Highly efficient synthesis of enantiopure diacetylated C 2-symmetric diols by ruthenium- and enzyme-catalyzed dynamic kinetic asymmetric transformation (DYKAT)

Article abstract of DOI:10.1002/chem.200600257

Highly efficient synthesis of enantiopure diacetates of 2,4-pentanediol and 2,5-hexanediol starting from commercially available mixtures of the diols (dl/ meso ≈1:1) has been realized by combining a fast ruthenium-catalyzed epimerization with an enzymatic transesterification. The in situ coupling of these two processes produces the diacetates in high yield in > 99 % enantiomeric excess.

Full text of DOI:10.1002/chem.200600257

FULL PAPER
DOI: 10.1002/chem.200600257
Highly Efficient Synthesis of Enantiopure Diacetylated C2-Symmetric Diols
by Ruthenium- and Enzyme-Catalyzed Dynamic Kinetic Asymmetric
Transformation (DYKAT)
BelØn Martín-Matute,[a, b] Michaela Edin,[a] and Jan-E. Bäckvall*[a]
Abstract: Highly efficient synthesis of enantiopure diacetates of 2,4-pentanediol
and 2,5-hexanediol starting from commercially available mixtures of the diols (dl/
meso ꢀ1:1) has been realized by combining a fast ruthenium-catalyzed epimeriza-
tion with an enzymatic transesterification. The in situ coupling of these two pro-
cesses produces the diacetates in high yield in >99% enantiomeric excess.
Keywords: diols · dynamic kinetic
asymmetric
enzymes
racemization
transformations
kinetic resolution
·
·
·
Introduction
conditions.[5,11b] The ruthenium-catalyzed racemization of
the sec-alcohols involves hydrogen transfer via a ketone and
the mechanism of hydrogen transfer with catalysts 1 and 2
has recently been investigated.[5b,14]
Great achievements in catalytic asymmetric synthetic trans-
formations using transition metals, enzymes, and organoca-
talysts have been reported in recent years.[1] However, kinet-
ic resolution (KR) of racemic mixtures is still the most
common way to prepare enantiomerically pure compounds
on an industrial scale.[2] Enzymatic KR suffers, as does all
resolution, from being limited to a maximum theoretical
yield of 50%. In situ racemization of the slow-reacting
enantiomer leads to deracemization by dynamic kinetic res-
olution (DKR), and makes a 100% yield of enantiopure
product possible. The number of examples of chemoenzy-
matic DKR that combine an enzymatic KR with an in situ
racemization method has increased during the past few
years.[3–11] In particular, we have performed DKR of a varie-
ty of sec-alcohols by combining KR with an in situ racemiza-
tion catalyzed by Shvoꢀs[12] ruthenium complex (1).[3] More
recently, we have employed a more effective racemization
catalyst (2)[13] that, in combination with an enzyme, allows
us to perform DKR of sec-alcohols under very mild reaction
An enzyme-catalyzed kinetic asymmetric transformation
(KAT) of a commercial mixture of diols (dl/meso ꢀ1:1)
would give, in the best case, a maximum theoretical yield of
25% of one enantiomer.[15] Therefore, the development of
new synthetic methods to obtain enantiopure diols in 100%
yield is highly desirable. The combination of KATs with an
in situ racemization/epimerization leads to a dynamic kinetic
asymmetric transformation (DYKAT) and makes a theoreti-
cal yield of 100% possible. As part of our ongoing program
we have recently reported the DYKAT of diols by employ-
ing Candida antarctica lipase B (CALB) for the resolution
and a ruthenium complex (1 or 2) for the epimerization.[16,17]
Despite the high enantiomeric excess obtained in the
DYKAT of 1,3- and 1,4-symmetrical diols, this process
showed low diastereoselectivity for 2,4-pentanediol (3) and
2,5-hexanediol (4) in their transformation to 5 and 6, respec-
tively (Scheme 1). In our previous studies, we concluded
that the reason for the poor diastereoselectivity
((R,R):meso=38:62) for the former diol is a rapid 1,3-acyl-
[a] Dr. B. Martín-Matute, Dr. M. Edin, Prof. Dr. J.-E. Bäckvall
Department of Organic Chemistry, Arrhenius Laboratory
Stockholm, University, 10691 Stockholm (Sweden)
Fax : (+46)8-154-908
E-mail: jeb@organ.su.se
[b] Dr. B. Martín-Matute
Current address: Facultad de Ciencias
Department of Organic Chemistry
Universidad Autónoma de Madrid, 28059 Madrid (Spain)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2006, 12, 6053 – 6061
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6053

Scheme 1. Shvo- and CALB-catalyzed DYKAT of diols.
migration in the syn-1,3-diol monoacetate intermediate,[18]
and for the latter diol (4) the low selectivity ((R,R):meso=
86:14) is due to a low enantioselectivity of the enzyme for
the monoacylated intermediates, resulting in the formation
of anti-Kazlauskas[19] products (i.e., acylation at an (S)-alco-
hol site). It is highly desirable to find a solution to the prob-
lem with the poor diastereoselectivity in Scheme 1 due to
the importance of optically pure C2-symmetric diols, as sour-
ces of chiral auxiliaries and ligands. Furthermore, 1,4-diols
have been used in the preparation of enantiopure trans-2,5-
disubstituted pyrrolidines.[20]
Here we report a detailed study of the enzyme-catalyzed
acylation (kinetic asymmetric transformation) of 2,5-hexane-
diol (4) and its monoacetates, and also of those intermedi-
ates produced under dynamic conditions, that is, in the pres-
ence of the ruthenium catalyst. The use of a fast racemiza-
tion/epimerization Ru complex (2), together with a better
understanding of the enzyme-catalyzed acylations, has made
it possible to develop a very efficient DYKAT providing a
method for the synthesis of (R,R)-2,5-hexanediol diacetate
(6). Also, by using Ru complex 2, the DYKAT of 2,4-penta-
nediol (3) can be performed at a lower temperature (508C).
These new reaction conditions outrun the acyl migration,
avoiding the formation of unwanted meso-diacetate 5. Both
(R,R)-diacetates, 5 and 6, can be prepared in excellent enan-
tiomeric excess (ee), diastereomeric excess (de), and yield
(Scheme 2).
Scheme 3. DYKAT of 2,5-hexanediol (4).
acylated (at a moderate rate) at the (S)-alcohol site (anti-
Kazlauskas acylation) to yield diacetate meso-6. The diol
(S,S)-4 is essentially nonreactive (vide infra) in the system
and it would have to be epimerized to meso-4 before any
transformation can occur.
The ratio of the rates of formation of (R,R)-7 and meso-6
from (R,S)-7 is shown in Equation (1).
formation of ðR,RÞ-7
formation of meso-6
k1½Ruꢁ
ð1Þ
¼
k2½Acyl-enzymeꢁ
The DYKAT process would become efficient if the above
ratio could be significantly increased. We envisioned two ap-
proaches: 1) the use of a more selective enzyme, which
gives less (S)-acylation of the monoacetate (R,S)-7 (smaller
k2) than CALB, or 2) the use of a faster epimerization cata-
lyst, so that competing (S)-acylation of (R,S)-7 does not
occur. A simple and fast experiment to test different en-
zymes is to study the enzyme-catalyzed acylation of diol
meso-4 (Scheme 4). We were looking for an enzyme that
would efficiently catalyze the first acylation to form monoa-
cetate (R,S)-7, but would not undergo acylation of the re-
maining (S)-alcohol affording meso-diacetate (meso-6) (k2 !
k1).
Scheme 2. DYKAT of diols catalyzed by an enzyme and ruthenium com-
plex 2.
Results and Discussion
Scheme 4. Enzyme-catalyzed KAT of meso-4.
Kinetic studies of 2,5-hexanediol (4): We first turned our at-
tention to the DYKAT of 2,5-hexanediol. A simplified over-
view of the mechanism is shown in Scheme 3. Diol (R,R)-4
reacts fast with the acyl donor in the presence of the
enzyme to give monoacetate (R,R)-7, which is immediately
acylated again yielding diacetate (R,R)-6. The meso-diol will
form monoacetate (R,S)-7 very fast. This reaction can follow
two pathways: 1) it can be epimerized with the Ru catalyst
to yield monoacetate (R,R)-7, or 2) it can be enzymatically
Pure meso diol (meso-4) was prepared as previously de-
scribed (see Supporting Information).[15d,18] From our experi-
ence,[3] we knew that only two R-selective lipases[21] have
been successfully combined with an in situ racemization/epi-
merization for DKR of alcohols: Candida antarctica lipase
B (CALB) and Pseudomonas cepacia lipase (PS-C “Ama-
no II”). Figure 1 shows the kinetics of CALB- and PS-C-cat-
6054
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6053 – 6061

Dyanmic Kinetic Transformations
FULL PAPER
Figure 1. CALB- (top) and PS-C-catalyzed (bottom) reaction of meso-4.
Scheme 5. Sequential esterifications of a) rac-4 and b) meso-4.
~
&
”:% meso-4, :% (R,S)-7, :% meso-6.
and ent-(R,S)-7 were prepared as described in the Support-
ing Information.
First we studied the consecutive kinetic resolutions of rac-
4 (Scheme 6). For molecules containing two alcohols that
alyzed reactions of diol meso-4 as a function of time. Both
KRs were performed at 708C by using three equivalents of
p-chlorophenyl acetate as the acyl donor in toluene. It was
observed that CALB showed very high activity for both acy-
lations: after 1 h the diol meso-4 had been completely trans-
formed to a 4:1 mixture of monoacetate (R,S)-7 and diace-
tate meso-6, and after 44 h the reaction mixture consisted of
only 10% of monoacetate (R,S)-7 and 90% of meso-6. In
contrast, both acylations, in particular the second, were
much slower for the reaction catalyzed by PS-C “Amano II”;
after 45 min essentially no diacetate had been formed
(<1%) and GC analysis showed a mixture of diol meso-4
and monoacetate (R,S)-7 in a ratio of 42:58. After 44 h
there was still very little diacetate meso-6 (7%) and the
main product was monoacetate (R,S)-7 (Figure 1, bottom).
Because of the better results obtained with PS-C lipase
for the reaction of meso-4 (Figure 1), it was of interest to
compare selectivity and activity of both enzymes for all the
enzyme-catalyzed transformations involved in the acylation
of 2,5-hexanediol 4. The esterification of diol 4 is not a
simple single-step reaction, but involves two sequential es-
terifications in which each step can exhibit selectivity. Fur-
thermore, the relative configuration of the two stereogenic
centers may also affect the selectivity of each of the acetyla-
tions (Scheme 5). Therefore, the meso- and rac-diols were
studied separately.
Scheme 6. KR of (R,R)-4 and (S,S)-4.
can be acylated by the enzyme, the reaction does not stop at
the chiral monoacetate stage, but rather undergoes a second
acylation to yield diacetates. This complicates the determi-
nation of the selectivity of the enzyme for the first acylation.
In the case of rac-4, the second acylation is very fast, and
when aliquots were taken at different reaction times, a mix-
ture of diol, monoacetate, and diacetates was detected by
chiral GC analysis. In the case of PS-C, the reaction was per-
formed at 708C. After 15 min, the reaction was stopped.
The diols were separated from the mixture of monoacetates
and diacetates, which were collected together, by silica gel
chromatography. The ee of the diols (18% ee) was measured
after chemical acylation to the diacetates. The monoacetate/
diacetate fraction was analyzed by GC, and showed a 1.9:1
mixture of monoacetate (R,R)-7 in >99% ee and diacetate
(R,R)-6 in >99% ee. This shows that the E1 (PS-C) is >200
for the monoacylation of rac-4. Similar studies were per-
formed for the reaction catalyzed by CALB. In this case,
due the higher activity of this enzyme, the reaction was per-
formed at room temperature for only 2.5 min. Also in this
case a mixture of enantiomerically pure monoacetate (R,R)-
The selectivity of the enzymes was measured by quantify-
ing the enantiomeric ratio (E) for the different steps of the
enzymatic transformations.[22] The E value only applies to
enantiomers, therefore, pure racemic diol (rac-4), and race-
mic mixtures of monoacetates had to be prepared. Diol rac-
4, diol meso-4, a racemic mixture of monoacetates (R,R)-7
and (S,S)-7, and a racemic mixture of monoacetates (R,S)-7
Chem. Eur. J. 2006, 12, 6053 – 6061
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6055

J.-E. Bäckvall et al.
7 (>99% ee) and diacetate R,R)-6 (>99% ee) was obtained,
therefore E1 (CALB) is also >200. The results are shown in
Scheme 6 and Table 1. The conclusion is that both enzymes
show very high enantioselectivity for the first acylation.
was determined in toluene at 708C. A racemic mixture of
monoacetates (R,R)-7 and (S,S)-7 was subjected to enzymat-
ic acylation in the presence of p-chlorophenyl acetate
(3 equiv) as the acyl donor (Scheme 8). Both of the en-
zymes, PS-C and CALB showed very high selectivity, in par-
ticular PS-C, which was found to have an E2 value of 268.
The corresponding E2 value for CALB was 94. The results
are summarized in Table 3.
Table 1. KR of (R,R)-4 and (S,S)-4.[a]
Enzyme
t [min]/
T [8C]
ee [%]
of 4[b]
ee [%]
of 7[c,d]
E1
PS-C “Amano II”
CALB
22/70
2.5/RT
18 (S,S)
8 (S,S)
>99 (R,R)
>99 (R,R)
>200
>200
[a] For reaction conditions, see Experimental Section. [b] Determined by
GC after chemical diacetylation. [c] Determined considering that also di-
acetate (R,R)-6 was produced in 99% ee. [d] ee determined by GC.
As in the case of rac-4, the enzymatic reaction of meso-4
may not stop at the chiral monoacetate stage (7), but can
undergo a second acylation yielding diacetate meso-6 (Sche-
me 5b). Scheme 7 shows the desymmetrization of meso-4 to
Scheme 8. KR of (R,R)-7 and (S,S)-7.
The selectivity of the second acylation of the meso series
(E’2; Scheme 9) was determined by KR of the monoacetate
rac-(R,S)-7 (racemic mixture of (R,S)-7/ent-(R,S)-7). The
KR was carried out at 708C for both enzymes (Table 4).
The E’2 values obtained are rather low, 26 for PS-C and 7
for CALB. It is of interest to compare these results with
those obtained for the kinetic resolution of (R,R)-7/
(Scheme 8), in which the enantiomeric ratio (E2) is large;
A
Scheme 7. Desymmetrization of meso-4.
monoacetates 7. Since this is
Table 3. KR of rac-7 ((R,R)-7/
(S,S)-7).[a]
A
not
a kinetic resolution we
Enzyme
t [min]/
T [8C]
ee [%]
of 7[b]
ee [%]
of 6[b]
Conversion [%][c]
E2
cannot obtain an enantiomeric
ratio (E value). However, we
can define a pseudo-E value
(E’1).[23] When PS-C was em-
ployed as the enzyme, the de-
symmetrization was performed
at 708C. However, due to a
fast second acylation when em-
PS-C “Amano II”
CALB
30/70
30/70
31 (S,S)
93 (S,S)
>99 (R,R)
93 (R,R)
24
50
268
94
[a] For reaction conditions, see Experimental Section. [b] Determined by GC. [c] Determined applying the for-
mula: conversion=ees/(ees+eep).
AHCTREUNG
ploying CALB (cf. Figure 1, top), the desymmetrization was
performed at room temperature in this case. In both cases,
the formation of diacetate was not detected after 15 min;
GC analysis showed a mixture of only monoacetate and diol
in each case. The pseudo E’1 values are shown in Table 2,
and in both cases they are >199. Thus, for both enzymes,
the desymmetrization of meso-4 is highly selective.
Scheme 9. KR of monoacetates rac-(R,S)-7.
Table 4. KR of monoacetates rac-(R,S)-7.[a]
Table 2. Desymmetrization of meso-4.[a]
Enzyme
t [min]/
T [8C]
ee [%]
of 7[b]
Conversion [%][c]
E’2
Enzyme
t [min]/ ee [%]
T [8C]
of (R,S)-7[b]
Conversion [%][c] Pseudo E’1
PS-C “Amano II”
CALB
30/70
30/70
27
63
23
51
26
7
PS-C “AmanoII” 15/70
CALB 15/23
>99
>99
39
26
>199
>199
[a] For reaction conditions, see Experimental Section. [b] Determined by
1
[a] For reaction conditions, see Experimental Section. [b] Determined by
GC. [c] Determined by H NMR spectroscopy.
1
GC. [c] Determined H NMR spectroscopy.
268 for PS-C and 94 for CALB. These results show that the
relative stereochemistry of the neighboring acetate group is
of importance for the enantioselectivity in the enzymatic
acylation of the diol monoacetates to diacetates; in contrast
Kinetic studies of 2,5-hexanediol monoacetates (7): The
second enantiomeric value (E2) for the transformation of
(rac)-4 to diacetate involves monoacetates as substrates and
6056
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6053 – 6061

Dyanmic Kinetic Transformations
FULL PAPER
with the very slow acylation of monoacetate (S,S)-7 in
Scheme 8, acylation of the (S)-alcohol in monoacetate
(R,S)-7 seems to be facile.
Kinetic studies of keto intermediates formed in the DYKAT
of 2,5-hexanediol (4): The above studies show that in gener-
al Pseudomonas cepacia lipase shows higher selectivity than
Candida antarctica lipase B. However, the enzyme- and
ruthenium-catalyzed DYKAT gives only slightly higher de
for the reaction catalyzed by Pseudomonas cepacia lipase
compared to the reaction catalyzed by Candida antarctica
lipase B (vide infra). We therefore next considered the inter-
mediates that are formed in the reaction mixture under
DYKAT conditions. It is important to note that during the
racemization of sec-alcohols catalyzed by ruthenium com-
plexes, a hydrogen-transfer process takes place, and ketones
are formed as reaction intermediates. The concentration of
ketones in the reaction mixtures increases as the reaction
temperature increases.[24] In the case of 1,4-diols, the oxida-
tion of one of the hydroxyl groups gives rise to g-hydroxyke-
tones 8. The enantioselectivity of the enzymatic acylation of
hydroxyketone 8 may influence the overall enantioselectivi-
ty and diastereoselectivity of the DYKAT process, since the
ketoacetate formed finally will be converted to diacetate
(after ketone reduction and subsequent acylation). Thus, we
decided to study the kinetic resolution of compounds 8. In
this case, the conversion of the enzyme-catalyzed KR was
measured by NMR spectroscopy. Since it was difficult to
measure the conversion due to the formation of cyclic hemi-
acetals 9 (Scheme 11), the kinetic resolutions were run three
It is also of interest to determine the relative rate of acy-
lation of (5R)- and (5S)-alcohols when there is a (2R)-ace-
tate present, that is, to determine the relative rate of enzy-
matic acylation of (R,R)-7 and (R,S)-7. These two monoace-
tates will be produced in higher concentrations than the
other monoacetates in the reaction mixture of the DYKAT
process, because of the excellent enantioselectivity (E1 and
pseudo E’1) in the acylation of the diols. Monoacetates
(R,R)-7 and (R,S)-7 were prepared according to a previously
described procedure[18] with a slight modification (see Sup-
porting Information).
The enzymatic acylation of monoacetate (R,R)-7 was
compared with that of monoacetate (R,S)-7 for the approxi-
mate first-order acetylation reaction (Scheme 10). The ki-
Scheme 10. PS-C-catalyzed acylation of (R,R)-7 and of (R,S)-7.
Scheme 11. KR of 5-hydroxy-2-hexanone (8).
netic studies when employing PS-C as the enzyme gave a
ratio k3/k2 of 45 (Figure 2), which can be compared to the E
values obtained for the monoacetates (E2 and E’2). We have
previously carried out the analogous kinetic experiment for
the CALB-catalyzed reaction and a ratio k3/k2 of 25 was ob-
served.[18] Once again, the PS-C lipase gave a more selective
reaction than CALB.
times for each enzyme, which gave more reliable E values
(Table 5). It is interesting to note that both enzymes gave
Table 5. KR of 5-hydroxy-2-hexanone (8).[a]
[d]
Enzyme
t [min]/T [8C]
ee [%]
of 10[b]
Conversion [%][c]
E3
PS-C “Amano II”
CALB
15/70
15/70
11
70
10
47
1.3
9
[a] See Experimental Section. [b] Determined by GC. [c] Determined by
1H NMR spectroscopy. [d] Average of three runs.
very low E values for the kinetic resolution of g-hydroxyke-
tone 8 and PS-C showed no selectivity at all (E=1.3). A
similar negative effect of a keto function in the d-position of
a 2-alkanol was recently observed,[17b,25] and an E value of
2.2
for the CALB-catalyzed
KR of Ph(CO)-
AHCTREUNG
the corresponding 1,4-diol, PhCH(OH)(CH2)2CH(OH)Me,
G
was 285.[17b] This drastic drop of the enantioselectivity on
changing from 1,4-diol to 4-hydroxyketone is intriguing.
Computational and modeling studies carried out in our
group have shown that hydrogen bonding to the ketone is
~
^
Figure 2. PS-C rate of acetylation of (R,R)-7 ( ) and of (R,S)-7 ( ). Re-
actions performed at 708C in toluene using p-chlorophenylacetate as acyl
donor (3 equiv) and 30 mg of enzyme per mmol of 7.
Chem. Eur. J. 2006, 12, 6053 – 6061
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6057

J.-E. Bäckvall et al.
responsible for the increased relative rate of acylation of the
table), or when 6 mol% of tBuOK was employed (entry 6).
In the last case, although 6 was obtained in 97% yield, more
of the undesired meso-6 was produced, probably due to
chemical acylation with the excess of base. When the cata-
lyst loading was reduced to 2 mol%, the product was ob-
tained quantitatively, but, once again, more of undesired
meso-6 was produced, possibly due to a slower racemization
rate (entry 7). The DYKAT was also tested with ruthenium
complex 2 and CALB as the enzyme, yielding diacetate 6 in
excellent yields and much shorter reaction times (entries 8
and 9) than when the DYKAT was performed with catalyst
1 (cf. entry 1). However, the amount of CALB has to be re-
duced to 3 mg per mmol of diol 4 in order to obtain a high
(R,R)/meso-6 ratio (94:6) (entry 9). This ratio was not fur-
ther improved when less than 3 mg of CALB were used.
(S)-enantiomer of 8.[26]
DYKAT of 2,5-hexanediol (4): The DYKAT of 2,5-hexane-
diol (4) was tested with both enzymes. During our initial
studies on the DYKAT of diol 4, we used Shvoꢀs ruthenium
complex (1) as the racemization catalyst and the reactions
were run under the conditions required for dynamic kinetic
resolution (DKR) by using ruthenium complex 1, that is,
708C and with p-chlorophenylacetate as the acyl donor.[27]
The results with complex 1 as the catalyst are shown in
Table 6, entries 1 and 2. Thus, when CALB was employed as
the enzyme, diacetate 6 was obtained in moderate yield as a
mixture of (R,R)-6 (99% ee) and meso-6 (77:23).[28] PS-C
gave slightly better results (Table 6, entry 2), yielding diace-
tate 6 in higher yield (72%) and a better (R,R)-6/meso-6
ratio (89:11). In both cases, very long reaction times were
required. More recently we have found that ruthenium com-
plex 2 efficiently catalyzes racemization of sec-alcohols at
room temperature, and, importantly, it is compatible with
the use of isopropenyl acetate when performing DKR.[5]
Catalyst 2 needs to be activated by a catalytic amount of
tBuOK. We decided to combine this fast racemization cata-
lyst (2) with PS-C “Amano II”, since this enzyme gives
better results in the kinetic studies than CALB (vide supra).
The reactions using catalyst 2 were performed at 508C.
When using 30 mg of PS-C per mmol of diol 4 (entry 3), a
good (R,R)-/meso-6 ratio was obtained (90:10) with 91%
yield in 40 h. To decrease the reaction time, the DYKAT
was performed with double amount of enzyme (60 mg per
mmol of diol 4; entry 4). Diacetate 6 was obtained in excel-
lent yield (96%) and good (R,R)/meso-6 ratio (92:8) in 20 h.
When the amount of enzyme was further increased the reac-
tion time was reduced to 6 h (entry 5) and 6 was obtained
quantitatively with the same (R,R)/meso-6 ratio as in
entry 4. The results were not improved when the amount of
enzyme employed was further increased (not shown in the
Mechanistic consideration of DYKAT of 2,5-hexanediol (4):
The (R,R)/meso ratio does not dramatically change when
the amount of PS-C “Amano II” is increased. However,
when the amount of CALB is increased, a considerable de-
crease of the (R,R)/meso ratio is observed. These results are
in accordance with the faster undesired anti-Kazlauskas acy-
lation of monoacetate (R,S)-7 to meso-6 for the CALB-cata-
lyzed reaction compared to the PS-C-catalyzed reaction (see
Figure 1 and Scheme 3).
Ruthenium catalyst 2 is a faster racemization/epimeriza-
tion catalyst than Shvoꢀs complex (1).[13] When the DYKAT
is catalyzed by complex 2 and PS-C, the fast epimerization
outruns the unwanted anti-Kazlauskas acylation of (R,S)-7
to meso-6 (i.e., k1 >k2, Scheme 3). With CALB as the
enzyme and 2 as the ruthenium catalyst, this situation is ach-
ieved when the amount of enzyme is reduced with a factor
20 (Table 6, entry 9).
The reaction temperature also has an important effect on
the outcome of the DYKAT of diol 4. When the reactions
were analyzed by GC at about 50% conversion, different re-
action intermediates were detected depending on the reac-
Table 6. DYKAT of diol 4 (dl/meso ꢀ1:1).[a]
Entry
Enzyme [mg per mmol of diol 4]
Catalyst [mol%]
tBuOK [mol%]
T [8C]
t [h]
Yield of 6 [%][c]
ee [%][c]
N
1[b]
2[b]
3
4
5
6
7
8
9
CALB (60)
PS-C (60)
PS-C (30)
PS-C (60)
PS-C (80)
PS-C (80)
PS-C (80)
CALB (60)
CALB (3)
1 (5)
1 (5)
2 (5)
2 (5)
2 (5)
2 (5)
2 (2)
2 (5)
2 (5)
–
–
5
5
5
6
2
5
5
70
70
50
50
50
50
50
50
50
48
40
40
20
6
6
20
6
61
72
91
99
99
99
77:23
89:11
90:10
92:8
96
99
>99 (95)[d]
97[e]
99[e]
97
>99
>99
>99
>99
>99
92:8
83:17
84:16
80:20
94:6
21
>99
[a] Unless otherwise noted, Ru-catalyst 2 (2–5 mol%), CALB or PS-C, Na2CO3 (1 mmol) and KOtBu (2–6 mol%) were stirred in toluene (2 mL) at RT
for 6 min before adding the diol 4 (1 mmol). After 4 min isopropenyl acetate (3 mmol) was added and the mixture was stirred at 508C under an argon at-
mosphere. [b] A solution of diol 4 (0.2 mmol) and p-chlorophenyl acetate (0.6 mmol) in toluene (0.5 mL) was added to a mixture of Ru-catalyst 1
(5 mol%) and the enzyme, CALB or PS-C. The mixture was stirred at 708C under an argon atmosphere. [c] Determined by chiral GC. [d] In parenthesis
1
isolated yield. [e] Yield determined by H NMR spectroscopy.
6058
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6053 – 6061

Dyanmic Kinetic Transformations
FULL PAPER
tion temperature: for those DYKAT transformations per-
formed at 708C, the only intermediates observed were ace-
toxy ketones 10, and for those performed at 508C, the main
intermediates were acetoxy alcohols (R,R)-7 and (R,S)-7. In
the latter case traces of acetoxy ketones 10 were also detect-
ed. During the ruthenium-catalyzed racemization, a hydro-
gen-transfer process takes place, and ketones are formed as
reaction intermediates. The concentration of ketones in the
reaction mixtures increases as the reaction temperature in-
creases.[24] Although the formation of hydroxyketone inter-
mediates 8 decreases with decreasing the temperature, it
cannot be completely suppressed. The enzyme-catalyzed
acylation of intermediates 8 takes place fast and with very
low selectivity (E3 =1.3 and 9 for PS-C and CALB, respec-
tively; see Table 5). As a consequence, a small amount of
meso-6 is always produced during the DYKAT (Scheme 12).
Scheme 14. DYKAT of 2,4-pentanediol (3).
Conclusion
We have performed kinetic studies of all the enzyme-cata-
lyzed steps involved in the KAT of 2,5-hexanediol (4), and
also of those intermediates formed under dynamic condi-
tions (i.e., in the presence of a transfer-hydrogenation ruthe-
nium complex) using two different enzymes, Candida antarc-
tica lipase B (CALB) and Pseudomonas cepacia lipase (PS-
C “Amano II”). Both enzymes
show very high selectivity for
the acylation of diols to mono-
acetates and for the acylation
of
(R,R)/(S,S)-monoacetate.
A
However, the enantioselectivi-
ty of CALB- or PS-C-cata-
lyzed acylation of alcohols
bearing a carbonyl group or an
(R)-acetoxy group at the d-po-
sition is rather low. The reason
for this anomalous behavior is
Scheme 12. Formation of meso-6 from 5-hydroxy-2-hexanone 8.
currently being investigated in
our laboratories.[26] Important
DYKAT of 2,4-pentanediol (3): In the DYKAT of 2,4-penta-
nediol (3) previously reported by our group, diacetate 5 was
obtained in high yield (90%), but with the meso diacetate
as the major product ((R,R)/meso=38:62) (cf. Scheme 1).[16]
In that DYKAT, the Shvo ruthenium catalyst was used for
epimerization and therefore the reaction was performed at
708C. From deuterium labeling studies, it was subsequently
demonstrated that formation of meso-5 is mainly due to a
facile 1,3-acyl migration in the syn-1,3-diol monoacetate in-
termediates (Scheme 13).[18]
improvements of the DYKAT of symmetrical diols have
been achieved by using a highly efficient catalyst (2) to ach-
ieve very fast epimerization. In the DYKAT of 2,5-hexane-
diol 4 catalyzed by 2, the faster epimerization and the lower
reaction temperature reduce the concentration of intermedi-
ates that could evolve by anomalous enzyme-catalyzed S-
acylations, such as monoacetate (R,S)-7 and hydroxyketone
8. (R,R)-Diacetate 6 is now obtained in quantitative yield
with excellent ee (>99%) and high (R,R)/meso ratio (94:6).
DYKAT of 2,4-pentanediol (3) by using catalyst 2 was also
performed at 508C. At this temperature, the epimerization
outruns the acyl migration and a commercial stereoisomeric
mixture of the diol was transformed to 96% of (R,R)-diace-
tate 5 with an excellent ee (>99%) and an excellent (R,R)/
meso ratio (97:3).
Scheme 13. Intramolecular acyl migration in a syn-1,3-diol monoacetate.
Experimental Section
General: All reactions were carried out under dry argon atmosphere in
flame-dried glassware. Solvents were purified and dried with standard
procedures. To control the water activity, the enzymes were stored in a
sealed container with a saturated solution of LiCl for a minimum of 24 h.
Flash chromatography was carried out on 60 Š (35–70 mm) silica gel. 1H
and 13C NMR spectra were recorded at 400 or 300 MHz and at 100 or
75 MHz, respectively. Enantiomeric excess and diastereomeric excess
were determined by analytical gas chromatography employing a CP-Chir-
asil-Dex CB chiral capillary column.
With the new and more efficient ruthenium complex 2,
DYKAT of meso/dl-2,4-pentanediol (3) was dramatically im-
proved. At 508C the epimerization outruns the acyl migra-
tion and a commercial stereoisomeric mixture of the diol
(meso/dl=1:1) was transformed to (R,R)-diacetate 5 in high
yield (96%) with an excellent ee (>99%) and an excellent
(R,R)/meso ratio (97:3) (Scheme 14).
Chem. Eur. J. 2006, 12, 6053 – 6061
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6059

J.-E. Bäckvall et al.
DYKAT of 2,4-pentanediol (3): Ruthenium complex
2
(32 mg,
nol. 2003, 14, 131; e) F. F. Huerta, A. B. E. Minidis, J.-E. Bäckvall,
Chem. Soc. Rev. 2001, 30, 321–331.
[4] P. J. Um, D. G. Drueckhammer, J. Am. Chem. Soc. 1998, 120, 5605–
5610.
0.05 mmol), CALB (6 mg), and Na2CO3 (106 mg, 1 mmol) were placed in
a Schlenk flask. The flask was evacuated and filled with argon and then
toluene (2 mL) was added. Addition of tBuOK (0.5m in THF; 180 mL,
0.09 mmol) to the yellow suspension resulted in a color change to orange.
The mixture was then stirred for 6 min, and then 2,4-pentanediol
(110 mL, 1 mmol) was added. On the addition of the diol the mixture
turned red. After 4 min, isopropenyl acetate (330 mL, 3 mmol) was added
and the flask was placed in an oil bath at 508C. After 20 h the reaction
mixture was filtered and analyzed: 96% yield, (R,R)/meso=97:3 (achiral
GC, CP-Sil 8 CB column, constant column flow: 1.8 mLminꢂ1, hydrogen
carrier gas. Temperature program: 508C for 2 min, then up to 1808C with
58Cminꢂ1; then up to 3008C with 808Cminꢂ1 and keep for 5 min. Reten-
tion times: (R,R)-5=12.32, meso-5=13.36 min), >99% ee (chiral GC,
CP-Chirasil-Dex column, constant flow: 1.8 mLminꢂ1, hydrogen carrier
gas. Temperature program: 908C for 3 min, then up to 1158C with
38Cminꢂ1; then up to 2008C with 808Cminꢂ1 and keep for 8 min. Reten-
tion times: (R,R)-5=6.67, meso-5=7.96 min. (S,S)-5 (not detected in re-
action mixtures)=6.20).[29]
[5] a) B. Martín-Matute, M. Edin, K. Bogµr, J.-E. Bäckvall, Angew.
Chem. 2004, 116, 6697–6701; Angew. Chem. Int. Ed. 2004, 43, 6535–
6539; b) B. Martín-Matute, M. Edin, K. Bogµr, F. B. Kaynak, J.-E.
Bäckvall, J. Am. Chem. Soc. 2005, 127, 8817–8825.
[6] a) J. H. Choi, Y.-H. Kim, S. H. Nam, S. T. Shin, M.-J. Kim, J. Park,
Angew. Chem. 2002, 114, 2479–2482; Angew. Chem. Int. Ed. 2002,
41, 2373–2376; b) J. H. Choi, Y. K. Choi, Y. H. Kim, E. S. Park, E. J.
Kim, M. J. Kim, J. Park, J. Org. Chem. 2004, 69, 1972–1977.
[7] For a DKR employing acid zeolites for racemization of benzylic al-
cohols, see: S. Wuyts, K. De Temmerman, D. E. De Vos, P. A.
Jacobs, Chem. Eur. J. 2005, 11, 386–397.
[8] For a related DKR of amines see: a) M. T. Reetz, K. Schimossek,
Chimia 1996, 50, 668–669; b) J. Paetzold, J.-E. Bäckvall, J. Am.
Chem. Soc. 2005, 127, 17620–17621.
[9] a) G. K. M. Verzijl, J. G. de Vries, Q. B. Broxterman, Tetrahedron:
Asymmetry 2005, 16, 1603–1610; b) B. A. C. van As, J. van Buijte-
nen, A. Heise, Q. B. Broxterman, G. K. M. Verzijl, A. R. A. Pal-
mans, E. W. Meijer, J. Am. Chem. Soc. 2005, 127, 9964–9965.
[10] T. H. Riermeier, P. Gross, A. Monsees, M. Hoff, H. Trauthwein, Tet-
rahedron Lett. 2005, 46, 3403–3406.
General procedure for the DYKAT of 2,5-hexanediol (4): A solution of
tBuOK (0.5m in THF; 100 mL, 0.05 mmol) was added to a 10 mL Schlenk
flask. The THF was carefully removed under vacuum and the flask filled
with argon. PS-C “Amano II” (80 mg), Na2CO3 (106 mg, 1 mmol), and
ruthenium catalyst
2 (32 mg, 0.05 mmol) were quickly added. The
Schlenk flask was evacuated and filled with argon. After addition of tol-
uene (2 mL) the mixture turned dark orange. After 6 min 4 (123 mL,
1 mmol) was added, and after 4 min isopropenyl acetate (330 mL,
3 mmol) was added. The reaction mixture was stirred for 6 h at 508C,
and then filtered, concentrated, and analyzed: >99% yield, (R,R)/
meso=92:8, >99% ee (chiral GC, CP-Chirasil-Dex column). Purification
by column chromatography (SiO2; pentane/diethyl ether 98:2) afforded
(R,R)-2,5-diacetoxyhexane as a colorless oil (193 mg, 95% yield, (R,R)/
meso=92:8, >99% ee (chiral GC, CP-Chirasil-Dex column, constant
flow: 1.8 mLminꢂ1, hydrogen carrier gas. Temperature program: 708C for
[11] For S-selective DKR, see: a) M.-J. Kim, Y. Chung, Y. Choi, H. Lee,
D. Kim, J. Park, J. Am. Chem. Soc. 2003, 125, 11494–11495; b) L.
BorØn, B. Martín-Matute, Y. Xu, A. Córdova, J.-E. Bäckvall, Chem.
Eur. J. 2006, 12, 225–232.
[12] N. Menasche, Y. Shvo, Organometallics 1991, 10, 3885–3891.
[13] G. Csjernyik, K. Bogµr, J.-E. Bäckvall, Tetrahedron Lett. 2004, 45,
6799–6802.
[14] a) C. P. Casey, S. W. Singer, D. R. Powell, R. K. Hayashi, M. Kavana,
J. Am. Chem. Soc. 2001, 123, 1090–1100; b) C. P. Casey, J. B. Johns-
son, J. Am. Chem. Soc. 2005, 127, 1883–1894; c) C. P. Casey, G. A.
Bikzhanova, Q. Cui, I. A. Guzei, J. Am. Chem. Soc. 2005, 127,
14062–14071; d) J. S. M. Samec, A. H. ꢂll, J. E. Bäckvall, Chem.
Commun. 2004, 2748–2749
[15] For studies on enzymatic kinetic resolutions of diols, see: a) A.
Mattson, N. ꢃhrner, K. Hult, T. Norin, Tetrahedron: Asymmetry
1993, 4, 925–930; b) M.-J. Kim, J. S. Lee, J. Org. Chem. 1993, 58,
6483–6485; c) H. Nagai, T. Morimoto, K. Achiwa, Synlett 1994,
289–290; d) G. Caron, R. J. Kazlauskas, Tetrahedron: Asymmetry
1994, 5, 657–664; e) J. S. Wallace, B. W. Baldwin, C. Morrow, J. Org.
Chem. 1992, 57, 5231–5231.
3 min, then up to 1408C with 28Cminꢂ1
; then up to 1808C with
1008Cminꢂ1 and keep for 10 min. Retention times: (R,R)-6=19.51,
meso-6=17.70 min, (S,S)-6 (not detected in reaction mixtures)=14.6).
The NMR spectra were in agreement with those previously reported in
the literature.[27b]
Acknowledgements
[16] Symmetrical diols: B. A. Persson, F. F. Huerta, J.-E. Bäckvall, J.
Org. Chem. 1999, 64, 5237–5240.
Financial support from the Swedish Research Council and the Swedish
Foundation for Strategic Research is gratefully acknowledged.
[17] Unsymmetrical diols: a) M. Edin, J. Steinreiber, J.-E. Bäckvall,
Proc. Natl. Acad. Sci. USA 2004, 101, 5761–5766; b) B. Martín-
Matute, J.-E. Bäckvall, J. Org. Chem. 2004, 69, 9191–9195; c) M.
Edin, B. Martín-Matute, J.-E. Bäckvall, Tetrahedron: Asymmetry
2006, 17, 708–715.
[18] a) M. Edin, J.-E. Bäckvall, J. Org. Chem. 2003, 68, 2216–2222; b) A
related acyl transfer has been observed in lipase-catalyzed hydrolysis
of diacetates of meso-1,3-diols (2-substituted 1,3-propane diols):
K. K.-C Liu, K. Nozaki, C.-H. Wong, Biocatalysis 1990, 3, 169–177.
[19] R. J. Kazlauskas, A. N. E. Weissfloch, A. T. Rappaport, L. A.
Cuccia, J. Org. Chem. 1991, 56, 2656–2665.
[20] a) R. P. Short, R. M. Kennedy, S. Masamune, J. Org. Chem. 1989, 54,
1755–1756; b) M. E. Zwaagstra, A. Meetsma, B. L. Feringa, Tetrahe-
dron: Asymmetry 1993, 4, 2163–2172; c) M.-J. Kim, I. S. Lee, Synlett
1993, 767–768.
[21] R and S selectivity are used for typical sec-alcohols in which the
large group has the higher priority in the sequential rule for deter-
mining the configuration (according to the Cahn–Ingold–Prelog
system).
[1] a) J. Halpern, B. M. Trost, Proc. Natl. Acad. Sci. USA 2004, 101,
5347; b) Catalytic Asymmetric Synthesis, 2nd ed. (Ed.: I. Ojima),
Wiley-VCH, New York, 2000; c) Comprehensive Asymmetric Cataly-
sis, (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer,
Berlin, 1999; d) R. Noyori in Asymmetric Catalysis in Organic Syn-
thesis, Wiley, New York, 1994; e) Enzyme Catalysis in Organic Syn-
thesis: A Comprehensive Handbook, Vols. I–III, 2nd ed. (Eds.: K.
Drauz, H. Waldmann), Wiley-VCH, Weinheim, 2002; f) K. Faber,
Biotransformations in Organic Chemistry, 4th ed, Springer, Berlin,
2000; g) P. I. Dalko, L. Moisan, Angew. Chem. 2004, 116, 5248–
5286; Angew. Chem. Int. Ed. 2004, 43, 5138–5175.
[2] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Kebeler, R.
Stürmer, T. Zelinski, Angew. Chem. 2004, 116, 806–846; Angew.
Chem. Int. Ed. 2004, 43, 788–824.
[3] a) O. Pàmies, J.-E. Bäckvall, Trends Biotechnol. 2004, 22, 130–135;
b) O. Pàmies, J.-E. Bäckvall, Chem. Rev. 2003, 103, 3247–3261; c) O.
Pàmies, J.-E. Bäckvall, Curr. Opin. Biotechnol. 2003, 14, 407–413;
d) M.-J. Kim, Y. Anh, J. Park, Curr. Opin. Biotechnol. 2002, 13, 578–
587; corrigendum: M.-J. Kim, Y. Anh, J. Park, Curr. Opin. Biotech-
[22] The enantiomeric ratio E defines the discrimination between two
enantiomers by the enzyme. See: a) C.-S. Chen, Y. Fujimoto, G. Gir-
daukas, C. J. Sih, J. Am. Chem. Soc. 1982, 104, 7294–7299; b) C.-S.
6060
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6053 – 6061

Dyanmic Kinetic Transformations
FULL PAPER
Chen, S.-H. Wu, G. Girdaukas, C. J. Sih, J. Am. Chem. Soc. 1987,
109, 2812–2817.
[23] By looking on the meso compound as two enantiomers that are cou-
pled to one another, one can define an enantioselectivity value E (in
this case E’1).
[24] B. Martín-Matute, J. Jansson, M. Edin, J.-E. Bäckvall, unpublished
results.
[25] P. Crotti, V. D. Bussolo, V. Favero, F. Minutolo, M. Pineschi, Tetrahe-
dron: Asymmetry 1996, 7, 1347–1356.
[27] a) A. L. E. Larsson, B. A. Persson, J.-E. Bäckvall, Angew. Chem.
1997, 109, 1256–1258; Angew. Chem. Int. Ed. Engl. 1997, 36, 1211–
1212; b) B. A. Persson, A. L. E. Larsson, M. Le Ray, J.-E. Bäckvall,
J. Am. Chem. Soc. 1999, 121, 1645–1650.
[28] The (R,R)/meso ratio slightly varies depending on the enzyme
batch. Thus, in our previous communication in 1999 (see refer-
ence [16]) an (R,R)/meso ratio of 86:14 was obtained.
[29] O. Itoh, Y. Ichikawa, H. Katano, K. Ichikawa, Bull. Chem. Soc. Jpn.
1976, 49, 1353–1358.
[26] M. Bocola, B. Martín-Matute, M. Edin, J.-E. Bäckvall, unpublished
results.
Received: February 23, 2006
Published online: June 26, 2006
Chem. Eur. J. 2006, 12, 6053 – 6061
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6061