Chemoenzymatic synthesis of enantiopure 1-phenyl-2-haloethanols and their esters

Article abstract of DOI:10.3109/10242422.2010.501406

Several secondary haloalcohols have been resolved by esterification catalyzed by lipases A and B from Candida antarctica (CALA and CALB, respectively). Immobilized CALB (Novozym 435) gave better selectivities and faster reaction rates than CALA (Novozyme

Full text of DOI:10.3109/10242422.2010.501406

Biocatalysis and Biotransformation, July–August 2010; 28(4): 272–278
ORIGINAL ARTICLE
Chemoenzymatic synthesis of enantiopure 1-phenyl-2-haloethanols
and their esters
SINA MARIA LYSTVET1, BÅRD HELGE HOFF1, THORLEIF ANTHONSEN1 &
ELISABETH EGHOLM JACOBSEN2
1Department of Chemistry, Norwegian University of Science and Technology,Trondheim, Norway and 2Department of
Chemistry and Material Sciences, South Trøndelag University College,Trondheim, Norway
Abstract
Several secondary haloalcohols have been resolved by esterification catalyzed by lipases A and B from Candida antarctica
(CALA and CALB, respectively). Immobilized CALB (Novozym 435) gave better selectivities and faster reaction rates
than CALA (Novozyme 735) with all of the substrates. Vinyl butanoate was the favoured acyl donor compared to vinyl
acetate due to remarkably faster reactions. The alcohols (R)-2a, (R)-2c and the butanoates (S)-3a-c have been isolated
with 98–99% enantiomeric excess (ee) in good yields.
Keywords: Kinetic resolution, CALA, CALB, enantiopure secondary alcohols, enantiopure secondary esters
Introduction
Reymond 2004).Similar alcohols have been obtained
Enantiopure secondary haloalcohols are valuable
building blocks for synthesis of bioactive molecules
and materials. In particular, fluoro compounds have
become valuable as generic pharmaceuticals, since bin-
ding interactions, metabolic stability and reactivities
are improved when fluorine or fluoroalkyl groups are
introduced. (Hagmann 2008). Several strategies exist
for preparation of enantiopure alcohols, lipase cata-
lyzed resolution is one option. However, success of
this approach depends on the enantioselectivity of
the enzyme. Therefore, development and investiga-
tion of kinetic resolutions are still important.
A series of 1-(4-benzyloxyphenyl)ethanols, 2a-e,
have been prepared by reduction of the ketones 1a-e
and resolved by CALA and CALB.These molecules
can easily be converted into valuable building blocks
by modifications of the alcohol function, nucleo-
philic reaction at the halogens, conversion to epox-
ides, electrophilic aromatic substitution and
deprotection and functionalization of the phenolic
unit. Previously, some of these alcohols have been
used in the synthesis of bioactive compounds; how-
ever, not always in 100% enantiopurity (Azumai
et al. 1989; Goswami et al. 2000; 2001; Renaudet &
in 94–99% ee by different resolution and asymme-
trization methods (Azumai & Minamii 1988;
Gamble et al. 1998; Goswami & Goswami 2005;
Fuglseth et al. 2008a). Structurally related com-
pounds have also been resolved using CALB (Kour-
ist et al. 2005) and CALA (Fuglseth et al. 2006).
With the aim to develop more efficient and new
lipase catalyzed methods for halogenated secondary
alcohols and to investigate substituent effects on the
enantiomeric ratio (E), we have studied CALA and
CALB as catalysts in resolution of alcohols 2a-e using
both vinyl acetate and vinyl butanoate as acyl donors.
CALA has been found to exhibit an almost uni-
form activity towards various straight chain primary
alcohols and carboxylic acids, but shows low activity
towards very short chained acids. It is selective towards
trans-fatty acids and is stable in the acidic pH range.
It is also considered to be an excellent biocatalyst for
kinetic resolutions of amino acids and amino esters (de
Maria et al. 2005). ω-Cyanohydrins have also been
resolved by CALA with quite high enantioselectivity
(de Gonzalo et al. 2004). The enzyme has also been
found to exhibit high activity towards a diversity of
sterically hindered alcohols, including both secondary
Correspondence: Elisabeth Egholm Jacobsen, Department of Chemistry and Material Sciences, South Trondelag University College, N-7004 Trondheim,
/
Norway. Tel: + 47 988 43 559, Fax: + 47 73 55 96 31. E-mail: elisabeth.e.jacobsen@hist.no or elis-jac@online.no
(Received 9 May 2010; accepted 11 June 2010)
ISSN 1024-2422 print/ISSN 1029-2446 online © 2010 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)
DOI: 10.3109/10242422.2010.501406

Enantiopure 1-phenyl-2-haloethanols and -esters 273
and tertiary alcohols (Kirk & Christensen 2002;
Schmidt et al. 2005). Secondary alcohols with aro-
matic groups on both sides of the stereocentre have
also been resolved with high enantiomeric ratio cata-
lyzed by CALA (Tjosås et al. 2008). Recently, the
X-ray structure of the lipase was published. It has been
found that the CALA structure is quite different from
CALB and it is suggested that it represents the first in
a new lipase sub-family (Ericsson et al. 2008).
CALB is known to effectively catalyze reactions with
a wide range of straight chain secondary alcohols.
However, the enzyme shows a limitation in substrate
size and electronic effects of the smallest group con-
nected to the stereocentre, which is placed in the ‘ste-
reospecificity pocket’ in the active site of the enzyme
(Anthonsen & Hoff 1998; Jacobsen et al. 2000).
are in Hertz (Hz). Mass spectra were recorded using
a MAT 95 XL (ThermoQuest Finnigan, San Jose,
California, USA) with EI probe (DIP) (2e and 3d)
and ESI (3a-3c) as ionization sources. Melting points
were determined using Sanyo Gallenkamp (USA)
and Büchi (Switzerland) melting point instruments.
GC-MS analyses were performed using a Varian
CP-3800 Gas Chromatograph and a Varian 5 MS
(i.d. 0.25 mm, 30 m, film density 0.25 μm) and He
carrier gas. Enantiomeric ratios, E, were calculated
based on ping-pong bi-bi kinetics using the computer
program E & K Calculator 2.1b0 PCC (Anthonsen et
al. 1996).
Chiral HPLC analyses of 2a-d, 3a-d and 4a were
performed with two instruments; a Varian HPLC
coupled to a Varian 2550 UV detector with a Varian
Prostar 4000 autosampler and an Agilent 1100 Series:
G1379ADegasser,G1311AQuatPumpwithG1313A
coupled to a Bruker UV detector and an ALS auto-
injector. The HPLC column was a Daicel Chiracel
OD-H (i.d. 4.6 mm, 25 cm, film density 5 μm). Hexane
and isopropanol (96:4) were used as eluents, flow rate
1 mL minꢀ1. Retention times of alcohol enantiom-
ers: 2a: 38.4 min ((S)-2a), 43.3 min ((R)-2a), RS ꢂ
2.0, 2b: 32.5 min ((S)-2b), 35.9 min ((R)-2b), RS ꢂ
1.7, retention times butanoate enantiomers: 3a: 10.0
min ((S)-3a), 11.4 min ((R)-3a), RS ꢂ 2.2, 3b: 9.6
min ((S)-3b), 11.1 min ((R)-3b), RS ꢂ 2.4, 3c: 9.3 min
((S)-3c), 10.5 min ((R)-3c), RS ꢂ 1.6, 3d: 7.1 min
((R)-3d), 7.8 min ((S)-3d), RS ꢂ 1.8, acetate 4a:
13.4 min ((S)-4a), 14.8 min ((R)-4a), RS ꢂ 1.7.
Chiral GLC analyses of the alcohols 2c and 2d,
butanoate 3d and the acetates 4c and 4d were per-
formed using aVarian 3380 gas chromatograph (FID
detector) with a Varian CP-8410 autoinjector and a
Varian CP Chirasil-Dex column (i.d. 0.32 mm, 25 m,
film density 0.25 μm), 7 psi and 60 mL min–1. H2
5.0 as carrier gas. Temperature program alcohol 2c:
80–188°C, 10.0°C min–1, 188°C (6), 188–189°C,
1.0°C min–1 (2), 189–193°C, 1.0°C min–1, retention
times 2c enantiomers: 14.1 min ((R)-2c), 14.3 min
((S)-2c), RS ꢂ 1.8, temperature program alcohol 2d:
80–155°C, 2.5°C min–1, 155–158°C, 1.0°C min–1, 158–
162°C, 0.1°C min–1, retention times 2d enantiomers:
43.1 min ((R)-2d), 43.8 min ((S)-2d), RS ꢂ 2.1,
temperature program butanoate 3d: 80–148°C,
8.0°C min–1, 148°C (79), retention times 3d
enantiomers: 75.8 min ((S)-3d), 78.2 min ((R)-3d),
RS ꢂ 1.65, temperature program acetate 4d: 80–
155°C, 2.5°C minꢀ1, 155–158°C, 1.0°C min–1, 158–
162°C, 0.1°C min-1, retention times 4d enantiomers:
42.8 min ((S)-4d), 44.0 min ((R)-4d), RS ꢂ 3.3.
Materials and methods
General
Enzymes were gifts from Novozymes AS (Bagsværd,
Denmark) andViaZym BV (Delft,The Netherlands).
Enzyme specifications:Novozym 435 (CALB) immo-
bilized on macroporous acrylic resin, activity 10 000
PLUgꢀ 1,batchnr.LC200204,Novozym735(CALA)
immobilized on MCM-41, Lot nr LDN00021, (both
Novozymes A/S), VZ 1031-3 (CALA) immobilized
on macroporous beads, activity 725 U gꢀ1, lot nr.
080100 from ViaZym BV.
Analytical grade chemicals were purchased from
Sigma-Aldrich (Buchs, Switzerland). The solvents
were from VWR International AS, (Oslo, Norway)
except isopropanol, which was purchased from Arcus
(Oslo, Norway). Hexane was dried over Al2O3. Dichlo-
romethane (CH2Cl2) was dried by distillation over
calcium hydride (CaH2). Purification and separation
of products were performed using VersaFlashTM from
Supelco, Sigma-Aldrich (Buchs, Switzerland) column
VersaPakTM (silica, spherical 23 ꢁ 110 mm, 23 g),
Cat.No. 97758-U, max. pressure 60 psi), except for
separation of 2a from 3a, where VersaPakTM Silica
Cartridge (40 ꢁ 150 mm, Cat. No. 97706-U, max
pressure 80 psi) was used. The eluent for separation
of alcohols and butanoates was pentane:ethylacetate,
4:1. By reduction of 1a, the product 2a was purified
with hexane:ethylacetate, 4:1.
Analyses
Optical rotations ([α]D) were determined at 23°C
using a Perkin-Elmer 243 B instrument, concentra-
tions are given in g/100 mL. NMR was recorded with
a Bruker Avance DPX 300 and Bruker Avance DPX
400 operating at 300 MHz and 400 MHz for 1H and
75 and 100 MHz for 13C, respectively. Chemical shifts
are in ppm relative to TMS and coupling constants
Determination of absolute configurations
The absolute configuration of the faster reacting enan-
tiomer in lipase catalyzed resolution was determined

274
S. M. Lystvet et al.
by the known enantiopreference of CALB (Hoff &
Anthonsen 1999) and by comparing the elution orders
of the enantiomers with GLC elution orders of similar
enantiopure compounds synthesized from (S)-epichlo-
rohydrin (Jacobsen et al. 2000). All enzyme prepara-
tions showed the same enantiopreference. Moreover,
the elution orders on Daicel Chiracel OD-H of the
faster reacting enantiomers were the same.
to an ice bath (75 mL) and white crystals were pre-
cipitated at pH 8. A few drops of HCl (10%) were
added for neutralization. The crystals were filtered
off and re-crystallized from MeOH, then dried
and concentrated. The product was purified by
VersaFlash column chromatography (hexane:
EtOAc, 4:1).
1-(4-(Benzyloxy)phenyl)-2-bromoethanol (2a).Yield
1.60 g (67%), purity (HPLC) 100%, mp 76–78°C.
1H-NMR (CDCl3): δ 2.6 (s, 1H, OH), ABX: nA 3.53,
nB 3.60, nX 4.88, JAB ꢂ 10.4, JAX ꢂ 3.4, JBX ꢂ 9.0,
(–CH2-CHOH-), 5.07 (s, 2H, Ph-CH2), 6.99 (m, 2H,
Ar-CH), 7.23–7.45 (m, 5ꢃ2H, Ar-H). 13C-NMR
(CDCl3): δ 40.3, 70.0, 73.4, 115.0 (2C), 127.2 (2C),
127.4 (2C), 128.0, 128.6 (2C), 132.6, 136.8, 158.9.
Syntheses
Compounds 1a, 1b and 1d were synthesized as des-
cribed previously (Fuglseth et al. 2008b).
1-(4-(Benzyloxy)phenyl)-2-chloroethanone (1c) (Gamble
et al. 1998). 1-(4-Benzyloxyphenyl)ethanone (5.49 g,
19.0 mmole) and p-TsOH (2.31g, 12.1 mmole) were
suspended in MeOH (500 mL) at 40°C. 1,3-Dichloro-
5,5-dimethylhydantoin (3.59 g, 18.2 mmole) was
added in portions over 1 h, followed by agitation of the
reaction mixture at 40°C for 22 h. MeOH was distilled
off until crystallization started, followed by slow addi-
tion of water (200 mL). The suspension formed was
stirred for 45 min followed by isolation of the crystal-
line material by filtration.The crude product was dried
and re-crystallized from EtOAc/EtOH, which gave
4.03 g (63%) of a white solid, mp. 111–112°C. Melting
point, 1H and 13C-NMR were in correspondence with
reported data (Gamble et al. 1998).
1-(4-(Benzyloxy)phenyl)-2-chloroethanol (2b). Yield
1.46 g (97%), purity (HPLC) 94%, mp 80–81°C. (S)-
2b: 78–79°C. 1H-NMR (CDCl3): δ 2.59 (br.s, 1H,
OH), 3.63 (dd, 1H, CH2, J ꢂ 8.8 Hz, 11.2 Hz), 3.71
(dd, 1H, CH2, J ꢂ 3.6 Hz, 11.2 Hz), 4.85 (dd, 1H,
CH, J ꢂ 3.4 Hz, 8.8 Hz), 5.07 (s, 2H, Ph-CH2), 6.98
(m, 2H, Ar-H), 7.25–7.48 (m, 7H, Ar-H). 13C-NMR
(CDCl3): δ 50.9 (1C, CH2-Cl), 70.7 (1C, Ph-CH2),
73,7 (1C, CHOH), 115.0 (2C, Ar-CH), 127.3 (2C,
Ar-CH), 127.4 (2C, Ar-CH), 128.0 (1C, Ar-CH),
128.6 (2C, Ar-CH), 132.3 (1C, Ar-CH), 136.8 (1C,
Ar-CH), 158.9 (1C, Ar-CH).
1-(4-(Benzyloxy)phenyl)-2,2-dichloroethanone
(1e).
1-(4-(Benzyloxy)phenyl)-2-fluoroethanol (2c). Yield 0.89
g (88%), purity (GLC) 90%, mp 69–71°C. 1H-NMR
(CDCl3): δ 2.41(d, 1H, OH), 4.28–4.59 (m, 2H, CH2,
J ꢂ 3.5 Hz, 8.3 Hz, 9.6 Hz, 40.2 Hz), 4.92–5.05
(m, 1H, CH), 5.07 (s, 2H, Ph-CH2), 7.0 (m, 2H,
Ar-CH), 7.25–7.43 (m, 7H, Ar-H). 13C-NMR
(CDCl3): δ 70.0, (1C, Ph-CH2), 72.5 (1C, CH), 87.1
(1C, CH2), 115.0 (2C, Ar-CH), 127.4 (2C, Ar-CH),
127.6 (2C, Ar-CH), 128.0 (1C, Ar-CH), 128.6 (1C,
Ar-CH), 130.5 (1C, Ar-C-CHOH), 136.8 (1C,
Ar-C-CH2-O), 158.9 (1C, Ar-C-OBn).
1-(4-Benzyloxyphenyl)ethanone (4.30, 19.0 mmole)
and p-TsOH∗H2O (5.42 g, 28.5 mmole) were sus-
pended in acetonitrile (700 mL) and N-chlorosuccin-
imide (5.4 g, 40.0 mmole) was added. The mixture
was refluxed for 20 h.The solvent was removed under
reduced pressure and the residue was taken up in ethyl
acetate (100 mL). The organic fraction was washed
with water (4 ꢁ 25 mL) and brine (25 mL) and dried
over MgSO4. After removal of solvent the crude prod-
uct was re-crystallized from ethyl acetate/hexane giv-
ing 1.47 g (26%) of 2 as a white solid. HRMS (EI):
1
294.0210, (calcd. 294.0214). H NMR (CDCl3) δ:
1-(4-(Benzyloxy)phenyl)ethanol (2d). Yield 1.04 g
(96%), purity (HPLC): 100%, mp 84.8–86.5°C. 1H
NMR (CDCl3): δ 1.51 (d, 3H, CH3, J ꢂ 6.4 Hz),
1.78 (1H, br. s, OH), 4.88 (q, 1H, CH, J ꢂ 6.4 Hz),
5.09 (s, 2H, Ph-CH2-O), 7.0 (m, 2H, Ar-H), 7.28–
7.48 (m, 7H, Ar-H). 13C NMR (CDCl3): δ 22.62,
25.0, 69.9, 70.0, 114.8 (2C), 126.7 (2C), 127.4
(2C), 127.9, 128.6 (2C), 137.0, 138.2, 158.2.
5.09 (s, 2H), 6.55 (s, 1H), 6.98 (m, 2H), 7.28–7.37
(m, 5H) and 8.00 (m, 2H). 13C NMR (CDCl3) δ:
68.0, 70.5, 115.2 (2C), 124.3, 127.7 (2C), 128.6,
128.9 (2C), 132.5 (2C), 135.9, 163.9 and 184.7.
Synthesis of alcohols 2a-e by reduction of ketones 1a-e
The ketones 1a-e (4–8 mmole) were dissolved in
methanol (60 mL). Sodium borohydride (10–12
mmole) was added to the slurry, which was stirred
at RT for 3 h. The reaction mixture was transferred
1-(4-(Benzyloxy)phenyl)-2,2-dichloroethanol (2e).Yield
0.32 g (65%), purity (HPLC) 99%, mp 77.5–77.7°C.

Enantiopure 1-phenyl-2-haloethanols and -esters 275
MS (EI): M•ꢃ 296.7 (calcd. 297.1765), m/z 91 (26%,
C7H7), m/z 213 (9%, M– CHCl2). 1H NMR (CDCl3):
δ 2.82 (d, 1H, OH, J ꢂ 3.8 Hz), 4.92 (dd, 1H, CH, J
ꢂ 3.8, 5.5 Hz), 5.07 (s, 2H, Ph-CH2), 5.77 (d, 1H,
CH, J ꢂ 5.6 Hz), 6.99 (m, 2H, Ar-H), 7.30–7.45 (m,
7H, Ar-H). 13C NMR (CDCl3): d 70.0, 76.6, 78.5,
114.7 (2C), 127.5 (2C), 128.1, 128.4 (2C), 128.6
(2C), 129.7, 136.7, 159.2.
CH2, J ꢂ 4.7 Hz, 10.8 Hz), 3.64 (dd, 1H, CH2, J ꢂ
8.4 Hz, 10.8 Hz), 5.06 (s, 2H, Ph-CH2), 5.94 (dd,
1H, CH, J ꢂ 4.7 Hz, 8.3 Hz), 7.0 (m, 2H, Ar-H),
7.26–7.42 (m, 2ꢃ5H, Ar-H). 13C-NMR (CDCl3) δ
13.6, 18.4, 34.4, 36.2, 70.0, 74.3, 114.9, 127.4, 127.9,
128.0, 128.6, (2C), 130.2, 136.7, 159.1, 172.5.
(R)-1-(4-(Benzyloxy)phenyl)-2-chloroethanol, (R)-
2b. Yield:0.22 g, (85%), purity (HPLC): 98%, ee: 94%,
[α]D23 ꢂ –22.7 (c ꢂ 1.06, MeOH), mp: 75.0–75.8°C.
Kinetic resolutions
Transesterification reactions. These were performed in
a Minitron Incubator Shaker at 30°C agitating at
200 rpm. Racemic alcohol 2a-e (1–2 mmole) and
vinyl acetate or vinyl butanoate (3–10 mmole) were
mixed in CH2Cl2:hexane, (1:3, 40–200 mL) and the
reactions started by addition of immobilized lipase
(0.5–1.0 g) and incubated for 4–103 h before the
enzyme was filtered off and the solvent removed.
Two replicates of each reaction were performed.The
enantiomers were separated by VersaFlash column
chromatography. In small scale reactions 0.1–0.2
mmol of substrate, 0.3–1.0 mmole of acyl donor and
20–40 mg of enzyme in 3 mL solvent were used.Two
or more replicates of each reaction were performed.
Chiral GLC and HPLC analyses gave ees- and eep-
values from which the degree of conversion was cal-
culated according to c ꢂ ees/(ees ꢃ eep). In control
experiments under the same reaction conditions but
without enzyme, no acylation was observed.
(S)-1-(4-(Benzyloxy)phenyl)-2-chloroethyl butanoate,
(S)-3b. Yield: 0.19 g, (58%), purity: 100%, ee
23
(GLC): 99%, [α]D ꢂ ꢃ 66.9 (c ꢂ 0.8, MeOH),
mp: 54.4–54.9°C. MS (ESI): MNaꢃ 355:357, 3.7:1.
1
GC-MS: m/z 296 (M– HCl), m/z 91 (C7H7). H
NMR (CDCl3): δ 0.94 (t, 3H, CH3, J ꢂ 7.4 Hz),
1.61–1.73 (sextet, 2H, CH2, J ꢂ 7.4 Hz), 2.27–2.42
(m, 2H, CH2), 3.68 (dd, 1H, CH2, J ꢂ 4.6 Hz, 11.6
Hz), 3.76 (dd, 1H, CH2, J ꢂ 8.2 Hz, 11.6 Hz), 5.05
(s, 2H, Ph-CH2), 5.92 (dd, 1H, CH, J ꢂ 4.5 Hz, 8.2
Hz), 6.95 (m, 2H, Ar-CH), 7.24–7.44 (m, 7H,
Ar-CH). 13C-NMR (CDCl3): δ 13.6 (1C, CH3),
18.4 (1C, CH2), 36.2 (1C, CH2), 46.5 (1C, CH2),
70.0 (1C, Ph-CH2), 74.5 (1C, CH), 115.0 (2C,
Ar-CH), 127.4 (2C, Ar-CH), 128.0 (1C, Ar-CH),
128.1 (2C, Ar-CH), 128.6 (2C, Ar-CH), 129.7 (1C,
Ar-CH), 136.7 (1C, Ar-CH), 159.1 (1C, Ar-CH),
172.5 (1C, CꢂO).
Hydrolysis of butanoate 3b. 3b (0.05 g, 0.17 mmole)
was performed in acetone (1 mL).The solution was
added to potassium phosphate buffer (pH 7.00, 0.05
M, 35 mL) and the reaction was started by addition
of Novozym 435 (20 mg). The reaction was stirred
at RT. Samples (0.5 mL) were withdrawn, extracted
with Et2O, dried over anhyd. MgSO4 and analysed.
Chiral GLC and HPLC analysis as for transesterifi-
cation reactions.
(R)-1-(4-(Benzyloxy)phenyl)-2-fluoroethanol, (R)-2c.
Yield: 0.20 g, (83%), purity (GLC): 99%, ee: 98%,
[α]D ꢂ –30.1 (c ꢂ 1, MeOH), [α]D ꢂ –34.2
(c ꢂ 0.6, CHCl3), mp 74.8–75.0°C.
23
23
(S)-1-(4-(Benzyloxy)phenyl)-2-fluoroethyl butanoate,
(S)-3c. Yield: 0.29 g, (93%), purity (HPLC): 100%,
23
ee: 99% , [α]D ꢂ ꢃ81.3 (c ꢂ 1.3, MeOH), mp:
47.0–47.5°C. MS (ESI): MNaꢃ 339.3. GC-MS: No
peak at m/z 316, m/z 296 (M– HF) m/z 91 (C7H7).
1H-NMR (CDCl3): δ 0.87 (t, 3H, CH3), 1.55–1.65
(sextet, 2H), 2.23–2.36 (m, 2H), 4.35–4.60 (m, 2H,
1H, JAX ꢂ 7.6 Hz, JBX ꢂ 3.5 Hz, JFAB ꢂ 47.9 Hz),
4.98 (s, 2H), 5.84–5.97 (ddd, 1H, JAX ꢂ 7.5 Hz, JBX
ꢂ 3.5 Hz, JFX ꢂ 15.6 Hz), 6.90 (m, 2H), 7.16–7.37
(m, 7H). 13C NMR (CDCl3): δ 13.5, 18.4, 36.2, 70.1,
73.4, 84.1, 115.0 (2C), 127.4 (2C), 127.9, 128.0,
128.3 (2C), 128.6 (2C), 136.7, 159.1, 172.7.
Characterization data of alcohol enantiomers
were similar to racemic alcohols.
(R)-1-(4-(benzyloxy)phenyl)-2-bromoethanol, (R)-2a.
Yield: 0.10 g (90%), purity (HPLC) 100%, ee:
23
(Chiral GLC) 99%, [α]D ꢂ –26.6 (c ꢂ 1.00,
CHCl3), mp 70.9–72.1°C.
(S)-1-(4-(Benzyloxy)phenyl)-2-bromoethyl butanoate,
(S)-3a. Yield: 0.13 g, (74%), purity (HPLC): 95%,
23
ee: 99%, [α]D ꢂ ꢃ 62.3 (c 1.88, MeOH), mp
1
75.6–76.1°C. MS (ESI): MNaꢃ 399 and 401. H-
(S)-1-(4-(Benzyloxy)phenyl)etanol, (S)-2d. Yellow
oil with white crystals.Yield: 0.09 g, (73%), purity:
NMR (CDCl3) δ 0.95 (t, 3H, CH3, J ꢂ 7.4 Hz),
1.62–1.72 (sextet, 2H -CH2, J ꢂ 7.4 Hz), 2.28–2.42
(m, 2H, CH2, J ꢂ 7.5 Hz, 14.6 Hz), 3.56 (dd, 1H,
23
96%, ee: 92%, [α]D ꢂ –25.7 (c ꢂ 0.74, MeOH),
[α]D23 ꢂ –33.1 (c ꢂ 1.69, CHCl3), mp 66.1–66.5°C.

276
S. M. Lystvet et al.
(R)-1-(4-(Benzyloxy)phenyl)ethyl butanoate, (R)-3d.
Oil. Yield: 0.13 g, (78%), purity: 90%, ee: 86%,
vinyl acetate as acyl donor were slow; showing less
than 20% conversion after 7 days. These reactions
were not scaled up and the acetates 4a-d were only
detected by chromatography, not isolated.
23
[α]D ꢂ ꢃ87.3 (c ꢂ 1.1, MeOH). MS (EI): M•ꢃ
298, m/z 91 (C7H7). GC-MS: m/z 298, m/z 211 (M–
1
C4H7O2), m/z 91 (C7H7). H-NMR (CDCl3, 400
In the reactions of 2a-c with Novozym 435 and
vinyl butanoate, E-values above 200 were obtained.
An E-value of 27 was obtained in the CALB cata-
lyzed reaction of 2d. No product was formed when
2e was the substrate (XꢂCl,Cl). The reaction times
to 50% conversion decreased from ∼ 2 weeks to 2
days when 40 mg enzyme was used instead of 20 mg.
Surprisingly, the reaction time to 50% conversion in
the reaction of 2d with 40 mg Novozyme 435 was
only 0.5 h, although the E-value was only 27.When
enzyme selectivity changes in kinetic resolutions of
a compound as the reaction conditions change or
changes in substrate structure, this may be due to
increased or decreased reaction rate of one of the
enantiomers. It can also be caused by a change of
reaction rate of both of the enantiomers. Figure 1
shows progress curves of the conversion of each
enantiomer of 2b (XꢂCl) and 2d (XꢂH).The reac-
tion rate of each of the enantiomers can be deter-
mined by calculating the concentration of the two
product ester enantiomers at different degrees of
conversion from equations (1) and (2) below.
MHz): δ 0.95 (t, 3H, CH3, J ꢂ 7.4 Hz), 1.54 (d, 3H,
CH3, J ꢂ 6.6 Hz), 1.62–1.72 (sextet, 2H, CH2, J ꢂ
7.4 Hz), 2.29–2.34 (dt, 2H, CH2), 5.09 (s, 2H,
Ph-CH2-O), 5.89 (q, 1H, CH, J ꢂ 6.6 Hz), 6.98 (m,
2H, Ar-H), 7.26–7.48 (m, 7H, Ar-H). 13C NMR
(CDCl3): δ 13.6, 18.4, 22.0, 36.5, 70.0, 71.6, 114.7
(2C), 127.4 (2C), 127.5 (2C), 127.9, 128.6 (2C),
134.2, 136.9, 158.4, 172.9.
Results and discussion
Synthesis of the ketones 1a-e was performed by known
methods achieving moderate yields.The ketones were
reduced with sodium borohydride in methanol in
almost quantitative yield to give the alcohols 2a-e,
which were kinetically resolved in transesterification
reactions in dichloromethane:hexane (1:3) catalyzed
by Novozym 435 (CALB) and Novozym 735 (CALA)
with both vinyl acetate and vinyl butanoate as acyl
donors. The alcohol 2a was also esterified with vinyl
butanoate using CALA fromViaZym BV as a catalyst
(Scheme 1 and Table I).
In order to improve the enantioselectivity of
CALB towards 2d, the butanoate of 2d (3d) was
hydrolyzed in phosphate buffer. Previously, we have
reported that a change of reaction type from trans-
esterification reaction of alcohol to hydrolysis of the
corresponding ester improves the enantioselectivity
remarkably (Jacobsen et al. 2000). All reactions with
cn.F ꢂ c x (1 ꢃeep)
cn.S ꢂ c x (1 ꢀ eep)
(1)
(2)
where cν.F and cν.S are the conversions of the faster
and the slower reacting enantiomer (% converted of
total concentration of substrate enantiomer, i.e. the
relative amount), respectively, c is conversion of the
racemic substrate and eep is enantiomeric excess of
R1 R2
X
PhCH2O
1 R1,R2 = O, 2 R1,R2 = H,OH, 3 R1,R2 = H, OBu, 4 R1,R2 = H, OAc
a) X = Br, b) X = Cl, c) X = F,d) X = H, e) X = Cl,Cl
Lipase
Acyl donor
O
OBu
OH
OH
NaBH4
MeOH
X
X
X
X
+
CH2Cl2:hexane
PhCH2O
PhCH2O
PhCH2O
PhCH2O
(S)-3a X = Br,(S)-3b X = Cl
(S)-3c X = F,(R)-3d X = H
(S)-3e X = Cl,Cl
(R)-2a X = Br,(R)-2b X = Cl
(R)-2c X = F, (S)-2d X = H
(R)-2e X = Cl,Cl
2a X = Br,2b X = Cl
2c X = F,2d X = H
2e X = Cl,Cl
1a X = Br,1b X = Cl
1c X = F,1d X = H
1e X = Cl,Cl
Scheme 1. Reduction of ketones 1a-e with subsequent lipase catalysed resolution of alcohols 2a-e.

Enantiopure 1-phenyl-2-haloethanols and -esters 277
Table I. E-values of transesterification reactions of the alcohols 2a-e with vinyl butanoate as acyl donor to produce the butanoates 3a-d
in CH2Cl2:hexane (1:3) catalyzed by lipases A and B from Candida antarctica and hydrolysis of the butanoate of 2d (3d) in buffer catalyzed
by Novozym 435.The gem-dichloroalcohol 2e did not react with either of the enzymes. Rx. time in h to reach 50% conversion (c). Optical
rotations [α]D were determined at 23°C in MeOH; for concentrations, see Experimental section.
Substrate
Catalyst
E-value
Alcohol, % ee
(R)-2a, 99
Ester, % ee
(S)-3a, 99
Rx time (h)
[α]D
[α]D
2a
2a
2a
2b
2b
2c
2c
2d
Novozym 435
Novozym 735
ViaZym VZ 1031-3
Novozym 435
Novozym 735
Novozym 435
Novozym 735
Novozym 435
75
103
24
71
52
4
ꢄ 200
23
13
ꢄ 200
38
ꢀ26.6, (CHCl3)
ꢃ62.3
(R)-2b, 94
(R)-2c, 98
(S)-2d, 92
(S)-3b, 99
(S)-3c, 99
(R)-3d, 86
ꢀ22.7, (MeOH)
ꢀ30.1, (MeOH)
ꢃ66.9
ꢃ91.3
ꢃ87.3
ꢄ 200
22
27
40
27
ꢀ25.7, (MeOH)
ꢀ33.1, (CHCl3)
2d
3d
Novozym 735
Novozym 435
7
48
35
5
the product. The results are plotted in Kaleidagraph
3.0 and the progress curves of different reactions can
be compared. Figure 1 shows that the decrease in
E-value from the CALB catalyzed resolution of 2b
with vinyl butanoate (E ꢅ 200) compared to the
equivalent resolution of 2d (E ꢂ 27) was due to the
higher esterification rate of the slower reacting
enantiomer (the S-enantiomer) of 2d.
In the hydrolysis of 3d catalyzed by Novozym
435 an increase of E-value from 27 (in the transes-
terification reaction) to 48 was observed. In irrevers-
ible kinetic resolutions, as in the reactions with vinyl
acyl donors, high enantiomeric excesses of the
remaining substrate are always obtained, but with a
sacrifice of yield.
with CALA and CALB. The gem-dichloroalcohol 2e
did not react with either of the enzymes. Both size and
unfavourable electronic effects of the small group are
thought to be the reason why 2e is not a suitable
substrate for CALB. It was anticipated that CALA
would show low E-values for all of the substrates, with
one small and one large group bound to the stereo-
centre. However, we anticipated that the gem-dichloro
group in 2e would fit better into CALA, which would
discriminate between the two enantiomers due to the
larger size of the ‘small’ group. Unfavourable elec-
tronic effects may be the reason that this molecule
was not suitable for CALA. However, when a meth-
oxy group was introduced in the 3-position on the
phenyl ring in compounds otherwise similar to 2a
(XꢂBr) and 2b (XꢂCl), the E-value exceeded 200
and 80, respectively, in the transesterification with
CALA (Novozym 735) in dichlomethane:hexane
(1:1) (Fuglseth et al. 2006). The size difference is in
the largest group of the molecule compared to 2a and
2b and shows that proper substrates for CALA are
not restricted to those with two large groups con-
nected to the stereocentre. The CALA preparation
from ViaZym BV gave 40% lower E-values in the
reaction with 2a compared to the enzyme preparation
from Novozymes; however, the reaction was much
faster.The difference in reaction rate may be explained
by difference of activity, however, the difference of
enantioselectivity is not simple to rationalize. Differ-
ent immobilization material and methods between the
twopreparationscouldbeoneofthereasons(Hanefeldt
et al. 2009). Further studies of the differences in
CALA enzyme preparations are under way.
CALA gave E-values of 20–40 in all of the reac-
tions.When XꢂH, as in 2d, the E-value was low both
100
(S)-3b
80
(R)-3d
60
40
(S)-3d
(R)-3b
20
0
40
60
80
100
0
20
The alcohols (R)-2a, (R)-2c and the butanoates
(S)-3a-c were isolated in moderate-to-good yields,
with ee's of 99%. These results show a remarkable
increase of ee in (R)-2a and (R)-2c, which previously
have been synthesized by asymmetrization in 94% ee
(Gamble et al. 1998; Goswami et al. 2000; 2001).
Time (h)
Figure 1. Conversion of each enantiomer in esterification of 2b
and 2d with Novozym 435 and vinyl butanoate plotted against
time. Filled and open circles: Product ester (S)-3b and (R)-3b,
respectively. Filled and open squares: Product ester (R)-3d and
(S)-3d, respectively.

278
S. M. Lystvet et al.
de Maria PD, Carboni-Oerlemans C, Tuin B, Bargeman G,
van der Meer A, van Gemert R. 2005. Biotechnological applica-
tions of Candida antarctica lipase A: state-of-the-art. J Mol Catal
B Enzymatic 37:36–46.
Ericsson DJ, Kasrayan A, Johansson P, BergforsT, Sandström AG,
Bx00E4;ckvall J-E, Mowbray SL. 2008. X-ray structure
of Candida antarctica lipase A shows a novel lid structure and
a likely mode of interfacial activation. J Mol Biol 376:
109–119.
Fuglseth E, Anthonsen T, Hoff B. 2006. New chiral building
blocks from acetovanillone using lipases A and B from Candida
antarctica. Tetrahedron: Asymmetry 17:1290–1295.
Fuglseth E, Sundby E, Bruheim P, Hoff BH. 2008a. Asymmetric
reduction using (R)-MeCBS and determination of absolute
configuration of para-substituted 2-fluoroarylethanols.Tetrahedron:
Asymmetry 19:1941–1946.
(R)-2b, (R)-2d and (S)-3d were obtained with ee's
of 86–94%. All enantiomers were isolated from the
CALB catalyzed reactions.The butanoates (S)-3a-c
will give access to the corresponding S-alcohols (S)-
2a-c in 99% ee when the ester groups are hydrolyzed,
either by a lipase catalyzed reaction or base cata-
lyzed. The proof of absolute configurations of the
enantiomers was based on the known stereoprefer-
ence of CALB (Hoff & Anthonsen 1999) and by
comparison of GLC elution orders with similar
enantiopure compounds synthesized from (S)-
epichlorohydrin (Jacobsen et al. 2000).
Fuglseth E, Krane Tvedt TH, Hoff B. 2008b. Electrophilic and
nucleophilic side chain fluorination of para-substituted acetop-
henones. Tetrahedron 64:7318–7323.
Gamble MP, Smith ARC,Wills M. 1998. A novel phosphinamide
catalyst for the asymmetric reduction of ketones by borane.
J Org Chem 63:6068–6071.
Goswami A, Goswami J. 2005. Corrigendum to DMSO-triggered
enhancement of enantioselectivity in Novozyme[435]-catalyzed
transesterification of chiral 1-phenylethanols. Tetrahedron Lett
46:4411–4413.
Conclusion
Efficient processes for synthesis of several enantio-
pure halogenated secondary alcohols and esters have
been developed using CALB as a catalyst and vinyl
butanoate as an acyl donor. All chiral compounds
have been extensively characterized.
Goswami A, Bezbaruah RL, Goswami J, Borthakur N, Dey D.
Hazarika AK. 2000. A convenient stereoselective synthesis of
(R)-(-)-denopamine and (R)-(-)-salmeterol. Tetrahedron:
Asymmetry 12:3343–3348.
Hagmann WK. 2008. The many roles for fluorine in medicinal
chemistry. J Med Chem 51:4359–4369.
Acknowledgement
We thank Novozymes AS, Denmark, for gifts of
Novozym 435 and 735, and ViaZym BV, The
Netherlands, for gift of ViaKit CAL.
Hanefeldt U, Gardossi L, Magner E. 2009. Understanding enzyme
immobilisation. Chem Soc Rev 38:453–468.
Hoff BH, Anthonsen T. 1999. Gas chromatographic enantiomer
separation of C-3 and C-4 synthons: prediction of absolute
configuration from elution order and enzymatic resolution.
Chirality 11:760–767.
Declaration of interest: The authors report no
conflicts of interest. The authors alone are respon-
sible for the content and writing of the paper.
Jacobsen EE, Hoff BH, Anthonsen T. 2000. Enantiopure deriva-
tives of 1,2-alkanediols: substrate requirements for lipase B from
Candida antarctica. Chirality 12:654–659.
References
Anthonsen HW, Hoff BH, Anthonsen T. 1996. Calculation of
enantiomer ratio and equilibrium constants in biocatalytic ping-
pong bi-bi resolutions. Tetrahedron: Asymmetry 7:2633–2638.
Anthonsen T, Hoff BH. 1998. Resolution of derivatives of 1,2-
propanediol with lipase B from Candida antarctica. Effect of
substrate structure, medium, water activity and acyl donor on
enantiomeric ratio. Chem Phys Lipids 93:199–207.
Azumai T, Kurimoto I,Toda S, Minamii M. 1989. Preparation of
optically active phenol derivatives as intermediates for liquid
crystals. Japanese patent JP 03047143.
Azumai T, Minamii M. 1988. Optically active 4-benzyloxyaryl
ethers as intermediates for liquid crystals. Japanese patent JP
01221341.
de Gonzalo G, Lavandera I, Brieva R, Gotor V. 2004. Enzymatic
acylation reactions onω-hydroxy cyanohydrins. Tetrahedron
60:10525–10532.
Kirk O, Christensen MW. 2002. Lipases from Candida antarctica:
unique biocatalysts from a unique origin. Org Proc Res Dev 6:
446–451.
Kourist R, Gonzalez-Sabin J, Liz R, Rebolledo R. 2005. Kinetic
resolution of 1-biaryl- and 1-(pyridylphenyl)alkan-1-ols cata-
lyzed by the Lipase B from Candida antarctica. Adv Synth Catal
347:695–702.
Renaudet O, Reymond JL. 2004. Synthesis of ether oligomers.
Org Lett 6:397–400.
Schmidt M, Barbayianni E, Fotakopoulou I, Hoehne M,
Constantinou-Kokotou V, Bornscheuer UT, Kokotos G. 2005.
Enzymatic removal of carboxyl protecting groups. 1. Cleavage
of the tert-butyl moiety. J Org Chem 70:3737–3740.
Tjosås F, Anthonsen T, Jacobsen EE. 2008. Biocatalalytic resolu-
tion of saphenic acid. Substrate preferences for lipases A and B
from Candida antarctica. ARKIVOK vi:81–90.