Enantiopure poly(glycidyl methacrylate-co-ethylene glycol dimethacrylate): A new material for supported catalytic asymmetric hydrogen transfer reduction

Article abstract of DOI:10.1016/S0957-4166(01)00129-X

A novel copolymer containing chiral epoxy residues was prepared. Free radical initiated suspension copolymerization of (R)- or (S)-glycidyl methacrylate with ethylene glycol dimethacrylate afforded crosslinked copolymer 1 in high yield. Optically active polymers containing amino alcohol functionalities were then formed from 1 through epoxide ring opening with a number of achiral and homochiral amines. It was shown that ruthenium complexes based on these new polymeric amino alcohol ligands were effective catalysts for the asymmetric hydrogen transfer reduction of acetophenone.

Full text of DOI:10.1016/S0957-4166(01)00129-X

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 811–815
Enantiopure poly(glycidyl methacrylate-co-ethylene glycol
dimethacrylate): a new material for supported catalytic
asymmetric hydrogen transfer reduction
Alice Rolland,a Damien He´rault,a Franc¸ois Touchard,a Christine Saluzzo,a Raphae¨l Duvalb and
Marc Lemairea,*
aLaboratoire de Catalyse et Synthe`se Organique, UMR 5622, UCBL, CPE, 43 Bd du 11 novembre 1918,
69622 Villeurbanne Cedex, France
bSocie´te´ CHIRALSEP, Parc d’activite´s de la Boissie`re, 11 rue de la Boissie`re, 76170 La Frenaye, France
Received 7 March 2001; accepted 15 March 2001
Abstract—A novel copolymer containing chiral epoxy residues was prepared. Free radical initiated suspension copolymerization
of (R)- or (S)-glycidyl methacrylate with ethylene glycol dimethacrylate afforded crosslinked copolymer 1 in high yield. Optically
active polymers containing amino alcohol functionalities were then formed from 1 through epoxide ring opening with a number
of achiral and homochiral amines. It was shown that ruthenium complexes based on these new polymeric amino alcohol ligands
were effective catalysts for the asymmetric hydrogen transfer reduction of acetophenone. © 2001 Elsevier Science Ltd. All rights
reserved.
1. Introduction
cobalt(III) complexes and their copolymerization with
ethylene glycol dimethacrylate. The synthesis of a fam-
ily of enantiomerically pure amino alcohol copolymers
obtained by epoxy ring cleavage of poly(glycidyl
methacrylate-co-ethyleneglycol dimethacrylate) and
their subsequent use as ligands in the asymmetric het-
erogeneous catalytic hydrogen transfer reduction of
acetophenone is also reported.
Natural optically active polymers such as proteins and
genes have played a major role in molecular recognition
and catalytic investigations owing to their specific chiral
structure. The preparation of optically active polymers
represents an interesting challenge as it involves asym-
metric polymerization1 or polymerization of an enan-
tiopure monomer.2
In 1975, Svec et al.3 reported the preparation of the
racemic crosslinked copolymer, poly(glycidyl methacry-
late-co-ethylene dimethacrylate). The appeal of this
material lies in the straightforward modification of the
epoxy group. Epoxide derived products have found use
in ion exchange chromatography,4–6 as ion exchange
resins,7–9 as gas chromatography stationary phases,10 as
gas sorbents,11,12 protecting groups13 and enzyme
immobilization agents.14 However, to our knowledge,
an enantiopure equivalent has not yet been synthesized.
2. Results and discussion
2.1. Hydrolytic kinetic resolution
Formation of the enantiopure glycidyl methacrylate
monomer was initially examined. An alternative to the
Sharpless method15 to obtain optically active epoxides
is hydrolytic kinetic resolution. One method uses a
hydrolytic enzyme to access optically active epoxy
esters. Ladner and Whitesides16 have shown that lipase
from porcine pancreas was able to hydrolyze racemic
esters of several saturated epoxy alcohols with useful
levels of enantioselectivity. Applied to racemic glycidyl
methacrylate, under mild conditions (pH 7.8, T=25°C)
this lipase (E.C. 3.1.1.3 sigma type II) permits the
isolation of (R)-glycidyl methacrylate with an e.e. of
70% at 57% conversion (E=6.5) (Scheme 1). Neverthe-
less, this system gave only the (R)-enantiomer.
Herein, we present results relating to the synthesis of
optically active glycidyl methacrylate monomers by
hydrolytic kinetic resolution involving lipase or salen
* Corresponding author. Tel.: 33-(0)4-72-43-14-07; fax: 33-(0)4-72-43-
14-08; e-mail: marc.lemaire@univ-lyon1.fr
0957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00129-X

812
A. Rolland et al. / Tetrahedron: Asymmetry 12 (2001) 811–815
Scheme 1.
Scheme 2.
In order to form (S)-glycidyl methacrylate, we used the
readily accessible chiral cobalt–salen complexes devel-
oped by Jacobsen et al.17 as catalysts for the asymmet-
ric hydrolysis of saturated terminal epoxides. Thus,
hydrolysis of racemic glycidyl methacrylate in the pres-
ence of 0.5 mol% of Co(III)-[(R,R)-salen](OAc) com-
plex, led to the (S)-enantiomer with excellent
enantioselectivity of up to 99% and 35% yield (S=15.6)
(Scheme 1). It is noteworthy that under the same
conditions, use of the Co(III)–[(S,S)-salen](OAc) com-
plex afforded the (R)-enantiomer in high selectivity and
workable yield.
agent, particle size3 depends mainly on the rate of
stirring during the polymerization reaction. Thus, the
compromise chosen was to perform the copolymeriza-
tion using 70 wt% of ethylene glycol dimethacrylate
with a mixture of cyclohexanol and dodecanol (91/9
wt/wt) as an inert solvent mixture and a stirring rate of
200 rpm in a 500 mL glass reactor equipped with a
stirring anchor. These conditions gave rise to spherical
particles with a specific surface area of between 80 and
100 m2 g−1 (determined by B.E.T.20), 87% of the sample
had particle size from 106 to 300 mm (Table 1).
2.3. Functionalization of poly(glycidyl methacrylate-co-
ethylene glycol dimethacrylate)
2.2. Polymerization
The epoxy functions of polymer 1 were subsequently
submitted to ring opening with benzylamine, N-benzyl-
We performed the copolymerization of (S)-glycidyl
methacrylate with ethylene glycol dimethacrylate, using
radical suspension polymerization with AIBN initiator3
(Scheme 2).
methylamine,
dimethylbenzylamine,
and methylamine to afford the polyamino alcohols
2a–2f, according to the procedure of Lindsay and
Sherrington8 which favors regioselective attack at the
(R)-a-methylbenzylamine,
(R)-N,a-
(S)-N,a-dimethylbenzylamine
In order to use this copolymer as a polymer supported
catalyst, we focussed our attention on the size of spher-
ical polymeric particles, the specific surface area and
also the pore sizes. Previously, Svec et al. showed that
specific surface area18 is predominantly affected by the
concentration of the inert phase (mixture of alcohols
acting as solvent) and the concentration of the
crosslinking agent. Whereas porosity and pore size19 are
mainly affected by the volume and the nature of the
porogenic component of the inert phase (aliphatic alco-
hol) and also by the concentration of the crosslinking
Table 1. Distribution of the particle size of copolymer
Fraction %
−¥B106 mm 106B¥B300
mm
300B¥B 500 500 mmB¥
mm
:0
87
13
:0
Scheme 3.

A. Rolland et al. / Tetrahedron: Asymmetry 12 (2001) 811–815
Table 2. Functionalization ratio of the epoxy polymer 1
813
Run
Polymer
R1
R2
R3
Amine configuration
Functionalization
Ratio (%) mmol/g
1
2
3
4
5
6
2a
2b
2c
2d
2e
2f
H
Me
H
Me
Me
H
H
H
Me
Me
Ph
H
Ph
Ph
Ph
Ph
Me
Me
69
71
70
59
60
36
1.20
1.22
1.21
0.94
0.95
0.76
R
R
S
Table 3. Ruthenium catalyzed transfer hydrogenation of acetophenone
Run
Polymer
R1
R2
R3
Time (h)
% Conversion
% e.e. (R)
1
2
3
4
5
6
2a
2b
2c
2d
2e
2f
H
Me
H
Me
Me
H
H
H
Me
Me
Ph
H
Ph
Ph
Ph
Ph
Me
Me
3
22
22
22
22
1
94
7
94
11
7
70
29
45
34
23
65
95
less hindered position of the epoxide ring (Scheme 3).
To determine the level of functionalization of the final
polymers 2, elemental analysis for nitrogen content was
completed. The results indicate that, in all cases, the
contents of amino alcohol units are, respectively, 70%
for 2a–2c, 60% for 2d and 2e and 36% for 2f (Table 2)
of the initial epoxide functional group.
which gave products with e.e.s of 70 and 65%, respec-
tively. In contrast, amino alcohol ligands from sec-
ondary amine (runs 2 and 4) led to the lowest
conversions with enantioselectivities of about 30%. The
chirality of the amino a-substituent had no influence on
the enantioselectivity.
An attempt to recycle the ruthenium complex 2a led to
a marked decrease in activity from 94 to 27% and
lowered selectivity from 70 to 54%.
2.4. Asymmetric hydrogen transfer reduction of
acetophenone
Amino alcohols are known to be ligands for asymmet-
ric homogeneous or heterogeneous catalysis for CꢀC,
CꢀO and CꢀH bond formations;21 this prompted us to
use polymers 2 as catalytic ligands in the asymmetric
hydrogen transfer reduction reaction, which has been
studied in our laboratory and is attractive because it
avoids the use of pressurized hydrogen which requires
special equipment.
3. Conclusions
We have synthesized enantiopure poly(glycidyl
methacrylate-co-ethylene glycol dimethacrylate) and
subsequently transformed it into amino alcohol deriva-
tives for use as a ruthenium ligand in the asymmetric
hydrogen transfer reduction of acetophenone leading to
encouraging results in terms of activity and enantiose-
lectivity.
Table 3 summarizes the results of the hydrogen transfer
reduction of acetophenone (Scheme 4), using ruthenium
complexes of amino alcohol polymers 2 synthesized
from [RuCl2(p-cymene)]2 precursor and iso-propanol as
a hydrogen source. The reaction was carried out under
argon with a ratio of acetophenone:Ru:ligand:t-BuOK
of 20:1:4:5.
In order to ascertain the scope and limitations of these
new catalytic systems different substrates and metals
will be investigated for transfer hydrogenation reac-
tions. The catalytic properties of such polymeric ligands
As shown in Table 3, amino alcohol ligands derived
from primary amine (runs 1, 3 and 6) gave the best
conversions and the best e.e.s. However, a-substituted
primary amine (run 3) led to a lower e.e. of 45%
compared to unsubstituted amines (runs 1 and 6),
Scheme 4.

814
A. Rolland et al. / Tetrahedron: Asymmetry 12 (2001) 811–815
in other reactions will also be examined. Additionally
appropriate derivatization 1 will allow its use in solid-
phase combinatorial chemistry and chiral chromato-
graphy.
room temperature, the solvent was removed in vacuo.
The black residue obtained was cooled to 0°C, and
treated with ( )-glycidyl methacrylate (19 g, 133 mmol)
followed by very slow addition of bidistilled water (1.2
g, 66 mmol, 0.55 equiv.). The reaction mixture was
allowed to stir for 24 h at room temperature.
4. Experimental
4.1. General
The product was separated from the diol by silica gel
column chromatography (Merck 60, 40–60 mm). (S)-
glycidyl methacrylate (6.65 g, 46.55 mmol, yield 35 %)
e.e.=99.8%; [h]D=+29.3 (c=0.564, CH2Cl2).
For the hydrolytic kinetic resolution of glycidyl
methacrylate and hydrogen transfer reduction of acet-
ophenone, enantiomeric excess and conversion values
were determined by GC on a Supelco b dex 225 (30
m×0.25 mm) chiral column, using a Shimadzu GC-14A
apparatus equipped with FID connected to a Shimadzu
C-R6A integrator. 1H and 13C NMR spectra were
recorded with an AM200 (1H: 200 MHz, 13C: 50 MHz)
using TMS as internal standard and CDCl3 as solvent.
4.4. Copolymerization
A solution of AIBN (204 mg) in a mixture of (S)-gly-
cidyl methacrylate (6.13 g) and ethylene glycol
dimethacrylate (14.30 g) (30/70 wt/wt) was added to a
solution of cyclohexanol (24.64)/dodecanol (2.43) (91/9
wt/wt). This mixture was then added to a solution of
polyvinylpyrrolidone (1.5 g) in water (150 mL). The
reaction mixture, formed in this way, stirred at 200
rpm, was heated at 70°C for 2 h then, at 80°C for 6 h.
After 2 h at room temperature, the spherical particles
formed were washed 4–5 times with ethanol 95%, dried
by means of a vacuum oven and sifted to afford 1.
Calcd: C, 60.17; H, 7.06; O, 32.77. Found: 59.75; H,
7.40; O, 32.85%. Functional: 2.11 mmol/g.
Polarimetric measurements were performed on
Perkin–Elmer 241 instrument, at ambient temperature,
at 589 nm, at a concentration of grams per 100 mL.
a
Elemental analyses were carried out by CNRS (Service
Central d’Analyse-De´partement Analyse Ele´mentaire),
´
Solaize, France. Granulometric analyses were obtained
using a Coulter LS230 (small volume module). B.E.T.
measurements were performed on an automatic home
made ‘Institut de Recherche sur la Catalyse’ apparatus,
at a (N2) temperature of −196°C. Before any measure-
ment, the support was heated to 240°C for 3 h in
vacuo. The Roberts’ model was used to determine the
pore size.
4.5. Typical polyamino alcohol synthesis
Under argon, 3 equivalents of amine were added to a
suspension the chiral copolymer (1 mmol/g of epoxide)
in anhydrous DMF (2 mL) under mechanical stirring
(180 rpm). Then the reaction mixture was heated at
100°C for 22 h.
4.2. Hydrolytic kinetic resolution with lipase
Elemental analyses:
2a: Calcd: C, 62.64; H, 7.24; O, 28.45; N, 1.67. Found:
C, 61.05; H, 7.15; O, 30.1; N, 1.15%; functional: 69%,
1.20 mmol/g.
2b: Calcd: C, 64.07; H, 7.48; O, 26.10; N, 2.36. Found:
C, 62.39; H, 7.67; O, 28.23; N, 1.71%; functional: 71%,
1.22 mmol/g.
2c: Calcd: C, 64.07; H, 7.48; O, 26.10; N, 2.36. Found:
C, 61.70; H, 7.50; O, 29.15; N, 1.66%; functional: 70%,
1.21 mmol/g.
2d Calcd: C, 64.57; H, 7.63; O, 25.49; N, 2.30. Found:
C, 62.40; H, 7.75; O, 28.50; N, 1.35%; functional: 59%,
0.94 mmol/g.
2e Calcd: C, 64.57; H, 7.63; O, 25.49; N, 2.30. Found:
C, 61.75; H, 7.41; O, 29.45; N, 1.38%; functional: 60%,
0.95 mmol/g.
2f Calcd: C, 58.85; H, 7.62; N, 2.78. Found: C, 58.45;
H, 7.25; N, 1.01%; functional: 36%, 0.76 mmol/g.
A mixture of racemic glycidyl methacrylate (10 g, 70.3
mmol) and bidistilled water (80 mL) was added to
lipase (10 g) (E.C.3.1.1.3, sigma type II, from porcine
pancreas). The reaction mixture was allowed to stir
vigorously maintaining the pH of the reaction at 7.8 by
addition of 1.4 M NaOH with a pH controller, for
24 h (conversion 57%). After filtration of the crude
product over silica gel (300 g), which was washed once
with DCM (300 mL), the organic layer was washed
with 10% aqueous NaHCO3 (25 mL) and twice with
bidistilled water (15 mL). The resulting organic extracts
were dried over a mixture of Na2SO4/NaHCO3 (90/10
wt/wt) then concentrated to yield (S)-glycidyl
methacrylate (3.11 g, 21.9 mmol, 31.2%) e.e.=70%;
[h]D=−20.6 (c=0.564, CH2Cl2); 1H NMR: 1.95 (s, 3H),
2.6–2.7 (m, 1H), 2.8–2.9 (m, 1H), 3.25 (m, 1H), 4.0–4.1
(m, 1H), 4.4–4.5 (m, 1H), 5.62 (s, 1H), 6.18 (s, 1H); 13C
NMR: 18.3, 44.6, 49.4, 65.2, 126.2, 135.9, 167.0.
4.6. Typical reduction procedure of acetophenone
4.3. Hydrolytic kinetic resolution with salen
Under argon, the polyamino alcohol 2 and [RuCl2(p-
cymene)]2 (2/Ru ratio=4:1) in suspension in iso-
propanol (2 mL for 3.7 mg of [RuCl2(p-cymene)]2) were
stirred and heated at 80°C for 30 min. The reaction
mixture and the polymer underwent a color change
Acetic acid (76 mL, 1.336 mmol) was added to a solu-
tion of (1R,2R)-1,2-diaminocyclohexane N,N% bis (3,5-
ditertbutylsalicylidene) cobalt(II) (0.457 g, 0.668 mmol)
dissolved in toluene (12 mL). After stirring for 1 h at

A. Rolland et al. / Tetrahedron: Asymmetry 12 (2001) 811–815
815
10. Onjia, A.; Milonjic, S. K.; Jovanovic, N. N.; Jovanovic,
S. M. React. Funct. Polym. 2000, 43, 269–277.
11. Belyakova, L.; Kiselev, A.; Platonova, N.; Kalal, J.; Svec,
F.; Hradil, J. Angew. Makromol. Chem. 1981, 96, 69–84.
12. Hradil, J.; Svec, F.; Kalal, J.; Belyakova, L.; Kiselev, A.;
Platonova, N.; Shevchenko, T. React. Polym. 1982, 1,
59–65.
from colorless to red and were cooled to room temper-
ature. Acetophenone was added (acetophenone/Ru
ratio: 20/1), then a solution of potassium tert-butoxide
(0.03 mol/L) is added (Ru/base ratio: 1/5). The reaction
times and e.e. are summarized in Table 3.
13. Frechet, J.; Bald, E.; Svec, F. React. Polym. 1982, 1,
21–26.
References
14. Petro, M.; Svec, F.; Frechet, J. Biotechnol. Bioeng. 1996,
49, 355–363.
15. Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.;
Masamune, H.; Sharpless, K. B. J. Am. Chem. Soc. 1987,
109, 5765–5780.
16. Ladner, W. E.; Whithesides, G. M. J. Am. Chem. Soc.
1984, 106, 7250–7251.
17. Tokunaga, M.; Larrow, J. F.; Kakiuchi, F. E.; Jacobsen,
N. Science 1997, 277, 936–938.
18. Horak, D.; Svec, F.; Bleha, M.; Kalal, J. Angew. Makro-
mol. Chem. 1981, 95, 109–115.
19. Horak, D.; Svec, F.; Ilavsky, M.; Bleha, M.; Baldrian, J.;
Kalal, J. Angew. Makromol. Chem. 1981, 95, 117–127.
20. Brunauer, S.; Emmet, P. H.; Teler, J. J. Am. Chem. Soc.
1938, 60, 309–319.
1. Okamoto, Y.; Nakano, T. Chem. Rev. 1994, 94, 349–372.
2. Chiellini, E.; Solaro, R.; Galli, G.; Ledwith, A. Adv.
Polym. Sci. 1984, 62, 144–169.
3. Svec, F.; Hradil, J.; Coupek, J.; Kalal, J. Angew. Makro-
mol. Chem. 1975, 48, 135–143.
4. Svec, F.; Frechet, J. J. Chromatogr. A 1995, 702, 89–95.
5. Svec, F.; Frechet, J. Biotechnol. Bioeng. 1995, 48, 476–
480.
6. Pohl, C.; Stillian, J.; Jackson, P. J. Chromatogr. A 1997,
789, 29–41.
7. van Berkel, P. M.; Sherrington, D. C. Polymer 1996, 37,
1431–1435.
8. Lindsay, D.; Sherrington, D. C. React. Polym. 1985, 3,
327–339.
9. Hainey, P.; Sherrington, D. C. React. Funct. Polym. 2000,
43, 195–210.
21. Fache, F.; Schulz, E.; Tommasino, M.; Lemaire, M.
Chem. Rev. 2000, 100, 2159–2231.
.
.