Enantioselective, Stereoconvergent Resolution Copolymerization of Racemic cis-Internal Epoxides and Anhydrides

Article abstract of DOI:10.1002/anie.202011259

Unprecedented enantioselective resolution copolymerization of racemic cis-internal epoxides and anhydrides was mediated by dinuclear aluminum complexes with multiple chirality, affording optically active polyesters with two contiguous stereogenic centers, and the unreacted substrates in good enantioselectivity. Unexpected stereoconvergence is observed in this resolution copolymerization, where the selectivity factor for the enantioselective formation of copolymer significantly exceeds the kinetic resolution coefficient based on the unreacted epoxide at various conversions. Catalytic activity and copolymer enantioselectivity are strongly influenced by the phenolate ortho-substituents of the ligand set, as well as the axial linker and its chirality. An enantiopure binaphthol-linked bimetallic Al^III complex allows stereoconvergent access to the stereoregular semi-crystalline polyesters and a concomitant kinetic resolution of the epoxide substrates.

Full text of DOI:10.1002/anie.202011259

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: Enantioselective, Stereoconvergent Resolution Copolymerization
of Racemic cis-Internal Epoxides and Anhydrides
Authors: Xiao-Bing Lu, Guang-Hui He, Bai-Hao Ren, Shi-Yu Chen, and
Ye Liu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202011259
Link to VoR: https://doi.org/10.1002/anie.202011259

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
Enantioselective, Stereoconvergent Resolution Copolymerization
of Racemic cis-Internal Epoxides and Anhydrides
[a]
[a]
[a]
[a]
[a]
Guang-Hui He , Bai-Hao Ren , Shi-Yu Chen , Ye Liu and Xiao-Bing Lu *
[
a]
G.-H. He, B.-H. Ren, S.-Y. Chen, Y. Liu, Prof. Dr. X.-B. Lu
State Key Laboratory of Fine Chemicals
Dalian University of Technology
2
Linggong Road, Dalian, 116024 (China)
E-mail: xblu@dlut.edu.cn
Supporting information for this article is given via a link at the end of the document.
Abstract: Enantioselective copolymerization of unsymmetric
disubstituted epoxides (e.g. racemic cis-internal epoxides) is a very
challenging issue, due to their poor reactivity, and complicated stereo-
and affords two separable products, an enantiopure isotactic
polymer and the unreacted enantiomer in the solution. A
representative example is the binaphthol-linked dinuclear cobalt
complex mediated enantioselective resolution polymerization of
racemic terminal epoxides to afford isotactic polyethers and
enantiopure epoxides, which exhibited high activity and
enantioselectivity with a selectivity factor (s-factor) in the range of
60−370.[5-7]
When this powerful strategy is applied to the enantioselective
resolution copolymerization of racemates with achiral
nucleophiles, optically active copolymers with main-chain chirality
can be formed.[8] Employing this strategy, several groups
independently explored the kinetic resolution copolymerization of
/
regio-selectivity. Herein, we present an unprecedented
enantioselective resolution copolymerization of racemic cis-internal
epoxides and anhydrides mediated by dinuclear aluminum complexes
with multiple chirality, affording optically active polyesters with two
contiguous stereogenic centers, and the unreacted substrates in good
enantioselectivity. An unexpected stereoconvergence is observed in
this resolution copolymerization, where the selectivity factor (s-factor)
for the enantioselective formation of copolymer significantly exceeds
the kinetic resolution coefficient (krel) based on the unreacted epoxide
at various conversions. The catalytic activity and copolymer
enantioselectivity are revealed to be strongly influenced by the
phenolate ortho-substituents of the ligand set, as well as the axial
linker and its chirality. The use of enantiopure binaphthol-linked
bimetallic (R,R,R,R,R)-Al(III) complex without any substituent at the
2
racemic terminal epoxides and CO to afford the alternating
polycarbonates with more than 95% head-to-tail linkages;
however, the kinetic resolution coefficients (krel) for all tested chiral
catalyst systems were less than 25.[
9-12]
Contrary to the limited
phenolate ortho-positions of the ligand allows not only
a
success of this type of kinetic resolution copolymerization, the
asymmetric copolymerization of meso-epoxides and CO or cyclic
stereoconvergent access to the highly stereoregular semi-crystalline
polyesters with an unprecedented levels of enantioselectivities of up
to 99% ee (with a s-factor of 437), but also a concomitant kinetic
resolution of the epoxide substrates (with a krel of 118). Mechanistic
studies performed by combining experimental observations and
density functional theory calculations suggest that both the
enantiomorphic site of the ligand set and the propagating chain-end
control mechanisms determined the enantioconvergent access to
enantioenriched copolymers.
2
anhydrides achieved significant progress through the use of
enantiopure bimetallic catalyst systems (Scheme 1A), where the
resulting polycarbonates or polyesters possessed completely
alternating microstructures, low polydispersity index, and up to
99% enantioselectivity.[13-16] Prior to these studies, Williams and
coworkers significantly contributed toward developing distinct
bimetallic catalysts for the copolymerization of epoxides with CO
2
or anhydrides.[17-19] More recently, our group achieved highly
enantioselective resolution copolymerization of racemic terminal
epoxides and anhydrides mediated by binaphthol- or
hydrogenated binaphthol-linked bimetallic complexes, exhibiting
both krels and s-factors of greater than 300 (Scheme 1B).[20]
Introduction
In comparison with simple terminal epoxides with a chiral
center, unsymmetric disubstituted epoxides contain two chiral
carbons, as well as trans- and cis-structure. Although
enantioenriched cis-epoxides can be efficiently synthesized via
the asymmetric epoxidation of cis-olefins,[21] the use of readily
Most of the naturally occurring polymers are optically active,
and some of these exhibit unique characteristics, such as
molecular recognition ability, resulting from their distinctive chiral
structure.[
1,2]
In the realm of synthetic chemistry, asymmetric
polymerization is one of the most challenging methods to prepare
enantioenriched polymers with configurational main-chain
chirality.[3] In terms of the reaction process, asymmetric
polymerization can be classified into two categories: asymmetric
synthesis polymerization of optically inactive prochiral monomers
via enantioselective catalysis; and enantiomer-selective
polymerization of only one enantiomer of the racemic monomers
with completely configurational retention in the kinetic
resolution.[ As most of the racemates are readily available and
inexpensive, the latter method has attracted significant interest
available and inexpensive racemic epoxides through the
catalytically kinetic resolution provides a competitive alternative
route to afford an enantiomerically pure form.[22,23] However, in
sharp contrast to the extensive literatures on the kinetic resolution
of racemic terminal epoxides, only few studies on the catalytically
kinetic resolution of racemic cis-internal epoxides have been
reported. In 2004, Umani-Ronchi et al. reported an effective
methodology for the kinetic resolution of 2,3-disubstituted
aromatic epoxides through a CC bond forming reaction using 2-
methylindole as the resolving reagent and chiral Cr(salen)
4]
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
complexes as catalysts, affording
the indolyl derivatives and the
unreacted
enantiomeric excess (ee).
Kobayashi and
discovered that the niobium
complex of tetradentate
binaphthol derivative, meso-
epoxides
in
high
Later,
[
24]
coworkers
a
a
desymmetrization catalyst, was also
effective in the regio- and
stereoselective ring-opening of
unsymmetrical cis-epoxides through
a kinetic resolution process using
anilines as nucleophiles.[25]
practically useful kinetic resolution
A
Figure 1. Structure of employed Al(Ⅲ) complexes
of racemic epoxides was achieved
by using chiral phosphoric acid as a catalyst and carboxylic acids
as activators at an extremely low reaction temperature of −50
oC.[26] More recently, Chen et al. reported a catalyst-site-controlled
Results and Discussion
diastereoselective polymerization methodology that directly
polymerized the diastereomeric mixtures of eight-membered
Enantioselective resolution copolymerization of racemic
-oxabicyclo[4,1,0]hept-2-ene (CHDO) and phthalic
anhydride (PA). We initiated our studies by using the alternating
copolymerization of racemic 7-oxabicyclo[4,1,0]hept-2-ene
7
diolide
monomers
into
stereosequenced
crystalline
[
27]
polyhydroxyalkanoates with adjustable properties.
Herein, we present an unprecedented stereoconvergent
resolution copolymerization of racemic cis-internal epoxides with
anhydrides using multi-chiral and bimetallic synergistic catalysis,
where the s-factors for the enantioselective formation of
copolymers are significantly higher than the krels based on the
unreacted epoxides, as the catalyst systems predominantly
dictate the configuration of the newly formed stereogenic centers
via stereoselective ring-opening, regardless of the extant
stereochemistry of the substrate (Scheme 1C). This phenomenon
is distinct to the principle in a traditional kinetic resolution process,
where the krel is defined as being equal to the s-factor.
(
CHDO) and phthalic anhydride (PA) as a model reaction, and an
enantiopure dinuclear Al(Ⅲ) complex 1 based catalyst system,
which has been reported to be highly active and enantioselective
for copolymerizing meso-epoxides and cyclic anhydrides.[
When activated with bis(triphenylphosphine)-iminium chloride
15]
(
(
PPNCl), the biphenol-linked dinuclear Al(Ⅲ) complex 1 with
R,R,R,R)-configuration showed moderate activity with a turnover
frequency (TOF) of 123 h-1 at ambient temperature (Table 1, entry
1
). This enantioselective resolution copolymerization afforded
enantioenriched (+)-(2R,3S)-CHDO with 76% enantiomeric
excess (ee) at 49% conversion, and an optically active polyester
= -129o (c = 1 in CHCl
with a specific rotation value of [α] ),
3
20
D
Scheme 1. Synthetic strategies of chiral polyesters from various epoxides by enantioselective copolymerization with anhydrides
2
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
Scheme 2. Stereochemical transformation involved in the ring-opening copolymerization of racemic cis-CHDO.
corresponding to 82% ee for (R,R)-configuration based on the
resulting diol after the copolymer hydrolysis. Notably, both
linked complex (R,R,R,R,R)-2a bearing tert-butyl groups on the
phenolate ortho-positions (Entry 2), while slightly less-hindered
(R,R,R,R,R)-2b bearing isopropyl groups showed a high activity
(
2S,3R)-CHDO and (2R,3S)-CHDO have two available chiral
carbons for the nucleophilic ring-opening reaction, resulting in
four divergent ring-opening pathways with different
-1
of up to 308 h , where the krel and s-factor values were 25 and 43,
respectively (Entry 3). Replacement with a further less-hindered
(R,R,R,R,R)-2c without any substituent increased both the krel
and s-factor values to a certain extent (Entry 4). For binaphthol-
linked complexes 3a–3f, similar effects of the steric hindrance of
the phenolate ortho-position substituents on the catalytic activity,
as well as the krel and s-factor were also observed (Entries 5–8),
with the highest activity of 363 h-1 obtained for the (R,R,R,R,R)-
3d system (Entry 8). In this case, a conversion of 43.5% was
achieved within 18 minutes at ambient temperature, and the
polyester obtained in 98% ee exhibited a high s-factor of 226,
which was significantly greater than the krel value of 44, indicating
an unexpected stereoconvergence for copolymer formation. The
change of the phenolate para-position substituents from tert-butyl
to fluoro or methoxyl groups had limited effect on the catalytic
activity and polyester enantioselectivity (Entries 9 and 10). The
use of the catalyst with (S,S,S,S,S)-configuration afforded
enantioenriched (-)-(2S,3R)-CHDO and polyester with (S,S)-
configuration (Entry 11). Surprisingly, when (R,R,S,R,R)-3f was
employed, the (S,S)-configured polyester was obtained in 88% ee
(Entry 12). This result implied the strong chirality inducing ability
of the chiral axial linker. For further investigating the influence of
the axial linker on the chiral induction, complex 4 bearing (R)-
binaphenol linker and achiral ethylenediamine backbone was
prepared and tested for the racemic CHDO/PA copolymerization.
The polymerization predominantly afforded (R,R)-enriched
regioselectivities or stereoselectivities in the copolymerization
with cyclic anhydrides. However, only two types of configured
chiral diol-units ((R,R)-units and (S,S)-units) could be obtained
through the hydrolysis of the resultant polyesters (Scheme 2).
Considering the unknown and complexity of the four possible
chiral transformation processes, the krel =ln[(1-c)(1-eeepoxide)]/ln[(1-
c)(1+eeepoxide)]
and
s-factor=ln[1-c(1+eepolyester)]/ln[1-c(1-
eepolyester)] were determined to separately evaluate the unreacted
epoxide and copolymer enantioselectivities.
The biphenol-linked dinuclear Al(III) complex (R,R,R,R)-1 has
two diastereoisomers with (R)- or (S)-biphenol stereochemistry,
(
R,R,R,R,R)- and (R,R,S,R,R)-stereoisomers, of which the former
is the matched configuration for better activity and
enantioselectivity in recent studies on the asymmetric
copolymerization involving
meso-epoxides.[14] Therefore,
enantiopure dinuclear Al(Ⅲ) complexes 2 and 3 bearing the
configuration-matched binaphthol and hydrogenated-binaphenol
linkers were synthesized for CHDO/PA copolymerization. To our
delight, most catalyst systems exhibited excellent activities (TOF
-1
>
300 h ) and enhanced enantioselectivities compared to those
with the biphenol-linked Al(Ⅲ) complex 1 catalyst system under
same conditions. The steric hindrance of the phenolate ortho-
position substituents of the ligand strongly influenced the catalytic
activity, the krel, and polymer enantioselectivity (corresponding to
the s-factor). For example, no or very low activity was observed
with the catalyst system based on hydrogenated-binaphenol-
3
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
Table 1. Enantioselective copolymerization of racemic cis-CHDO and PA[a]
Conv.
Specific
rotation [i]
TOF[c]
eeepo[d]
eepro[f]
M
[h]
n
Temp.
(oC)
Time
(h)
【b】
[e]
s-factor[g]
Ð [h]
Entry
Cat.
k
rel
-1
(h )
(%)
(%)
(kg/mol)
o
(%)
( )
1
2
3
4
5
6
7
8
9
(R,R,R,R)-1
(R,R,R,R,R)-2a
(R,R,R,R,R)-2b
(R,R,R,R,R)-2c
(R,R,R,R,R)-3a
(R,R,R,R,R)-3b
(R,R,R,R,R)-3c
(R,R,R,R,R)-3d
(R,R,R,R,R)-3e
(R,R,R,R,R)-3f
(S,S,S,S,S)-3f
(R,R,S,R,R)-3f
(R)-4
25
25
25
25
25
25
25
25
25
25
25
25
25
0
1
4
49.0
<2
123
--
76(2R,3S)
--
19
--
82(R,R)
--
24
--
5.6
--
1.31
--
-129
--
0.4
1
49.3
45.9
45.4
49.7
42.5
43.5
45.2
45.7
49.3
39.7
42.7
42.7
46.9
42.4
47.2
78.1
61.6
53.2
54.7
73.4
308
115
38
311
354
363
283
286
247
66
71
36
4
80(2R,3S)
74(2R,3S)
60(2R,3S)
77(2R,3S)
65(2R,3S)
70(2R,3S)
71(2R,3S)
77(2R,3S)
87(2S,3R)
47(2S,3R)
61(2R,3S)
70(2R,3S)
85(2R,3S)
71(2R,2S)
83(2R,3S)
--
25
32
11
19
30
44
28
52
51
10
19
67
142
118
70
--
88(R,R)
95(R,R)
75(R,R)
94(R,R)
98(R,R)
98(R,R)
96(R,R)
94(R,R)
93(S,S)
88(S,S)
90(R,R)
96(R,R)
98(R,R)
99(R,R)
97(R,R)
>99(R,R)
>99(R,R)
60(R,R)
62(R,R)
20(R,R)
43
97
13
110
216
226
119
78
86
28
38
105
280
437
186
--
5.7
5.3
5.0
5.8
4.8
5.0
5.2
5.4
5.9
4.4
4.7
4.6
5.5
4.6
15.3
3.0
3.0
2.7
2.8
3.1
1.43
1.24
1.18
1.20
1.22
1.12
1.14
1.30
1.11
1.20
1.09
1.05
1.08
1.23
1.26
1.33
1.21
1.16
1.12
1.13
-146
-172
-114
-166
-182
-183
-175
-168
+168
+144
-154
-177
-185
-188
-180
-190
-190
-89
3
0.4
0.3
0.3
0.4
0.4
0.5
1.5
1.5
3
10
11
12
13
14
(R,R,R,R,R)-3f
(R,R,R,R,R)-3f
(R,R,R,R,R)-3d
(R,R,R,R,R)-3d
(R,R,R,R,R)-2b
5
1
5[j]
6[j]
0
30
24
6
1
0
4
1
7[k]
25
25
25
25
25
0
79
--
1
1
2
2
2
8[l]
9[l]
0[l]
1[l]
2[l]
24
96
120
96
336
--
--
--
--
(S,S,S,S,S)-2b
(S,S,S,S,S)-3d
(S,S,S,S,S)-2b
--
--
--
--
--
--
--
--
-91
--
--
--
--
-45
[
a]Conditions: The reaction was performed in toluene (2.7 mL, 25 mmol) in a 20 mL flask; toluene/CHDO/anhydride/catalyst/PPNCl = 500/250/125/1/2 (molar
1
ratio), except for entries 15–22. [b] Calculated using H NMR spectroscopy using CHDO as the limiting reagent. [c] Turnover frequency (TOF) = moles of product
(polyester)/moles of catalyst per hour. [d] Measured by chiral HPLC, and the (2R,3S)-CHDO was the major enantiomer remaining when (R,R,R,R,R)-catalyst
was used. [e] krel = ln[(1-c)(1-eeepo)]/ln[(1-c)(1+eeepo)], where c is the conversion and eeepo is the enantiomeric excess of the recovered epoxide expressed
as %/100. [f] Measured by hydrolyzing the polymer and analyzing the resulting diol by chiral GC, and the (R,R)-diol was the major enantiomer when (R,R,R,R,R)-
catalyst was used. [g] s-factor = ln[1-c(1+eepro)]/ln[1-c(1-eepro)], where c is the conversion and eepro is the enantiomeric excess of the resulting diol expressed
as %/100. [h] Determined by using gel permeation chromatography in THF, calibrated using polystyrene. [i] Specific rotation values of the polymers were
o
determined in chloroform at 20 C (c = 1) by using a polarimeter; the wavelength of the polarized light was 589.3 nm. [j] Toluene/CHDO/anhydride/catalyst/PPNCl
=
2000/250/125/1/2 (molar ratio). [k] Toluene/CHDO/anhydride/catalyst/PPNCl = 2000/1000/500/1/2 (molar ratio). [l] Enantiopure (2S,3R)-CHDO was used for
entries 18–22, toluene/epoxides/anhydride/catalyst/PPNCl = 4000/200/200/1/2 (molar ratio)
polyester and retained the unreacted epoxide with (2R,3S)-
isomer excess (Entry 13). Both the krel and s-factor values were
higher than those observed with the (R,R,S,R,R)-3f catalyst
system. Notably, the catalytic activity of (R)-4 or (R,R,S,R,R)-3f
4
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
100
1
00
2
2
2
1
1
80
40
00
60
20
krel
280
240
200
160
120
80
80
60
40
s-factor
8
0
0
6
40
Simulated s-factor = 35
8
0
0
0
2
0
Simulated Krel
=
35
20
0
Simulated s-factor = 200
Experimental data
4
40
Experimental data
0
0
15
20
25
30
35
40
45
50
0
20
40
60
80
100
0
20
40
60
80
100
%
conversion
% conversion
% conversion
Figure 2. Left: Determination of krel and s-factor using four data points at different conversions mediated by (R,R,R,R,R)-3d/PPNCl. Middle: Comparison of
experimental data with the simulated plot of substrate enantioselectivity. Right: Comparison of experimental data with the simulated plot of product
enantiooselectivity. Experimental data are described in Supplementary tables S1 and S2).
was significantly lower than that of the (R,R,R,R,R)-3f catalyst
system. Overall, in the enantiopure dinuclear Al(Ⅲ) complex
mediated racemic CHDO/PA copolymerization, the axial chirality
determined the chiral induction direction, and the
cyclohexanediamine backbones with the same configuration was
beneficial for the catalytic activity and product enantioselectivity.
When copolymerization was performed at 0 oC in toluene
using (R,R,R,R,R)-3f/PPNCl as the catalyst, the resulting
polyester showed a high enantioselectivity of 96% ee, and
exhibited an enhanced s-factor (Table 1, entry 14). The dilution of
both the catalyst and monomer concentrations in the reaction
system using toluene resulted in a substantial loss in the
polymerization rate, but significantly increased both krel and s-
factor values (Entries 15 and and 16). The highest s-factor of 437
and krel of 118 was observed for the (R,R,R,R,R)-3d/PPNCl
exceeded the krel values by a factor of approximately five (Figure
2, left). This result indicated an unexpected enantioconvergent
phenomenon in the enantioselective resolution copolymerization.
In this case, both enantiomers incorporated into the polyesters
afforded the (R,R)-diol units. Moreover, the s-factor increased
with the conversion, while the enantioselectivity of the resulting
copolymer was maintained at 98% ee. For a comparative analysis,
the krel and s-factor values at various conversions were calculated
for
the
(R,R,R,R,R)-2b/PPNCl-mediated
CHDO/PA
copolymerization. Although, no apparent increase in the s-factor
value was observed with the conversion, a slightly higher s-factor
than krel was also identified in this process (Supporting information,
figure S34).
Furthermore, for an in-depth understanding of the
enantioselective copolymerization process,
a
monitoring
o
catalyst system at 0 C, and the resultant copolymer possessed a
experiment was conducted with a 1:1 feed ratio of racemic CHDO
and PA at various time points. The experimental data for the
enantioselectivity of the unreacted epoxide at various conversions
fits well with the simulated plot for krel of 35 (Figure 2, middle),[28,29]
while the data for copolymer enantioselectivity apparently
overwhelmed the plot for s-factor of 35, or even 200 (Figure 2,
high enantiopurity of up to 99% ee for (R,R)-configuration, with a
2
D
0
o
3
specific rotation of [α] = -188 (c = 1 in CHCl ). Notably, when the
CHDO/catalyst ratio was increased to 1000, the krel of 70 and s-
factor of 186 were obtained at 47.2% conversion at room
temperature, affording the copolymer with an enhanced molecular
weight and a high level enantiopurity of 97% ee (Entry 17).
Although the accurate assignment of the microstructure
right; the simulated
krel and s-factor values adopted the
experimental values at approximate 50% conversion). Three
important differences were observed through the comparative
study: (1) the s-factor for evaluating the enantioselective
copolymerization significantly exceeded the krel based on the
enantioselectivity of the unreacted epoxide at various
conversions; (2) the decrease in copolymer enantioselectivity
started from approximately 54% conversion, slightly exceeding
the maximum efficiency (50%) for the traditional kinetic resolution
processes; (3) after complete conversion of the epoxide, the
resulting copolymer had an enantioselectivity of 29% ee, which
was clearly different from 0% ee obtained by the traditional kinetic
resolution process.[30] These results demonstrated the
enantioconvergent phenomenon during the enantioselective
resolution copolymerization of racemic CHDO and PA, in which
approximately 54% conversion was the point of inflexion for the
decrease of polyester enantioselectivity.
(
regio/stereoregularity) of CHDO/PA copolymer is not clear,
significant differences in carbonyl, methine and methylene region
between the atactic and isotactic polymers are easily identified on
the basis of 13C NMR spectroscopy studies (Figure S23). Notably,
the isotactic CHDO/PA copolymer (poly(1,3-cyclohexadiene
phthalate)) with main-chain chirality was found to be a typical
semicrystalline material, possessing a melting endothermic peak
o
-1
at 213 C with ∆퐻 = 18.3 J g . Its crystallizability was further
푚
identified by powder X-ray diffraction (XRD). Diffraction peaks at
o
o
o
o
o
1
0.9 , 15.3 , 18.1 , 22.2 and 23.3 were observed for the
isotactic-enriched poly (1,3-cyclohexadiene phthalate) sample,
demonstrating its semicrystalline structure, while no diffraction
signal was found in the atactic copolymer (Figure S26).
Comprehensive understanding of enantioconvergent
copolymerization behavior (s-factor > krel). As the calculations
of krel and s-factor are based on the conversion and the
enantioselectivities of the unreacted epoxide and resultant
copolymer, their values might deviate from the actual ones, due
to the small measurement errors. [23] Therefore, four sets of data
for krel and s-factor were calculated at different conversions in the
To clarify the asymmetric induction of the dinuclear Al(III)
complex in the ring-opening process of the unmatched CHDO
isomer, enantiopure (2S,3R)-CHDO was prepared for the
copolymerization with PA at various reaction conditions (Table 1,
entries 18–22). Excellent enantioselectivity was observed with the
use of both the matched chiral (R,R,R,R,R)-2b and achiral
salenAlCl catalyst 5, where the copolymerization of (2S,3R)-
enantioselective CHDO/PA copolymerization mediated by
o
(
R,R,R,R,R)-3d/PPNCl at 25 C. The krel values ranging from 41
o
to 50 were obtained between 20 to 48% conversion while the
corresponding s-factor values varied from 180 to 265, which
CHDO with PA at 25 C afforded highly isotactic polyester with
2
D
0
o
more than 99% ee, and a specific rotation value of [α] = -190 (c
5
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
≠
Scheme 3. Gibbs free energy barrier (△G ) of different ring-opening pathways mediated by (R,R,R,R,R)-3d. Path A: Attack of unfavorable (2R,3S)-CHDO
by (R,R)-configured chain end; Path B: Attack of favorable (2S,3R)-CHDO by (R,R)-configured chain end; Path C: Attack of unfavorable (2R,3S)-CHDO by
-
1
(
S,S)-configured chain end; Path D: Attack of favorable (2S,3R)-CHDO by (S,S)-configured chain end. The energies are given in kcal mol .
=
1 in CHCl
reaction of a single stereoisomer was highly regioselective and
the ring opened at the more active allylic position via a S
3
). This result indicated that the copolymerization
influence the stereochemistry of the fresh unit incorporated into
the polymer. In prior reports on enantioselective epoxide
N
2
copolymerization,
the
enantiomorphic-site
mechanism
pathway, resulting in a stereospecific ring-opening reaction.
When the reaction was carried out using the unmatched chiral
predominantly affected the stereochemistry of the resultant
copolymers.[7] To gain insights into the enantioconvergent
copolymerization mechanism, the copolymer chain growth
processes were simulated using the density functional theory
(DFT) calculations. The (R,R,R,R,R)-3d was selected as a model
catalyst, and CHDO/PA copolymerization as a model reaction.
The epoxide was coordinated to one Al(III) center and
subsequently ring-opened by the benzoate bearing a chiral chain-
end bound to the other Al(III) center, affording four pathways for
attacking the (2R,3S)-CHDO or (2S,3R)-CHDO by the (R,R)- or
(S,S)-chain end in the ring-opening process. Each pathway
afforded two possible transition states (TSs) for attacking the (R)-
or (S)-methine position (Scheme 3).
It was observed that the benzoate bearing a different chiral
ester chain-end unit resulted in a significant difference in the
Gibbs free energy barrier for (2R,3S)-CHDO ring-opening at
either the (R)- or (S)-methine position. For example, when the
coordinated (2R,3S)-CHDO was ring-opened by the benzoate
bearing an (R,R)-ester end unit, the Gibbs free energy barriers for
attacking (S)-methine (Scheme 3, Path A) and (R)-methine C─O
bonds (Scheme 3, Path Aꞌ) of the epoxide were 14.3 and 19.7
kcal/mol, respectively. This implied that it was easier for the ring-
opening reaction of the next enchained (2R,3S)-isomer to occur
at (S)-methine to produce a new (R,R)-ester unit. In the case of
(2S,3R)-CHDO ring-opened by the benzoate bearing an (R,R)-
ester end unit, the Gibbs free energy barriers for attacking (S)-
methine (Scheme 3, Path B) and (R)-methine (Scheme 3, Path
(
S,S,S,S,S)-2b or (S,S,S,S,S)-3d under same conditions,
approximately 20% (S,S)-configuration units were detected in the
resulting polyester. Moreover, when performed at 0 °C, the
enantiomeric ratio of (S,S)- and (R,R)- figuration units reached to
40:60. These results demonstrated that the multi-chiral
environment of the dinuclear Al(III) complex significantly
influenced the ring-opening position of the configuration-
unmatched CHDO incorporated into the copolymer.
Based on the data shown above, it could be concluded that in
the
enantioselective copolymerization, the unfavorable stereoisomer
2R,3S)-CHDO was the only source for producing (S,S)-units
(R,R,R,R,R)-3d-mediated
racemic
CHDO/PA
(
distributed into the (R,R)-configuration-enriched polyester.
Besides, the almost complete consumption of the favorable
(
2S,3R)-isomer occurred at approximately 54% conversion.
Therefore, the incorporation of the favorable (2S,3R)-isomer
became extremely insufficient at more than 54% conversion, a
point of inflection where the copolymer enantioselectivity started
to decrease. After the completely consumption of (2S,3R)-CHDO,
the copolymerization only involved the unmatched (2R,3S)-
CHDO isomer at a very low rate. Consequently, the ring-opening
regioselectivity for the unmatched (2R,3S)-CHDO could be
different at various conversions, possibly due to the chain-end
stereochemical control. It is generally known that both the ligand
set of the discrete catalyst and the propagating chain-end
6
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
Scheme 4. Enantioconvergent behavior in the enantioselective copolymerizations of CHDO and PA by combining the enantiomorphic-site and chain-end control.
Bꞌ) of the epoxide were 12.3 and 16.7 kcal/mol, respectively, with
the same tendency to produce a new (R,R)-ester unit.
of the next favorable (2S,3R)-isomer until it was completely
consumed. After the favorable (2S,3R)-isomer was consumed,
the benzoate bearing the (S,S)-configured ester end was
beneficial for the ring-opening of the next enchained (2R,3S)-
isomer at its (R)-methine C−O bond to produce (S,S)-ester units,
thereby changing the ring-opening tendency to cause an
irreversible collapse of the copolymer stereoselectivity. Therefore,
the enantioconvergence (s-factor > krel) was attributed to the
same asymmetric ring-opening direction of both isomers, prior to
the complete consumption of the favorable stereoisomer.
Alternately, when the coordinated (2R,3S)-CHDO was ring-
opened by the benzoate bearing an (S,S)-ester end unit, the
Gibbs free energy barriers for attacking (S)-methine (Scheme 3,
Path Cꞌ) and (R)-methine (Scheme 3, Path C) were calculated to
be 20.5 and 11.8 kcal/mol, respectively, implying the tendency to
produce a new (S,S)-ester unit. Whereas, in the case of (2S,3R)-
CHDO ring-opened by the benzoate bearing an (S,S)-ester end
unit, the Gibbs free energy barrier for attacking (S)-methine
(
Scheme 3, Path D) and (R)-methine (Scheme 3, Path Dꞌ) were
calculated to be 7.6 and 16.1 kcal/mol, respectively, implying the
inclination toward ring-opening at (S)-methine to afford the (R,R)-
unit. Briefly, with the use of (R,R,R,R,R)-3d, it was easier for the
ring-opening of the favorable (2S,3R)-CHDO to occur at (S)-
methine to produce a new (R,R)-unit regardless of the chirality of
the chain end, whereas the chain-end chirality significantly
affected the ring-opening direction of the unfavorable (2R,3S)-
CHDO. Therefore, it could be concluded that both the
enantiomorphic-site of the ligand set in the(R,R,R,R,R)-3d and
propagating (R,R)-ester chain-end control mechanisms
determined the regioselective ring-opening at (S)-methine C-O
bond for either (2S,3R)- or (2R,3S)-CHDO incorporated into the
copolymer before the 54% conversion.
Substrate scope of enantioselective racemic cis-
disubstituted epoxide/anhydride copolymerization. With
highly active and enantioselective (R,R,R,R,R)-3d catalyst
system in hand, we evaluated the substrate scope of both the
racemic cis-internal epoxides and anhydrides with toluene as a
solvent. Because of its relatively high reactivity, PA was selected
as a model cyclic anhydride for investigating the catalytic activity
and enantioselectivity. Various racemic cis-internal epoxides,
including alicyclic 8-oxabicyclo[5.1.0]oct-2-ene (7EO), aromatic
1,2-dihydronaphthalene oxide (ONa) and cis-beta-methylstyrene
oxide (OMSy), and aliphatic cis-2,3-epoxyhexane (2-6-O) and cis-
2,3-epoxyoctane (2-8-O), exhibited reactivities at ambient
temperature, affording (R,R)-configuration-enriched polyesters
with 84~99% ee (Table 2, entries 1-7). Additionally, the
enantioconvergent copolymerization behavior was observed in
these systems. For example, the enantioselective resolution
copolymerization of 7EO and PA showed a krel value of 31 at room
temperature, and a corresponding s-factor of 88 at 49.5%
Based on the above-mentioned studies,
a reasonable
explanation for the enantioconvergent behavior (s-factor > krel
)
observed in the enantioselective copolymerization of racemic
CHDO with PA mediated by complex (R,R,R,R,R)-3d catalyst
systems could be proposed (Scheme 4). In this system, the
effective kinetic resolution process proceeded smoothly to
conversion (Entry 1). A decrease in the reaction temperature from
o
preferentially consume (2S,3R)-isomer and retain (2R,3S)-isomer. 25 to 0 C resulted in great increases in both krel (61) and s-factor
The favorable (2S,3R)-isomer was stereospecifically ring-opened
at (S)-methine C-O bond via S 2 mechanism, affording poly(1,3-
(126) (Entry 2). The resultant polyester had a high enantiopurity
of 96% ee. Similar enantioconvergent behavior was also
observed in the ONa/PA copolymerization system, where krel of
21 and s-factor of 65 were obtained at room temperature (Entry
3), and the corresponding values of 40 and 131, respectively,
N
cyclohexadiene phthalate) with (R,R)-configuration. Importantly,
the simultaneous influence of both the enantiomorphic-site control
and the propagating (R,R)-ester chain-end control resulted in the
incorporation of a very low amount of the unfavorable (2R,3S)-
isomer into the copolymer as an (R,R)-unit via ring-opening at the
o
were found at 0 C, affording polyesters with an enantiopurity of
97% ee (Entry 4). In the case of OMSy, a moderate kinetic
resolution process proceeded with a krel of 9 and s-factor of 39 at
25 oC (Entry 5). Surprisingly, (R,R,R,R,R)-3d was found to be
ineffective in catalyzing the enantioselective resolution
copolymerization of PA with 2-6-O or 2-8-O. However, to our
delight, the (R,R,R,R,R)-3b catalyst system was highly
(
S)-methine C-O bond. Pathways Aꞌ and Bꞌ involving ring-opening
at the (R)-methine C-O bonds of both (2S,3R)- and (2R,3S)-
CHDO isomers were inhibited due to their higher Gibbs free
energy barriers. Any unavoidable stereochemical error that
appeared in the chain-end could be corrected by the enchainment
7
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
Table 2. Substrates scope of enantioconvergent copolymerization[a]
[h]
Specific rotation[i]
Epoxide/
Temp.
(oC)
Time
(h)
Conv. [b]
(%)
TOF[c]
eeepo[d]
eepro [f]
M
n
[
e]
s-factor[g]
Ð [h]
Entry
k
rel
-1
o
Anhydride
(h )
(%)
(%)
(kg/mol)
4.6
[ ]
1
2 [j]
3
7EO/PA
7EO/PA
25
0
8
49.5
46.1
41.7
40.5
52.6
33.2
43.9
41.3
45.9
15
3
83(2R,3S)
79(2R,3S)
60(2R,3S)
62(2R,3S)
70(2R,3S)
45(2R,3S)
72(2R,3S)
60(2R,3S)
70(2R,3S)
31
61
21
40
9
93(R,R)
96(R,R)
94(R,R)
97(R,R)
84(R,R)
97(R,R)
99(R,R)
94(R,R)
96(R,R)
88
126
65
1.14
1.10
1.09
1.08
1.08
1.11
1.05
1.29
1.14
-49
-51
40
3
4.0
ONa/PA
25
0
35
1
5.5
-73
4[j]
5
ONa/PA
72
48
12
24
12
10
131
39
5.0
-75
OMSy/PA
2-6-O/PA
2-8-O/PA
CHDO/NA
CHDO/DMPA
25
25
25
25
25
3
4.8
-9
6[k]
7[k]
8
7
31
52
23
22
106
470
64
7.1
+52
+30
-320
-148
5
10.0
3.7
9
9
11
124
4.6
[
a] Conditions: The reaction was performed in toluene in a 20 mL flask; toluene/epoxides/anhydride/catalyst/PPNCl = 500/250/125/1/2 (molar ratio), except for
entries 6 and 8. [b] Calculated using 1H NMR spectroscopy with epoxides as the limiting reagents. [c] Turnover frequency (TOF)= moles of product
polyester)/moles of catalyst per hour. [d] Measured by chiral HPLC; the absolute configuration was confirmed by determining the specific rotation or from
(
literature results. [e] krel= ln[(1-c)(1-eeepo)]/ln[(1-c)(1+eeepo)], where c is the conversion and eeepo is the enantiomeric excess of the recovered epoxide expressed
as %/100. [f] Measured by hydrolyzing the copolymer and analyzing the resulting diol by chiral GC or HPLC, and the (R,R)-diol was the major enantiomer when
(R,R,R,R,R)-catalyst was used. [g] s-factor = ln[1-c(1+eepro)]/ln[1-c(1-eepro)], where c is the conversion and eepro is the enantiomeric excess of the resulting diol
i
expressed as %/100. [h] Determined by using gel permeation chromatography in THF, calibrated with polystyrene. [i] Specific rotation of the polymers was
determined in chloroform at 20 oC (c
Toluene/CHDO/anhydride/catalyst/PPNCl = 1000/250/125/1/2 (molar ratio). [k] The reaction was mediated by complex (R,R,R,R,R)-3b.
=
1) by using
a
polarimeter. The wavelength of the polarized light was 589.3 nm. [j]
enantioselective for this copolymerization. The reaction of
aliphatic epoxides 2-6-O showed moderate krel value of 31, but a
high s-factor value of 106 was obtained, with 97% ee for the
affording polyester (Entry 6). Notably, both the high Krel (52) and
s-factor (470) values were obtained for the reaction of 2-8-O with
PA, affording the polyester with up to 99% ee (Entry 7). Moreover,
naphthyl anhydride (NA), and 4,5-dimethylphthalic anhydride
(R,R,R,R,R)-3d without any substituent at the phenolate ortho-
positions of the ligand exhibited unprecedented levels of
enantioselectivity of up to 99% ee for the (R,R)-configuration
copolymer formation with an s-factor of 437, affording the
unreacted (2R,3S)-CHDO-enriched epoxides with good
enantiopurity, and the highly stereoregular semi-crystalline
polyesters with a perfectly alternating structure. Various racemic
unsymmetric cis-disubstituted epoxides with anhydrides were
tested, and all of these copolymerization systems showed an
unanticipated enantioconvergent behavior, i.e., the s-factor
significantly exceeded the krel at various conversions. The DFT
theoretical calculations suggested that both the enantiomorphic-
site of the ligand set and propagating chain-end control
mechanisms influenced the enantioconvergent behavior in the
enantioselective resolution copolymerization.
(
DMPA) were employed for the enantioselective resolution
copolymerization using racemic CHDO as a model racemic cis-
internal epoxide monomer at ambient temperature (Entries 8 and
9). As anticipated, the resultant copolymers exhibited perfectly
alternating structure and enhanced enantioselectivities of 94%
and 96% ee, respectively.
Although enantioconvergent catalysis is emerging as a
powerful method for synthesizing single enantiomerically-
enriched organic compounds from the racemic substrates with a
maximum yield of 100%,[31-35] such a strategy has not been
applied to the preparation of optically active polymers from a
racemic starting material. The present study is the first report of
the enantioselective, stereoconvergent copolymerization of
racemic cis-internal epoxides with anhydrides to afford
Conclusion
In summary, we report the first highly enantioselective
resolution copolymerization of racemic cis-internal epoxides with
anhydrides via multi-chiral synergistic catalysis using enantiopure
dinuclear aluminum catalyst systems. The steric hindrance of the
phenolate ortho-position substituents of the ligand and axial
chirality significantly affected both the catalytic activity and
enantioselectivity. For the alternating copolymerization of racemic
enantioenriched
semi-crystalline
polyesters
with
high
enantioselectivities of up to 99% ee at more than 50% conversion.
7-oxabicyclo[4,1,0]hept-2-ene (CHDO) and phthalic anhydride
It is anticipated that the enantioconvergent catalytic
(
PA), the binaphthol- linked bimetallic Al(III) complex
8
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
polymerization can possibly allow the production of highly
isotactic polymers through kinetic resolution polymerization even
with an inferior kinetic resolution coefficient.
[21] E. Rose, B. Andrioletti, S. Zrig, M. Quelquejeu-Ehteve, Chem.
Soc. Rev. 2005, 34, 573583.
[22] M. Tokunaga, J. F. Larrow, F. Kakiuchi, E. N. Jacobsen,
Science 1997, 277, 936938.
[
[
23] E. N. Jacobsen, Acc. Chem. Res. 2000, 33, 421431.
24] M. Bandini, P. G. Cozzi, P. Melchiorre, Umani-Ronchi, Angew.
Chem. Int. Ed. 2004, 43, 8487; Angew. Chem. 2004, 116,
Acknowledgements
8
689.
This work is supported by the National Natural Science
Foundation of China (NSFC, Grant 21690073 and 21920102006),
and Program for Changjiang Scholars and Innovative Research
Team in University (IRT-17R14).
[
[
[
[
[
[
[
25] K. Arai, S. Lucarini, M. M. Salter, K. Ohta, Y. Yamashita, S.
T. Kobayashi, J. Am. Chem. Soc. 2007, 129, 81038111.
26] M. R. Monaco, D. Fazzi, N. Tsuji, M. Leutzsch, S. Liao, W.
Thiel, B. List, J. Am. Chem. Soc. 2016, 138, 1474014749.
27] X. Tang, A. H. Westlie, E. M. Watson, E. Y. X. Chen, Science
Keywords: stereoconvergence
•
asymmetric ring-opening
2019, 366, 754758.
copolymerization• chain-end control • polyester • cis-internal
epoxides
28] J. R. Lamb, M. Mulzer, A. M. LaPointe, G. W. Coates, J. Am.
Chem. Soc. 2015, 137, 1504915054.
29] M. Tokunaga, J. Kiyosu, Y. Obora, Y. Tsuji, J. Am. Chem.
Soc. 2006, 128, 44814486.
30] E. Vedejs, M. Jure, Angew. Chem. Int. Ed. 2005, 44,
[
[
[
1] E. Yashima, K. Maeda, H. Lida, Y. Furusho, K. Nagai, Chem.
Rev. 2009, 109, 61026211.
2] C. Rest, R. Kandanelli, G. Fernández, Chem. Soc. Rev. 2015,
3
9744001; Angew. Chem. 2005, 117, 40404069.
31] V. Bhat, E. R. Welin, X. Guo, B. M. Stoltz, Chem. Rev. 2017,
17, 45284561.
44, 25432572.
3] J. C. Worch, H. Prydderch, S. Jimaja, P. Bexis, M. L. Becker,
1
A. P. Dove, Nat. Rev. Chem. 2019, 3, 514535.
[
[
2
[
[
32] E. L. Lucas, E. R. Jarvo, Nat. Rev. Chem. 2017, 1, 0065.
33] S. L. Bartlett, J. S. Johnson, Acc. Chem. Res. 2017, 50,
2842296.
34] H. Yin, G. C. Fu, J. Am. Chem. Soc. 2019, 141, 1543315440.
35] Z. Wang, H. Yin, G. C. Fu, Nature. 2018, 563, 379383.
[
[
4] Y. Okamoto, T. Nakano, Chem. Rev. 1994, 94, 349372.
5] W. Hirahata, R. M. Thomas, E. B. Lobkovsky, G. W. Coates, J.
Am. Chem. Soc. 2008, 130, 1765817659.
[
6] S. M. Ahmed, A. Poater, M. I. Childers, P. C. B. Widger, A. M.
LaPointe, E. B. Lobkovsky, G. W. Coates, L. Cavallo, J. Am.
Chem. Soc. 2013, 135, 1890118911.
[
[
[
[
[
7] M. I. Childers, J. M. Longo, N. J. Van Zee, G. W. Coates, Chem.
Rev. 2014, 114, 81298152.
8] X.-B. Lu, W.-M. Ren, G.-P. Wu, Acc. Chem. Res. 2012, 45,
1
7211735.
9] X.-B. Lu, Y. Wang, Angew. Chem. Int. Ed. 2004, 43,
5743577; Angew. Chem. 2004, 116, 36583661.
10] C. T. Cohen, T. Chu, G. W. Coates, J. Am. Chem. Soc. 2005,
27, 1086910878.
3
1
11] X.-B. Lu, L. Shi, Y.-M. Wang, R. Zhang, Y.-J. Zhang, X.-J.
Peng, Z.-C. Zhang, B. Li, J. Am. Chem. Soc. 2006, 128,
16641674.
[
[
12] W.-M. Ren, Y. Liu, G.-P. Wu, J. Liu, X.-B. Lu, J. Polym. Sci.,
Part A: Polym. Chem. 2011, 49, 48944901.
13] Y. Liu, W.-M. Ren, J. Liu, X.-B. Lu, Angew. Chem. Int. Ed.
2013, 52, 1159411598; Angew. Chem. 2013, 125,
1180811812.
[
[
[
14] Y. Liu, W.-M. Ren, C. Liu, S. Fu, M. Wang, K.-K. He, G.-G. Li,
R. Zhang, X.-B. Lu, Macromolecules 2014, 47, 77757788.
15] J. Li, Y. Liu, W.-M. Ren, X.-B. Lu, J. Am. Chem. Soc. 2016,
138, 1149311496.
16] J. Li, B.-H. Ren, S.-Y. Chen, G.-H. He, Y. Liu, W.-M. Ren, H.
Zhou, X.-B. Lu, ACS Catal. 2019, 9, 19151922.
[
17] M. R. Kember, P. D. Knight, P. T. R. Reung, C. K. Williams,
Angew. Chem. Int. Ed. 2009, 48, 931933; Angew. Chem.
2
009, 121, 949951.
18] J. A. Garden, P. K. Saini, C. K. Williams, J. Am. Chem. Soc.
015, 137, 1507815081.
[
[
2
19] C. Romain, Y. Zhu, P. Dingwall, S. Paul, H. S. Rzepa, A.
Buchard, C. K. Williams, J. Am. Chem. Soc. 2016, 138,
41204131.
[
20] J. Li, B.-H. Ren, Z.-Q. Wan, S.-Y. Chen, Y. Liu, W.-M. Ren,
X.-B. Lu, J. Am. Chem. Soc. 2019, 141, 89378942.
9
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202011259
RESEARCH ARTICLE
Entry for the Table of Contents
Stereoconvergence is observed in the enantioselective resolution copolymerization of racemic cis-internal epoxides and anhydrides
mediated by dinuclear aluminum complexes with multiple chirality, where the selectivity factor for chiral copolymer formation
significantly exceeds the kinetic resolution coefficient based on the unreacted substrate at various conversions.
1
0
This article is protected by copyright. All rights reserved.