Iterative tandem catalysis of secondary diols and diesters to chiral polyesters

Article abstract of DOI:10.1002/chem.200700818

The well-known dynamic kinetic resolution of secondary alcohols and esters was extended to secondary diols and diesters to afford chiral polyesters. This process is an example of iterative tandem catalysis (ITC), a polymerization method where the concurrent action of two fundamentally different catalysts is required to achieve chain growth. In order to procure chiral polyesters of high enantiomeric excess value (ee) and good molecular weight, the catalysts employed need to be complementary and compatible during the polymerization reaction. We here show that Shvo's catalyst and Novozym 435 fulfil these requirements. The optimal polymerization conditions of 1,1′-(1,3-phenylene) diethanol (1,3-diol) and diisopropyl adipate required 2mol% Shvo's catalyst and 12 mg Novozym 435 per mmol alcohol group in the presence of 0.5 M 2,4-dimethyl-3- pentanol as the hydrogen donor. With these conditions, chiral polyesters were obtained with peak molecular weights up to 15kDa, an ee value up to 99% and with 1-3% ketone end groups. Also with the structural isomer, 1,4-diol, a chiral polyester was obtained, albeit with lower molecular weight (8.3 kDa) and slightly lower ee (94%). Aliphatic secondary diols also resulted in enantio-enriched polymers but at most an ee of 46% was obtained with molecular weights in the range of 3.33.7 kDa. This low ee originates from the intrinsic low enantioselectivity of Novozym 435 for this type of secondary aliphatic diols. The results presented here show that ITC can be applied to procure chiral polyesters with good molecular weight and high ee from optically inactive AA-BB type monomers.

Full text of DOI:10.1002/chem.200700818

FULL PAPER
DOI: 10.1002/chem.200700818
Iterative Tandem Catalysis of Secondary Diols and Diesters to Chiral
Polyesters
Bart A. C. van As, Jeroen van Buijtenen, Tristan Mes, Anja R. A. Palmans,* and
E. W. Meijer*[a]
Abstract: The well-known dynamic ki-
netic resolution of secondary alcohols
and esters was extended to secondary
diols and diesters to afford chiral poly-
esters. This process is an example of
The optimal polymerization conditions
of 1,1’-(1,3-phenylene) diethanol (1,3-
diol) and diisopropyl adipate required
2 mol% Shvoꢀs catalyst and 12 mg No-
vozym 435 per mmol alcohol group in
the presence of 0.5m 2,4-dimethyl-3-
pentanol as the hydrogen donor. With
these conditions, chiral polyesters were
obtained with peak molecular weights
up to 15 kDa, an ee value up to 99%
and with 1–3% ketone end groups.
Also with the structural isomer, 1,4-
diol, a chiral polyester was obtained,
albeit with lower molecular weight
(8.3 kDa) and slightly lower ee (94%).
Aliphatic secondary diols also resulted
in enantio-enriched polymers but at
most an ee of 46% was obtained with
molecular weights in the range of 3.3–
3.7 kDa. This low ee originates from
the intrinsic low enantioselectivity of
Novozym 435 for this type of secon-
dary aliphatic diols. The results pre-
sented here show that ITC can be ap-
plied to procure chiral polyesters with
good molecular weight and high ee
from optically inactive AA–BB type
monomers.
iterative tandem catalysis (ITC),
a
polymerization method where the con-
current action of two fundamentally
different catalysts is required to ach-
ieve chain growth. In order to procure
chiral polyesters of high enantiomeric
excess value (ee) and good molecular
weight, the catalysts employed need to
be complementary and compatible
during the polymerization reaction. We
here show that Shvoꢀs catalyst and No-
vozym 435 fulfil these requirements.
Keywords:
enzyme catalysis
·
kinetic resolution · polymerization ·
racemization · ruthenium
Introduction
Recently, we introduced the concept of iterative tandem
catalysis (ITC). ITC is a novel polymerization strategy by
which chain growth is effectuated by a combination of two
(or more) intrinsically different catalytic processes that are
both compatible and complementary.[8] Using ITC, a race-
mic mixture of monomers can be quantitatively turned into
a homochiral polymer. Proof of principle was provided by
the ITC of (S)-6-methyl-e-caprolactone ((S)-6-MeCL). We
showed that the lipase-catalyzed ring-opening of w-substitut-
ed lactones, such as 6-MeCL, results in a ring-opening prod-
uct bearing a secondary alcohol. Since lipases generally only
accept the R enantiomer of a secondary alcohol as the nu-
cleophile, propagation halts after the ring-opening of an (S)-
6-MeCL molecule. Combining ruthenium-catalyzed racemi-
zation with lipase-catalyzed ring-opening enabled the two-
pot oligomerization of (S)-6-MeCL.[8a] An efficient one-pot
polymerization proved difficult using catalyst 1 (Figure 1).
Recently, we achieved the synthesis of well-defined chiral
poly((R)-6-methyl-e-caprolactone) in one pot, with an ee up
to 96% and Mn up to 25 kDa.[8b] This was realized by replac-
ing racemization catalyst 1 by Shvoꢀs catalyst, 2.
Tandem catalysis has attracted much interest in the past
decade as an alternative to traditional step-by-step synthe-
sis.[1–5] Reduction of waste, costs and energy are the primary
driving forces for this development. Aprominent example
of tandem catalysis is the dynamic kinetic resolution (DKR)
of racemic secondary alcohols or amines into enantiopure
esters or amides in essentially 100% yield.[6,7] Extension of
tandem catalysis to the field of polymer chemistry would
allow for the synthesis of macromolecules of higher structur-
al complexity.
[a] Dr. B. A. C. van As, Dr. J. van Buijtenen, T. Mes,
Dr. A. R. A. Palmans, Prof. Dr. E. W. Meijer
Department of Chemical Engineering and Chemistry
Laboratory of Macromolecular and Organic Chemistry
Technische Universiteit Eindhoven, P.O. Box 513
5600 MB Eindhoven (The Netherlands)
Fax : (+31)40-24-1036
E-mail: E.W.Meijer@tue.nl
Chem. Eur. J. 2007, 13, 8325 – 8332
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8325

catalyst 2 was chosen as the racemization catalyst because
we recently found it showed an excellent compatibility with
Novozym 435 in the ITC of 6-MeCL.[8b] As model com-
pounds, 1,1’-(1,3-phenylene)diethanol (1,3-diol) and diiso-
propyl adipate (DIA) were employed as the secondary diol
and the diester, respectively (Scheme 1). In order to explore
the generality of ITC in polycondensation type reactions,
the 1,4-substituted structural isomer 1,1’-(1,4-phenylene)die-
thanol (1,4-diol) and aliphatic secondary diols were evaluat-
ed as well.
Figure 1. Racemization catalysts employed in ITC.
To broaden the applicability of ITC and to have easy
access to a variety of chiral polyesters, we aimed at perform-
ing polycondensation type reactions with AA (secondary
diols) and BB (diesters) type monomers, in analogy to the
DKR of secondary alcohols. In contrast to the ITC of 6-
MeCL, an AA–BB type system allows oligomerization to
take place by the action of the lipase only since the R-secon-
dary alcohols of the diols are readily esterified. However,
polymers of reasonable molecular weight can only be at-
tained when, at the same time, the remaining S alcohol
groups undergo an inversion of the configuration by the rac-
emization catalyst. Therefore, also in this polycondensation-
type reaction, concurrent action of both the acylation and
the racemization catalyst is required to procure polymers of
good molecular weight, making an AA–BB polymerization
under DKR conditions another example of ITC.
Scheme 1. ITC of 1,1’-(1,3-phenylene)diethanol (1,3-diol) and diisopropyl
adipate (DIA).
Results and Discussion
Recently, initial efforts on achieving such an extension of
DKR were reported by Hilker et al. comprising 1,1’-(1,4-
phenylene)diethanol and dimethyl adipate as the monomers
and using Novozym 435 and complex 1 as the acylation and
racemization catalysts, respectively.[9] Enantiomerically en-
riched oligomeric material (ratio R/S 33:1) was indeed ob-
tained, although the molecular weight was moderate at best
(3.0–4.0 kDa). It appeared that dehydrogenation of the alco-
hol end group to the ketone (creating chain stoppers) and
lipase deactivation (requiring
ITC of 1,3-diol and DIA: For the ITC of 1,3-diol and DIA,
we selected Shvoꢀs catalyst 2 as the racemization catalyst
and Novozym 435 as the acylation catalyst (Table 1,
entry 1). The reaction was performed in toluene at 708C
using 1 mol% of Ru catalyst
2 and 23 mg Novozym
435 mmol alcohol groups. In order to shift the equilibrium
to polymers, a reduced pressure of 280 mbar was applied to
remove isopropanol. 2,4-Dimethyl-3-pentanol (DMP) was
added as a hydrogen donor to suppress dehydrogenation of
Table 1. Results of ITC of secondary diols and DIA.[a]
the addition of fresh lipase
every 12 h) were primarily re-
sponsible for the lack of mo-
lecular weight built up.
Here, we report on our ef-
forts to improve the catalytic
performance of ITC in an
Entry[a] Monomer Ru
[mol%][b] [mgmmolÀ1
Nov435
DMP
[m]
t
Conv. ki ”103
eep
Mp
Ketone
[%]
[b]
[d]
]
[h]
[%][c]
[hÀ1
]
[%] [kDa]
1
2
3
4
5
6
7
8
1,3-diol
1,4-diol
3a
3b
3c
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1,3-diol
1
2
2
2
2
0.5
1
1
2
2
4
2
1
1
2
6
23
23
23
23
23
23
12
46
23
23
23
23
23
23
23
23
0.7
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.1
0.1
0.2
0.7
3.9
0.5
0.5
215
171
390
390
310
121
121
121
121
118
118
118
223
168
98
98
97
98
94
98
97
98
98
99
95
99
98
96
57
78
86
15.8
9.1
6.5
4.6
9.1
18.3
25.8
26.9
48.5
33.0
26.9
38.3
12.0
1.7
92
94
43
41
46
93
98
92
95
99
99
98
10.7
8.3
3.3
3.7
3.7
7.2
5.2
7.7
12.3
8.4
15.4
14.0
4.5
n.d.
n.d.
n.d.
0.5
2.4
n.d.
n.d.
n.d.
2.2
2.2
2.3
1.9
2.2
3.1
1.8
0.8
n.d.
n.d.
n.d.
AA–BB
polycondensation
system by selecting two cata-
lysts that are compatible and
complementary during the
polymerization process. Novo-
zym 435, Candida antarctica
9
10
11
12
13
14
15[e]
16[e]
lipase
B immobilized on a
95
n.d.
n.d.
n.d.
resin, was selected as the acyla-
tion catalyst since it is well
studied in DKR processes and
has proven itself as a robust
biocatalysts with a high enan-
12.5
13.9
98
n.d.=not determined. [a] Conditions, unless otherwise noted: 0.87 mmol diol, 0.87 mmol DIA, 3 Š MS, 2 mL
dry toluene, T=708C; p=280 mbar (argon). [b] The catalyst loadings are calculated with respect to the total
amount of alcohol groups. [c] Conversion is based on the total conversion of alcohol groups. [d] The initial
rate constant ki was determined by linear regression from the logarithmic plot of (1Àconversion of alcohol
groups) versus time. [e] 6 mL toluene was used instead of 2 mL.
tioselectivity for benzylic
R
secondary alcohols.[9,10] Shvoꢀs
8326
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8325 – 8332

Chiral Polyesters
FULL PAPER
end groups which results in inactive ketone chain ends. This
strategy was successfully employed by us previously.[8]
Figure 2a shows the time-conversion plot for the acylation
of the alcohol groups. Initially, the reaction proceeds quick-
ly, with a conversion of 61% reached within 24 h. At this
point, all R secondary alcohol functionalities have been acy-
lated and the racemization reaction has now become rate-
limiting. Although slower, conversion of the alcohol groups
to ester groups proceeded steadily beyond this point: after
215 h, a conversion level of 98% was observed. Figure 2b
shows the development of Àln(1Àconversion of alcohol
groups) versus time once the 50% conversion threshold has
been passed. An approximately linear relationship is ob-
served which is indicative for first order kinetic behavior.[11]
This linear relationship allows for the calculation of a
pseudo first order rate constant, in this case ki =15.8”
10À3 hÀ1.
Figure 3 shows the 1H NMR spectrum obtained for this
polymer. The end groups that appear at d 4.9 and 5.0 ppm
(designated with F and G in Figure 3) have almost fully dis-
appeared and the intrachain b-hydrogens are observed at d
5.8 ppm (designated with A). Analysis of the crude polymer
by gel permeation chromatography (GPC) indicated that
indeed polymeric material had been formed and a peak mo-
lecular weight Mp of 10.7 kDa was calculated (Figure 4). To
confirm the formation of an enantiomerically enriched poly-
ester, the polymer was degraded under basic conditions
(overnight stirring at room temperature in 0.5m NaOH in
ethanol), restoring the 1,3-diol. Subsequent analysis by
chiral GC indicated an ee of 92% for the alcohol groups in
the polymer. Atotal amount of 0.5% of ketone functionali-
ties was calculated from the chiral GC trace, which is sur-
prisingly low considering the dehydrogenation potential of
the Shvo catalyst.[12] Based on these results, we concluded
that ITC of 1,3-diol and DIAusing Novozym 435 and race-
mization catalyst 2 is feasible
Figure 2. a) Time-conversion plot for the ITC of 1,3-diol and DIAusing
Novozym 435 and Shvoꢀs catalyst 2; 1 mol% Ru; 23 mg Novozym 435
per mmol alcohol functionality; 0.7m DMP; T=708C; solvent: toluene;
p=280 mbar. The conversion was determined by 1H NMR. b) Logarith-
mic plot of À(1Àconversion) versus time.
to 97% conversion in 171 h. According to GPC, a polymer
was obtained with a peak molecular weight of 8.3 kDa. The
amount of ketone functionalities was 2.4%, while the ee of
this polymer was 94%.
and an enantio-enriched poly-
ester had been formed.
ITC of other diols and DIA:
In order to investigate the
scope of ITC employing AA–
BB type monomers, we investi-
gated the use of 1,4-diol and
aliphatic diols (Figure 5). In all
polymerizations, DIAwas used
as the diester. The reactions
were performed in toluene at
708C with 2 mol% of catalyst
2
and
23 mg
Novozym
435 mmol alcohol group. 0.5m
DMP was added as a hydrogen
donor under reduced pressure
(280 mbar) to remove isopro-
panol. The polymerization of
1,4-diol and DIA(Table 1,
entry 2) proceeded smoothly
Figure 3. 1H NMR spectrum of the polymer obtained from the ITC of 1,3-diol with DIA.
Chem. Eur. J. 2007, 13, 8325 – 8332
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8327

E. W. Meijer, A. R. A. Palmans et al.
case for aliphatic diols. Therefore, we first investigated the
enantioselectivity of Novozym 435 towards aliphatic diols in
a kinetic resolution experiment. 2,9-Decanediol (3c) was
subjected to enzymatic acylation as depicted in Scheme 3.
Figure 4. GPC trace of the crude polymer obtained from the ITC of 1,3-
diol with DIA, Mp =10.7 kDa. The peaks between 17 and 20 min arise
from catalyst residues.
Scheme 3. Lipase-catalyzed acylation of 2,9-decanediol with vinyl acetate
as the acyl donor.
To ensure a fast reaction, vinyl acetate was selected as the
acyl donor. If the enantioselectivity of Novozym 435 in the
acylation of the first secondary alcohol group is sufficiently
high, then a negligible amount of the SS diastereomer
would react away. The time-conversion plot of this reaction
is shown in Figure 6. The RR and RS diastereomers quickly
Figure 5. 1,4-Diol and aliphatic diols employed in ITC.
Aliphatic diols 3a–c were synthesized from the appropri-
ate dienes by first converting the latter to the diepoxide by
treatment with m-chloroperbenzoic acid. Reduction with
LiAlH4 furnished selectively the aliphatic secondary diols in
overall yields of 38 to 59% (Scheme 2). The amount of pri-
mary alcohol groups could be limited to 2–3% during the
reduction by carefully keeping the reaction mixture around
08C. Since primary alcohols are readily acylated by the
lipase, this impurity does not pose a problem in ITC. Subse-
quently, compounds 3a–c were subjected to ITC using DIA
as the diester (Table 1, entries 3 to 5). Gratefully, we con-
cluded that for all monomers indeed polymerization took
place and conversions of 94–98% were reached, although
longer reaction times were necessary to reach this conver-
sion level (310 to 390 h). The peak molecular weights of
these polymers were in the range of 3.3–3.7 kDa; considera-
bly lower than using aromatic diols. Presumably, these rela-
tively low molecular weights can be attributed to a relatively
high extent of dehydrogenation of end groups. Overlapping
Figure 6. Time-conversion plot for the lipase-catalyzed acylation of 2,9-
*
decanediol using vinyl acetate as the acyl donor; SS-diol ( ), RS-diol
~
*
(
), RR-diol ( ). Conditions: 25 mg Novozym 435, 1.15 mmol 3c,
4.64 mmol vinyl acetate, 5 mL toluene, T=708C. Conversions are calcu-
lated from chiral GC spectra.
react away, with—as expected—the highest reaction rate for
the RR diastereomer. However, the SS diastereomer also
shows considerable reactivity. This indicates that the enan-
tioselectivity for Novozym 435 is considerably lower for 3c
than for regular secondary alcohols, such as 2-octanol.
1
peaks in the H NMR spectra unfortunately hampered the
quantification of the amount of ketone end groups.
Surprisingly, the ee values of the polymer obtained from
3a–c were low with values ranging from 41 to 46%
(Table 1). Although the enantioselectivity of Novozym 435
towards linear secondary alcohols is very high (for 2-octanol
an E value of 340 was reported)[13] this is not necessarily the
Optimization study: Encouraged by the promising result of
the ITC of 1,3-diol and DIA, we selected this polymeri-
zation reaction for an optimization study. We focused on ob-
taining high molecular weight
polymers with high optical
purity, while minimizing the re-
action time. Since the molecu-
lar weight in a polycondensa-
tion is highly dependent on
stoichiometry, the percentage
Scheme 2. Synthesis of aliphatic secondary diols from the corresponding dienes.
of ketone end groups is the
8328
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8325 – 8332

Chiral Polyesters
FULL PAPER
only reasonable parameter to consider, since ketone end
groups act as chain stoppers. As DMP is used as a hydrogen
donor in order to suppress dehydrogenation, the concentra-
tion of DMP is thought to be of importance. The reaction
rates are expected to depend on the catalyst loadings. Since
the racemization rate is the rate limiting step after acylation
of all R alcohols (see above), increasing the amount of cata-
lysts 2 will increase the reaction rate. Increasing the amount
of Novozym 435 will also have a beneficial effect on the re-
action rate in the initial state of the reaction but is also ex-
pected to lower the enantioselectivity. Therefore, we evalu-
ated the effect of DMP concentration and the catalyst load-
ings employed on the kinetics, the optical purity and the
percentage of end groups in the isolated polymers. In all ex-
periments, first order kinetic behavior was observed after
50% of the alcohol groups were acylated. The calculated
rate constants are related to this stage of the reaction. The
results of this study are summarized in Table 1 (entries 6–
16).
Figure 7. Effect of the ruthenium loading on the kinetics of the polymeri-
zation (Table 1, entries 6, 8, 9, 10 and 11). The two experiments with
2 mol% Ru are at different DMP concentration and, therefore, display
different reaction rates.
Figure 7): the calculated rate constants were 18.3, 26.9 and
48.5”10À3 hÀ1, respectively. As shown by entries 10 and 11,
surprisingly, a further increase to 4 mol% leads to slightly
slower kinetics (right part of Figure 7): the calculated rate
constants were 33.0 and 26.9”10À3 hÀ1, respectively.[14]
Athreefold dilution of the system by using more solvent
resulted in a sharp decrease in reaction rate (entry 15); a
rate constant of only 12.5”10À3 hÀ1 was calculated. This is
attributed to slower racemization kinetics due to the lower
alcohol concentration.[15] An alternative explanation would
be that the equilibrium between the inactive dimeric ruthe-
nium precursor and the active 16-electron Ru0- and 18-elec-
tron RuII complexes is affected by the dilution.[16] However,
the experiment with a threefold dilution while keeping the
ruthenium concentration constant with respect to the undi-
luted experiments with 2 mol% (thus effectively increasing
the ruthenium loading to 6 mol%) shows that this is not the
case, as the kinetics of this experiment are equally slow with
a ki of 13.9”10À3 hÀ1 (entry 16).
Effect of the lipase loading: The amount of lipase that is
used in ITC may affect both the kinetics as well as the opti-
cal purity of the polymer obtained. In DKR experiments
using racemization catalyst 2, usually racemization is the
rate-limiting process, since the racemization activity of cata-
lyst 2 is much lower than the acylation activity of the lipase.
If this is also the case in ITC, the observed kinetics would
not change when increasing or decreasing the lipase catalyst
loading. The optical purity might decrease if the amount of
lipase is increased. The rate of acylation of S secondary al-
cohols is very low, and insignificant with respect to overall
rate of acylation. However, the effect of more lipase can
have a significantly negative effect on the ee of the polymer.
By performing the ITC at lipase catalyst loadings of 12 and
46 mg per mmol alcohol functionality, the effect on these
parameters was investigated. Entries 7 and 8 show that
indeed the lipase loading does not have an influence on the
net rate of reaction; the calculated reaction rate constants
for the experiments with 12 and 46 mg Novozym 435 per
mmol alcohol functionality were 25.8 and 26.9”10À3 hÀ1, re-
spectively. Apparently, the racemization is rate-limiting even
when using only 12 mg Novozym 435 per mmol alcohol
functionality. The optical purity of the polymer, however,
does depend on the amount of lipase used. Entries 7 and 8
clearly show that an increase in the amount of enzyme from
12 to 46 mg per mmol alcohol functionality leads to a de-
crease in the ee of the polymer from 98 to 92%.
Effect of the DMP concentration: DMP is added to the
polymerization system to suppress dehydrogenation of alco-
hols group. Therefore, it is of interest to investigate the
effect of different concentrations of DMP on the amount of
ketone end groups in the isolated polymers. Although DMP
is a sterically hindered hydrogen donor, it acts a competitive
substrate for the racemization catalyst. Thus, it is expected
that DMP has an influence on the racemization kinetics. By
changing the DMP concentration from 0.1 to 3.9m, these ef-
fects were investigated (entries 8 to 10 and 12 to 14). In-
creasing the DMP concentration from 0.1 to 0.5m lead to
significantly faster kinetics with ki values of 33.0, 38.3 and
48.5”10À3 hÀ1, respectively (entries 10, 12 and 9; left part of
Figure 8). Afurther increase from 0.5 to 3.9 m resulted in a
sharp decrease in reaction rate (entries 8, 13 and 14; right
part of Figure 8); the calculated ki values were 26.9, 12.0
and 1.7”10À3 hÀ1, respectively.[17] Apparently, the addition of
a limited amount of DMP results in faster kinetics; possibly
the change in polarity affects the equilibrium between inac-
tive dimeric ruthenium precursor and the active monomeric
Effect of the ruthenium loading: Since the racemization is
rate-limiting after 50% conversion, we expected that the ki-
netics are dependent on the ruthenium loading. To investi-
gate this, ITC was carried out with ruthenium loadings of
0.5 mol% to 4 mol% (entries 6 and 8 to 11). Also, the effect
of the concentration of ruthenium was investigated by dilut-
ing the system (entries 15 and 16). The results are visualized
in Figure 7. Entries 6, 8 and 9 show that an increase from
0.5 to 2 mol% results in a faster rate of reaction (left part of
Chem. Eur. J. 2007, 13, 8325 – 8332
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8329

E. W. Meijer, A. R. A. Palmans et al.
presented here illustrate that chiral polyesters are readily
accessible from optically inactive monomers by the concur-
rent use of different catalysts, a process referred to as itera-
tive tandem catalysis. The reaction times are still relatively
long which results from the slow racemization activity of the
Shvoꢀs catalyst under the conditions employed. Neverthe-
less, ITC is an effective way to prepare chiral polymers from
a variety of optically inactive monomers.
Experimental Section
Figure 8. Effect of the amount of hydrogen donor DMP on the kinetics
of the polymerization (Table 1, entries 11, 13, 10, 9, 14 and 15). The two
experiments with 0.5m DMP are at different catalyst loadings and, there-
fore, display different reaction rates.
Materials: 1,1’-(1,3-Phenylene)diethanol (1,3-diol) and 1,1’-(1,4-phenyl-
ene)diethanol (1,4-diol) were synthesized by reduction of the correspond-
ing ketone with NaBH4 according to a literature procedure.[18] Diisoprop-
yl adipate (DIA) was purchased from TCI Europe. Novozym 435 was
purchased from Novozymes A/S. All solvents were purchased from Bio-
solve and stored on dry molecular sieves 4 Š) prior to use to remove
traces of water. Shvoꢀs catalyst 2 was purchased from Strem Chemicals.
All other chemicals were purchased from Aldrich and used as received
unless otherwise noted.
Analytical methods: 1H and 13C NMR spectra were measured in CDCl3
at 400, 300 or 200 MHz (1H NMR) and 100, 75 or 50 MHz (13C NMR)
using a Varian Mercury Vx 400, 300 or 200 spectrometer. Chiral gas chro-
matography (GC) was performed on a Shimadzu 6C-17 AGC equipped
complexes, or in higher activity of these active complexes.
At higher concentrations, substrate competition by DMP
prevails, and net kinetics slow down severely. Finally, the
amount of ketone functionalities appeared to be independ-
ent of the DMP concentration; percentages of ketone
always were in the 0.5–3.1% range and no clear trend could
be observed.
In summary, the amount of lipase has no significant influ-
ence on the overall kinetics of ITC. The optical purity of the
polymer, however, was optimal at 12 mg lipase per mmol al-
cohol functionality. The overall kinetics were fastest when
using 2 mol% ruthenium catalyst reducing the reaction time
to ꢀ120 h. The DMP concentration had no influence on the
amount of ketone functionalities in the isolated polymers;
the overall kinetics of the polymerization, however, were
fastest at 0.5m. From these results, we concluded that the
optimal conditions for the ITC of 1,3-diol and DIAare
2 mol% ruthenium catalyst 2, 12 mg Novozym 435 per
mmol alcohol functionality and 0.5m DMP.
with
a Chrompack Chirasil-DEX CB (i.d.=0.25 mm, Df =0.25 mm)
column and an FID. Samples were injected using a Shimadzu AOC-20i
autosampler. Injection and detection temperatures were set at 250 and
3008C, respectively. Conversions were determined by the internal stan-
dard method using 1,3,5-tri-tert-butylbenzene as the internal standard.
The ee value was calculated as follows: ee=(RÀS)/
A
represent the area of the GC peaks of the R and S enantiomer, respec-
tively. Samples containing 1,3-diol were analyzed directly while samples
containing 1,4-diol were derivatized with butyric anhydride for 16 h at
808C in toluene prior to analysis. GC-MS spectra were taken with a Shi-
madzu GC-17 Aemploying a Zebron-ZB-5 column (i.d. =0.25 mm, Df =
0.25 mm). Injector and detector temperatures were set at 3008C. Gel per-
meation chromatography (GPC) was carried out on a Shimadzu HPLC
system equipped with a Shimadzu LC-10 AD VP pump, a Shimadzu
RID-10 Adifferential refractometer detector and two PL gel columns
(mixed C and mixed D, 10 mm, 300”7.5 mm, Polymer Laboratories),
using THF as the eluent. All molecular weights are given relative to poly-
styrene standards.
Typical procedure for ITC: Novozym 435 (40 mg),
2
(18 mg,
Conclusion
0.017 mmol), 3 Š molar sieves and a magnetic stirring bar were put in a
15 mL Schlenk tube. The tube was put overnight in a vacuum oven
(10 mm Hg) at 508C in presence of P2O5. The oven was backfilled with
nitrogen and the tube was removed from the oven. 1,3-Diol (144 mg,
0.87 mmol), DIA(199 mg, 0.87 mmol), dry toluene (2 mL) and 2,4-di-
methyl-3-pentanol (120 mg) were added to a 5 mL round bottom flask
and were stirred at 458C for 16 h in presence of 3 Š molar sieves to
remove traces of water. The mixture was allowed to cool to room tem-
perature and subsequently transferred to the Schlenk tube. Five vacuum-
argon cycles were performed to remove oxygen. The mixture was stirred
at 708C for 120 h at reduced pressure (280 mbar). During reaction, ali-
quots (ꢀ0.05 mL) were drawn from the reaction mixture using a syringe,
which was flushed with argon prior to use. The sample was diluted with
CDCl3 analyzed by 1H NMR. After reaction, the enzyme was removed
by filtration over a class 3 glass filter. The filter was flushed with di-
chloromethane. The crude product was obtained after removal of the or-
ITC of 1,1’-(1,3-phenylene)diethanol (1,3-diol) and DIAre-
sulted in a chiral polyester using Shvoꢀs catalyst 2 and Novo-
zym 435. An optimization study was performed, which lead
to the optimal conditions of 2 mol% 2, 12 mg Novozym 435
per mmol alcohol group in the presence of 0.5m DMP as
the hydrogen donor. With these conditions, chiral polymers
were obtained with peak molecular weights up to 15 kDa, ee
values up to 99% and with 1–3% ketone functionalities in
ꢀ120 h. Also with the structural isomer 1,4-diol, a chiral
polyester was obtained, albeit with lower molecular weight
(8.3 kDa) and slightly lower ee (94%), which can probably
be attributed to the lack of optimization for this particular
monomer. ITC using aliphatic diols also succeeded, but did
not lead to enantiopure polymers. At most, an ee of 46%
was obtained with low molecular weights in the range of
3.3–3.7 kDa. The latter was attributed to the low of selectivi-
ty of Novozym 435 for these secondary diols. The results
ganics by rotary evaporation as
a
brownish oil (120 mg). 1H NMR
(CH3)(OCO)), 2.53
(CDCl3, 258C, TMS): d=7.21 (m, Ar-H), 5.80 (m, CH
A
ACHTREUNG
(brs, Ar-CO-CH2), 2.30 (brs, OCOCH2), 1.55 (brs, OCOCH2CH2),
1.45 ppm (d, CH3).
1,2,9,10-Diepoxydecane: 1,9-Decadiene (10.6 g, 77 mmol) was dissolved
in chloroform (100 mL) and added dropwise in 1 h using a dropping
8330
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8325 – 8332

Chiral Polyesters
FULL PAPER
funnel to a suspension of 77% m-chloroperbenzoic acid (34 g, 197 mmol)
in chloroform (150 mL). The mixture was stirred overnight at room tem-
perature. Then, the solution was filtered over Celite and the filtrate was
washed with Na2S2O3 solution (150 mL, 10%), Na2CO3 solution (3”
150 mL, saturated) and brine (150 mL). The organic layer was dried with
MgSO4, filtered and concentrated in vacuo. The resulting clear liquid
508C per min to 2008C, isothermal at 2008C for 10 min, temperature gra-
dient 25 to 2258C): SS-diol 21.5 min, RS-diol 22.5 min, RR-diol 23.1 min.
1H NMR of the polymer obtained from the ITC of 2,8-nonanediol and
DIA(CDCl 3, 258C, TMS): d=4.98 (m, COOCH
(CH3)(OCO)), 4.05 (t, CH2(OCO)), 3.79 (m, CH
OCOCH2), 2.10 (s, COCH3), 1.78–1.22 (m, OCOCH2CH2 and COOCH-
(CH3)(CH2)2), 1.19 ppm (d, CH3).
ACHTREUNG
A
G
A
ACHTREUNG
1
A
ACHTREUNG
(12.4 g, 98%) was characterized by H NMR and used in the second step
without further purification. 1H NMR (CDCl3, 258C, TMS): d=2.90 (m,
CH2-O-CH), 2.75 (dd, CH2-O-CH), 2.45 (dd, CH2-O-CH), 1.2–1.8 ppm
(CH2-R).
Kinetic resolution of 2,9-decanediol using vinyl acetate as the acyl donor:
Novozym 435 (25 mg), 2,9-decanediol (200 mg, 1.15 mmol), vinyl acetate
(400 mg, 4.64 mmol), tri-tert-butylbenzene (40 mg, 0.16 mmol), dry tolu-
ene (5 mL), dry 2,4-dimethyl-3-pentanol (1 mL) and a magnetic stirring
bar were put in a 10 mL vial. The mixture was stirred at 708C. At 5, 10,
24, 40, 80, 160, 360 and 1500 min samples (ꢀ0.02 mL) were withdrawn
from the reaction mixture using a syringe, which was flushed with argon
prior to use. The sample was diluted with dichloromethane and the
enzyme was removed from the sample by filtration over cotton wool.
The samples were analyzed by chiral GC. For determination of the con-
versions of the separate enantiomers, the samples were derivatized using
trifluoroacetic anhydride prior to analysis. Temperature program underiv-
atized samples: 1258C isothermal for 20 min, temperature gradient 208C
per min to 2008C, isothermal at 2008C for 25 min, temperature gradient
25 to 2258C). GC retention times: diol 23.50 min, S-ester-S-alcohol
1,2,7,8-Diepoxyoctane: 1H NMR (CDCl3, 258C, TMS): d=2.90 (m, CH2-
O-CH), 2.75 (dd, CH2-O-CH), 2.45 (dd CH2-O-CH), 1.2–1.8 ppm (CH2-
R).
1
1,2,8,9-Diepoxynonane: H NMR (CDCl3, 258C, TMS): d=2.90 (m, CH2-
O-CH), 2.75 (dd, CH2-O-CH), 2.45 (dd CH2-O-CH), 1.2–1.8 ppm (CH2-
R).
2,9-Decanediol (3c): 1,2,9,10-Diepoxydecane (12.4 g, 76 mmol) was dis-
solved in diethyl ether (300 mL) and added dropwise in 1 h using a drop-
ping funnel to an ice cooled suspension of LiAlH4 (2.22 g, 59.25 mmol) in
diethyl ether (300 mL). The reaction was performed under a nitrogen at-
mosphere. The resulting suspension was stirred overnight. After comple-
tion of the reaction (as confirmed by 1H NMR), the mixture was
quenched with water and aqueous HCl was added (200 mL, 1m). The or-
ganic layer was separated and the aqueous layer was washed with di-
chloromethane (2”200 mL). The organic layers were combined and con-
centrated in vacuo. The resulting liquid was distilled (b.p. 1008C at p=
0.03 mbar). The product was characterized by 1H NMR, 13C NMR and
GC/MS. According to 1H NMR, the product contained 97% secondary
alcohol groups and 3% primary alcohol groups impurity (7.5 g, 60%).
Prior to GC analysis, the sample was derivatized using trifluoroacetic an-
hydride. 1H NMR (CDCl3, 258C, TMS): d=3.78 (m, CH3CH(OH)), 3.63
23.80 min, SS-diester 23.97 min, R-ester-R-alcohol
+ R-ester-S-alcohol
24.01 min, RS-diester 24.24 min, RR-diester 24.49 min. The peaks were
tentatively assigned based on deduction and mass balance consistency.
Temperature program derivatized samples: 1058C isothermal for 30 min,
temperature gradient 508C per min to 2008C, isothermal at 2008C for
10 min, temperature gradient 25 to 2258C). GC retention times: SS-diol
23.74 min, RS-diol 24.15 min, RR-diol 24.96 min.
Typical procedure for the hydrolysis of chiral polymers: Chiral polymer
(40 mg) was dissolved in EtOH (2 mL). NaOH (42 mg, 1.0 mmol) was
added, and the mixture was stirred for 16 h at room temperature. The
mixture was poured into a solution of 300 mg NH4Cl in 10 mL H2O. A
few drops of concentrated HCl were added, until pH ꢀ7 as evidenced by
the use of pH paper. The EtOH was removed by concentration in vacuo
and the mixture was extracted with ethyl acetate (3”20 mL). The organ-
ics were combined, washed with brine and dried over MgSO4. After con-
centration, the resulting mixture (40 mg) was analyzed by 1H NMR to
confirm full degradation of the polymeric material and subsequently ana-
lyzed by chiral GC for determination of the ee of the diol (the 1,4-diol
was measured after derivatization with butyric anhydride). GC retention
times: 1,3-Diol: diketone 7.76 min, R-alcohol-ketone 19.44 min, S-alco-
hol-ketone 20.06 min, SS-diol 32.1 min, RS-diol 32.8 min, RR-diol
34.9 min (temperature program 1508C isothermal for 45 min, tempera-
ture gradient 508C per min to 2008C, isothermal at 2008C for 40 min,
temperature gradient 25 to 2258C). 1,4-Diester: SS-diester 38.1 min, RS-
diester 38.4 min, RR-diester 38.6 min (temperature program 1608C iso-
thermal for 35 min, temperature gradient 408C per min to 2008C, isother-
mal at 2008C for 10 min).
(t, CH2CH2OH, impurity), 1.2–1.6 (m, CH(CH2)6), 1.18 ppm (d,
R
CH3CH(OH)); 13C NMR (CDCl3, 258C, TMS): d=67 (CH3CH(OH)), 39
(CH2), 30 (CH2), 26 (CH2), 24 ppm (CH3)); GC-MS (FW =174.28): m/z
(%): 155 (1); GC-FID retention times (temperature program: 1058C iso-
thermal for 30 min, temperature gradient 508C per min to 2008C, isother-
mal at 2008C for 10 min, temperature gradient 25 to 2258C): SS-diol
1
23.74 min, RS-diol 24.15 min, RR-diol 24.96 min. H NMR of the polymer
obtained from the ITC of 2,9-decanediol and DIA(CDCl 3, 258C, TMS):
d=4.98 (m, COOCH
A
G
N
A
ACHTREUNG
1.78–1.22 (m, OCOCH2CH2 and COOCH
N
N
CH3).
2,7-Octanediol (3a): Yield: 2.55 g (38%). B.p. 828C at p=0.03 mbar. Ac-
cording to 1H NMR, the product contained 98% secondary alcohol
groups and 2% primary alcohol groups. Prior to GC analysis, the sample
was derivatized using trifluoroacetic anhydride. 1H NMR (CDCl3, 258C,
TMS): d=3.78 (m, CH3CH(OH)), 3.63 (t, CH2CH2OH, impurity), 1.2–1.6
(m, CH
(CH2)4), 1.18 ppm (d, CH3CH(OH)); 13C NMR (CDCl3, 258C,
G
TMS): d=68 (CH3CH(OH)), 39 (CH2), 26 (CH2), 24 ppm (CH3); GC/MS
(FW =146.23): m/z (%): 126 (1); GC-FID retention times (temperature
program: 958C isothermal for 20 min, temperature gradient 508C per
min to 2008C, isothermal at 2008C for 10 min, temperature gradient 25
Acknowledgements
1
to 2258C): SS-diol 12.1 min, RS-diol 12.4 min, RR-diol 13.0 min. H NMR
of the polymer obtained from the ITC of 2,7-octanediol and DIA
The authors acknowledge National Research School Combination Catal-
ysis Controlled by Chemical Design (NRSC-C) for funding and Andreas
Heise, Gerard Verzijl and Rinus Broxterman from DSM and Emmo
Meijer from Unilever for stimulating discussions.
(CDCl3, 258C, TMS): d=4.98 (m, COOCH
(OCO)), 4.05 (t, CH2(OCO)), 3.79 (m, CH
OCOCH2), 2.10 (s, COCH3), 1.78–1.22 (m, OCOCH2CH2 and COOCH-
(CH3)(CH2)2), 1.19 ppm (d, CH3).
A
N
A
N
ACHTREUNG
A
ACHTREUNG
2,8-Nonanediol (3b): Yield: 3.14 g (49%). B.p. 958C at p=0.03 mbar.
According to 1H NMR, the product contained 98% secondary alcohol
groups and 2% primary alcohol groups. Prior to GC analysis, the sample
was derivatized using trifluoroacetic anhydride. 1H NMR (CDCl3, 258C,
TMS): d=3.78 (m, CH3CH(OH)), 3.63 (t, CH2CH2OH, impurity), 1.2–1.6
[1] J. C. Wasilke, S. J. Obrey, R. T. Baker, G. C. Bazan, Chem. Rev.
2005, 105, 1001–1020.
[2] D. E. Fogg, E. N. dos Santos, Coord. Chem. Rev. 2004, 248, 2365–
2379.
(m, CH
(CH2)5), 1.18 ppm (d, CH3CH(OH)); 13C NMR (CDCl3, 258C,
A
[3] A. Bruggink, R. Schoevaart, T. Kieboom, Org. Process Res. Dev.
TMS): d=68 (CH3CH(OH)), 39 (CH2), 30 (CH2), 26 (CH2), 23 ppm
(CH3); GC/MS (FW =160.25): m/z (%): 141 (1); GC-FID retention times
(temperature program: 958C isothermal for 20 min, temperature gradient
2003, 7, 622–640.
[4] S. F. Mayer, W. Kroutil, K. Faber, Chem. Soc. Rev. 2001, 30, 332–
339.
Chem. Eur. J. 2007, 13, 8325 – 8332
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8331

E. W. Meijer, A. R. A. Palmans et al.
[5] a) L. F. Tietze, Chem. Rev. 1996, 96, 115–136; b) L. F. Tietze, N.
Rackelmann, Pure Appl. Chem. 2004, 76, 1967–1983.
[10] A. O. Magnusson, M. Takwa, A. Hamburg, K. Hult, Angew. Chem.
2005, 117, 4658–4661; Angew. Chem. Int. Ed. 2005, 44, 4582–4585.
[11] The rate of racemization (which equals the rate of production of R-
secondary alcohols if racemization is rate-limiting) is assumed to be
first order in the concentration of the S alcohol. If indeed racemiza-
tion is rate-limiting and all alcohols groups are in the S configura-
tion, (and assuming that the changing polarity of the system plays
no significant role) the total rate of reaction (which equals the pro-
duction of R-secondary alcohols) will be first order in concentration
of the S alcohol. Therefore, once the 50% conversion threshold has
been crossed, a linear relationship should hold between Àln(1Àcon-
version) and time and a pseudo first order rate constant can be cal-
culated.
[12] J. H. Choi, N. Kim, Y. J. Shin, J. H. Park, J. Park, Tetrahedron Lett.
2004, 45, 4607–4610.
[13] C. Orrenius, F. Haeffner, D. Rotticci, N. Ohrner, T. Norin, K. Hult,
Biocatal. Biotransform. 1998, 16, 1–15.
[14] The two experiments with 2 mol% Ru are at different DMP con-
centration and, therefore, display different reaction rates
[15] The kinetics of racemization are assumed to be first order in alcohol
concentration.
[16] C. P. Casey, S. W. Singer, D. R. Powell, R. K. Hayashi, M. Kavana, J.
Am. Chem. Soc. 2001, 123, 1090–1100.
[6] 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) O. Pami›s, J.-E. Bäckvall, Chem. Rev. 2003, 103, 3247–
3261; c) O. Pami›s, J.-E. Bäckvall, Trends Biotechnol. 2004, 22, 130–
135; d) J. H. Choi, Y. K. Choi, Y. H. Kim, E. S. Park, E. J. Kim, M. J.
Kim, J. W. Park, J. Org. Chem. 2004, 69, 1972–1977; e) D. Lee, E. A.
Huh, M.-J. Kim, H. M. Jung, J. H. Koh, J. Park, Org. Lett. 2000, 2,
2377–2379; f) J. H. Koh, H. M. Jung, M.-J. Kim, J. Park, Tetrahedron
Lett. 1999, 40, 6281–6284; g) G. K. M. Verzijl, J. G. De Vries, Q. B.
Broxterman, Tetrahedron: Asymmetry 2005, 16, 1603–1610.
[7] a) M. T. Reetz, K. Schimossek, Chimia 1996, 50, 668–669; b) M. J.
Kim, W. H. Kim, K. Han, Y. K. Choi, J. W. Park, Org. Lett. 2007, 9,
1157–1159; c) S. Gastaldi, S. Escoubet, N. Vanthuyne, G. Gil, M. P.
Bertrand, Org. Lett. 2007, 9, 837–839; d) M. Stirling, J. Blacker,
M. I. Page, Tetrahedron Lett. 2007, 48, 1247–1250; e) J. Paetzold, J.-
E. Bäckvall, J. Am. Chem. Soc. 2005, 127, 17620–17621; f) O.
Pꢂmies, A. H. Ell, J. S. M Samec, N. Hermanns, J.-E. Bäckvall, Tetra-
hedron Lett. 2002, 43, 4699–4702.
[8] a) B. A. C. van As, J. van Buijtenen, A. Heise, Q. B. Broxterman,
G. K. M. Verzijl, A. R. A. Palmans, E. W. Meijer, J. Am. Chem. Soc.
2005, 127, 9964–9965; b) J. van Buijtenen, B. A. C. van As, J. Meul-
dijk, A. R. A. Palmans, J. Vekemans, L.A. Hulshof, E. W. Meijer,
Chem. Commun. 2006, 3169–3171.
[9] I. Hilker, G. Rabani, G. K. M. Verzijl, A. R. A. Palmans, A. Heise,
Angew. Chem. 2006, 118, 2184–2186; Angew. Chem. Int. Ed. 2006,
45, 2130–2132.
[17] The two experiments with 0.5m DMP are at different catalyst load-
ings and, therefore, display different reaction rates.
[18] J. S. Wallace, B. W. Baldwin, C. J. Morrow, J. Org. Chem. 1992, 57,
5231–5239.
Received: May 30, 2007
Published online: July 20, 2007
8332
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8325 – 8332