Chiral polyesters by dynamic kinetic resolution polymerization

Article abstract of DOI:10.1002/anie.200503496

(Figure Presented) Problem resolved: Enzymatic polymerization was combined with an in situ Rucatalyzed racemization process. The reaction mixture comprises a racemic mixture of a chiral diol and dimethyl adipate. Because of the high stereoselectivity of the enzyme Candida antarctica lipase B (CALB), the S-configured stereocenters of the chiral diol are practically nonreactive. Polycondensation was achieved by concurrent racemization of the S- into the R-configured centers.

Full text of DOI:10.1002/anie.200503496

Communications
Asymmetric Polymerization
reason is that in a typical polycondensation, significant
molecular weights can be realized only at an almost quanti-
tative monomer conversion. We anticipated that for chiral
monomers this can be achieved by a process analogous to the
DKR of small molecules.
Here we report on our investigation of a novel concept for
the synthesis of chiral polyesters, a lipase-catalyzed dynamic
kinetic resolution polymerization of racemic monomers. As
shown in Scheme 1, a mixture of stereoisomers of a secondary
DOI: 10.1002/anie.200503496
Chiral Polyesters by Dynamic Kinetic Resolution
Polymerization**
Iris Hilker, Gouher Rabani, Gerard K. M. Verzijl,
Anja R. A. Palmans, and Andreas Heise*
Chirality is one of the most intriguing features of natural
compounds. It determines, for instance, whether a molecule
has a beneficial biological function. Consequently, it is of
paramount importance in drug discovery to master asym-
metric synthesis. For this reason, a number of methods have
been developed in modern organic chemistry to obtain
enantiopure compounds from racemic mixtures, for example,
chemoenzymatic dynamic kinetic resolution (DKR).[1] In this
process, the enzymatic kinetic resolution of a racemic
compound is combined with the in situ chemical racemization
of the chiral center of the substrate. While the theoretical
yield in a normal kinetic resolution process is limited to 50%,
which corresponds to total conversion of the preferred
enantiomer, in DKR quantitative substrate conversion and
high optical purity (> 99% ee) can be obtained.
Despite the numerous asymmetric methods in organic
synthesis, concepts for making chiral synthetic polymers are
still limited. One obvious reason is the need for optically pure
monomers. Possible sources are naturally occurring optically
pure monomers such as l-lactide, which, however, limits the
range of available monomer building blocks.
An alternative is the direct resolution polymerization
from synthetic racemic monomers.[2] While most chemical
polymerization catalysts are nonstereoselective and therefore
not suited for the direct resolution of racemic monomer
mixtures, enzymes can be be employed successfully in kinetic
resolution polymerizations. In the recent past we and others
have shown this for the ring-opening polymerization (ROP)
of chiral caprolactones.[2c–e] This process yields polymers with
molecular weights of up to 5000 gmolÀ1 and with over
98% ee. However, due to the maximum conversion of 50%
in kinetic resolutions it cannot be applied in the polyconden-
sation of racemic diols and dicarboxylic acid derivatives. The
Scheme 1. Concept of dynamic kinetic resolution polymerization
(DKRP) from a racemic diol and a dicarboxylic acid derivative.
CALB=Candida antarctica lipase B.
diol is enzymatically polymerized with a difunctional acyl
donor (dicarboxylic acid derivative). Because of its enantio-
selectivity the lipase converts only the hydroxy groups at the
R-configured centers. In situ racemization of the hydroxy-
substituted stereocenters from the S to the R configuration
allows the polymerization to proceed to high conversion. We
recently reported the combination of racemization and
enzymatic ring opening of chiral lactones; the two reaction
steps were conducted alternating in separate reaction vessels
and low-molecular-weight oligomers were obtained (degree
of polymerization 3–5).[3] In contrast, our goal here is the
combination of DKR with a polymerization process in one
pot, thereby building up polymers of significant molecular
weight and with high optical purity.
The applied catalyst system consists of the Noyori-type
ruthenium catalyst 3 and an immobilized Candida antarctica
Lipase B (Novozym 435). This catalyst combination has been
used successfully for the DKR of monofunctional com-
pounds.[4] Since this combination tolerates a wide range of
acyl donors, we expected that it would also be suitable with
difunctional acyl donors for the formation of polyconden-
sates. Based on the high enantioselectivity of CALB in the
esterification of secondary benzylic alcohols (reactivity ratio
R/S ꢀ 1 ” 106),[5] we chose the racemic a,a’-dimethyl-1,4-
benzenedimethanol (1) as the diol component and dimethyl
adipate (DMA, 2) as the acyl donor (Scheme 2).
[*] Dr. I. Hilker, Dr. G. Rabani, G. K. M. Verzijl, Dr. A. Heise
DSM Research
P.O. Box 18, 6160 MD Geleen (The Netherlands)
Fax: (+31)46-476-0508
E-mail: andreas.heise@dsm.com
Dr. A. R. A. Palmans
Laboratory of Macromolecular and Organic Chemistry
Technische Universiteit Eindhoven
The selectivity of the catalyst combination with 2 as the
acyl donor was confirmed in a DKR of racemic (monofunc-
tional) 1-phenylethanol. The R-selective acylation proceeded
smoothly within 18 h to provide the diester, and subsequent
hydrolysis yielded pure (R)-1-phenylethanol with 99% ee
(chiral GC). Subsequently, the DKR polymerization was
conducted with 1 and 2 in a 1:1 molar ratio. In this process it is
important that the acyl donor is present in stoichiometric
P.O. Box 513, 5600 MB Eindhoven (The Netherlands)
[**] This work was supported in part by the Marie Curie Action RTN
program “Biocatalytic Approach to Material Design” (BIOMADE; I.H.
contract no. MRTN-CT-2004-505147) and by the NWO (grant to
A.R.A.P.).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
2130
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2130 –2132

Angewandte
Chemie
assess both the diols present as unreacted monomers and
incorporated in the polymer, the samples were hydrolyzed
prior to chiral-GC analysis. Before the start of the reaction,
the mixture shows the expected S,S/R,R/R,S diastereomer
ratio (1:1:2) corresponding to a R/S group ratio of 1:1
(Figure 1). During the course of the reaction all hydroxy
groups at R-configured centers are gradually incorporated
into the polymerization product while re-equilibration by the
racemization catalyst transforms free hydroxy-functionalized
S-configured centers into the reactive R configuration.
Consequently one expects the amount of S,S- and R,S-diol
to decrease and the relative R/S group ratio to increase.
Inspection of Figure 1 shows that after 30 h of reaction the S,S
enantiomer has indeed almost completely disappeared while
the ratio of R,R to R,S monomer is approximately 3:1 (R/S
ꢀ 7:1). After 70 h and 92% conversion of hydroxy groups, no
further conversion was observed yielding a final R,R/R,S ratio
of 16:1 (R/S ꢀ 33:1). It is reasonable to assume that the
racemization of the free diols as well as the esterification of
the alcohols in R configuration occurs quickly. In comparison,
long chains with a terminal hydroxy-substituted S-configured
center would racemize much more slowly. This suggests that
eventually all polycondensation products are end-capped
with S stereocenters (chain stoppers). As a result, the
racemization occurs at the end-groups of polymer chains of
gradually increasing length.
An average degree of polymerization can be calculated
from the final enantiomer composition if one assumes that all
polymer chains are end-capped with an R,S diastereomer on
both ends, in other words, each polymer chain contains two
such units. For the detected R,R/R,S ratio of 16:1 this means
an average degree of polymerization of 67 (32 ” RR + 2 ” RS
+ 33 ” 2), which corresponds to an maximum theoretical
number average molecular weight of ca. 9300 gmolÀ1. The
average molecular weight obtained from SEC analysis is
Mw = 3400 gmolÀ1 (Mn = 2100 gmolÀ1) based on polystyrene
standards (Figure 2). The discrepancy with the calculated
molecular weight can be explained partly by the difference in
the hydrodynamic volume between the SEC standards and
the polyester. Furthermore, the calculation does not take
further side reactions into account. 1H NMR spectra indicate
that approximately 3% of the hydroxy groups are oxidized to
Scheme 2. Reaction sequence for the one-pot DKR polymerization.
amounts, whereas the use of an excess is common practice in
conventional DKR. Control experiments proved that neither
the ruthenium catalyst nor Novozym 435 alone led to polymer
formation. As expected, with Novozym 435 only 50% con-
version and thus low-molecular-weight oligomers were
obtained since 1 consists of a statistical mixture of R,R, S,S,
and R,S diastereoisomers (1:1:2). The extremely low reac-
tivity of the hydroxy groups at the S-configured centers limits
chain growth by acting as chain stoppers in the enzymatic
(poly)esterification process. This proves that without trans-
formation of hydroxy-substituted stereocenters, higher
molecular weight products are not possible.
Polymers of higher molecular weight were obtained only
when the immobilized lipase and the racemization catalyst
were used in combination. The reaction was allowed to
proceed over four days, during which time the molecular
weight was seen to gradually increase to 3000–4000 gmolÀ1
according to size exclusion chromatography (SEC, polystyr-
ene standards). Moreover, the optical rotation of the reaction
mixture increased from À0.68 to 1288 during the process,
which further confirms the enrichment of R,R enantiomer in
the reaction product. These results suggest that the enzymatic
polymerization and the metal-catalyzed racemization work
concurrently.
Further evidence for the in situ racemization was obtained
by monitoring the change of the enantiomeric composition of
1 during the reaction by chiral GC (Figure 1). In order to
Figure 1. Change of overall stereoisomer composition of 1 during the
course of DKR polymerization determined by chiral GC after polyester
hydrolysis. Dashed line indicates the addition of 3. Fresh lipase was
added every 12 h.
Figure 2. SEC plot of polymer obtained after DKR polymerization
(Mw =3400 gmolÀ1; Mn =2100 gmolÀ1). tRet =retention time.
Angew. Chem. Int. Ed. 2006, 45, 2130 –2132
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2131

Communications
the ketone, which creates monofunctional chain stoppers.
This side reaction is caused by the racemization catalyst
(hydrogen transfer) and consequently lowers the obtained
molecular weight of the polymer. Minimizing all side
reactions should eventually result in molecular weights close
to the calculated one.
In conclusion we have shown that DKR can be combined
with enzymatic polymerization for the synthesis of chiral
polyesters from racemic secondary diols. This concept offers
an efficient route for the one-pot synthesis of chiral polymers
from nonnatural monomers. Further research will aim at
understanding the mechanism and suppressing side reactions
in order to increase the molecular weight of the polymeric
product.[6]
Received: October 3, 2005
Published online: February 27, 2006
Keywords: chiral resolution · enzyme catalysis · polymerization
.
[1] a) A. L. E. Larsson, B. A. Persson, J.-E. Bäckvall, Angew. Chem.
1997, 109, 1256; Angew. Chem. Int. Ed. Engl. 1997, 36, 1211; b) O.
Pami›s, J.-E. Bäckvall, Chem. Rev. 2003, 103, 3247; c) O. Pami›s,
J.-E. Bäckvall, TIBTECH 2004, 22, 130; 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; e) B. Martín-Matute, M. Edin, K. Bogµr, J.-
E. Bäckvall, Angew. Chem. 2004, 116, 6697; Angew. Chem. Int.
Ed. 2004, 43, 6535; f) G. K. M. Verzijl, J. G. De Vries, Q. B.
Broxterman, Tetrahedron: Asymmetry 2005, 16, 1603.
[2] a) S. Namekawa, S. Suda, H. Uyama, S. Kobayashi, Int. J. Biol.
Macromol. 1999, 25, 145; b) H. Kikuchi, H. Uyama, S. Kobayashi,
Macromolecules 2000, 33, 8971; c) T. F. Al-Azemi, L. Kondaveti,
K. S. Bisht, Macromolecules 2002, 35, 3380; d) J. Peeters, M. Veld,
F. Scheijen, A. Heise, A. R. A. Palmans, E. W. Meijer, Biomac-
romolecules 2004, 5, 1862; e) B. A. C. van As, P. Thomassen, B.
Kalra, R. A. Gross, E. W. Meijer, A. R. A. Palmans, A. Heise,
Macromolecules 2004, 37, 8973.
[3] 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.
[4] G. K. M. Verzijl, J. G. De Vries, Q. B. Broxterman, PCT Int. Appl.
WO 010396 A120011129, 2001.
[5] A. O. Magnusson, M. Takwa, A. Hamburg, K. Hult, Angew.
Chem. 2005, 117, 4658; Angew. Chem. Int. Ed. 2005, 44, 2.
[6] Supporting Information available: Experimental procedures,
monomer conversion plot.
2132
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2130 –2132