Kinetic resolution of primary 2-methyl-substituted alcohols via Pseudomonas cepacia lipase-catalysed enantioselective acylation

Article abstract of DOI:10.1039/a908023f

The enantioselectivities of lipases from Pseudomonas cepacia (PFL, Amano PS, etc.) towards a series of primary 2-methyl-substituted alcohols using vinyl acetate as the acyl donor in transesterifications in organic solvents were studied. In terms of enantioselectivity, the best results were found for 3-aryl-2-methylpropan-1-ols with enantiomeric ratios (E-values) over 100 in most cases, whereas other 3-substituted primary 2-methylpropan-1-ols generally displayed lower enantioselectivities: 3-cycloalkyl-2-methylpropan-1-ols (E ≈ 20) and 2-methylalkan-1-ols (E ≈ 10). Moving the aryl group closer or further away from the chiral centre resulted in low enantioselectivities: 2-arylpropan-1-ols (E < 10), 2-methyl-4-(2-thienyl)butan-1-ol (E = 12), 2-methyl-5-(2-thienyl)pentan-1-ol (E = 3.2) and 2-methyl-6-(2-thienyl)hexan-1-ol (E = 3.8).

Full text of DOI:10.1039/a908023f

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetic resolution of primary 2-methyl-substituted alcohols via
Pseudomonas cepacia lipase-catalysed enantioselective acylation
Ove Nordin, Ba-Vu Nguyen, Carin Vörde, Erik Hedenström and Hans-Erik Högberg
1
Department of Chemistry and Process Technology, Mid Sweden University,
SE-851 70 Sundsvall, Sweden
Received (in Lund, Sweden) 4th October 1999, Accepted 22nd November 1999
The enantioselectivities of lipases from Pseudomonas cepacia (PFL, Amano PS, etc.) towards a series of primary
2-methyl-substituted alcohols using vinyl acetate as the acyl donor in transesterifications in organic solvents
were studied. In terms of enantioselectivity, the best results were found for 3-aryl-2-methylpropan-1-ols with
enantiomeric ratios (E-values) over 100 in most cases, whereas other 3-substituted primary 2-methylpropan-1-ols
generally displayed lower enantioselectivities: 3-cycloalkyl-2-methylpropan-1-ols (E ≈ 20) and 2-methylalkan-1-ols
(E ≈ 10). Moving the aryl group closer or further away from the chiral centre resulted in low enantioselectivities:
2-arylpropan-1-ols (E < 10), 2-methyl-4-(2-thienyl)butan-1-ol (E = 12), 2-methyl-5-(2-thienyl)pentan-1-ol (E = 3.2)
and 2-methyl-6-(2-thienyl)hexan-1-ol (E = 3.8).
One of the most attractive approaches for obtaining chiral
nonracemic compounds is via enantioselective catalysis. Within
this area biocatalysis has evolved as one of the most efficient
methods and both whole-cell systems and isolated enzymes are
widely used as catalysts.1–5
A large number of hydrolytic enzymes, especially lipases,
have been used successfully for the kinetic resolution of racemic
secondary alcohols via enantioselective acylation or hydrolysis
of their esters.5–7 However, a very limited number of enzymes
are capable to resolve racemates of primary alcohols or their
esters. Thus, only lipases from Pseudomonas and Porcine
Pancreas (PPL) are known to efficiently resolve these
substrates.5
Enantiomerically pure 2-methyl-branched, primary alcohols
are valuable synthetic building blocks for, e.g., pheromone
synthesis8,9 and many methods have been developed for their
preparation (Scheme 1). They can be obtained via chemical
Fig. 1 Empirical rules for the enantiopreferences of Pseudomonas
lipases.25,26 Models A and B describe the situation for secondary and
primary alcohols, respectively, assuming the latter employ the same
pockets. M and L symbolise medium sized and large substituents, e.g.
methyl and phenyl, respectively. The rules for models B and C do not
hold if either M or L is equal to OR.21
Scheme 1 Some examples of synthetic strategies for the preparation
of enantiomerically pure 2-methylalkan-1-ols. Reagents and conditions:
i, base then remove Xc, then reduction; ii, baker’s yeast reduction
provides one enantiomer; iii, lipase catalysed kinetic resolution then
product separation provides the pure enantiomers (one as ester).
PPL-catalysed kinetic resolutions of racemic 2-methyl-
substituted alcohols or esters,18–20 Pseudomonas-derived lipases
are the biocatalysts of choice for these transformations
today.5,21–24
methods, e.g. auxiliary-based ones such as the widely used
diastereoselective alkylation of amide enolates which, after
removal of the chiral auxiliary followed by reduction, gives
nonracemic 2-methyl-substituted alcohols.10,11 Biocatalysts
have also been found very useful for the preparation of such
alcohols. Thus baker’s yeast reduction of (E)-3-aryl-2-
methylpropenals provides moderate to good yields of (S)-3-
aryl-2-methylpropan-1-ols.12–17 However, if both enantiomers
are needed, lipase-catalysed kinetic resolutions are preferred.
Although a few examples are available of moderately successful
Predictive rules have been derived for the enantiopreference
of Pseudomonas lipases towards secondary alcohols and also
for primary alcohols branched in the 2-position (A and B,
respectively, in Fig. 1) without oxygen atoms at the stereogenic
centre.25,26 These rules postulate that primary alcohols dock
into the same pockets as secondary alcohols. Assuming that this
is the case, additional models, including box models, have been
presented, describing the active site of Pseudomonas lipases.27–29
For Pseudomonas cepacia lipase (PCL, see introductory remark
under Experimental section), the X-ray crystal structure of the
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376
This journal is © The Royal Society of Chemistry 2000
367

View Article Online
open conformation has recently been reported.30,31 Molecular-
modelling studies of transition-state analogues bound to the
active site of PCL, have been performed starting from the X-ray
structure of PCL and two large hydrophobic pockets have been
identified.21 From these studies it appears that primary alcohols
bind to PCL in a different mode than do secondary ones.21
Thus, for 2-methyl-3-phenylpropan-1-ol (2g, see structures in
Table 2) the large benzyl group binds to an alternative hydro-
phobic pocket in a narrow groove not used by secondary alco-
hols (see Fig. 1, C).21 The enantioselectivity of PCL toward the
S-enantiomer arises from binding of the methyl group at the
stereogenic centre to a small hydrophobic pocket.21 This inter-
action is absent with the slow reacting R-enantiomer.21 It must
be pointed out that if the enzyme employs different docking
modes for primary and secondary alcohols, box models derived
from reactions of both types of alcohols must be used with
caution.
butyl-2-thienyl)propan-1-ol 1c via Friedel–Crafts acylation
followed by Huang-Minlon reduction (Scheme 2c).14 2-Methyl-
3-(5-propyl-2-thienyl)propan-1-ol 2d, its precursor 2c, and 3-(5-
ethyl-3-thienyl)-2-methylpropan-1-ol34 2e were obtained in a
similar way from compounds 2a and 2b.
We have long been interested in the preparation of enantio-
merically pure 2-methyl-1-alkyl building blocks for the
syntheses of highly pure stereoisomers of pine sawfly
pheromones.9,32–34 For 2-methylalkanols, we have studied
baker’s yeast-mediated reductions14 and Candida rugosa lipase-
catalysed kinetic resolution by esterification of 2-methyl-
alkanoic acids followed by reduction.35,36 Although the latter
sequence is quite satisfactory for obtaining both enantiomers
of a 2-methylalkan-1-ol, we wished to develop alternative
strategies. Therefore we decided to study the use of PCL as
the catalyst for the kinetic resolution of a number of 2-methyl-
substituted alcohols via transesterification of vinyl acetate. The
results of our investigations are summarised here (preliminary
accounts of parts of this work: Högberg et al.,9 Nguyen et al.22
and Nordin et al.23).
Results and discussion
Preparation of substrates
In order to explore the enantioselectivity of the PCL-catalysed
transesterification as a function of the substrate structure we
have studied a number of new substrates of types 1–6 (Chart 1).
Scheme 2 Preparation of substrates containing thienyl and cycloalkyl
moieties. For R1 and R2, arrows indicate points of attachment.
Reagents and conditions: i, LDA then MeI; ii, LAH then H2O; iii, Anion
of diethyl methylmalonate then NaOH, EtOH, H2O then H3Oϩ, H2O
then heat (decarboxylation); iv, R3COCl, SnCl4, CH2Cl2; v, N2H4,
NaOH, diethylene glycol, 120 to >200 ЊC; vi, 2 eq. LDA then MeOTs.
2-Methyl-4-(2-thienyl)butan-1-ol 3a and 2-methyl-6-(2-
thienyl)hexan-1-ol 5a were prepared from 2-(2-bromoethyl)-
thiophene and 2-(4-bromobutyl)thiophene, respectively, using
the methylmalonate method described above (Scheme 2b).
Deprotonation of 5-(2-thienyl)pentanoic acid with 2 mole
equiv. of LDA followed by alkylation with methyl toluene-
p-sulfonate in the presence of 1,3-dimethylpropylene urea
(DMPU) gave, after acidification, 2-methyl-5-(2-thienyl)pent-
anoic acid which, when subjected to LAH reduction, furnished
2-methyl-5-(2-thienyl)pentan-1-ol 4a (Scheme 2d).
Some additional substrates were also prepared (see Schemes
2 and 3) and used in this study. Thus 3-cyclopentyl- and
3-cyclohexyl-2-methylpropan-1-ol, 2l and 2m, respectively, were
prepared using the methylmalonate route (Scheme 2b). Sec-
ondly, 2-methoxybenzaldehyde was condensed with propanal.
After reduction of the double bond of the aldol condensation
product followed by borohydride reduction of the resulting
aldehyde, 3-(o-methoxyphenyl)-2-methylpropan-1-ol 2h was
obtained (Scheme 3a). Racemisation of 2-methylalkanols can
be performed by treatment with sodium and benzophenone in
hot toluene.24 Application of this method to (S)-3-(2-furyl)-
2-methylpropan-1-ol (S)-2f, which can be obtained from 3-(2-
furyl)-2-methylpropenal via baker’s yeast reduction,12 furnished
3-(2-furyl)-2-methylpropan-1-ol 2f (Scheme 3b). Finally, con-
jugate addition of thiophenol to methacrolein furnished an
Chart 1 2-Alkylalkan-1-ols studied as substrates for PCL-catalysed
transesterifications.
The syntheses of these are summarised below and details of
the preparative procedures can be found in the Experimental
section.
For comparison with other 2-methyl-ω-(2-thienyl)alkan-1-ols
we prepared some thiophene derivatives 1a–c, 2b, 3a, 4a and 5a
as described in Scheme 2. Thus ethyl thiophene-2-acetate was
alkylated37 with methyl iodide and reduced to give 2-(2-thienyl)-
propan-1-ol 1a (Scheme 2a). 2-(3-Thienyl)propan-1-ol 1b was
obtained in a similar way from ethyl thiophene-3-acetate. The
two 2-methyl-3-thienylpropan-1-ols 2a and 2b were prepared
via alkylation of diethyl methylmalonate followed by hydrolysis
to the diacid, decarboxylation and LAH reduction (Scheme
2b). 2-(2-Thienyl)propan-1-ol 1a was transformed into 2-(5-
368
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376

View Article Online
Table 1 PCL-catalysed kinetic resolution of 2-methylalkan-1-ols by transesterification with vinyl acetate
Substrate n =
Enzyme
Solvent
c/%
eep/%
ees/%
E
Ref.
5
6
8
Amano PS
Amano PS
Amano PS
PFL (Fluka)
Amano PS
Amano PS
Amano AK
PFL (Fluka)
PFL (Fluka)
PFL (Fluka)
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
CH2Cl2
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
80.2
74.6
78.0
39.5
70.0
40.6
39.3
38.3
39.4
42.7
28.7
33.7
27.4
60.0
42.4
73.4
72.5
73.8
74.4
72.4
98
99
96.2
5.9
8.7
5.7
5.8
9.9
10.7
9.9
10.4
10.9
10.6
24
24
24
23
24
23
23
23
23
23
8
10
10
10
10
12
14
98.1
aldehyde which, after borohydride reduction, gave 2-methyl-3-
R = C7H15) provides (S)-2-methyldecan-1-ol albeit in both
unsatisfactory ee and yield.14 To circumvent this problem we
have utilised a thiophene ring as a masked -(CH2)4- group,
which is easily revealed by Raney nickel reduction (Scheme 4,
left).14
(phenylthio)propan-1-ol 3b (Scheme 3c).38
It is known that 2-methyl-3-phenylpropan-1-ol 2g can be effi-
ciently resolved by PCL-catalysed kinetic acylation with vinyl
acetate.9,39,40 Therefore, it occurred to us that the heterocyclic
analogue 2-methyl-3-(2-thienyl)propan-1-ol 2a could behave in
a similar way. An efficient approach to both of the pure enantio-
mers of 2-methylalkanols would then be available, using either
of the pure enantiomers of 2-methyl-3-(2-thienyl)propan-1-ol
2a as the starting material (Scheme 4, right). Using racemic
Scheme 3 Preparation of substrates without thienyl or cycloalkyl
moieties. Reagents and conditions: i, Propanal, NaOH, EtOH, H2O; ii,
Pd, H2 then NaBH4; iii, Na, toluene, reflux; iv, 2-methylprop-2-enal,
Cupric acetate [catalyst]; v, NaBH4.
Kinetic resolutions by transesterifications catalysed by PCL
2-Methylalkan-1-ols. The kinetic resolution of 2-methyl-
alkan-1-ols with alkanol chain lengths from 5 to 14 by trans-
esterification with vinyl acetate in the presence of PCL gives
E-values1–4 (i.e., reaction rate ratios of the enantiomers in a
racemic starting material) of between 6 and 11 (see Table 1).23,24
We studied the possibility of increasing the E-values in this
reaction for some long-chain 2-methylalkanols using PCL
[PFL (Fluka), Amano PS or AK] by changing the solvent, by
immobilising the enzyme, by changing the temperature and
water activity. However, only small effects were registered.23
Because these low E-values were unsatisfactory for preparative
purposes, an alternative approach had to be used for the prep-
aration of such 2-methylalkan-1-ols of high enantiomeric
excesses (ees).
Scheme 4 Synthetic strategies for the preparation of the enantiomers
of 2-methyldecan-1-ol using thiophene as a masked -(CH2)4- group
either via baker’s yeast reduction (left) or via PCL-catalysed kinetic
resolution by transesterification of vinyl acetate (right).
2-methyl-3-(2-thienyl)propan-1-ol 2a as the substrate in a PFL-
catalysed transesterification of vinyl acetate in chloroform with
an initial water activity of aw = 0.32, we found, to our satis-
faction, that this reaction was highly enantioselective (E ≈ 200).
The same reaction has also been studied by others using PCL.15
They showed that both enantiomers can be used as build-
ing blocks using a similar Friedel–Crafts acylation–Huang-
Minlon–Raney nickel reduction sequence to that described in
Scheme 4, to prepare both enantiomers of the antimicrobial
and cytostatic alkaloid niphatesine C.41
Substrates of type 2 (see Chart 1). We have earlier shown
that baker’s yeast reduction of 2-methyldec-2-enal (Scheme 1,
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376
369

View Article Online
Table 2 PCL-catalysed kinetic resolution of 3-substituted 2-methylpropan-1-ols by transesterification with vinyl acetate
Substrate
Enzyme
Solvent
c/%
eep/%
ees/%
E
Ref.
2a
2a
2a
2b
2c
2d
2e
2f
PFL (Fluka)
Amano PS
Amano PS
Amano PS
Amano PS
Amano PS
Amano PS
Amano PS
PFL
Amano PS
Amano PS
Amano PS
Amano PS
PFL
CHCl3
CHCl3
TBME
CHCl3
TBME
TBME
TBME
CHCl3
CH2Cl2
CHCl3
TBME
TBME
EtOAcb
CHCl3
CHCl3
TBME
TBME
39.3
37.9
42.2
38.9
40.2
43.5
38.5
31.0
98.2
97.3
97.5
96.6
97.3
98.5
95.1
97.1
200
130
170
108
144
300
75
105
172
116
90
a, 9
a, 9
a, 9
a
a, 22
a, 22
a
a
2g
2g
2g
2h
2i
42
a, 9
a, 9
a
13
43
43
a
40.9
40.5
41.2
45
96.6
96.6
94.2
98
67
79
ӷ100c
150
45
2j
2k
2l
PFL
Amano PS
Amano PS
41.2
39.4
81.9
83.2
18
18
2m
a
a This work. b Anhydrous ethyl acetate (EtOAc) was used both as solvent and acyl donor. c E = 29 was reported by Bianchi et al.13 but from the eep, ees
and conversion given by the authors, E = 240 was calculated by using Sih’s and Rakels’ formulae.44,45
In order to establish how the nature of the substituent R in
compounds of type 2 (see Chart 1) affects the E-value a number
of these were prepared as described above and resolved accord-
ing to the standard procedure (see Experimental section). The
results are summarised in Table 2 and compared with some
earlier results13,43 obtained by other groups.
When comparing the E-values presented in Table 2 it was
evident that there were two categories of substrates. One group
was the 3-aryl-2-methylpropan-1-ols, e.g. 2a–h which were
resolved with high E-values (E ≈ 70–300). The E-values
obtained were roughly independent of whether the ring was
2-furyl, 3-thienyl or phenyl. They all gave E ≥ 100. An interest-
ing observation was that substitution in the 5-position of the
thienyl ring in the 2-methyl-3-(3-thienyl)propan-1-ols gave a
lower E-value for 2e compared with the unsubstituted 2b,
E = 75 and E = 108, respectively. On the other hand, the 3-(2-
thienyl)propanol 2a, which gave E = 170, on substitution in the
5-position showed the opposite effect on E. Thus, substrate 2d
gave E = 300.
The second category of substrates in Table 2 consisted of the
3-cycloalkyl-2-methylpropan-1-ols 2l and 2m which had cyclo-
alkane rings instead of aromatic rings in their structures. In
these cases, the E-values were much lower, E ≈ 18, but not as
low as those observed for 2-methylalkan-1-ols, E between 6 and
11 (Table 1). This difference between the substrates containing
cycloalkyl and aryl substituents indicated that the aromatic
systems interacted in a favourable way with, e.g., aromatic
moieties in the enzyme.
in the optimal position for achieving high E-values. Therefore a
series of new substrates were studied.
Substrates of type 1 (see Chart 1). Some substrates of type 1
have been studied before and generally give rather low E-values
in PCL-catalysed kinetic resolutions.42,43 However, some
sterically demanding substrates, e.g. naphthyl derivatives 1e
and 1f, give improved E-values.43 The new compounds studied
here also gave similar results (see Table 3). It is interesting to
note that for substrates of type 1, the 2-thienyl-substituted
compound 1a gave a better E-value than the corresponding
3-thienyl and phenyl derivatives 1b and 1d, i.e. similar to the
corresponding substrates of type 2. Increasing the bulk of the
substrate by using 5-butyl-2-thienyl compound 1c also gave a
higher E-value.
Substrates of type 3 (see Chart 1). Some derivatives of type
3 (Chart 1) with X = O and with very bulky R-groups [R =
TBDMS or ButPh2Si (TBDPS)] have been efficiently resolved
with vinyl acetate and PCL.46,47 Since oxygen directly linked to
the stereocentre alters the enantioselectivity of primary alco-
21,25
hols,
it might also have unpredictable effects when located
further away. Because a sulfur atom, on the other hand, is simi-
lar in size and polarity to a CH2 group we considered it worth-
while to study the phenylthio derivative of 3 (X = S, R = Ph), i.e.
compound 3b (Scheme 3d). Resolution of this produced the
acetate in 67.3% ee at 34.0% conversion, corresponding to an
E-value of 3.4. To our knowledge the case with X = CH2 and
R = aryl has not been studied previously. Therefore, we pre-
pared 2-methyl-4-(2-thienyl)butan-1-ol (3a, Scheme 2b) and
subjected it to transesterification under the standard condi-
Thus, for substrates of type 2 containing aromatic rings, the
E-values were good to excellent in most cases. However, this
does not mean that the aromatic ring in the substrate is located
370
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376

View Article Online
Table 3 PCL-catalysed kinetic resolution of 2-substituted 2-methylethanols by transesterification with vinyl acetate
Substrate
Enzyme
Solvent
c/%
eep/%
ees/%
E
Ref.
1a
1b
1c
1d
1e
1f
Amano PS
Amano PS
Amano PS
PFL
PFL
PFL
CHCl3
CHCl3
CHCl3
CH2Cl2
CHCl3
CHCl3
39.9
39.9
31.0
12.6
68.8
20.5
6.2
42.2
2.3
1.2
8.1
1.8
9.0
6.0
a, 22
a
a, 22
42
43
43
a This work.
tions. At 40% conversion the acetate (S)-3aAc was produced
(eep = 76.6%) and the remaining alcohol (R)-3a (ees = 45.6%)
which gave E = 12. Thus, the E-values obtained with substrates
of type 3 (Chart 1) were clearly inferior to those of type 2.
Therefore, studies with these substrates were not pursued
further.
alcohols 6a–d were obtained from the racemic alcohols via
kinetic resolution by transesterification24,49 and the alcohol 6e
was obtained after hydrolysis of the corresponding racemic
chloroacetate.50 The E-values for the two 2-isopropyl substi-
tuted alcohols 6c49 and 6e5 are 29 and 11, respectively.
2-Isopropyl-5-methylhex-5-en-1-ol 6f could serve as a build-
ing block for the total synthesis of germacra-1,6-dien-5-ol, a
major constituent of the defence secretion of the pine sawfly
Neodiprion sertifer and of the needle resin of Scots pine.51
Therefore, we subjected the racemic alcohol 6f to PCL-
catalysed transesterification with vinyl acetate in TBME, as
shown in Scheme 5. The E-value was found to be E = 14 for
Substrates of type 4 (see Chart 1). Only a very limited number
of compounds of this type with R = Aryl has been studied.
Both 2-methyl-4-phenylthiobutan-1-ol as well as the corre-
sponding phenylseleno derivative,
4 (R = Ph) and X = S
and X = Se, respectively, have been successfully resolved.48 No
E-values were given but have been estimated to be around 20.25
We prepared 4a (R = 2-thienyl, X = CH2, Scheme 2d) and
resolved it with PCL and vinyl acetate in tert-butyl methyl ether
(TBME) as solvent. At 42% conversion, the acetate (S)-4aAc
(eep = 42.8%) and the remaining alcohol (R)-4a, (ees = 27.7%)
were obtained, corresponding to E = 3.2.
Substrates of type 5 (see Chart 1). No example of this case
with R = aryl is, to our knowledge, known. We prepared
(Scheme 2b) and kinetically resolved compound 5a using PCL
and vinyl acetate. The acetate produced at 43.5% conversion
was of 45.9% ee, corresponding to an E-value of 3.8.
Substrates of type 6 (see Chart 1). Some examples have
been reported of Pseudomonas lipase-catalysed resolution of
primary 2-alkyl-substituted alcohols having larger substituents
than methyl in the 2-position. The fast reacting enantiomer of
these alcohols are shown in Chart 2. Using lipase catalysis, the
Scheme 5 PCL-catalysed transesterification of the 2-isopropyl alcohol
6f with vinyl acetate as acyl donor and TBME as solvent. Enantiomeric
ratio E = 14. The R configuration of the fast reacting enantiomer (ϩ)-6f
was established after conversion to 7.
this reaction and at 40% conversion the acetate (R)-6fAc
(eep = 80%) was obtained and at 60% conversion the alcohol
(S)-6f (ees = 95%) remained, corresponding to E =14. The R
configuration of the fast reacting enantiomer (ϩ)-6f was
established after conversion to (R)-(ϩ)-tetrahydrolavandulol 7
with known configuration52 (Scheme 5). The preparation of
the racemic substrate 6f will be published elsewhere.
Chart 2 Some 2-alkylalkan-1-ols with larger substituents than methyl
that have been prepared by Pseudomonas lipase-catalysed kinetic reso-
lution. The fast reacting enantiomer in each case is shown.
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376
371

View Article Online
Conclusions
hot bath (Büchi). Bps were registered as air bath temperatures
1
in Büchi Kugelrohr apparatus. H and 13C NMR spectra were
When comparing racemic substrates containing aromatic
rings from the groups 1–5, it was evident that the location of
this ring in a substrate was crucial for achieving a successful
resolution. Indeed, only substrates of type 2 were more effi-
ciently resolved than straight-chain 2-methylalkan-1-ols.
Thus, 2-methyldecan-1-ol gave E ≈ 11 whereas the series of
2-methyl-(5-substituted 2-thienyl)alkan-1-ols with the same
number of carbon atoms, i.e. compounds 1c, 2d and 5a, gave
E ≈ 8, 300 and 4, respectively. The same trend was observed
in the series of 2-methyl-ω-(2-thienyl)alkan-1-ols 1a, 2a, 3a,
4a and 5a (E ≈ 2, 200, 12, 3 and 4). For both substrates of
type 1 and 2 the enantioselectivities varied roughly in the
same direction for the substituents 3-thienyl < phenyl < 2-
thienyl. Thus both the nature of the aromatic ring and,
especially, its position relative to the stereogenic centre are
important for achieving high E-values.
The alternative large pocket in PCL (see Fig. 1, C) has been
proposed to bind the benzyl group of the primary alcohol 2g.21
This pocket is lined with the hydrophobic amino acid residues
Tyr-23, Leu-27, Tyr-29, Phe-146, Ile-290 and Leu-293.21 It is
interesting to note that both tyrosine and phenylalanine are
aromatic amino acids. A tentative explanation for our results is
that the aromatic amino acids in this pocket most likely are
loosely bound via π–π interactions to suitably located aryl
groups in a substrate alcohol, e.g. phenyl or thienyl in 2g or 2a,
respectively. These types of π–π interactions may give addi-
tional stability to ‘normal’ hydrophobic interactions present.
Such π–π interactions are absent for the 3-cycloalkyl-
2-methylpropan-1-ols 2l and 2m and, of course, also for the
2-methylalkanols, hence should lead to lowered E-values. The
observed values were E ≈ 20 and E ≤ 11, respectively. Similarly,
moving the aromatic ring away from the π–π interaction sites of
the protein could explain the low E-values observed for the
substrate groups 1, 3, 4 and 5.
recorded on a JEOL EX 270 spectrometer or a Bruker DMX
250 spectrometer. Optical rotation was measured on a Perkin-
Elmer 241 polarimeter. Mass spectra were recorded using GC-
MS (Varian 3300 GC and an ion-trap detector, Finnigan ITD
800 or Varian 3800 GC and an ion-trap detector, Varian Saturn
2000). IR spectra were recorded on a Perkin-Elmer 1600 FT-IR
spectrometer. Exact masses (HRMS) were obtained using a
VG-70E mass spectrometer. Refractive indices were measured
on a Pleuger refractometer.
Transesterification reactions at initial aw ؍ 0.32
The procedure described earlier23 was slightly modified.
Amano PS (10.0 mg) and a racemic alcohol 1–6 (1.0 mmol)
were mixed with solvent (1.8 ml) in the reaction flask containing
a magnetic stirring bar. The mixture in the open flask was
stirred for 24 h in a sealed container over saturated MgCl2
(aw = 0.3253). After 24 h pre-equilibrated solvent was added to
the reaction flask to compensate for loss by evaporation. Vinyl
acetate (0.34 ml, 3.7 mmol) was added to the reaction flask,
which was immediately sealed with a septum. The mixture was
stirred at 400 rpm at rt. The conversion was followed by GC.
When the reaction had reached the desired conversion (≈ 40%)
the mixture was filtered through a filter of low porosity and the
solid collected was washed with n-pentane (3 ml). The com-
ponents were separated by MPLC.
Reduction of the enantiomerically enriched acetates
The chemically pure, enantiomerically enriched esters (R)-
1aAc, (S)-1bAc, (R)-1cAc, (S)-2a–hAc, (S)-2l,mAc, (S)-3aAc,
(R)-3bAc, (S)-4aAc, (S)-5aAc and (R)-6fAc were obtained as
described above (0.3–0.4 mmol). One of them was dissolved in
anhydrous diethyl ether (1 ml) and added to a stirred suspen-
sion of LiAlH4 (30 mg) in anhydrous diethyl ether (3 ml) under
argon. The mixture was stirred at rt for 1 h and then quenched
with water–THF (1:1; 60 µl) followed by 15% NaOH (30 µl)
and water (20 µl). After refluxing for 10 min the mixture was
filtered, washed with diethyl ether and dried (MgSO4). After
evaporation of the solvent either pure (R)-1a, (S)-1b, (R)-1c,
(S)-2a–h, (S)-2l,m, (S)-3a, (R)-3b, (S)-4a, (S)-5a or (R)-6f was
obtained in quantitative yield. For optical rotation values for
these compounds, see Table 4.
We have demonstrated that among substrates carrying
2-thienyl substituents in the series 1–5, only those of type 2,
e.g. 2-methyl-3-(2-thienyl)propan-1-ols 2a, 2c, and 2d, were
resolved with very high enantioselectivity. Thus, among the
substrates investigated, only the latter substrates can serve as
useful starting materials for Raney nickel reduction to enantio-
merically pure, long-chain 2-methyl alcohols.
Experimental
Determination of conversion
Pseudomonas cepacia was previously named Pseudomonas
fluorescens. Thus, in earlier reports PCL is called PFL. The
old abbreviation PFL is still used by some authors. P. cepacia
has been reclassified again and is now called Burkholderia
cepacia.5,21 Different abbreviations and trade names have been
used by different authors and manufacturers for PCL: Lipase P,
Lipase PS, Amano P, Amano PS, Amano P-30, PFL, etc.5 In
this paper we use the old name PCL for this lipase and, if
necessary, Amano PS and PFL (Fluka) are used to specify the
reaction conditions. Amano PS (PCL) from Pseudomonas
cepacia was obtained from Amano Pharmaceutical Company.
The specific activity was 30.0 U mgϪ1 and the enzyme was
stored at 4 ЊC over dried silica. Unless otherwise stated, starting
materials and solvents were used as received from commercial
suppliers. Dry THF was distilled from a mixture of potassium
and benzophenone and dry diethyl ether was distilled from
LiAlH4 before use. Dry diisopropylamine was distilled from
CaH2. Air-sensitive reagents were handled with gas-tight
syringes and the reactions were performed under argon.
Preparative liquid chromatography (medium-performance
liquid chromatography, MPLC) was performed on straight-
phase silica gel (Merck 60, 230–400 mesh) employing a gradient
of an increasing concentration of diethyl ether in n-pentane as
eluent. Mps were measured in glass capillaries in a Tottoli type
The conversions in the transesterification reactions were deter-
mined by capillary GC. The conversions were calculated from
the area of the ester peak relative to that of the alcohol peak
after calibration against the racemic ester.
Determination of enantiomeric excess
The ees of the alcohols 1–6, except 2f and 3b, were determined
by analysing the diastereomeric mixture of their correspond-
ing 2-alkylacyl-1-phenylethylamides.54 Baseline separation was
readily obtained using a Carbowax-coated capillary column
and He as carrier gas. The ees of the alcohols 2f and 3b were
determined by analysing both the 1H and 19F NMR spectra12,14
of their corresponding ‘α-methoxy-α-(trifluoromethyl)phenyl-
acetate’ (MTPA) esters.
2-(2-Thienyl)propan-1-ol 1a. 2-(2-Thienyl)propanoic acid37
was reduced using general procedure A (see below). MPLC and
distillation (82 ЊC/0.02 mbar†) gave the title compound (95%
yield, >99% by GC), nD20 1.5903 (HRMS: Calc. for C7H10OS: M,
142.045. Found: Mϩ, 142.043); νmax (neat, NaCl) 3371, 3355,
3105, 2930, 2360, 1382, 1077, 824, 695, 432 cmϪ1; δH (270 MHz;
† 1 bar = 105 Pa.
372
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376

View Article Online
Table 4 Optical rotations of the fast reacting alcohol enantiomers
evaporated in vacuo. Bulb-to-bulb distillation (132–135 ЊC/0.65
mbar) gave the title compound (0.39 g, 86% yield, >99% pure
by GC). The spectroscopic data were in good agreement with
those reported for (S)-2-methyl-3-(2-thienyl)propan-1-ol (S)-
2a.14
Alcohol
eep/%
[α]D25 a
[α]D (literature)
(R)-1a
(S)-1b
(R)-1c
(S)-2a
(S)-2b
(S)-2d
(S)-2f
(S)-2g
(S)-2h
(S)-2i
(S)-3a
(R)-3b
(S)-4a
(S)-5a
(R)-6f
(S)-6fg
31.0
12.6
68.8
98
Ϫ6.2 c 19b
Ϫ2.0 c 11b
Ϫ11.4 c 57b
Ϫ19.0 neat
Ϫ16.0 c 0.8c
Ϫ17.3 c 1.9c
Ϫ15.4 c 0.7b
Ϫ10.1 c 0.8b
Ϫ10.8 c 1.1c
Ϫ17.6 c 0.7c
Ϫ16.4 c 36b
Ϫ19.9 c 1.4e
Ϫ7.3 c 1.1c
Ϫ6.4 c 1.3c
ϩ8.3 c 1.0c
Ϫ14.3 c 1.2c
2-Methyl-3-(3-thienyl)propan-1-ol 2b. The title compound
was prepared by general procedure A (see above) from
2-methyl-3-(3-thienyl)propanoic acid obtained by general pro-
cedure B (see below). The spectroscopic data were in good
agreement with those reported for (S)-2-methyl-3-(3-thienyl)-
propan-1-ol (S)-2b.14
Ϫ19.3 neat14
Ϫ14.9 c 1.8c,d
Ϫ17.4 c 1.5c,14
96.6
98.5
81.9
83.2
94.2
97.1
76.6
67.3
42.8
45.9
79.9
95
Ϫ18.5 c 8.0c,12
2-Methyl-3-(5-propyl-2-thienyl)propan-1-ol 2d. This was
prepared by the method described for (S)-2d14 but from
2-methyl-3-(2-thienyl)propan-1-ol 2a via SnCl4-catalysed
propionylation followed by Huang-Minlon reduction. The
spectroscopic data was in good agreement with literature data.14
Ϫ18.9 c 3.9e,65
ϩ24.6 neatf,60
a The optical rotation was measured on the alcohols obtained after
reduction of the produced esters. b Chloroform as solvent. c Methanol
as solvent. d The optical rotation reported previously by us14 was cor-
rected. e Dichloromethane as solvent. f The literature value60 is for the
R-enantiomer of 4a. g Remaining substrate (slow reacting enantiomer).
3-(5-Ethyl-3-thienyl)-2-methylpropan-1-ol 2e. This was
obtained from the batch prepared by Karlsson and
Hedenström.34
3-(2-Furyl)-2-methylpropan-1-ol 2f. The procedure described
for racemisation of primary 2-methylalkanols was used.24 Thus,
(S)-2-methyl-3-(2-furyl)propan-1-ol12 (2.5 g, 17.9 mmol), [α]D25
Ϫ12.1‡ (c 1.35, CHCl3), and Na (25 mg, 1.1 mmol) were added
to a solution of toluene (50 ml) and benzophenone (0.18 g,
1.0 mmol) in a round-bottom flask. The reaction mixture
was refluxed for 24 h and then water (50 ml) was added. The
mixture was extracted with diethyl ether (3 × 50 ml). The com-
bined extracts were washed with water (100 ml) and dried with
MgSO4. The solvent was evaporated and the residue was sub-
jected to MPLC (120 g silica), giving the title compound (1.98
g, 79% yield, 99.8% by GC), [α]D25 Ϫ0.3 (c 1.5, CHCl3, corre-
sponds to 2% S). The spectroscopic data were identical with
those reported for (S)-2f.12
CDCl3) 1.38 (3H, d, J = 7 Hz), 1.68 (1H, s), 3.22 (1H, m),
3.68 (2H, m), 6.85–7.00 (2H, m), 7.18 (1H, m); δC (67.8
MHz; CDCl3) 18.60, 38.11, 68.93, 123.48, 123.86, 126.79,
147.42.
2-(3-Thienyl)propan-1-ol 1b. 2-(3-Thienyl)propanoic acid55
was reduced using general procedure A (see below). Distillation
(77 ЊC/0.02 mbar) gave the alcohol 1b (99.6% purity by GC),
nD20 1.5405 (HRMS: Calc. for C7H10O: M, 142.045. Found: Mϩ,
142.045); νmax (neat, NaCl) 3357, 3348, 2962, 2928, 2875, 2360,
2342, 1377, 688, 656 cmϪ1; m/z (relative intensity) 142 (Mϩ,
24%), 111 (100), 97 (6), 85 (6), 77 (27), 67 (16), 59 (13), 51 (5);
δH (270 MHz; CDCl3) 1.29 (3H, d, J = 7 Hz), 1.42 (1H, s), 3.07
(1H, m), 3.68 (2H, m), 7.03 (2H, m), 7.30 (1H, m); δC (67.8
MHz; CDCl3) 17.47, 37.94, 68.32, 120.45, 125.89, 126.76,
144.55.
3-(o-Methoxyphenyl)-2-methylpropan-1-ol 2h. Prepared by
KBH4 reduction of 3-(o-methoxyphenyl)-2-methylpropion-
aldehyde57 as described for 3b below. Bulb-to-bulb distillation
(108–110 ЊC/0.06 mbar) gave the alcohol at 80% yield and 97%
pure by GC, nD20 1.604 (HRMS: Calc. for C11H16O2: M, 180.115.
Found: Mϩ, 180.114); m/z (relative intensity) 180 (Mϩ, 32%),
162 (12), 147 (15), 121 (95), 93 (50), 91 (100), 77 (20); νmax (neat,
NaCl) 2950, 2870, 1707, 1459, 1417, 1380, 1293, 1244, 1180,
1096 cmϪ1; δH (270 MHz; CDCl3) 0.95 (3H, d, J = 7 Hz), 1.95
(1H, m), 2.05 (1H, br s), 2.54 (1H, dd, J = 14 and 7 Hz), 2.71
(1H, dd, J = 14 and 7 Hz), 3.39 (2H, apparent t, J = 5 Hz), 3.83
(3H, s), 6.85–6.92 (2H, m), 7.10–7.22 (2H, m) ppm; δC (67.8
MHz; CDCl3) 16.95, 33.14, 36.73, 55.45, 66.94, 110.44, 120.65,
127.28, 128.75, 131.19, 157.46.
2-(5-Butyl-2-thienyl)propan-1-ol 1c. Following the method
described for (S)-2d14 but starting with 2-(2-thienyl)propan-1-
ol 1a and butyryl chloride, SnCl4-catalysed acylation followed
by Huang-Minlon reduction, MPLC and distillation (95 ЊC/
0.02 mbar) gave the title compound (54% overall yield, 99%
pure by GC), nD20 1.5159 (HRMS: Calc. for C11H18OS: M,
198.108. Found: Mϩ, 198.111); νmax (neat, NaCl) 3363, 2872,
2853, 2360, 1465, 1451, 1379, 1042, 1030, 798 cmϪ1; m/z
(relative intensity) 198 (Mϩ, 9%), 167 (100), 155 (2), 137 (2), 125
(18), 111 (21), 105 (3), 91 (11), 79 (5), 77 (5), 59 (5); δH (270
MHz; CDCl3) 0.91 (3H, t, J = 7 Hz), 1.31 (3H, d, J = 7 Hz), 1.38
(2H, m), 1.61 (1H, s), 1.63 (2H, m), 2.76 (2H, t, J = 8 Hz), 3.15
(1H, m), 3.65 (2H, m), 6.61 (1H, d, J = 3 Hz), 6.67 (1H, d, J = 3
Hz); δC (67.8 MHz; CDCl3) 13.79, 18.42, 22.21, 29.81, 33.73,
38.30, 68.88, 123.45 (intense), 144.10, 144.51.
General procedure B
Representative procedure for preparation of 2-methylalkanoic
acids.
General procedure A
3-Cyclopentyl-2-methylpropanoic acid. A slightly modified
version of the method described for 2-methyldodecanoic
acid.58 Sodium (0.54 g 23.4 mmol) was dissolved in absolute
ethanol (18 ml) under an atmosphere of argon. When all the
sodium had dissolved, diethyl methylmalonate (4.2 ml, 24.8
mmol) was added and the solution was refluxed for 5 min.
After cooling of the mixture to room temperature, (bromo-
methyl)cyclopentane (3.46 g, 21.2 mmol) was added followed
by reflux for 5 h. In order to neutralise the excess of base a few
drops of glacial acetic acid were added at room temperature.
Representative procedure for the reduction of 2-methylalkanoic
acids.
2-Methyl-3-(2-thienyl)propan-1-ol 2a. 2-Methyl-3-(2-thienyl)-
propanoic acid56 (0.5 g, 2.9 mmol) dissolved in dry diethyl ether
(10 ml) was added dropwise, under argon, to a stirred suspen-
sion of LiAlH4 (0.3 g, 8 mmol) in dry diethyl ether (20 ml) at rt.
After stirring of the mixture for 2 h at rt, water (0.3 ml) was
added dropwise, followed by 15% NaOH (0.3 ml) and water
(0.2 ml). The mixture was refluxed for 20 min and cooled to rt.
After filtration, drying (MgSO4), and filtration, the solvent was
‡ Specific optical rotations are given in units of 10Ϫ1 deg cm2 gϪ1
.
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376
373

View Article Online
The solvent was evaporated off and the residue was dissolved
in water (100 ml) and extracted with diethyl ether (3 × 50 ml).
The combined organic phases were washed successively with
10% Na2CO3 (25 ml) and brine (50 ml), and dried (MgSO4).
After solvent evaporation the crude diester was added to a
solution of KOH in 95% ethanol (2 M; 50 ml) and refluxed for
4 h (or longer if needed). After cooling to room temperature,
the solvent was evaporated off and the residue was dissolved
in water (50 ml) and washed with diethyl ether (2 × 25 ml).
Acidification with conc. hydrochloric acid at 0 ЊC and extrac-
tion with diethyl ether (3 × 50 ml) was followed by successive
washings of the combined ether phases with water (25 ml) and
brine (15 ml). After drying over MgSO4 the mixture was
evaporated to yield the crude diacid (4.42 g, 22.1 mmol), which
was decarboxylated neat in an open flask at 190 ЊC for 3 h.
After cooling, the product was dissolved in 10% aq. Na2CO3
(100 ml) and washed with diethyl ether (30 ml) followed by
acidification at 0 ЊC with conc. hydrochloric acid. Extraction
with diethyl ether (3 × 50 ml) and washing with brine (30 ml)
followed by drying over MgSO4, filtration and concentration in
vacuo gave the crude acid (2.23 g, 14.3 mmol). After bulb-to-
bulb distillation (135 ЊC/0.65 mbar), the title compound was
obtained (2.0 g, 60% yield, >99% purity by GC), nD20 1.603
(HRMS: Calc. for C9H16O2: M, 156.115. Found: Mϩ, 156.116);
m/z (relative intensity) 157 (Mϩ ϩ 1, 53%), 139 (91), 137 (28),
111 (45), 109 (22), 83 (100), 74 (55) 69 (44), 67 (51), 55 (65);
δH (270 MHz; CDCl3) 1.08 (2H, m), 1.17 (3H, d, J = 7 Hz),
1.34–1.63 (5H, m), 1.71–1.86 (4H, m), 2.49 (1H, m), 11.7 (1H,
br s); δC (67.8 MHz; CDCl3) 17.27, 25.03, 25.09, 32.53, 32.69,
37.92, 38.74, 40.00, 183.70.
2-Methyl-4-(2-thienyl)butan-1-ol 3a. Reduction of 2-methyl-
4-(2-thienyl)butanoic acid59 according to general procedure A
(see above), followed by distillation (115 ЊC/0.02 mbar), gave
the title alcohol (81% yield, 99% purity by GC), nD20 1.5272;
νmax (neat, NaCl) 3355, 3106, 2926, 2360, 1460, 1440, 1034,
848, 821, 692 cmϪ1; m/z (relative intensity) 170 (Mϩ, 15%),
152 (8), 137 (20), 123 (17), 110 (71), 97 (100), 85 (5), 73 (5),
59 (16), 53 (24); δH (270 MHz) 0.98 (3H, d, J = 7 Hz), 1.44
(1H, s), 1.52 (1H, m), 1.67–1.90 (2H, m), 2.89 (2H, m), 3.50
(2H, m), 6.79 (1H, m), 6.91 (1H, m), 7.10 (1H, m); δC (67.8
MHz) 16.35, 27.39, 35.15 (intense), 68.00, 122.84, 123.95,
126.67, 145.43.
2-Methyl-3-(phenylthio)propanal.38 Methacrolein (21 ml, 0.25
mol) was stirred with Cu(OAc)2 (50 mg) in a round-bottom flask
on an ice-bath. Thiophenol (20 ml, 0.19 mol) was added drop-
wise for 30 min. The mixture was heated to 40 ЊC for 2 h. A
distillation set-up was connected to the reaction flask and the
product was distilled (110 ЊC/2 mmHg) from the reaction mix-
ture to give the title compound in 92% yield. Spectroscopic data
were identical with those in the literature.38
2-Methyl-3-(phenylthio)propan-1-ol 3b.38
A mixture of
NaBH4 (5.5 g) and 2M NaOH (5.5 ml) in water (50 ml) was
added dropwise to a solution of 2-methyl-3-(phenylthio)-
propanal (31 g, 0.18 mol) in methanol (100 ml). After the addi-
tion was complete (1.5 h) the reaction mixture was poured into
water and extracted with diethyl ether twice and the combined
extracts were washed successively with saturated aq. NaHCO3
and brine, and dried with MgSO4. After solvent evaporation
and distillation (115–116 ЊC/1 mmHg) the title alcohol was
obtained (23 g, 84% yield, 99% by GC). The spectroscopic data
were identical with those in the literature.38
3-Cyclopentyl-2-methylpropan-1-ol 2l. Using general pro-
cedure A (see above), 3-cyclopentyl-2-methylpropanoic acid
(0.8 g, 5.1 mmol) gave the alcohol 2l (0.45 g, 63% yield, >99%
purity by GC) after bulb-to-bulb distillation (80 ЊC/0.35 mbar),
nD20 1.603 [HRMS: (Mϩ Ϫ H2O)§ Calc. for C9H16: m/z 124.125.
Found: m/z 124.125]; m/z (relative intensity) 141 (Mϩ Ϫ 1,
0.1%), 123 (2), 109 (7), 95 (21), 82 (56), 69 (100), 67 (97); νmax
(neat, NaCl) 3341, 2950, 2868, 2357, 1447, 1379, 1034, 984
cmϪ1; δH (270 MHz; CDCl3) 0.91 (3H, d, J = 7 Hz), 0.99–1.19
(3H, m), 1.29–1.39 (2H, m), 1.46–1.88 (8H, m), 3.39–3.50 (2H,
m); δC (67.8 MHz; CDCl3) 16.80, 25.11, 32.56, 33.37, 34.86,
37.50, 39.75, 68.68.
2-Methyl-5-(2-thienyl)pentan-1-ol 4a.60 Using
a slightly
modified version of the method described for the preparation
of 2-methyldodecanoic acid,61 2-methyl-5-(2-thienyl)pentanoic
acid59 was prepared. n-Butyllithium in hexane (1.6 M; 20.0
ml, 32.0 mmol) was added slowly (60 min) to a solution of
dry diisopropylamine (4.5 ml, 32 mmol) in dry THF (25 ml)
at Ϫ5 ЊC. After stirring of the mixture for 30 min at 0 ЊC, 5-
(2-thienyl)pentanoic acid (3.0 g, 16 mmol) in dry THF (14 ml)
was added slowly during 50 min at 0 ЊC. After stirring of the
mixture for 30 min at 0 ЊC, dry DMPU (1.8 ml, 15 mmol) was
added and the solution was stirred at rt for 1 h. After cooling
(Ϫ5 ЊC) of the mixture, methyl toluene-p-sulfonate (3.0 g, 16
mmol) was added and the mixture was stirred at rt overnight.
A mixture of the starting acid and the desired product
(30:70) was obtained. After acidification with 10% HCl,
extraction with diethyl ether (4 × 40 ml), drying (MgSO4), and
evaporation off of the solvent the resulting mixture of the
two acids was reduced according to general procedure A (see
above). The produced alcohols 4a and 5-(2-thienyl)pentan-1-
ol were separated with MPLC using 113 g of silica gel. Bulb-
to-bulb distillation (130 ЊC/0.74 mbar) gave the alcohol 4a
(1.3 g, 6.6 mmol, 41% overall yield, >99% pure according to
GC), nD20 1.5202 [lit.,60 nD16.5 1.5245 for the R-(ϩ) enantiomer];
νmax (neat, KBr) 3346, 2932, 2872, 1460, 1440, 1380, 1238,
1038, 850, 820 cmϪ1; m/z (relative intensity) 184 (30%), 166
(21), 156 (14), 137 (12), 123 (24), 111 (17), 110 (73), 98 (34),
97 (100), 53 (16); δH (250 MHz; CDCl3) 0.93 (3H, d, J = 6.7
Hz), 1.10–1.25 (1H, m), 1.42–1.57 (1H, m), 1.45 (1H, s), 1.58–
1.82 (3H, m), 2.83 (2H, t, J = 7.6 Hz), 3.43 (1H, dd, J = 10.5
and 6.3 Hz) 3.49 (1H, dd, J = 10.5 and 5.8 Hz), 6.78 (1H, dd,
J = 3.4 and 1.0 Hz), 6.92 (1H, dd, J = 5.1 and 3.4 Hz), 7.11
(1H, dd, J = 5.1 and 1.2 Hz); δC (62.9 MHz; CDCl3) 16.51,
29.23, 30.16, 32.54, 35.57, 68.20, 122.84, 124.03, 126.67,
145.49.
3-Cyclohexyl-2-methylpropan-1-ol 2m. Using general pro-
cedure B (see above) the corresponding acid was prepared from
(bromomethyl)cyclohexane (46.3 mmol) and diethyl methyl-
malonate (54 mmol). The acid was reduced using general pro-
cedure A (see above) and, after bulb-to-bulb distillation (90 ЊC/
0.02 mbar), the title compound was obtained at >99% purity,
nD20 1.466 [HRMS: (Mϩ Ϫ H2O)59 Calc. for C10H18: m/z, 138.141.
Found: m/z, 138.140]; m/z (relative intensity) 155 (Mϩ Ϫ 1, 9%),
137 (21), 123 (9), 109 (14), 96 (53), 95 (40), 83 (92), 81 (100), 69
(21), 67 (41); δH (270 MHz; CDCl3) 0.77–1.01 (3H, m), 0.88
(3H, d, J = 7 Hz), 1.10–1.35 (6H, m), 1.65–1.75 (6H, m), 3.41
(2H, m); δC (67.8 MHz; CDCl3) 16.84, 26.33, 26.44, 26.69,
32.60, 33.03, 34.27, 34.81, 41.10, 68.77.
2-Methyl-4-(2-thienyl)butanoic acid.59 This was prepared
from 2-(2-bromoethyl)thiophene according to general pro-
cedure B (see above). Distillation (140–143 ЊC/0.15 mbar) gave
the title acid, nD20 1.5189; νmax (neat, NaCl) 2975, 2936, 1705,
1465, 1441, 1339, 941, 849, 823, 695 cmϪ1; δH (270 MHz) 1.23
(3H, d, J = 7 Hz), 1.80 (1H, m), 2.12 (1H, m), 2.54 (1H, m), 2.90
(2H, t, J = 8 Hz), 6.80 (1H, d, J = 3 Hz), 6.91 (1H, m), 7.11 (1H,
m), 11.35 (1H, br s); δC (67.8 MHz) 16.80, 27.42, 35.20, 38.55,
123.16, 124.44, 126.74, 144.08, 182.84.
§ Note: We were unable to distinguish the Mϩ peak from the back-
ground so the detected peak is Mϩ Ϫ H2O.
374
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376

View Article Online
2-(4-Bromobutyl)thiophene.62,63 4-(2-Thienyl)butyl toluene-p-
sulfonate64 (16.3 g, 52.6 mmol) dissolved in dry acetone (66 ml)
was added to a suspension of anhydrous LiBr (20.63 g, 237
mmol) in dry acetone (165 ml) and heated to reflux for 36 h.
The mixture was cooled to rt followed by addition of water (300
ml) and extraction with diethyl ether (3 × 400 ml). The organic
phase was washed successively with water (300 ml) and brine
(300 ml), followed by drying (MgSO4) and solvent evaporation.
MPLC (114 g of silica gel) using an ethyl acetate–cyclohexane
gradient as eluent followed by bulb-to-bulb distillation (85–
87 ЊC/0.15 mbar) gave the title compound (10.3 g, 89%). Spec-
troscopic data were in good agreement with those in the
literature.62,63
References
1 K. Faber, in Biotransformation in Organic Chemistry, Springer-
Verlag, Berlin, 3rd edn., 1997.
2 S. M. Roberts, in Preparative Biotransformation, Wiley, New York,
1992.
3 Enzyme Catalysis in Organic Synthesis, ed. K. Drauz and H.
Waldmann, VCH, Weinheim, 1997, vol. 1 and 2.
4 L. Poppe and L. Novak, in Selective Biocatalysis, VCH, Weinheim,
1992.
5 R. J. Kazlauskas and U. T. Bornscheuer, in Biotechnology, ed. D. R.
Kelly, Wiley-VCH, Weinheim, 2nd edn., 1998, vol. 8a, Biotrans-
formations, pp. 37–191.
6 P. Berglund and K. Hult, in Stereoselective Biocatalysis, ed. R. Patel,
Marcel Dekker, New York, 1999, pp. 633–657.
7 F. Thiel, Chem. Rev., 1995, 95, 2203.
8 K. Mori, in The Total Synthesis of Natural Products, ed. K.
ApSimon, Wiley, New York, 1992, vol. 9.
9 H.-E. Högberg, P. Berglund, H. Edlund, J. Fägerhag, E.
Hedenström, M. Lundh, O. Nordin, S. Servi and C. Vörde, Catal.
Today, 1994, 22, 591.
10 H.-E. Högberg, in Houben-Weyl Methods of Organic Chemistry,
Stereoselective Synthesis, ed. G. Helmchen et al., Thieme, Stuttgart,
1995, vol. E21a, pp. 791–915.
11 A. G. Myers, B. H. Yang, H. Chen, L. McKintsry, D. J. Kopecky
and J. L. Gleason, J. Am. Chem. Soc., 1997, 119, 6496.
12 C. Fuganti, P. Grasselli, S. Servi and H.-E. Högberg, J. Chem. Soc.,
Perkin Trans. 1, 1988, 3061.
13 D. Bianchi, P. Cesti, P. Golini, S. Spezia, L. Filippini, C. Garavaglia
and L. Mirenna, J. Agric. Food Chem., 1992, 40, 1989.
14 H.-E. Högberg, E. Hedenström, J. Fägerhag and S. Servi, J. Org.
Chem., 1992, 57, 2052.
15 F. Bracher and T. Papke, Tetrahedron: Asymmetry, 1994, 5, 1653.
16 V. Sunjic, M. Majeric and Z. Hamersak, Croat. Chem. Acta, 1996,
69, 643.
17 M. Majeric, A. Avdagic, Z. Hamersak and V. Sunjic, Biotechnol.
Lett., 1995, 17, 1189.
18 G.-Q. Lin and W.-C. Xu, Bioorg. Med. Chem., 1996, 4, 375.
19 G. D. Gamalevich, A. V. Ignatenko, E. P. Serebryakov and N. E.
Voishvillo, Russ. Chem. Bull., 1995, 44, 743.
20 Y.-F. Wang, C.-S. Chen, G. Girdaukas and C. J. Sih, J. Am. Chem.
Soc., 1984, 106, 3695.
21 W. V. Tuomi and R. J. Kazlauskas, J. Org. Chem., 1999, 64, 2638.
22 B.-V. Nguyen, O. Nordin, C. Vörde, E. Hedenström and H.-E.
Högberg, Tetrahedron: Asymmetry, 1997, 8, 983.
23 O. Nordin, E. Hedenström and H.-E. Högberg, Tetrahedron:
Asymmetry, 1994, 5, 785.
24 S. Barth and F. Effenberger, Tetrahedron: Asymmetry, 1993, 4, 823.
25 A. N. E. Weissfloch and R. J. Kazlauskas, J. Org. Chem., 1995, 60,
6959.
26 R. J. Kazlauskas, A. N. E. Weissfloch, A. T. Rappaport and L. A.
Cuccia, J. Org. Chem., 1991, 56, 2656.
27 X. Grabuleda, C. Jaime and A. Guerrero, Tetrahedron: Asymmetry,
1997, 8, 3675.
28 K. Lemke, M. Lemke and F. Thiel, J. Org. Chem., 1997, 62, 6268.
29 E. Santaniello, P. Ferraboschi, P. Grisenti and A. Manzocci, in
Biocatalysis in Non-Conventional Media, ed. J. Tramper et al.,
Elsevier, Amsterdam, 1992, pp. 533–541.
30 K. K. Kim, H. K. Song, D. H. Shin, K. Y. Hwang and S. W. Suh,
Structure, 1997, 5, 173.
2-Methyl-6-(2-thienyl)hexanoic acid. This was prepared
from 2-(4-bromobutyl)thiophene using general procedure B
(see above). The acid was obtained (60% yield, 99% by GC)
after bulb-to-bulb distillation (137–138 ЊC/0.3 mbar), mp 39–
40 ЊC (HRMS: Calc. for C11H16O2S: M, 212.087. Found: Mϩ,
212.087); νmax (neat, NaCl) 2927, 2849, 1700, 1459, 1257,
1231, 965, 852, 828, 706 cmϪ1; m/z (relative intensity) 212
(Mϩ, 31%), 194 (3), 166 (2), 151 (2), 139 (7), 136 (9), 111
(28), 98 (32), 97 (100), 53 (12); δH (270 MHz; CDCl3) 1.18
(3H, d, J = 7 Hz), 1.43 (3H, m), 1.70 (3H, m), 2.46 (1H, m),
2.83 (2H, t, J = 7 Hz), 6.76 (1H, d, J = 4 Hz), 6.90 (1H, m),
7.09 (1H, d, J = 5 Hz), 11.7 (1H br s); δC (67.8 MHz; CDCl3)
16.80, 26.58, 29.67, 31.59, 33.14, 39.27, 122.82, 124.01, 126.65,
145.30, 183.20.
2-Methyl-6-(2-thienyl)hexan-1-ol 5a. Reduction of 2-methyl-
6-(2-thienyl)hexanoic using general procedure A (see above)
followed by distillation (109–109.5 ЊC/0.45 mbar), gave the
title alcohol (88% yield, 99% by GC), nD20 1.517 (HRMS:
Calc. for C11H18OS: M, 198.108. Found: Mϩ, 198.104);
νmax (neat, NaCl) 3323, 2925, 2851, 1461, 1438, 1377, 1233,
1035, 850, 821, 691 cmϪ1; m/z (relative intensity) 198 (Mϩ,
11%), 181 (4), 170 (25), 137 (7), 123 (19), 111 (19), 110 (32),
98 (43), 97 (100), 53 (13); δH (270 MHz; CDCl3) 0.91 (3H,
d, J = 7 Hz), 1.14 (1H, m), 1.31–1.47 (4H, m), 1.55–1.73
(3H, m), 2.83 (2H, t, J = 8 Hz), 3.43 (2H, m), 6.77 (1H, m),
6.90 (1H, m), 7.09 (1H, m); δC (67.8 MHz; CDCl3) 16.57,
26.43, 29.85, 32.02, 32.83, 35.69, 68.27, 122.77, 123.95, 126.65,
145.62.
(R)-(؉)-Tetrahydrolavandulol 7.52 (R)-(ϩ)-6f {[α]D25 ϩ8.3 (c 1.0,
MeOH), ee 79.9%} was obtained after reduction (general
method) of the corresponding acetate produced in the PCL
transesterification. This (30 mg) was hydrogenated in methanol
(2 ml) with Pd(C) (5 mg, 10% Pd) as catalyst. After 6 h the
catalyst was filtered off, and washed with methanol. After dry-
ing (MgSO4) and evaporation off of the solvent the title com-
pound was obtained, [α]D25 ϩ12 (c 0.4, MeOH) {lit.,52 [α]D20 ϩ12.8
(neat)}. Spectroscopic data were in good agreement with those
in the literature.52
31 J. D. Schrag, Y. Li, M. Cygler, D. Lang, T. Burgdorf, H.-J.
Hecht, R. Schmid, D. Schomburg, T. J. Rydel, J. D. Oliver, L. C.
Strickland, C. M. Dunaway, S. B. Larson, J. Day and A.
McPherson, Structure, 1997, 5, 187.
32 G. Bergström, A.-B. Wassgren, O. Anderbrant, J. Fägerhag,
H. Edlund, E. Hedenström, H.-E. Högberg, C. Geri, M. Auger, M.
Varama, B. S. Hansson and J. Löfqvist, Experientia, 1995, 51, 370.
33 G. Bergström, A.-B. Wassgren, O. Anderbrant, S. A. Ochieng,
F. Östrand, B. S. Hansson, E. Hedenström and H.-E. Högberg,
Naturwissenschaften, 1998, 85, 244.
34 S. Karlsson and E. Hedenström, Acta Chem. Scand., 1999, 53,
620.
35 H. Edlund, P. Berglund, M. Jensen, E. Hedenström and H.-E.
Högberg, Acta Chem. Scand., 1996, 50, 666.
36 P. Berglund and E. Hedenström, in Methods in Biotechnology
(Enzymes in Non-aqueous Media: IV Synthetic Aspects), ed. H.
Holland, Humana Press, Totowa, NJ, 2000, in press.
37 T. Kumamoto, K. Hosoya, S. Kanzaki and K. Masuko, Bull. Chem.
Soc. Jpn., 1986, 59, 3097.
38 B. M. Trost, D. E. Keeley, H. C. Arndt, J. H. Rigby and M. J.
Bogdanowicz, J. Am. Chem. Soc., 1977, 99, 3080.
Acknowledgements
We thank Amano Pharmaceutical Co, Ltd, Japan, for a gener-
ous gift of Amano PS lipase and Pierre Ljungquist at STFI
(Swedish Pulp and Paper Research Institute) for HRMS meas-
urements. This project was supported by the Swedish Council
for Forestry and Agricultural Research (SJFR), the Swedish
Natural Science Research Council (NFR) and the Commission
of the European Communities, Agriculture and Fisheries
(FAIR), specific RTD program, contract no. FAIR1-CT95-
0339, ‘Pine sawfly pheromones for sustainable management of
European forests’ (this study does not necessarily reflect the
Commisson’s views and in no way anticipates future policies in
this area).
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376
375

View Article Online
39 E. Santaniello, P. Ferraboschi and P. Grisenti, Enzyme Microb.
Technol., 1993, 15, 367.
40 D. L. Delinck and A. L. Margolin, Tetrahedron Lett., 1990, 31,
6797.
41 F. Bracher and T. Papke, J. Chem. Soc., Perkin Trans. 1, 1995,
2323.
53 R. H. Valivety, P. J. Halling and A. R. Macrae, Biochim. Biophys.
Acta, 1992, 1118, 218.
54 P. Berglund, M. Holmquist, E. Hedenström, K. Hult and H.-E.
Högberg, Tetrahedron: Asymmetry, 1993, 4, 1869.
55 P. M. Jackson, C. J. Moody and P. Shah, J. Chem. Soc., Perkin Trans.
1, 1990, 2909.
42 P. Ferraboschi, S. Casati, S. Degrandi, P. Grisenti and E.
Santaniello, Biocatalysis, 1994, 10, 279.
43 P. Ferraboschi, S. Casati, A. Manzocchi and E. Santaniello,
Tetrahedron: Asymmetry, 1995, 6, 1521.
56 P. Cagniant, CR Hebd. Seances Acad. Sci., 1951, 232, 734.
57 M. T. Bogert and G. Powell, J. Am. Chem. Soc., 1931, 53, 2747.
58 K. E. Schulte, W. Weißkopf and J. Kirschner, Hoppe-Seyler’s Z.
Physiol. Chem., 1951, 288, 69.
44 C.-S. Chen, Y. Fujimoto, G. Girdaukas and C. J. Sih, J. Am. Chem.
Soc., 1982, 104, 7294.
45 J. L. L. Rakels, A. J. J. Straathof and J. J. Heijnen, Enzyme Microb.
Technol., 1993, 15, 1051.
46 E. Santaniello, P. Ferraboschi and P. Grisenti, Tetrahedron Lett.,
1990, 31, 5657.
47 P. Grisenti, P. Ferraboschi, A. Manzocchi and E. Santaniello,
Tetrahedron, 1992, 48, 3827.
48 P. Ferraboschi, P. Grisenti, A. Manzocchi and E. Santaniello,
J. Org. Chem., 1990, 55, 6214.
49 K. Naemura, R. Fukuda, M. Murata, M. Konishi, K. Hirose and
Y. Tobe, Tetrahedron: Asymmetry, 1995, 6, 2385.
50 R. Guevel and L. A. Paquette, Tetrahedron: Asymmetry, 1993, 4,
947.
59 P. Cagniant, G. Merle and D. Cagniant, Bull. Soc. Chim. Fr., 1970
308.
60 N. Polgar and W. Smith, Chem. Ind. (London), 1961, 1959.
61 K. Morigaki, S. Dallavalle, P. Walde, S. Colonna and P. L. Luisi,
J. Am. Chem. Soc., 1997, 119, 292.
62 M. D. Dowle, R. Hayes, D. B. Judd and C. N. Williams, Synthesis,
1983, 73.
63 A. Padwa, D. L. Hertzog and W. R. Nadler, J. Org. Chem., 1994, 59,
7072.
64 R. M. Acheson and M. W. Cooper, J. Chem. Soc., Perkin Trans. 1,
1980, 1185.
65 R. Baker, M. J. O’Mahony and C. J. Swain, J. Chem. Soc., Perkin
Trans. 1, 1987, 1623.
51 O. Nordin, E. Hedenström and H.-E. Högberg, Acta Chem. Scand.,
1999, 53, 124.
52 J. E. Johansen and S. Liaaen-Jensen, Acta Chem. Scand., Ser. B,
1979, 33, 551.
Paper a908023f
376
J. Chem. Soc., Perkin Trans. 1, 2000, 367–376