Chemoenzymatic synthesis of α-halogeno-3-octanol and 4- or 5-nonanols. Application to the preparation of chiral epoxides

Article abstract of DOI:10.1016/S0957-4166(98)00470-4

A study of the microbiological reduction of different α- halogenoketones (4-chloro-3-octanone, 4-chloro-5-nonanone, 5-bromo-4-nonanone and 5-chloro-4-nonanone) with several strains of microorganism showed great difficulty in reducing ketone functions located in the middle of carbon chains. However, by choosing the appropriate microorganism, several enantiomerically pure diastereoisomers of the corresponding halohydrins have been obtained and were transformed into chiral epoxides.

Full text of DOI:10.1016/S0957-4166(98)00470-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 4441–4457
Chemoenzymatic synthesis of α-halogeno-3-octanol and 4- or
5-nonanols. Application to the preparation of chiral epoxides
Pascale Besse,∗ Tania Sokoltchik † and Henri Veschambre
Laboratoire de Synthèse, Electrosynthèse et Etude de Systèmes à Intérêt Biologique, UMR 6504 du CNRS, Université Blaise
Pascal, 63177 Aubière Cedex, France
Received 30 October 1998; accepted 17 November 1998
Abstract
A study of the microbiological reduction of different α-halogenoketones (4-chloro-3-octanone, 4-chloro-5-
nonanone, 5-bromo-4-nonanone and 5-chloro-4-nonanone) with several strains of microorganism showed great
difficulty in reducing ketone functions located in the middle of carbon chains. However, by choosing the appro-
priate microorganism, several enantiomerically pure diastereoisomers of the corresponding halohydrins have been
obtained and were transformed into chiral epoxides. © 1998 Elsevier Science Ltd. All rights reserved.
1. Introduction
Enantiomerically pure epoxides are important chiral building blocks in organic synthesis and can be
used as key intermediates in the synthesis of more complex enantiopure bioactive compounds due to
their ability to react with a broad variety of nucleophiles. In recent years, many chemical and biological
methods have been developed.1,2 Among the most used techniques, the Sharpless epoxidation of olefins3
is limited to allylic alcohols, and Jacobsen’s catalysts4 give mainly high enantiomeric excesses with some
cis-substituted olefins. Recently, Jacobsen5 reported a kinetic resolution of racemic primary epoxides via
catalytic hydrolysis using chiral cobalt based complexes.
Among the biocatalytic methods,6 two main direct pathways have been studied: the stereoselective
microbiological epoxidation of prochiral olefins and the enzymatic resolution of racemic epoxides
using hydrolytic enzymes. In the first method, the reaction is mostly catalyzed by monooxygenases
(cytochromes P450,7,8 ω-hydroxylases,9 methane monooxygenases,10,11 etc.). The preparative scale is
difficult to perform with isolated enzymes due to their complex nature and their dependance of a redox
∗
†
Corresponding author. Fax: 33 4 73 40 77 17; e-mail: besse@chimtp.univ-bpclermont.fr
Present address: Belarus State University of Technology, Department of Biotechnology and Organic Synthesis, Sverdlova
13A, 220690 Minsk, Republic of Belarus.
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00470-4

4442
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
cofactor, but whole-cell microorganisms can be used instead. Chloroperoxidase from Caldaryomyces
fumago is also able to transform alkenes into epoxides.12,13 However, as in the case of monooxygenases,
only the cis olefins yield epoxides with high enantiomeric excesses.
Hydrolytic enzymes may be more promising as enantioselective biocatalysts, because they are cofactor
independent.14 Although their mammalian origin has been known for a long time (mEH and cEH),
they have mainly been investigated during detoxification studies. More recently, several groups have
started a search for epoxide hydrolases from microbial sources: Furstoss and his co-workers were the
first to use the fungi Aspergillus niger15 and Beauveria bassiana16 to hydrolyze racemic epoxides;
then Faber’s group used bacteria, in particular Rhodococcus.17 More recently, Weijers found the same
activity in the yeast Rhodotorula glutinis.18 The drawback of this method is the theoretical limited yield
in epoxide (and in diol) of 50%. Although some improvements have been made to obtain a 100% yield in
enantiomerically pure diol,19 methods for obtaining a 100% yield in epoxide have not yet been realized.
Indirect strategies for the obtention of enantiomerically pure epoxides can also been planned, starting
from enantiomerically pure α-halohydrins for example. In 1993,20 we described a three-step chemoenzy-
matic synthesis of chiral 2,3-epoxides from α-halogenoketones. The microbiological reduction of these
ketones yielded, by choosing the appropriate microorganism, halohydrins with excellent enantiomeric
excesses. They were then chemically converted into epoxides (Scheme 1). Several α-halogenoketones
were tested but the R group was always a methyl or a phenyl group.
Scheme 1.
In this work, we wanted to know if our method could be generalized whatever the size of the R group.
We have chosen four aliphatic α-halogenoketones, each with a long carbon chain and with the keto
function at different positions on the chain: 4-chloro-3-octanone 1, 5-chloro-4-nonanone 2, 4-bromo-5-
nonanone 3 and 4-chloro-5-nonanone 4 (Fig. 1). We have studied their reduction with several strains of
microorganism in order to obtain enantiomerically pure halohydrins and then chiral 3,4- or 4,5-epoxides.
Figure 1.
2. Results and discussion
2.1. Synthesis of the substrates
Two different synthetic routes were used according to the symmetrical or unsymmetrical character
of the α-halogenoketone we wanted to test. The synthesis of 5-chloro-4-nonanone 2 has already beeen

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4443
described by Barluenga et al.21 The reaction of a carboxylic acid ester with bromochloromethyllithium,
and further treatment with lithium dibutylcuprate led, after hydrolysis, to compound 2 with 87% yield.
However, in spite of several attempts under different conditions, we did not succeed in obtaining 2 in
greater than 20% yield.
4-Chloro-3-octanone 1 and 5-chloro-4-nonanone 2, the unsymmetrical α-halogenoketones of our
study, were prepared in two steps via the formation of an intermediate diazoketone. The first step
consisted in the reaction of diazomethane with the appropriate acid chloride to yield a diazoketone.22
The reaction of this diazoketone with trialkylborane first and then with N-chlorosuccinimide (NCS)23
provided the corresponding α-haloketone (Scheme 2). The overall yields were around 30%.
Scheme 2.
The other haloketones, 4-bromo-5-nonanone 3 and 4-chloro-5-nonanone 4, were both obtained from
5-nonanone after reaction with bromine24 and sulfuryl chloride,25 respectively (Scheme 3).
Scheme 3.
2.2. Microbiological reductions of the substrates synthesized
The strains of microorganism selected were as follows: baker’s yeast and the yeast Rhodotorula
glutinis, the fungi Aspergillus niger, Beauveria bassiana, Cunninghamella echinulata var. elegans,
Geotrichum candidum, Mortierella isabellina and the bacterium Lactobacillus kefir. All these microorga-
nisms have already shown their abilities to reduce several 2-halogeno-3-ketones to yield enantiomerically
pure α-halohydrins.20,26 The yeast was used freeze-dried under non-fermenting conditions. Bioconver-
sions with the other microorganisms were carried out using washed resting cells. Each microorganism
was assayed for a 6, 24 or 48 h incubation time and with each substrate. The results are collected in
Table 1.
Most of the microorganisms tested were able to reduce the α-haloketones 1, 2, 3 and 4. However,
the GC analyses of some samples showed the presence of many by-products, some of them in large
quantities, particularly in the case of the reduction of 2, 3 and 4. Two by-products were identified as 4- or
5-nonanone and as the corresponding alcohols, but the others, which sometimes represented more than
50% of the mixture, have not been isolated. All the microorganisms which did not lead to at least 50% of
halohydrins, were not used further.
The next step consisted of quantitative assays in order to determine the absolute configuration and the
enantiomeric excess of each halohydrin formed. The results are given in Table 2. The chemical yields

4444
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
Table 1
Microbiological reductions of 1, 2, 3 and 4
recorded in the last column of Table 2 are the overall yields of diastereoisomers after work-up. The
proportions of each isomer are given in brackets.
The first comment concerned the low yields of halohydrins obtained whatever the substrate. They are
much lower than those obtained during the reduction of 3-bromo-2-octanone20 for example (50%). The
first explanation is the formation of many by-products, as described previously. We have also noticed
that the percentage in weight of the residue obtained after bioconversion and extraction was relatively
low (40%) compared to the initial weight of haloketone (loss of organic matter). The products of the
reaction were either totally degraded into H2O and CO2, or stocked in the cells. After bioconversion,
the cells were broken, suspended in ethanol and stirred for 24 h. From this ethanolic suspension, only

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4445
Table 2
Microbiological reductions of 1, 2, 3 and 4: quantitative assays
haloketone was isolated in a low yield (about 10% of the starting material). Therefore the haloketone or
the products formed are probably further metabolized. This hypothesis was confirmed by the presence
of several peaks on the GC chromatograms, a phenomenon that can explain the low yield of halohydrin
obtained.
The comparison of the results obtained for the four substrates shows that the further the position of
the ketone from the halide, the more difficult the microbiological reduction. The reduction of 4-chloro-
3-octanone 1 led to the four enantiomerically pure diastereoisomers: syn (3S,4S) and anti (3S,4R)-4-
chloro-3-octanol were obtained with baker’s yeast or R. glutinis. With A. niger, only the anti (3R,4S)
diastereoisomer was enantiomerically pure, whereas L. kefir gave only the syn (3R,4R) diastereoisomer.

4446
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
In the case of 5-chloro-4-nonanone 2, only three diastereoisomers were obtained enantiomerically pure:
the (4S,5S) and (4S,5R) with baker’s yeast, and the (4R,5S) with L. kefir. That is also the case for 4-chloro-
5-nonanone 4: the (4S,5S) and (4S,5R) diastereoisomers were obtained with L. kefir and the (4R,5S)
with baker’s yeast. In the end, only two diastereoisomers were obtained enantiomerically pure for 3:
the (4S,5S) and the (4S,5R) with R. glutinis. In contrast to the case of 1-phenyl-2-bromo-1-propanone
and 1-phenyl-2-chloro-1-propanone,26 where the change of the nature of the halide atom improved the
percentage of reduction, here changing the bromine atom into a chlorine has not greatly modified the
reduction of the haloketone. The same microorganisms led to the same isomers. Only the yield is better
in the case of 4.
Another surprising phenomenon in these results concerns the absolute configuration of the halohydrins
obtained with some microorganisms: R. glutinis, A. niger, L. kefir for 2, and L. kefir for 4. Indeed, in the
case of L. kefir with 4 for example, the syn diastereoisomer has an alcohol with an (S) configuration,
whereas the absolute configuration of the alcohol in the anti diastereoisomer is the opposite one (R), both
of them presenting excellent enantiomeric excesses. The reactions of reduction are not enantiogenic at
all. Moreover, the carbon atom bearing the halogen has, in both cases, the (S) configuration. The question
is where has the (R)-enantiomer of the haloketone gone. One hypothesis could be that, regarding the very
low yield in halohydrins (20% in our example), the isolated ones are not representative of all the reactions
that have taken place in the microorganism. Many by-reactions are observed (elimination of the halogen,
etc.) and we suppose that some isomers of the halohydrin can be preferentially metabolized or further
transformed.
2.3. Determination of the enantiomeric excesses of the halohydrins formed
Several methods were used to measure the enantiomeric excess of each isomer of the halohydrins
formed. For the products obtained by microbiological reduction of 1 and 4, enantiomeric excesses were
determined after converting each diastereoisomer of the racemic and of the optically active chlorohydrin
into the corresponding MTPA esters by Mosher’s method,27 either from 13C and 19F NMR spectra for 1,
or from 1H NMR spectra for 4 (see Experimental section).
For the halohydrins coming from the reduction of 2 and 3, the enantiomeric excesses were determined
by gas phase chromatography (see Experimental section). Two methods were used: (a) each isomer of
the halohydrin was analyzed directly using a chiral phase column coated with modified γ-cyclodextrins
(Lipodex E) (this method was used for the syn and anti diastereoisomers of 2 and for the syn isomer
of 3); (b) the anti diastereoisomer of 3 was treated with (S)-O-acetyllactic chloride28 and the derivative
obtained was analyzed on a non-chiral column.
2.4. Determination of the absolute configuration
None of the optically active halohydrins isolated from the reductions of 1, 2, 3 or 4 have been described
in the literature before. For each substrate, we assigned the absolute configuration of the halohydrins by
analogy with published data and by chemical correlations.
We have already studied the reduction of 3-bromo-2-octanone.20 With baker’s yeast an equivalent
mixture was obtained of (−)-(2S,3S) and (+)-(2S,3R)-3-bromo-2-octanol. Moreover, the diastereoisomer
which was first eluted from the chromatography column, and which exhibited the lower chemical shifts
1
for the CH–OH and the CH–Cl protons on the H NMR spectrum, corresponded to the syn (2S,3S)-3-
bromo-2-octanol. By comparison of the 1H NMR chemical shifts and of the sign of the specific rotation
of each isomer of 4-chloro-3-octanol, obtained by microbiological reduction with baker’s yeast, with

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4447
those of the isomers of 3-bromo-2-octanol, we tentatively assigned the (−)-(3S,4S) configuration to the
first eluted diastereoisomer and the (+)-(3S,4R) to the second one. To give evidence of the absolute
configuration proposed, a chemical correlation was assayed, corresponding to the dechlorination of each
diastereoisomer of the chlorohydrin (Scheme 4).
Scheme 4.
The treatment of each diastereoisomer of the chlorohydrin obtained with baker’s yeast with tributyltin
hydride in benzene, in the presence of 2,20-azobisisobutyronitrile (AIBN) as catalyst,29 afforded the 3-
octanol in high yield. The comparison of the sign of the specific rotation of the 3-octanol obtained by
dechlorination of the syn and anti 4-chloro-3-octanol, obtained with baker’s yeast, with that reported
in the literature30 allowed us to check the absolute configuration of the C-3 stereogenic centre. In both
cases, the dechlorination reaction yielded the (3S)-3-octanol and so the assignments of (−)-(3S,4S) for
the diastereoisomer first eluted and of (+)-(3S,4R) for the second one were confirmed.
The methods used to determine the absolute configuration of the chlorohydrins obtained after the
microbiological reduction of 5-chloro-4-nonanone 2 and 4-chloro-5-nonanone 4 by baker’s yeast and
Rhodotorula glutinis, respectively, are described in Scheme 5 for one of the diastereoisomers obtained in
each case.
For the chlorohydrin, first eluted after column chromatography from the products of the reduction of
5-chloro-4-nonanone 2 by baker’s yeast, the same dechlorination reaction as that described previously
was first carried out (Scheme 5A). The 4-nonanol obtained exhibits the (S) configuration (comparison
of the sign of its specific rotation with the data given in the literature31). So the asymmetric carbon
C4 of the chlorohydrin has an (S) configuration. In order to determine the absolute configuration of the
other asymmetric carbon (C5), a chemical correlation was made. The chlorohydrin was converted into
the corresponding epoxide by treatment with potassium carbonate.32 The cis or trans character can be
1
assigned to the epoxide formed by comparing its H NMR spectra with the data given in the relevant
literature33 for cis [δ=2.93 ppm (centred)] and trans [δ=2.68 ppm (centred)] racemic 4,5-epoxynonane.
In our case, we obtained a cis epoxide. Utaka et al.34 reported that a syn chlorohydrin gives a cis epoxide
while an anti chlorohydrin gives a trans epoxide. Moreover, the epoxide formation takes place with
inversion of configuration at the chlorine-bearing carbon atom. We can then deduce that this isomer
corresponds to the syn (4S,5S)-5-chloro-4-nonanol. The same reactions were carried out on the other
isomer of the 5-chloro-4-nonanol obtained with baker’s yeast. These reactions led to the formation of
(4S)-4-nonanol and trans 4,5-epoxynonane. This isomer corresponds to the anti (4S,5R)-5-chloro-4-
nonanol.
The absolute configurations of the diastereoisomers obtained from the reduction of 4 by Rhodotorula
glutinis were determined by the synthesis of the corresponding 4,5-epoxynonanes. The chlorohydrin
1
eluted first by column chromatography was converted into the epoxide. By comparing its H NMR
spectrum with data given in the literature,33 we can deduce the cis character of the epoxide formed.
The sign of its specific rotation is opposite (+5) to that of the cis-(4S,5R)-4,5-epoxynonane obtained
previously by cyclization of (4S,5S)-5-chloro-4-nonanol (Scheme 5A and B). Its absolute configuration
is then cis-(4R,5S)-4,5-epoxynonane. The epoxide formation takes place with inversion of configuration
at the chlorine-bearing carbon atom, so we can deduce that the chlorohydrin, eluted first by column
chromatography and obtained from the microbiological reduction of 4 with R. glutinis, corresponds to

4448
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
Scheme 5.
the syn (4S,5S)-4-chloro-5-nonanol. The same cyclization was carried out on the other isomer of 4-chloro-
5-nonanol obtained with R. glutinis, with formation of trans 4,5-epoxynonane. This epoxide presented
the same sign of specific rotation as the trans (4S,5S)-4,5-epoxynonane obtained from (4S,5R)-5-chloro-
4-nonanol. The second isomer is therefore the anti (4R,5S)-4-chloro-5-nonanol.
These assignments were checked by another chemical correlation. Each diastereoisomer of epoxide
was opened by LiAlH4 under reflux35 to yield a mixture of 4- and 5-nonanol. As the 5-nonanol is
inactive, the sign of the specific rotation of the mixture corresponds to that of 4-nonanol. By comparing
it with the data given in the literature,31 we assign the (4R) configuration to the alcohol coming from the
cis 4,5-epoxynonane (Scheme 5B) and the (4S) configuration to the alcohol coming from the trans 4,5-
epoxynonane. As the reaction of epoxidation takes place with inversion of configuration of the C4 carbon,
we confirm the assignment of the absolute configuration of the carbon C4 of both diastereoisomers of 4-
chloro-5-nonanol.
The absolute configuration of each diastereoisomer of the bromohydrin was assigned by analogy of the
1H NMR spectra and of the sign of specific rotation between 4-bromo-5-nonanol and 4-chloro-5-nonanol.

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4449
As the yields of the reduction of 3 were very low and as only two enantiomerically pure diastereoisomers
were obtained, further chemical correlations have not been investigated.
2.5. Synthesis of chiral epoxides
Our aim was to prepare the corresponding chiral epoxides. We were able to obtain the four isomers
of 4,5-epoxynonane enantiomerically pure, by treating with potassium carbonate the appropriate chloro-
hydrins arising from the reduction of 5-chloro-4-nonanone 2 or 4-chloro-5-nonanone 429 (Table 3).
Table 3
Synthesis of chiral 4,5-epoxynonane
The yields of epoxidation are quite good. The enantiomeric excesses of the epoxides were directly
deduced from those of the chlorohydrins. We have shown previously that no racemization was observed
during the cyclization.20,26 By the same method, it would be possible to prepare three isomers of 3,4-
epoxyoctane enantiomerically pure.
3. Conclusion
All these results show the possibility of reducing a ketone function even in the middle position
of a long carbon chain (contrary to what is reported by Faber36), and to obtain, by choosing the
appropriate microorganism, several diastereoisomers of the corresponding halohydrin enantiomerically
pure. However, the shift of the ketone position along the carbon chain implies a large decrease in the
yield of the reduction, mostly because of the formation of large quantities of by-products. This last
phenomenon leads to a limitation of the microorganisms that can be used to reduce this type of compound.
This work was carried out to validate the generality of our three-step chemoenzymatic synthesis of
chiral epoxides, shown before only on 2,3-epoxides: synthesis and microbiological reduction of an α-
halogenoketone, cyclization into epoxide from the diastereoisomers of chlorohydrins obtained previously
optically pure. This method is still applicable, but the synthesis of the 3,4- or 4,5-epoxides via this route
is less interesting than in the case of 3-halogeno-2-ketone, because of the low yield obtained.
4. Experimental section
4.1. General methods
Gas chromatography (GC) was performed using an instrument equipped with a flame ionization
detector and a 50 m×0.32 mm capillary column coated with Carbowax 20M for analytical analysis or

4450
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
a 25 m×0.25 mm capillary column coated with Lipodex E (modified γ-cyclodextrin) for determination
of enantiomeric excesses. The carrier gas was hydrogen at 65 kPa. Oven temperature: P1 (40°C for 5
min and then 40 to 200°C at 8°C/min) or P2 (80°C for 5 min and then 80 to 200°C at 5°C/min). For 1H
(400.13 MHz) and 13C (100.61 MHz) NMR spectra, the chemical shifts were relative to chloroform. For
19F (376.48 MHz), they were relative to CFCl3. Microanalyses were performed by the Service Central
d’Analyses du CNRS, Vernaison (France).
Microbiological methods: The microorganisms were all laboratory-grown except for freeze-dried
baker’s yeast which was a commercial product (VAHINE Monteux). Preculture and culture conditions for
the fungi Aspergillus niger ATCC 9142, Beauveria bassiana ATCC 9142, Cunninghamella elchinulata
var. elegans ATCC 9245, Geotrichum candidum CBS 233-76 and Mortierella isabellina NRRL 1757, for
the bacterium Lactobacillus kefir DSM 20587 and for the yeast Rhodotorula glutinis NRRL Y 1091 have
already been described elsewhere.26
4.2. Syntheses of the substrates
4.2.1. Synthesis of 4-chloro-3-octanone 1
4.2.1.1. Synthesis of diazoketone: CH3–CH2–CO–CH_N2. To a solution of diazomethane (2.9 g; 6.9
mmol; 3 equiv.) in diethyl ether, placed in an ice bath, was added dropwise 2.15 g (2.3 mmol; 1 equiv.)
of propionyl chloride. After the end of the addition, the solution was stirred at room temperature until
no nitrogen bubbling was observed. Diethyl ether was then removed on a steam bath. The residue was
purified by column chromatography on silica gel (eluent: pentane:ether 85:15). 1.45 g of pure diazoketone
was obtained (yield: 63%).
Retention time: 390 s (P1). 1H NMR (400.13 MHz) δ: 1.05 (t, 3H, J=7.5 Hz); 2.26 (q, 2H, J=7.5 Hz);
5.25 (s, 1H). 13C NMR (100.61 MHz) δ: 8.9 (CH3); 33.9 (CH2); 53.7 (CH); 195.8 (C_O).
4.2.1.2. Synthesis of 4-chloro-3-octanone 1. To a stirred solution of tributylborane (24 mL; solution 1
M in THF) was added dropwise a solution of 2.4 g (24.5 mmol) of diazoketone in 20 mL of dry THF
while maintaining the temperature at −40°C. After stirring for an additional hour at room temperature,
the solution was cooled to 0°C. N-Chlorosuccinimide (3.23 g; 25 mmol) was added in small portions to
maintain the temperature at 0°C. The solution was stirred at 0°C for an additional 15 min. 20 mL of cold
3 N NaOH solution was then added. The solution was stirred vigorously (15 min), then poured into cold
saturated brine solution (200 mL), and extracted with ether (three 50 mL portions). The residue remaining
after concentration (rotatory evaporator) of the dried (MgSO4) extract was purified by chromatography
(eluent: pentane:ether 98:2). 1.75 g of chloroketone was obtained (yield: 45%). Retention time: 450 s
1
(P2). H NMR (400.13 MHz) δ: 0.92 (t, 3H, J=7.1 Hz); 1.09 (t, 3H, J=7.2 Hz); 1.25–1.35 (m, 2H);
1.35–1.52 (m, 2H); 1.77–1.89 (m, 1H); 1.89–2.00 (m, 1H); 2.68 (q, 2H, J=7.2 Hz); 4.21 (dd, 1H, J=5.5
Hz, J=8.4 Hz). 13C NMR (100.61 MHz) δ: 7.8 (C-1); 13.8 (C-8); 22.1 (C-7); 28.2 (C-6); 31.9 (C-2); 33.6
(C-5); 63.6 (C-4); 206.2 (C-3). Anal. calcd for C8H15ClO: C, 59.07; H, 9.30; O, 9.84. Found: C, 59.06;
H, 9.17; O, 9.88.
4.2.2. Synthesis of 5-chloro-4-nonanone 2
4.2.2.1. Synthesis of diazoketone: CH3–CH2–CH2–CO–CH_N2. The same procedure as that
described previously for the synthesis of 1 was used. Butyryl chloride was added instead of propionyl
1
chloride. Yield: 66%. Retention time: 480 s (P1). H NMR (400.13 MHz) δ: 0.85 (t, 3H, J=6.9 Hz);
1.55 (sex, 2H, J=6.9 Hz); 2.15–2.35 (m, 2H); 5.25 (s, 1H). 13C NMR (100.61 MHz) δ: 13.4 (CH3); 18.5
(CH2); 42.7 (CH2); 54.0 (CH); 195.1 (C_O).

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4451
4.2.2.2. Synthesis of 5-chloro-4-nonanone 2. Same procedure as that described previously for the
1
synthesis of 1. Yield: 50%. Retention time: 480 s (P2). H NMR (400.13 MHz) δ: 0.91 (t, 3H, J=7.3
Hz); 0.93 (t, 3H, J=7.4 Hz); 1.25–1.40 (m, 3H); 1.40–1.52 (m, 1H); 1.63 (sex, 2H, J=7.3 Hz); 1.75–1.90
(m, 1H); 1.90–2.00 (m, 1H); 2.62 (t, 2H, J=7.2 Hz); 4.18 (dd, 1H, J=5.6 Hz, J=8.4 Hz). 13C NMR
(100.61 MHz) δ: 13.6, 13.9 (C-1, C-9); 17.1 (C-2); 22.1 (C-8); 28.2 (C-7); 33.5 (C-3); 40.4 (C-6); 63.8
(C-5); 205.6 (C-4). Anal. calcd for C9H17ClO: C, 61.18; H, 9.70; O, 9.06. Found: C, 61.15; H, 9.61; O,
8.99.
4.2.3. Synthesis of 4-bromo-5-nonanone 3
To a vigorously stirred solution of 4 g of 5-nonanone (28 mmol) in 40 mL glacial acetic acid was
added dropwise bromine (1.4 mL, 28 mmol) at room temperature. The mixture was then stirred for 1 h.
10 mL of water was then added to the mixture, which was extracted three times with ether. The organic
layer was dried over MgSO4. The residue was purified by column chromatography (eluent: pentane:ether
1
98:2). Yield: 50%. Retention time: 440 s (P2). H NMR (400.13 MHz) δ: 0.91 (t, 3H, J=7.3 Hz); 0.94
(t, 3H, J=7.4 Hz); 1.27–1.39 (m, 3H); 1.41–1.52 (m, 1H); 1.56 (qu, 2H, J=7.4 Hz); 1.84–2.01 (m, 2H);
2.58–2.75 (m, 2H); 4.22 (dd, 1H, J=6.4 Hz, J=8.1 Hz). 13C NMR (100.61 MHz) δ: 13.5; 13.9 (C-1,
C-9); 20.7 (C-2); 22.2 (C-8); 26.1 (C-6); 35.4 (C-7); 38.7 (C-3); 53.6 (C-4); 204.5 (C-5). Anal. calcd for
C9H17BrO: C, 48.88; H, 7.75; O, 7.23. Found: C, 48.81; H, 7.79; O, 7.26.
4.2.4. Synthesis of 4-chloro-5-nonanone 4
In a 50 mL round-bottomed flask was stirred a solution of 5-nonanone (2 g; 14 mmol) in 10 mL of
CCl4. To this solution was added dropwise 1.90 g (1.15 mL; 15 mmol) of sulfuryl chloride. After stirring
for an additional hour, the solution was washed with water and the organic layer was dried over MgSO4.
The residue, obtained after solvent evaporation, was purified on a silica gel column (eluent: pentane:ether
98:2). 1.5 g of pure chloroketone was obtained. Yield: 60%. Retention time: 490 s (P2). 1H NMR (400.13
MHz) δ: 0.87 (t, 3H, J=7.2 Hz); 0.90 (t, 3H, J=6.7 Hz); 1.28 (sex, 2H, J=7.5 Hz); 1.34–1.41 (m, 1H);
1.41–1.49 (m, 1H); 1.54 (qu, 2H, J=7.5 Hz); 1.70–1.80 (m, 1H); 1.80–1.92 (m, 1H); 2.60 (t, 2H, J=7
Hz); 4.16 (dd, 1H, J=5.4 Hz, J=8.4 Hz). 13C NMR (100.61 MHz) δ: 13.3; 13.7 (C-1, C-9); 19.3 (C-2);
22.1 (C-8); 25.7 (C-6); 35.6 (C-7); 38.2 (C-3); 63.4 (C-4); 205.3 (C-5). Anal. calcd for C9H17ClO: C,
61.18; H, 9.70; O, 9.06. Found: C, 61.09; H, 9.61; O, 9.08.
4.3. Microbiological reductions
Incubation times varied and are indicated for each microorganism. The products from the residues
were separated on a silica gel column, the eluent was pentane:ether 95:5. In each case, the yields are
overall yields for diastereoisomers after work-up.
4.3.1. Microbiological reductions of 4-chloro-3-octanone 1
Baker’s yeast: Incubation time: 24 h. The residue from 10 flasks consisted of 50% of (−)-(3S,4S)-4-
chloro-3-octanol and 50% of (+)-(3S,4R)-4-chloro-3-octanol. Yield: 13%.
(−)-(3S,4S)-4-Chloro-3-octanol: Retention time: 695 s (P2). 1H NMR (400.13 MHz) δ: 0.90 (t, 3H,
J=6.5 Hz); 0.95 (t, 3H, J=6.5 Hz); 1.15–1.30 (m, 1H); 1.30–1.45 (m, 2H); 1.45–1.55 (m, 1H); 1.55–1.70
(m, 2H); 1.80 (qu, 2H, J=6.5 Hz); 1.98 (s, 1H, exchangeable with D2O); 3.48–3.55 (m, 1H); 3.88 (td,
1H, J=3.5 Hz, J=7.1 Hz). 13C NMR (100.61 MHz) δ: 10.1 (C-1); 14.0 (C-8); 22.3 (C-7); 27.6 (C-6);
28.9 (C-2); 34.7 (C-5); 68.6 (C-4); 75.3 (C-3). Anal. calcd for C8H17ClO: C, 58.35; H, 10.40; O, 9.72.
Found: C, 58.36; H, 10.63; O, 9.76. [α]D25 −35 (c 0.03, CHCl3); ee≥98%.

4452
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
(+)-(3S,4R)-4-Chloro-3-octanol: Retention time: 765 s (P2). 1H NMR (400.13 MHz) δ: 0.90 (t, 3H,
J=7 Hz); 1.00 (t, 3H, J=7 Hz); 1.25–1.40 (m, 3H); 1.40–1.55 (m, 1H): 1.55–1.70 (m, 2H); 1.70–1.80 (m,
2H); 2.38 (s, 1H, exchangeable with D2O); 3.60 (dt, 1H, J=3.6 Hz, J=8.7 Hz); 3.94 (dt, 1H, J=3.6 Hz,
J=10.0 Hz). 13C NMR (100.61 MHz) δ: 10.3 (C-1); 14.0 (C-8); 22.3 (C-7); 25.4 (C-6); 29.0 (C-2); 32.3
(C-5); 68.8 (C-4); 76.3 (C-3). Anal. calcd for C8H17ClO: C, 58.35; H, 10.40; O, 9.72. Found: C, 57.96;
H, 10.36; O, 9.67. [α]D25 +25 (c 0.04, CHCl3); ee≥98%.
Rhodotorula glutinis: Incubation time: 6 h. The residue from 20 flasks consisted of 48% of (−)-(3S,4S)-
4-chloro-3-octanol and 52% of (+)-(3S,4R)-4-chloro-3-octanol. Yield: 26%.
(−)-(3S,4S)-4-Chloro-3-octanol: [α]D25 −35 (c 0.029, CHCl3); ee≥98%.
(+)-(3S,4R)-4-Chloro-3-octanol: [α]D25 +25 (c 0.032, CHCl3); ee≥98%.
Aspergillus niger: Incubation time: 24 h. The residue from 20 flasks consisted of 20% of 1, 24% of
(+)-(3R,4R)-4-chloro-3-octanol and 56% of (−)-(3R,4S)-4-chloro-3-octanol. Yield: 10%.
(+)-(3R,4R)-4-Chloro-3-octanol: [α]D25 +13 (c 0.01, CHCl3); ee=38%.
(−)-(3R,4S)-4-Chloro-3-octanol: [α]D25 −25 (c 0.05, CHCl3); ee≥98%.
Mortierella isabellina: Incubation time: 6 h. The residue from 10 flasks consisted of 50% of (−)-
(3S,4S)-4-chloro-3-octanol and 50% of (+)-(3S,4R)-4-chloro-3-octanol. Yield: 15%.
(−)-(3S,4S)-4-Chloro-3-octanol: [α]D25 −18 (c 0.02, CHCl3); ee=53%.
(+)-(3S,4R)-4-Chloro-3-octanol: [α]D25 +14 (c 0.02, CHCl3); ee=56%.
Lactobacillus kefir: Incubation time: 24 h. The residue from 12 flasks consisted of 10% of 1, 27% of
(+)-(3R,4R)-4-chloro-3-octanol and 63% of (−)-(3R,4S)-4-chloro-3-octanol. Yield: 18%.
(+)-(3R,4R)-4-Chloro-3-octanol: [α]D25 +35 (c 0.01, CHCl3); ee≥98%.
(−)-(3R,4S)-4-Chloro-3-octanol: [α]D25 −9 (c 0.01, CHCl3); ee=34%.
4.3.2. Microbiological reductions of 5-chloro-4-nonanone 2
Baker’s yeast: Incubation time: 24 h. The residue from 16 flasks consisted of 15% of 2, 48% of (−)-
(4S,5S)-5-chloro-4-nonanol and 37% of (+)-(4S,5R)-5-chloro-4-nonanol. Yield: 15%.
(−)-(4S,5S)-5-Chloro-4-nonanol: Retention time: 750 s (P2). 1H NMR (400.13 MHz) δ: 0.92 (t, 3H,
J=7.2 Hz); 0.94 (t, 3H, J=7.1 Hz); 1.24–1.46 (m, 4H); 1.46–1.62 (m, 4H); 1.81 (q, 2H, J=7.2 Hz); 1.96 (s,
1H, exchangeable with D2O); 3.58–3.68 (m, 1H); 3.90 (dt, 1H, J=3.7 Hz, J=6.7 Hz). 13C NMR (100.61
MHz) δ: 13.9; 14.0 (C-1, C-9); 19.1 (C-2); 22.4 (C-8); 28.9 (C-7); 34.7 (C-6); 37.0 (C-3); 69.2 (C-5);
73.7 (C-4). Anal. calcd for C9H19ClO: C, 60.49; H, 10.72; O, 8.95. Found: C, 59.83; H, 10.70; O, 8.83.
25
[α]D −36 (c 0.02, CHCl3); ee≥98%.
(+)-(4S,5R)-5-Chloro-4-nonanol: Retention time: 810 s (P2). 1H NMR (400.13 MHz) δ: 0.93 (t, 3H,
J=7.1 Hz); 0.97 (t, 3H, J=7.0 Hz); 1.28–1.42 (m, 4H); 1.42–1.67 (m, 4H); 1.69–1.78 (m, 2H); 2.38 (s,
1H, exchangeable with D2O); 3.72–3.80 (m, 1H); 4.01 (dt, 1H, J=3.7 Hz, J=9.6 Hz). 13C NMR (100.61
MHz) δ: 14.0, 14.1 (C-1, C-9); 19.2 (C-2); 22.3 (C-8); 29.1 (C-7); 32.3 (C-6); 34.6 (C-3); 69.3 (C-5);
74.6 (C-4). Anal. calcd for C9H19ClO: C, 60.49; H, 10.72; O, 8.95. Found: C, 60.05; H, 11.07; O, 8.78.
25
[α]D +14 (c 0.04, CHCl3); ee≥98%.
Rhodotorula glutinis: Incubation time: 6 h. The residue from 11 flasks consisted of 50% of (−)-(4S,5S)-
5-chloro-4-nonanol and 50% of (−)-(4R,5S)-5-chloro-4-nonanol. Yield: 8%.
25
(−)-(4S,5S)-5-Chloro-4-nonanol: [α]D −36 (c 0.01, CHCl3); ee≥98%.
(−)-(4R,5S)-5-Chloro-4-nonanol: [α]D25 −14 (c 0.01, CHCl3); ee≥98%.
Aspergillus niger: Incubation time: 48 h. The residue from 15 flasks consisted of 60% of (−)-(4S,5S)-
5-chloro-4-nonanol and 40% of (−)-(4R,5S)-5-chloro-4-nonanol. Yield: 8%.
25
(−)-(4S,5S)-5-Chloro-4-nonanol: [α]D −23 (c 0.02, CHCl3); ee=65%.
(−)-(4R,5S)-5-Chloro-4-nonanol: [α]D25 −10 (c 0.01, CHCl3); ee=71%.

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4453
Mortierella isabellina: Incubation time: 24 h. The residue from 10 flasks consisted of 85% of (−)-
(4S,5S)-5-chloro-4-nonanol and 15% of anti 5-chloro-4-nonanol. Yield: 5%.
25
(−)-(4S,5S)-5-Chloro-4-nonanol: [α]D −11 (c 0.01, CHCl3); ee=30%.
Lactobacillus kefir: Incubation time: 24 h. The residue from 12 flasks consisted of 10% of 2, 36% of
(−)-(4S,5S)-5-chloro-4-nonanol and 54% of (−)-(4R,5S)-5-chloro-4-nonanol. Yield: 6%.
25
(−)-(4S,5S)-5-Chloro-4-nonanol: [α]D −27 (c 0.01, CHCl3); ee=75%.
(−)-(4R,5S)-5-Chloro-4-nonanol: [α]D25 −14 (c 0.02, CHCl3); ee≥98%.
4.3.3. Microbiological reductions of 4-bromo-5-nonanone 3
Baker’s yeast: Incubation time: 24 h. The residue from 12 flasks consisted of 19% of 3, 73% of (+)-
(4R,5R)-4-bromo-5-nonanol and 8% of anti 4-bromo-5-nonanol. Yield: 9%.
(+)-(4R,5R)-4-Bromo-5-nonanol: Retention time: 890 s (P2). 1H NMR (400.13 MHz) δ: 0.91 (t, 3H,
J=7.1 Hz); 0.94 (t, 3H, J=7.3 Hz); 1.20–1.34 (m, 2H); 1.34–1.45 (m, 3H); 1.45–1.60 (m, 3H); 1.72–1.83
(m, 2H); 1.90–2.02 (m, 1H); 3.44–3.53 (m, 1H); 3.98 (td, 1H, J=4.0 Hz, J=8.7 Hz). 13C NMR (100.61
MHz) δ: 13.5; 14.1 (C-1, C-9); 21.1 (C-2); 22.7 (C-8); 27.8 (C-7); 35.6 (C-6); 37.9 (C-3); 65.3 (C-5);
73.9 (C-4). Anal. calcd for C9H19BrO: C, 48.44; H, 8.58; O, 7.17. Found: C, 48.39; H, 8.63; O, 7.10.
25
[α]D +7 (c 0.024, CHCl3); ee=40%.
Rhodotorula glutinis: Incubation time: 48 h. The residue from 12 flasks consisted of 25% of 3, 38% of
(−)-(4S,5S)-4-bromo-5-nonanol and 37% of (+)-(4R,5S)-4-bromo-5-nonanol. Yield: 18%.
25
(−)-(4S,5S)-4-Bromo-5-nonanol: [α]D −17 (c 0.02, CHCl3); ee≥98%.
(+)-(4R,5S)-4-Bromo-5-nonanol: Retention time: 810 s (P2). 1H NMR (400.13 MHz) δ: 0.90 (t, 3H,
J=7.1 Hz); 0.94 (t, 3H, J=7.0 Hz); 1.28–1.42 (m, 4H); 1.48–1.62 (m, 3H); 1.62–1.90 (m, 3H); 2.38 (s,
1H, exchangeable with D2O); 3.65–3.75 (m, 1H); 4.21 (td, 1H, J=3.2 Hz, J=10.4 Hz). 13C NMR (100.61
MHz) δ: 13.5, 14.1 (C-1, C-9); 21.3 (C-2); 22.7 (C-8); 28.2 (C-7); 32.9 (C-6); 35.2 (C-3); 65.2 (C-5);
74.9 (C-4). Anal. calcd for C9H19BrO: C, 48.44; H, 8.58; O, 7.17. Found: C, 48.41; H, 8.65; O, 7.21.
25
[α]D +17 (c 0.02, CHCl3); ee≥98%.
Mortierella isabellina: Incubation time: 6 h. The residue from 10 flasks consisted of 50% of (−)-
(4S,5S)-4-bromo-5-nonanol and 50% of (+)-(4R,5S)-4-bromo-5-nonanol. Yield: 18%.
25
(−)-(4S,5S)-4-Bromo-5-nonanol: [α]D −12 (c 0.01, CHCl3); ee=70%.
25
(+)-(4R,5S)-4-Bromo-5-nonanol: [α]D +12 (c 0.01, CHCl3); ee=70%.
4.3.4. Microbiological reductions of 4-chloro-5-nonanone 4
Baker’s yeast: Incubation time: 24 h. The residue from 10 flasks consisted of 10% of 4, 77% of (−)-
(4S,5S)-4-chloro-5-nonanol and 13% of (+)-(4R,5S)-4-chloro-5-nonanol. Yield: 28%.
(−)-(4S,5S)-4-Chloro-5-nonanol: Retention time: 837 s (P2). 1H NMR (400.13 MHz) δ: 0.94 (t, 3H,
J=7.1 Hz); 0.99 (t, 3H, J=7.2 Hz); 1.25–1.36 (m, 2H); 1.36–1.52 (m, 3H); 1.52–1.65 (m, 3H); 1.65–1.85
(m, 2H); 2.15 (s, 1H, exchangeable with D2O); 3.55–3.67 (m, 1H); 3.92 (td, 1H, J=4.1 Hz, J=8.8 Hz).
13C NMR (100.61 MHz) δ: 13.5, 14.0 (C-1, C-9); 19.9 (C-2); 22.6 (C-8); 27.9 (C-7); 34.3 (C-6); 36.9
(C-3); 68.7 (C-4); 74.0 (C-5). Anal. calcd for C9H19ClO: C, 60.49; H, 10.72; O, 8.95. Found: C, 60.53;
H, 10.71; O, 8.93. [α]D25 −6 (c 0.04, CHCl3); ee=17%.
(+)-(4R,5S)-4-Chloro-5-nonanol: Retention time: 899 s (P2). 1H NMR (400.13 MHz) δ: 0.92 (t, 3H,
J=7.1 Hz); 0.95 (t, 3H, J=7.3 Hz); 1.24–1.42 (m, 4H); 1.48–1.58 (m, 3H); 1.62–1.78 (m, 3H); 1.95 (s,
1H, exchangeable with D2O); 3.70–3.78 (m, 1H); 4.03 (td, 1H, J=3.8 Hz, J=9.5 Hz). 13C NMR (100.61
MHz) δ: 13.6, 14.0 (C-1, C-9); 20.1 (C-2); 22.7 (C-8); 28.1 (C-7); 32.2 (C-6); 34.6 (C-3); 68.9 (C-4);
74.9 (C-5). Anal. calcd for C9H19ClO: C, 60.49; H, 10.72; O, 8.95. Found: C, 60.72; H, 10.77; O, 9.00.
25
[α]D +13 (c 0.02, CHCl3); ee≥98%.

4454
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
Rhodotorula glutinis: Incubation time: 24 h. The residue from 10 flasks consisted of 50% of (−)-
(4S,5S)-4-chloro-5-nonanol and 50% of (+)-(4R,5S)-4-chloro-5-nonanol. Yield: 14%.
25
(−)-(4S,5S)-4-Chloro-5-nonanol: [α]D −32 (c 0.03, CHCl3); ee≥98%.
(+)-(4R,5S)-4-Chloro-5-nonanol: [α]D25 +10 (c 0.01, CHCl3); ee=78%.
Mortierella isabellina: Incubation time: 6 h. The residue from 15 flasks consisted of 50% of (−)-
(4S,5S)-4-chloro-5-nonanol and 50% of (+)-(4R,5S)-4-chloro-5-nonanol. Yield: 52%.
25
(−)-(4S,5S)-4-Chloro-5-nonanol: [α]D −4 (c 0.02, CHCl3); ee=12%.
(+)-(4R,5S)-4-Chloro-5-nonanol: [α]D25 +3 (c 0.03, CHCl3); ee=23%.
Lactobacillus kefir: Incubation time: 24 h. The residue from 14 flasks consisted of 30% of 4, 28% of
(−)-(4S,5S)-4-chloro-5-nonanol and 42% of (−)-(4S,5R)-4-chloro-5-nonanol. Yield: 20%.
25
(−)-(4S,5S)-4-Chloro-5-nonanol: [α]D −32 (c 0.02, CHCl3); ee≥98%.
(−)-(4S,5R)-4-Chloro-5-nonanol: [α]D25 −13 (c 0.05, CHCl3); ee≥98%.
4.4. Determination of the enantiomeric excesses
4.4.1. Diastereoisomers of 4-chloro-3-octanol
The enantiomeric excesses were determined by 13C or 19F NMR after converting an aliquot of each
diastereoisomer of the product to the corresponding α-methoxy-α-trifluoromethylphenyl acetate (MPTA
ester), synthesized according to Dale et al.,27 and was then purified by column chromatography (eluent:
pentane:ether 75:25). Yield: 90%.
Diastereoisomer 1 (syn): 1H NMR (400.13 MHz) δ: 0.85 (t, 3H, J=7.1 Hz); 0.90 (t, 3H, J=7.1 Hz);
1.25–1.43 (m, 3H); 1.48–1.60 (m, 1H); 1.60–1.90 (m, 4H); 3.60 (s, 3H); 3.97 (td, 1H, J=3.4 Hz, J=9.8
Hz); 5.19 (td, 1H, J=3.4 Hz, J=7.7 Hz); 7.35–7.50 (m, 3H); 7.55–7.67 (m, 2H). 13C NMR (100.61 MHz)
δ [italic shifts correspond to the (3S,4S) diastereoisomer]: 9.4, 9.8 (CH3); 13.9 (CH3); 22.2 (CH2); 23.7,
23.9 (CH2); 28.7, 28.8 (CH2); 33.6, 33.9 (CH2); 55.7 (OCH3); 62.0, 62.4 (CH–O); 79.3, 79.5 (CH–Cl);
122.0 (CF3); 124.8 (C–CF3); 127.4, 127.6 (Ar); 128.5 (Ar); 129.7 (Ar); 132.0 (Ar); 166.3 (C_O). 19F
NMR (376.48 MHz) δ: −0.001 (diastereoisomer R,R); 0.259 (diastereoisomer S,S).
1
Diastereoisomer 2 (anti): H NMR (400.13 MHz) δ: 0.83 (t, 3H, J=7 Hz); 0.93 (t, 3H, J=7 Hz);
1.18–1.43 (m, 3H); 1.50–1.80 (m, 4H); 1.80–1.95 (m, 1H); 3.61 (s, 3H); 4.11 (td, 1H, J=3.5 Hz, J=10.2
Hz); 5.15 (td, 1H, J=3.5 Hz, J=8.1 Hz); 7.35–7.50 (m, 3H); 7.50–7.58 (m, 1H); 7.58–7.70 (m, 1H).
13C NMR (100.61 MHz) δ [italic shifts correspond to the (3S,4R) diastereoisomer]: 9.5, 9.6 (CH3); 13.9
(CH3); 22.2 (CH2); 22.7, 23.4 (CH2); 28.6, 28.8 (CH2); 33.1, 33.3 (CH2); 36.2 (CH2); 55.6, 55.7 (OCH3);
62.3, 62.8 (CH–O); 80.0, 80.3 (CH–Cl); 122.0 (CF3); 124.8 (C–CF3); 127.5, 127.6 (Ar); 128.5 (Ar);
129.7 (Ar); 132.0, 132.2 (Ar); 166.3, 166.5 (C_O). 19F NMR (376.48 MHz) δ: 0.099 (diastereoisomer
R,S); 0.342 (diastereoisomer S,R).
4.4.2. Diastereoisomers of 5-chloro-4-nonanol
The enantiomeric excesses were determined by gas chromatography on a chiral capillary column
(Lipodex E). Oven temperature: 80°C. The retention times are as follows: 1316 s [diastereoisomer
(4S,5S)] and 1352 s [diastereoisomer (4R,5R)]; 1600 s [diastereoisomer (4S,5R)] and 1773 s [diastereo-
isomer (4R,5S)].
4.4.3. Diastereoisomers of 4-bromo-5-nonanol
The enantiomeric excesses were determined by gas chromatography: either on a Carbowax 20M after
derivatization of the syn diastereoisomer with lactic chloride28 [oven temperature: 100°C. Retention

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4455
times: 4870 s (4R,5R) and 5315 s (4S,5S)] or on a chiral capillary column (Lipodex E) for the anti
diastereoisomer: oven temperature: 90°C. Retention times: 1679 s (4S,5R) and 1800 s (4R,5S).
4.4.4. Diastereoisomers of 4-chloro-5-nonanol
The enantiomeric excesses were determined by 1H NMR after converting an aliquot of each diastereo-
isomer of the product to the corresponding MTPA-ester. Purification by column chromatography (eluent:
pentane:ether 95:5). Yield: 90%.
Diastereoisomer 1 (syn): 1H NMR (400.13 MHz) δ (italic shifts correspond to the (4S,5S) diastereo-
isomer): 0.84, 0.85 (t, 3H, J=7.0 Hz); 0.91, 0.92 (t, 3H, J=7.3 Hz); 1.05–1.21 (m, 1H); 1.21–1.45 (m,
4H); 1.45–1.56 (m, 1H); 1.62–1.79 (m, 4H); 3.58, 3.60 (s, 3H); 3.93–4.02 (m, 1H); 5.17–5.29 (m, 1H);
7.35–7.45 (m, 3H); 7.55–7.65 (m, 2H). 13C NMR (100.61 MHz) δ: 13.4 (CH3); 13.9 (CH3); 19.8 (CH2);
22.4 (CH2); 27.0, 27.4 (CH2); 30.0, 30.3 (CH2); 35.7, 36.1 (CH2); 55.7 (OCH3); 61.8, 62.0 (CH–O);
78.1, 78.2 (CH–Cl); 121.9 (CF3); 124.8 (C–CF3); 127.4, 127.5 (Ar); 127.7 (Ar); 129.7 (Ar); 132.0 (Ar);
166.3 (CO).
Diastereoisomer 2 (anti): 1H NMR (400.13 MHz) δ [italic shifts correspond to the (4R,5S) diastereo-
isomer]: 0.88 (t, 3H, J=7.0 Hz); 0.93 (t, 3H, J=7.1 Hz); 1.09–1.22 (m, 1H); 1.22–1.48 (m, 4H); 1.48–1.55
(m, 1H); 1.55–1.80 (m, 3H); 1.80–1.91 (m, 1H); 3.57, 3.63 (s, 3H); 4.13 (td, 1H, J=3.8 Hz, J=9.0 Hz);
5.10 (td, 1H, J=3.8 Hz, J=10.0 Hz); 7.35–7.45 (m, 3H); 7.55–7.65 (m, 2H). 13C NMR (100.61 MHz) δ:
13.4 (CH3); 14.1 (CH3); 19.7, 19.9 (CH2); 22.3 (CH2); 27.1, 27.2 (CH2); 29.1, 29.9 (CH2); 35.3, 35.5
(CH2); 55.6 (OCH3); 62.4, 62.9 (CH–O); 78.7, 78.8 (CH–Cl); 121.9 (CF3); 124.8 (C–CF3); 127.5 (Ar);
127.6 (Ar); 129.7 (Ar); 132.3 (Ar); 166.4 (CO).
4.5. Determination of the absolute configuration
4.5.1. Dechlorination of the chlorohydrins
This chemical correlation was used for both the diastereoisomers of 4-chloro-3-octanol and 5-chloro-
4-nonanol.
From 4-chloro-3-octanol: A solution of 50 mg (0.3 mmol) of each diastereoisomer of 4-chloro-3-
octanol, obtained by microbiological reduction with baker’s yeast, in 3 mL of benzene, was treated with
tributyltin hydride (90 µL, 35.6 mmol) and azoisobutyronitrile (AIBN 4 mg) under reflux for 2 h. The
solvent was evaporated off. The crude residue was chromatographed on silica (pentane:ether 90:10) to
afford 3-octanol. Yield: 90%. Retention time: 350 s (P2). 1H NMR (400.13 MHz) δ: 0.91 (t, 3H, J=7.1
Hz); 0.94 (t, 3H, J=7.1 Hz); 1.20–1.60 (m, 10H); 1.85 (s, 1H, exchangeable with D2O); 3.45–3.60 (m,
1H). Same 1H and 13C NMR spectra as those described by Bonini.30
– from the diastereoisomer syn: [α]D +11.5 (c 0.01, CHCl3). Lit.:30 (−)-(3R)-3-octanol: [α]D −12.5
(c 1.29, CHCl3).
25
25
– from the diastereoisomer anti: [α]D25 +11.1 (c 0.02, CHCl3). Lit.:30 (−)-(3R)-3-octanol: [α]D25 −12.5
(c 1.29, CHCl3).
From 5-chloro-4-nonanol: The same procedure as that described above was used. The starting material
was each diastereoisomer of 5-chloro-4-nonanol, obtained by the reduction with baker’s yeast, and the
final compound was 4-nonanol. Retention time: 408 s (P2). 1H NMR (400.13 MHz) δ: 0.91 (t, 3H, J=7.0
Hz); 0.95 (t, 3H, J=7.0 Hz); 1.26–1.34 (m, 8H); 1.34–1.56 (m, 4H); 1.75 (s, 1H, exchangeable with D2O);
3.55–3.67 (m, 1H). Same 1H and 13C NMR spectra as those described by Ohta.31
– from the diastereoisomer syn: [α]D +2 (c 0.05, CHCl3). Lit.:31 (+)-(4S)-4-nonanol: [α]D +0.57 (c
5.8, hexane), ee=60%.
25
25

4456
P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
– from the diastereoisomer anti: [α]D25 +2 (c 0.03, CHCl3). Lit.:31 (+)-(4S)-4-nonanol: [α]D +0.57 (c
5.8, hexane), ee=60%.
25
4.5.2. Synthesis of the chiral epoxides
General method: To a solution of 1 mmol of chlorohydrin in 5 mL of DMF was added 93 µL of distilled
water and 3 mmol of K2CO3. The mixture was stirred at room temperature for 2 days. Water was then
added and the mixture was extracted three times with ether. The organic layer was dried over MgSO4.
The solvent was removed on a water bath and the residue was chromatographed (eluent: pentane:ether
90:10).
From the syn (4S,5S) and anti (4S,5R)-5-chloro-4-nonanol obtained with baker’s yeast:
1
cis (4S,5R)-4,5-Epoxynonane: Yield: 50%. Retention time: 550 s (P2). H NMR (400.13 MHz) δ:
0.93 (t, 3H, J=7.2 Hz); 0.99 (t, 3H, J=7.2 Hz); 1.33–1.45 (m, 4H); 1.45–1.60 (m, 6H); 2.87–2.96 (m,
2H). 13C NMR (100.61 MHz) δ: 14.1 (C-1, C-9); 20.0 (C-2); 22.7 (C-8); 27.6 (C-7); 28.8 (C-6); 29.9
25
(C-3); 57.1; 57.2 (C-4, C-5). [α]D −5 (c 0.02, pentane).
trans (4S,5S)-4,5-Epoxynonane: Yield: 70%. Retention time: 530 s (P2). 1H NMR (400.13 MHz) δ:
0.90 (t, 3H, J=7.1 Hz); 0.95 (t, 3H, J=7.1 Hz); 1.24–1.45 (m, 4H); 1.45–1.60 (m, 6H); 2.61–2.70 (m,
2H). 13C NMR (100.61 MHz) δ: 14.1 (C-1, C-9); 19.4 (C-2); 22.6 (C-8); 28.2 (C-7); 31.9 (C-6); 34.3
25
(C-3); 58.8; 58.9 (C-4, C-5). [α]D −32 (c 0.02, pentane).
From the anti (4R,5S)-5-chloro-4-nonanol obtained with Lactobacillus kefir:
trans (4R,5R)-4,5-Epoxynonane: Yield: 72%. Same retention time and NMR spectra as those de-
25
scribed above. [α]D +32 (c 0.05, pentane).
From the syn (4S,5S) and anti (4R,5S)-4-chloro-5-nonanol obtained with Rhodotorula glutinis:
cis (4R,5S)-4,5-Epoxynonane: Yield: 48%. Same retention time and NMR spectra as those described
above. [α]D25 +5 (c 0.04, pentane).
trans (4S,5S)-4,5-Epoxynonane: Yield: 71%. Same retention time and NMR spectra as those de-
scribed previously. [α]D25 −25 (c 0.05, pentane).
4.5.3. Opening of 4,5-epoxynonane
General method: A solution of 1 mmol of epoxide in 6 mL dry ether was added to a suspension of
0.3 mmol of LiAlH4 in dry ether. The mixture was refluxed for 2 h, then decomposed with water and
extracted three times with ether. After evaporation of the solvent, the residue was chromatographed,
eluent: pentane:ether 70:30, and a mixture of 4- and 5-nonanol was obtained.
25
From cis (4R,5S)-4,5-epoxynonane: Yield: 60%. [α]D −0.6 (c 0.02, CHCl3). Lit.:31 (+)-(4S)-4-
nonanol: [α]D25 +0.57 (c 5.8, hexane), ee=60%.
From trans (4S,5S)-4,5-epoxynonane: Yield: 65%. [α]D +0.4 (c 0.02, CHCl3). Lit.:31 (+)-(4S)-4-
25
nonanol: [α]D25 +0.57 (c 5.8, hexane), ee=60%.
Acknowledgements
We gratefully acknowledge Martine Sancelme for technical assistance in microbiology and Jean
Gabriel Gourcy for the delicate synthesis of diazomethane.
References
1. Besse, P.; Veschambre, H. Tetrahedron 1994, 50, 8885–8927.

P. Besse et al. / Tetrahedron: Asymmetry 9 (1998) 4441–4457
4457
2. Pedragosa-Moreau, S.; Archelas, A.; Furstoss, R. Bull. Soc. Chim. Fr. 1995, 132, 769–800.
3. Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5974–5976.
4. Katsuki, T. Coord. Chem. Rev. 1995, 140, 189–218. Jacobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J. R.; Deng, L. J. Am.
Chem. Soc. 1991, 113, 7063–7064.
5. Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997, 277, 936–938.
6. Archelas, A.; Furstoss, R. Annu. Rev. Microbiol. 1997, 51, 491–525.
7. Ortiz de Montellano, P. R. In Cytochromes P-450; Ortiz de Montellano, P. R., Ed. Plenum Press: New York, 1986.
8. Guenguerich, F. P. Biochem. Mol. Biol. 1990, 25, 97–153.
9. May, S. W.; Abbott, B. J. J. Biol. Chem. 1973, 248, 1725–1730. Coon, M. J.; Stabel, H. W.; Autor, A. P.; Heidemeda, J. In
Biological Hydroxylation Mechanism; Boyd, G. S.; Smellie, R. M. S., Eds. Academic Press: New York, 1972.
10. Mahmoudian, M.; Michael, A. Appl. Microbiol. Biotechnol. 1992, 37, 23–31. de Bont, J. A. M.; Primrose, S. B.; Harder,
W. FEMS Microbiol. Lett. 1979, 6, 183–189. van Ginkel, C. G.; Welten, H. G. J.; de Bont, J. A. M. Appl. Microbiol.
Biotechnol. 1986, 24, 334–337. Small, F. J.; Ensign, S. A. J. Biol. Chem. 1997, 272, 24913–24920.
11. Takahashi, O.; Umezawa, J.; Furuhashi, K. Ann. N. Y. Acad. Sci. 1990, 613, 697–701. Gallagher, S. C.; Cammarck, R.;
Dalton, H. Eur. J. Biochem. 1997, 247, 635–641.
12. Geigert, J.; Lee, T. D.; Dalietos, D. J.; Hirano, D. S.; Neidleman, S. L. Biochem. Biophys. Res. Commun. 1986, 136,
778–782.
13. Allain, E. J.; Hager, L. P.; Deng, L.; Jacobsen, E. J. J. Am. Chem. Soc. 1993, 115, 4415–4416.
14. Archer, I. V. J. Tetrahedron 1997, 53, 15617–15662.
15. Chen, X.-J.; Archelas, A.; Furstoss, R. J. Org. Chem. 1993, 58, 5528–5532.
16. Pedragosa-Moreau, S.; Archelas, A.; Furstoss, R. J. Org. Chem. 1993, 58, 5533–5536.
17. Wandel, U.; Mischitz, M.; Kroutil, W.; Faber, K. J. Chem. Soc., Perkin Trans. 1 1995, 735–736. Kroutil, W.; Mischitz, M.;
Plachota, P.; Faber, K. Tetrahedron Lett. 1996, 37, 8379–8382. Mischitz, M.; Faber, K. Synlett 1996, 978–980.
18. Weijers, C. A. G. M. Tetrahedron: Asymmetry 1997, 8, 639–647.
19. Orru, R. V. A.; Mayer, S. F.; Kroutil, W.; Faber, K. Tetrahedron 1998, 859–874.
20. Besse, P.; Veschambre, H. Tetrahedron: Asymmetry 1993, 4, 1271–1285.
21. Barluenga, J.; Llavona, L.; Yus, M.; Concellon, J. M. J. Chem. Soc., Perkin Trans. 1 1991, 2890.
22. Fawzi, M. M.; Gutsche, C. D. J. Org. Chem. 1966, 31, 1390–1393.
23. Hooz, J.; Bridson, J. N. Can. J. Chem. 1972, 50, 2387–2390.
24. Perez, D.; Greenspoon, N.; Keinan, E. J. Org. Chem. 1987, 52, 5570–5574.
25. Bodot, H.; Dieuzeide, E.; Jullien, J. Bull. Soc. Chim. Fr. 1960, 1086–1097.
26. Besse, P.; Renard, M. F.; Veschambre, H. Tetrahedron: Asymmetry 1994, 5, 1249–1268.
27. Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34, 2543–2549.
28. Mosandl, A.; Gessner, M.; Günther, C.; Deger, W.; Singer, G. J. High Resol. Chromatogr. 1987, 67–70.
29. Denmark, S. E.; Parker Jr., D. L.; Dixon, J. A. J. Org. Chem. 1997, 62, 435–436.
30. Bonini, C.; Federici, C.; Rossi, L.; Righi, G. J. Org. Chem. 1995, 60, 4803–4812.
31. Ohta, S. Bull. Chem. Soc. Jpn. 1986, 59, 1181–1188.
32. Groves, J. T.; Ahn, K. H.; Quinn, R. J. Am. Chem. Soc. 1988, 110, 4217–4220.
33. Shimagaki, M.; Matsuzaki, Y.; Hori, I.; Nakata, T. Tetrahedron Lett. 1984, 25, 4779–4782.
34. Utaka, M.; Konishi, S.; Takeda, A. Tetrahedron Lett. 1986, 27, 4737–4740.
35. Eliel, E. L.; Rerick, M. N. J. Am. Chem. Soc. 1959, 82, 1362–1367.
36. Faber, K. In Biotransformations in Organic Chemistry; Springer Verlag, 1994; 2nd Edition, p. 161.