tully, L. Org. Process Res. Dev.

2008, 12, 392-398.

18 Pellissier, H. Adv. Synth. Catal. 2011, 353, 1613-1666.

19 Fu, G. C. Acc. Chem. Res. 2004, 37, 542-547.

20 Bellemin-Laponnaz, S.; Tweddell, J.; Ruble, J. C.; Breitling, F. M.; Fu, G. C. Chem.

Commun. 2000, 1009-1010.

21 Wang, Y.-F.; Chen, C.-S.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1984, 106, 3695-

3696.

22 Edin, M.; Bäckvall, J.-E.; Córdova, A. Tetrahedron Lett. 2004, 45, 7697-7701.

23 Rowe, B. J.; Spilling, C. D. Tetrahedron: Asymmetry 2001, 12, 1701-1708.

24 Pàmies, O.; Bäckvall, J.-E. Chem. Rev. 2003, 103, 3247-3262.

25 Kim, C.; Park, J.; Kim, M.-J. In Comprehensive Chirality; Carreira, E. M., Yamamoto, H.,

Eds.; Elsevier: Amsterdam, 2012; Vol. 7, p 156-180.

26 Verho, O.; Bäckvall, J.-E. J. Am. Chem. Soc. 2015, 137, 3996-4009.

27 Pàmies, O.; Bäckvall, J.-E. J. Org. Chem. 2001, 66, 4022-4025.

28 Laurell Nash, A.; Widyan, K.; Moberg, C. ChemCatChem 2014, 6, 3314-3317.

29 Lente, G. Chirality 2010, 22, 907-913.

70

30 Gruber, C. C.; Lavandera, I.; Faber, K.; Kroutil, W. Adv. Synth. Catal. 2006, 348, 1789-

1805.

31 Voss, C. V.; Gruber, C. C.; Faber, K.; Knaus, T.; Macheroux, P.; Kroutil, W. J. Am. Chem.

Soc. 2008, 130, 13969-13972.

32 Voss, C. V.; Gruber, C. C.; Kroutil, W. Synlett 2010, 991-998.

33 Lackner, A. D.; Samant, A. V.; Toste, F. D. J. Am. Chem. Soc. 2013, 135, 14090-14093.

34 Turner, N. J. Chem. Rev. 2011, 111, 4073-4087.

35 Köhler, V.; Wilson, Y. M.; Dürrenberger, M.; Ghislieri, D.; Churakova, E.; Quinto, T.;

Knörr, L.; Häussinger, D.; Hollmann, F.; Turner, N. J.; Ward, T. R. Nat. Chem. 2013, 5, 93-

99.

36 Fransson, L.; Laurell, A.; Widyan, K.; Wingstrand, E.; Hult, K.; Moberg, C. ChemCatChem

2010, 2, 683-693.

37 Wingstrand, E.; Laurell, A.; Fransson, L.; Hult, K.; Moberg, C. Chem. – Eur. J. 2009, 15,

12107-12113.

38 Effenberger, F. Angew. Chem., Int. Ed. Engl. 1994, 33, 1555-1564.

39 Brunel, J.-M.; Holmes, I. P. Angew. Chem., Int. Ed. 2004, 43, 2752-2778.

40 Chen, F.-X.; Feng, X. Curr. Org. Synth. 2006, 3, 77-97.

41 North, M.; Usanov, D. L.; Young, C. Chem. Rev. 2008, 108, 5146-5226.

42 Wang, W.; Liu, X.; Lin, L.; Feng, X. Eur. J. Org. Chem. 2010, 2010, 4751-4769.

43 Gregory, R. J. H. Chem. Rev. 1999, 99, 3649-3682.

44 North, M. Tetrahedron: Asymmetry 2003, 14, 147-176.

45 Moberg, C.; Wingstrand, E. Synlett 2010, 355-367.

46 North, M.; Parkins, A. W.; Shariff, A. N. Tetrahedron Lett. 2004, 45, 7625-7627.

47 Breckenridge, C. B.; Stevens, J. T. In Principles and Methods of Toxicology; 5th ed.; Hayes,

A. W., Ed.; CRC Press: 2007, p 727-840.

48 Beckmann, M.; Haack, K.-J. Chem. Unserer Zeit 2003, 37, 88-97.

49 Veum, L.; Pereira, S. R. M.; van der Waal, J. C.; Hanefeld, U. Eur. J. Org. Chem. 2006,

1664-1671.

50 Oldenburg, O.; Kribben, A.; Baumgart, D.; Philipp, T.; Erbel, R.; Cohen, M. V. Curr. Opin.

Pharmacol. 2002, 2, 740-747.

51 Donaldson, K. H.; Shearer, B. G.; Uehling, D. E. PCT Patent WO9965877A1, 1999.

52 Uehling, D. E.; Shearer, B. G.; Donaldson, K. H.; Chao, E. Y.; Deaton, D. N.; Adkison, K.

K.; Brown, K. K.; Cariello, N. F.; Faison, W. L.; Lancaster, M. E.; Lin, J.; Hart, R.;

Milliken, T. O.; Paulik, M. A.; Sherman, B. W.; Sugg, E. E.; Cowan, C. J. Med. Chem.

2006, 49, 2758-2771.

53 Sacco, E.; Bientinesi, R.; Tienforti, D.; Racioppi, M.; Gulino, G.; D'Agostino, D.; Vittori,

M.; Bassi, P. Expert Opin. Drug Discovery 2014, 9, 433-448.

54 Ishide, T. Curr. Med. Res. Opin. 2002, 18, 407-413.

55 Ursino, M. G.; Vasina, V.; Raschi, E.; Crema, F.; De Ponti, F. Pharmacol. Res. 2009, 59,

221-234.

56 Igawa, Y.; Michel, M. C. Naunyn-Schmiedeberg's Arch. Pharmacol. 2013, 386, 177-183.

57 Ellsworth, P.; Fantasia, J. Expert Opin. Invest. Drugs 2015, 24, 413-419.

58 Norsikian, S.; Holmes, I.; Lagasse, F.; Kagan, H. B. Tetrahedron Lett. 2002, 43, 5715-5717.

71

59 Holt, J.; Hanefeld, U. Curr. Org. Synth. 2009, 6, 15-37.

60 van der Gen, A.; Brussee, J. In Stereoselective Biocatalysis; Patel, R. N., Ed.; CRC Press:

2000, p 289-320.

61 Winkler, M.; Glieder, A.; Steiner, K. In Comprehensive Chirality; Carreira, E. M.,

Yamamoto, H., Eds.; Elsevier: Amsterdam, 2012; Vol. 7, p 350-371.

62 Dadashipour, M.; Asano, Y. ACS Catal. 2011, 1, 1121-1149.

63 Effenberger, F. In Stereoselective Biocatalysis; Patel, R. N., Ed.; CRC Press: 2000, p 321-

342.

64 Purkarthofer, T.; Pabst, T.; van den Broek, C.; Griengl, H.; Maurer, O.; Skranc, W. Org.

Process Res. Dev. 2006, 10, 618-621.

65 Hanefeld, U.; Li, Y.; Sheldon, R. A.; Maschmeyer, T. Synlett 2000, 1775-1776.

66 Hanefeld, U.; Straathof, A. J. J.; Heijnen, J. J. J. Mol. Catal. B: Enzym. 2001, 11, 213-218.

67 Veum, L.; Kuster, M.; Telalovic, S.; Hanefeld, U.; Maschmeyer, T. Eur. J. Org. Chem.

2002, 1516-1522.

68 Hamberg, A.; Lundgren, S.; Penhoat, M.; Moberg, C.; Hult, K. J. Am. Chem. Soc. 2006,

128, 2234-2235.

69 Hamberg, A.; Lundgren, S.; Wingstrand, E.; Moberg, C.; Hult, K. Chem. – Eur. J. 2007, 13,

4334-4341.

70 Kazlauskas, R. J.; Weissfloch, A. N. E.; Rappaport, A. T.; Cuccia, L. A. J. Org. Chem.

1991, 56, 2656-2665.

71 Rotticci, D.; Hæffner, F.; Orrenius, C.; Norin, T.; Hult, K. J. Mol. Catal. B: Enzym. 1998, 5,

267-272.

72 Anderson, E. M.; Larsson, K. M.; Kirk, O. Biocatal. Biotransform. 1998, 16, 181-204.

73 Schmid, R. D.; Verger, R. Angew. Chem., Int. Ed. 1998, 37, 1608-1633.

74 Berglund, P.; Hult, K. In Stereoselective Biocatalysis; Patel, R. N., Ed.; CRC Press: 2000, p

633-657.

75 Inagaki, M.; Hiratake, J.; Nishioka, T.; Oda, J. J. Am. Chem. Soc. 1991, 113, 9360-9361.

76 Inagaki, M.; Hatanaka, A.; Mimura, M.; Hiratake, J.; Nishioka, T.; Oda, J. Bull. Chem. Soc.

Jpn. 1992, 65, 111-120.

77 Inagaki, M.; Hiratake, J.; Nishioka, T.; Oda, J. J. Org. Chem. 1992, 57, 5643-5649.

78 Kanerva, L. T.; Rahiala, K.; Sundholm, O. Biocatalysis 1994, 10, 169-180.

79 Paizs, C.; Toşa, M.; Majdik, C.; Tähtinen, P.; Irimie, F. D.; Kanerva, L. T. Tetrahedron:

Asymmetry 2003, 14, 619-627.

80 Paizs, C.; Tähtinen, P.; Lundell, K.; Poppe, L.; Irimie, F.-D.; Kanerva, L. T. Tetrahedron:

Asymmetry 2003, 14, 1895-1904.

81 Paizs, C.; Tähtinen, P.; Toşa, M.; Majdik, C.; Irimie, F.-D.; Kanerva, L. T. Tetrahedron

2004, 60, 10533-10540.

82 Veum, L.; Kanerva, L. T.; Halling, P. J.; Maschmeyer, T.; Hanefeld, U. Adv. Synth. Catal.

2005, 347, 1015-1021.

83 Veum, L.; Hanefeld, U. Tetrahedron: Asymmetry 2004, 15, 3707-3709.

84 Veum, L.; Hanefeld, U. Synlett 2005, 2382-2384.

85 Shibasaki, M.; Kanai, M.; Matsunaga, S.; Kumagai, N. Acc. Chem. Res. 2009, 42, 1117-

1127.

72

86 Moberg, C. In Cooperative Catalysis; Peters, R., Ed.; Wiley-VCH Verlag GmbH & Co.

KGaA: 2015, p 35-66.

87 Ma, J.-A.; Cahard, D. Angew. Chem., Int. Ed. 2004, 43, 4566-4583.

88 Kanai, M.; Kato, N.; Ichikawa, E.; Shibasaki, M. Synlett 2005, 1491-1508.

89 Hamashima, Y.; Sawada, D.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 2641-

2642.

90 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.

91 Belokon, Y. N.; Green, B.; Ikonnikov, N. S.; North, M.; Tararov, V. I. Tetrahedron Lett.

1999, 40, 8147-8150.

92 Belokon’, Y. N.; Blacker, A. J.; Clutterbuck, L. A.; North, M. Org. Lett. 2003, 5, 4505-

4507.

93 Belokon, Y. N.; Ishibashi, E.; Nomura, H.; North, M. Chem. Commun. 2006, 1775-1777.

94 Belokon, Y. N.; Gutnov, A. V.; Moskalenko, M. A.; Yashkina, L. V.; Lesovoy, D. E.;

Ikonnikov, N. S.; Larichev, V. S.; North, M. Chem. Commun. 2002, 244-245.

95 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.

96 Belokon, Y. N.; Carta, P.; North, M. Lett. Org. Chem. 2004, 1, 81-83.

97 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.

98 Belokon’, Y. N.; Blacker, A. J.; Carta, P.; Clutterbuck, L. A.; North, M. Tetrahedron 2004,

60, 10433-10447.

99 Zhang, Z.; Wang, Z.; Zhang, R.; Ding, K. Angew. Chem., Int. Ed. 2010, 49, 6746-6750.

100 North, M. Angew. Chem., Int. Ed. 2010, 49, 8079-8081.

101 Belokon, Y. N.; Blacker, A. J.; Clutterbuck, L. A.; Hogg, D.; North, M.; Reeve, C. Eur. J.

Org. Chem. 2006, 4609-4617.

102 Marvel, C. S.; Brace, N. O.; Miller, F. A.; Johnson, A. R. J. Am. Chem. Soc. 1949, 71, 34-

36.

103 Hoffmann, H. M. R.; Ismail, Z. M.; Hollweg, R.; Zein, A. R. Bull. Chem. Soc. Jpn. 1990,

63, 1807-1810.

104 Scholl, M.; Lim, C.-K.; Fu, G. C. J. Org. Chem. 1995, 60, 6229-6231.

105 Okimoto, M.; Chiba, T. Synthesis 1996, 1188-1190.

106 Watahiki, T.; Ohba, S.; Oriyama, T. Org. Lett. 2003, 5, 2679-2681.

107 Baeza, A.; Nájera, C.; Retamosa, M. d. G.; Sansano, J. M. Synthesis 2005, 2787-2797.

108 Zhang, W.; Shi, M. Org. Biomol. Chem. 2006, 4, 1671-1674.

109 Zhang, J.; Du, G.; Xu, Y.; He, L.; Dai, B. Tetrahedron Lett. 2011, 52, 7153-7156.

110 Lundgren, S.; Wingstrand, E.; Penhoat, M.; Moberg, C. J. Am. Chem. Soc. 2005, 127,

11592-11593.

111 Wingstrand, E.; Lundgren, S.; Penhoat, M.; Moberg, C. Pure Appl. Chem. 2006, 78, 409-

414.

73

112 Lundgren, S.; Wingstrand, E.; Moberg, C. Adv. Synth. Catal. 2007, 349, 364-372.

113 Laurell Nash, A., Development and Studies of the Processes Involved in Minor Enantiomer

Recycling, School of Chemical Science and Engineering, KTH Royal Institute of

Technology, 2014.

114 Laurell Nash, A.; Hertzberg, R.; Wen, Y.-Q.; Dahlgren, B.; Brinck, T.; Moberg, C.

Manuscript

115 Baeza, A.; Nájera, C.; Sansano, J. M.; Saá, J. M. Tetrahedron: Asymmetry 2005, 16, 2385-

2389.

116 Baeza, A.; Nájera, C.; Sansano, J. M.; Saá, J. M. Tetrahedron: Asymmetry 2011, 22, 1282-

1291.

117 Casas, J.; Baeza, A.; Sansano, J. M.; Nájera, C.; Saá, J. M. Tetrahedron: Asymmetry 2003,

14, 197-200.

118 Baeza, A.; Casas, J.; Nájera, C.; Sansano, J. M.; Saá, J. M. Eur. J. Org. Chem. 2006, 1949-

1958.

119 Baeza, A.; Casas, J.; Nájera, C.; Sansano, J. M.; Saá, J. M. Angew. Chem., Int. Ed. 2003, 42,

3143-3146.

120 Baeza, A.; Nájera, C.; Sansano, J. M.; Saá, J. M. Chem. – Eur. J. 2005, 11, 3849-3862.

121 Laurell, A.; Moberg, C. Eur. J. Org. Chem. 2011, 3980-3984.

122 Menéndez, E.; Brieva, R.; Rebolledo, F.; Gotor, V. J. Chem. Soc., Chem. Commun. 1995,

989-990.

123 Hernández, L.; Luna, H.; Solís, A.; Vázquez, A. Tetrahedron: Asymmetry 2006, 17, 2813-

2816.

124 Kang, E. J.; Lee, E. Chem. Rev. 2005, 105, 4348-4378.

125 Wolfe, J. P.; Hay, M. B. Tetrahedron 2007, 63, 261-290.

126 Nazabadioko, S.; Pérez, R. J.; Brieva, R.; Gotor, V. Tetrahedron: Asymmetry 1998, 9, 1597-

1604.

127 Monterde, M. I.; Nazabadioko, S.; Rebolledo, F.; Brieva, R.; Gotor, V. Tetrahedron:

Asymmetry 1999, 10, 3449-3455.

128 Monterde, M. I.; Brieva, R.; Gotor, V. Tetrahedron: Asymmetry 2001, 12, 525-528.

129 Effenberger, F.; Gutterer, B.; Ziegler, T.; Eckhardt, E.; Aichholz, R. Liebigs Ann. Chem.

1991, 47-54.

130 Dukes, M.; Smith, L. H. J. Med. Chem. 1971, 14, 326-328.

131 Powell, C. E.; Slater, I. H.; LeCompte, L.; Waddell, J. E. J. Pharmacol. Exp. Ther. 1958,

122, 480-488.

132 Wang, Y.-F.; Chen, S.-T.; Liu, K. K. C.; Wong, C.-H. Tetrahedron Lett. 1989, 30, 1917-

1920.

133 van Almsick, A.; Buddrus, J.; Hönicke-Schmidt, P.; Laumen, K.; Schneider, M. P. J. Chem.

Soc., Chem. Commun. 1989, 1391-1393.

134 North, M.; Omedes-Pujol, M.; Williamson, C. Chem. – Eur. J. 2010, 16, 11367-11375.

135 Mühlhäuser, T., Enantioselective Preparation of O-Acylated Cyanohydrins from

Heteroatom-containing Aldehydes, Diploma work, KTH Royal Institute of Technology,

2013.

136 Lin, G.; Han, S.; Li, Z. Tetrahedron 1999, 55, 3531-3540.

74

137 Matsuo, N.; Ohno, N. Tetrahedron Lett. 1985, 26, 5533-5534.

138 Liu, Y.; Bae, B. H.; Alam, N.; Hong, J.; Sim, C. J.; Lee, C.-O.; Im, K. S.; Jung, J. H. J. Nat.

Prod. 2001, 64, 1301-1304.

139 Duval, R. A.; Poupon, E.; Brandt, U.; Hocquemiller, R. Biochim. Biophys. Acta, Bioenerg.

2005, 1709, 191-194.

140 Csuk, R.; Barthel, A.; Kluge, R.; Ströhl, D. Bioorg. Med. Chem. 2010, 18, 7252-7259.

141 Pérez, M.; Pérez, D. I.; Martínez, A.; Castro, A.; Gómez, G.; Fall, Y. Chem. Commun. 2009,

3252-3254.

142 Carter, N. B.; Nadany, A. E.; Sweeney, J. B. J. Chem. Soc., Perkin Trans. 1 2002, 2324-

2342.

143 Hiyama, T.; Oishi, H.; Suetsugu, Y.; Nishide, K.; Saimoto, H. Bull. Chem. Soc. Jpn. 1987,

60, 2139-2150.

144 Schlessinger, R. H.; Li, Y.-J. J. Am. Chem. Soc. 1996, 118, 3301-3302.

145 Li, Y.-J.; Ho, G.-M.; Chen, P.-Z. Tetrahedron: Asymmetry 2009, 20, 1854-1863.

146 Nishide, K.; Aramata, A.; Kamanaka, T.; Inoue, T.; Node, M. Tetrahedron 1994, 50, 8337-

8346.

147 Kondoh, A.; Arlt, A.; Gabor, B.; Fürstner, A. Chem. – Eur. J. 2013, 19, 7731-7738.

148 Prakash Rao, H. S.; Rafi, S.; Padmavathy, K. Tetrahedron 2008, 64, 8037-8043.

149 Pöchlauer, P.; Riebel, P.; Mayrhofer, H.; Wirth, I. PCT Patent US 6417377, 2002.

150 Syed, J.; Förster, S.; Effenberger, F. Tetrahedron: Asymmetry 1998, 9, 805-815.

151 Effenberger, F.; Syed, J. Tetrahedron: Asymmetry 1998, 9, 817-825.

152 Bühler, H.; Bayer, A.; Effenberger, F. Chem. – Eur. J. 2000, 6, 2564-2571.

153 Ohta, H.; Kimura, Y.; Sugano, Y. Tetrahedron Lett. 1988, 29, 6957-6960.

154 Herrmann, K.; Simchen, G. Synthesis 1979, 204-205.

155 Ikezaki, M.; Umino, N.; Gaino, M.; Aoe, K.; Iwakuma, T.; Oh-Ishi, T. Yakugaku Zasshi

1986, 106, 80-89.

156 Cooke, J. W. B.; Glover, B. N.; Lawrence, R. M.; Sharp, M. J.; Tymoschenko, M. F. PCT

Patent WO2002066418A2, 2002.

157 Zhang, C. PCT Patent WO2010118291A2, 2010.

158 Lawrence, R. M.; Millar, A. PCT Patent WO2001042195A1, 2001.

159 Dear, A. E.; Liu, H. B.; Mayes, P. A.; Perlmutter, P. Org. Biomol. Chem. 2006, 4, 3778-

3784.

160 Imanishi, M.; Itou, S.; Washizuka, K.; Hamashima, H.; Nakajima, Y.; Araki, T.;

Tomishima, Y.; Sakurai, M.; Matsui, S.; Imamura, E.; Ueshima, K.; Yamamoto, T.;

Yamamoto, N.; Ishikawa, H.; Nakano, K.; Unami, N.; Hamada, K.; Matsumura, Y.;

Takamura, F.; Hattori, K. J. Med. Chem. 2008, 51, 4002-4020.

161 Hama, T.; Liu, X.; Culkin, D. A.; Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 11176-11177.

162 Hama, T.; Culkin, D. A.; Hartwig, J. F. J. Am. Chem. Soc. 2006, 128, 4976-4985.

163 Hama, T.; Ge, S.; Hartwig, J. F. J. Org. Chem. 2013, 78, 8250-8266.

164 Dai, X.; Strotman, N. A.; Fu, G. C. J. Am. Chem. Soc. 2008, 130, 3302-3303.

165 Sato, M.; Miyaura, N.; Suzuki, A. Chem. Lett. 1989, 1405-1408.

166 Gooßen, L. J. Chem. Commun. 2001, 669-670.

75

167 Zimmermann, B.; Dzik, W. I.; Himmler, T.; Goossen, L. J. J. Org. Chem. 2011, 76, 8107-

8112.

168 Liu, X.-x.; Deng, M.-z. Chem. Commun. 2002, 622-623.

169 Lu, T.-Y.; Xue, C.; Luo, F.-T. Tetrahedron Lett. 2003, 44, 1587-1590.

170 Duan, Y.-Z.; Deng, M.-Z. Tetrahedron Lett. 2003, 44, 3423-3426.

171 Liu, C.; He, C.; Shi, W.; Chen, M.; Lei, A. Org. Lett. 2007, 9, 5601-5604.

172 Duan, Y.; Zhang, J.; Yang, J.; Deng, M. Chin. J. Chem. 2009, 27, 179-182.

173 Peng, Z.-Y.; Wang, J.-P.; Cheng, J.; Xie, X.-m.; Zhang, Z. Tetrahedron 2010, 66, 8238-

8241.

174 Lundin, P. M.; Fu, G. C. J. Am. Chem. Soc. 2010, 132, 11027-11029.

175 Molander, G. A.; Traister, K. M.; Barcellos, T. J. Org. Chem. 2013, 78, 4123-4131.

176 De Joarder, D.; Jennings, M. P. Tetrahedron Lett. 2013, 54, 3990-3992.

177 De Joarder, D.; Jennings, M. P. Eur. J. Org. Chem. 2015, 3303-3313.

178 Lei, A.; Zhang, X. Tetrahedron Lett. 2002, 43, 2525-2528.

179 Merino, P.; Marqués-López, E.; Herrera, R. P. Adv. Synth. Catal. 2008, 350, 1195-1208.

180 Kolodiazhnyi, O. I. Tetrahedron: Asymmetry 2005, 16, 3295-3340.

181 Pudovik, A. N.; Konovalova, I. V. Synthesis 1979, 81-96.

182 He, H.-W.; Yuan, J.-L.; Peng, H.; Chen, T.; Shen, P.; Wan, S.-Q.; Li, Y.; Tan, H.-L.; He,

Y.-H.; He, J.-B.; Li, Y. J. Agric. Food Chem. 2011, 59, 4801-4813.

183 Kategaonkar, A. H.; Pokalwar, R. U.; Sonar, S. S.; Gawali, V. U.; Shingate, B. B.; Shingare,

M. S. Eur. J. Med. Chem. 2010, 45, 1128-1132.

184 Kaboudin, B.; Karimi, M. Bioorg. Med. Chem. Lett. 2006, 16, 5324-5327.

185 Kaboudin, B.; Karimi, M. ARKIVOC (Gainesville, FL, U. S.) 2007, 124-132.

186 Kolodiazhnyi, O. I. Russ. Chem. Rev. 2011, 80, 883-910.

187 Li, Y.-F.; Hammerschmidt, F. Tetrahedron: Asymmetry 1993, 4, 109-120.

188 Colton, I. J.; Yin, D.; Grochulski, P.; Kazlauskas, R. J. Adv. Synth. Catal. 2011, 353, 2529-

2544.

189 Zhang, Y.; Yuan, C.; Li, Z. Tetrahedron 2002, 58, 2973-2978.

190 Pàmies, O.; Bäckvall, J.-E. J. Org. Chem. 2003, 68, 4815-4818.

191 Takaki, K.; Itono, Y.; Nagafuji, A.; Naito, Y.; Shishido, T.; Takehira, K.; Makioka, Y.;

Taniguchi, Y.; Fujiwara, Y. J. Org. Chem. 2000, 65, 475-481.

192 Müller, P.; Müller-Dolezal, H.; Stoltz, R.; Söll, H. Houben-Weyl Methods of Organic

Chemistry Vol. XII/2, 4th Edition: Phosphorus Compounds II (Derivatives of Phosphorous

and Phosphoric Acid); Thieme: Germany, 1964.

193 Okamoto, Y.; Azuhata, T. Bull. Chem. Soc. Jpn. 1984, 57, 2693-2694.

194 Breuer, E. In The Chemistry of Organophosphorus Compounds; Hartley, F. R., Ed.; John

Wiley & Sons, Ltd: 1996, p 653-729.

195 Sekine, M.; Satoh, M.; Yamagata, H.; Hata, T. J. Org. Chem. 1980, 45, 4162-4167.

196 Sekine, M.; Kume, A.; Hata, T. Tetrahedron Lett. 1981, 22, 3617-3620.

197 Katzhendler, J.; Ringel, I.; Karaman, R.; Zaher, H.; Breuer, E. J. Chem. Soc., Perkin Trans.

2 1997, 341-350.

198 Afarinkia, K.; Echenique, J.; Nyburg, S. C. Tetrahedron Lett. 1997, 38, 1663-1666.

199 Katritzky, A. R.; Suzuki, K.; Wang, Z. Synlett 2005, 1656-1665.

76

200 Katritzky, A. R.; Rachwal, S.; Hitchings, G. J. Tetrahedron 1991, 47, 2683-2732.

201 McConnell, R. L.; Coover Jr, H. W. J. Am. Chem. Soc. 1956, 78, 4450-4452.

202 Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120, 5315-5316.

203 Zhou, X.; Liu, X.; Yang, X.; Shang, D.; Xin, J.; Feng, X. Angew. Chem., Int. Ed. 2008, 47,

392-394.

204 Maeda, H.; Takahashi, K.; Ohmori, H. Tetrahedron 1998, 54, 12233-12242.

205 Baidya, M.; Brotzel, F.; Mayr, H. Org. Biomol. Chem. 2010, 8, 1929-1935.

206 Breugst, M.; Corral Bautista, F.; Mayr, H. Chem. – Eur. J. 2012, 18, 127-137.

207 Li, W.; Qin, S.; Su, Z.; Hu, C.; Feng, X. Comput. Theor. Chem. 2012, 989, 44-50.

208 Wang, C.; Xu, C.; Tan, X.; Peng, H.; He, H. Org. Biomol. Chem. 2012, 10, 1680-1685.

209 Satyanarayana, T.; Abraham, S.; Kagan, H. B. Angew. Chem., Int. Ed. 2009, 48, 456-494.

210 Katritzky, A. R.; Huang, T.-B.; Voronkov, M. V.; Steel, P. J. J. Org. Chem. 2000, 65, 8069-

8073.