Lewis pairs for ring-opening alternating copolymerization of cyclic anhydrides and epoxides

Article abstract of DOI:10.1039/c7gc03261g

A simple and highly active catalytic process for ring-opening alternating copolymerization (ROAC) of cyclic anhydrides and epoxides still remains a key challenge. Herein, we have described an effective group of versatile and low-toxic zinc dicarbyl/amine Lewis pairs for the ROAC. The facile route showed a high catalytic activity (TOF ≤ 210 h^-1 at 110 °C) and perfectly alternating selectivity (>99%). An unexpected highly regioselective ring-opening of asymmetric epoxides (PO, ECH and SO) was also achieved by the combination of zinc alkyls (or aryls) and amines. Of note, deprotonation side reaction of α-H of anhydrides with organic bases was uncovered, and subsequently was inhibited by using nonpolar solvents and Lewis acid/base pairs. Thus, an array of polyesters was synthesized by the coupling of various anhydrides (PA, CHA, SA and NA) and epoxides (CHO, PO, ECH and SO) using the same Lewis pairs. Furthermore, variable temperature ¹H NMR spectral and MALDI TOF MS analyses were performed to understand the possible mechanism and microstructure. The experimental results indicated that zwitterionic alkoxide and carboxylate intermediates alternately formed to enhance the ester repeat units in chain initiation and propagation. This work provides a simple and green catalytic strategy to prepare diversified polyesters from the ROAC process of cyclic anhydrides and epoxides with considerable catalytic activity and alternating selectivity.

Full text of DOI:10.1039/c7gc03261g

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: H. Ji, B. Wang, L.
Pan and Y. Li, Green Chem., 2017, DOI: 10.1039/C7GC03261G.
Volume 18 Number
7
7
April 2016 Pages 1821–2242
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Page 1 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
Lewis Pairs for RingꢀOpening Alternating Copolymerization of Cyclic
Anhydrides and Epoxides
†
, †
†
,†,‡
He-Yuan Ji, Bin Wang,* Li Pan, Yue-Sheng Li*
†
Tianjin Key Lab of Composite & Functional Materials, School of Materials Science and
Engineering, Tianjin University, Tianjin 300350, China
‡
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072,
China

Green Chemistry
Page 2 of 19
View Article Online
DOI: 10.1039/C7GC03261G
Abstract
Simple and highly active catalytic process for ringꢀopening alternating copolymerization (ROAC)
of cyclic anhydrides and epoxides still remains a key challenge. Herein, we described an effective
group of versatile and lowꢀtoxic zinc dicarbyl/amine Lewis pairs for the ROAC in this work. The
ꢀ1
facile route showed high catalytic activity (TOF ≤ 210 h at 110 ºC) and perfectly alternating
selectivity (> 99%). An unexpected highly regioselective ringꢀopening of asymmetric epoxides (PO,
ECH and SO) was also achieved by the combination of zinc alkyls (or aryls) and amines. Of unique,
deprotonation side reaction of αꢀH of anhydrides with organic bases was uncovered, and
subsequently was inhibited by using nonpolar solvents and Lewis acid/base pairs. Thus, an array of
polyesters was synthesized by the coupling of various anhydrides (PA, CHA, SA and NA) and
1
epoxides (CHO, PO, ECH and SO) under the same Lewis pairs. Further variable temperature H
NMR spectra and MALDI TOF MS analysis were performed to understand possible mechanism and
microstructure. The experimental results indicated that zwitterionic alkoxides and carboxylates
intermediate alternately formed to enhance ester repeat units in chain initiation and propagation.
This work provides a simple and green catalytic strategy to prepare diversified polyesters from the
ROAC process of cyclic anhydrides and epoxides with considerable catalytic activity and
alternating selectivity.

Page 3 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
Introduction
The development of simple and green catalytic process is a growing demand for the preparation of
1
various polymers. The versatility, commercial accessibility and high activity of several catalysts
has led to them being widely applied in preference of more traditional metal/ligand combinations,
2
1d,3
4ꢀ7
such as organic bases, hydrogenꢀbonding organocatalysts
and Lewis pairs. Indeed, the
advancement of catalytic systems has been represented by ringꢀopening polymerization (ROP) of
2
ꢀ6
cyclic esters.
Importantly, Lewis pairs, the combination of electronꢀpairꢀdonors and
electronꢀpairꢀacceptors, have been employed in the ROP of lactide or lactones, including
5
a
5b
6
5c
5d
6 5 2 2 6 5 2 2 6 5 3
Zn(C F ) /PMP, MgCl /NHC, Zn(C F ) /DMAP, MgCl /NHO and Al(C F ) /NHO. In all
cases, the presence of both a Lewis acid (LA) and a Lewis base (LB) has been essential in order to
enhance polymerization activity and product selectivity. With the powerful strategy, wellꢀdefined
polyesters and diblock copolyesters could be synthesized with the elimination of transesterification
side reactions. However, as an alternative route to polyesters, ringꢀopening alternating
copolymerization (ROAC) of cyclic anhydrides and epoxides catalyzed by aforementioned Lewis
pairs still remains in its vacant development.
In contrast to stepꢀgrowth polycondensation of diacids or diesters with diols, ROAC of cyclic
anhydrides and epoxides is an atomꢀeconomic and controlled approach to polyesters with the
8
advantage of low energy input. Meanwhile, the ROAC process also has another advantage to
architect miscellaneous polyesters over ROP of cyclic ester monomers, in which the structures and
8
properties of polyesters can be easily modulated by changing anhydrides or epoxides (Scheme 1).
Furthermore, many anhydrides and epoxides reported for the ROAC could derive from renewable
9
resources, like phthalic anhydride (PA), cyclohexanedicarboxylic anhydride (CHA), maleic
10
11
9c
anhydride (MA) and succinic anhydride (SA) , as well as cyclohexene oxide (CHO) and
12
epichlorohydrin (ECH). The corresponding polyesters are interesting examples sourced from
sustainable feedstocks.
A wide range of metal complexes has been used to catalyze the ROAC of anhydrides and
epoxides. In 2007, Coates et al. have ever made a big breakthrough by using βꢀdiiminate zinc
complex with withdrawing substituent on the ligand, yielding aliphatic polyesters with relatively
ꢀ1
11b
high reactivity (TOF = 31 h at 50 ºC). Since the seminal work, a diverse array of metalꢀbased

Green Chemistry
Page 4 of 19
View Article Online
DOI: 10.1039/C7GC03261G
9
c,11a,dꢀf,12b,c,13
9c,11a,c,e,f,12b,d,13a,h
complexes
manganese,
was
developed,
including
chromium,
cobalt,
1
1a,e,f,14
11a,e,f,12b,13a,h,15
16
16
15b,17
aluminum,
zinc, magnesium and iron.
In general, the
chromium and cobaltꢀbased salen catalysts (Scheme 1a) exhibited higher TOF values and
9
c,11a,e,f,12b,d,13a
outstanding regioselectivity.
Except for βꢀdiiminate zinc, other zinc/ligand complexes
have also been explored to catalyze the ROAC of anhydride/epoxide, showing excellent
1
6
chemoselectivity and moderate catalytic activity. Subsequently, to achieve satisfying catalytic
activity, chemoꢀ and regioselectivity, optimization of catalysts focused previously on the design of
metal centers and their ligand frameworks.
Scheme 1. The ROAC of cyclic anhydrides and epoxides to synthesize polyesters using various catalysts: (a)
metal complexes reported in previous work; (b) small organic molecules reported and (c) Lewis pairs used in
the present work.
Unfortunately, to the best of our knowledge, only a few studies laid emphasis on simple and
18
18a
green catalysts (Scheme 1b) for the ROAC approach in the literature. In 2015, Theato and Liu
reported that bis(triphenylphosphine)iminium chloride (PPNCl) could mediate the ROAC of
ꢀ1
norbornene anhydride (NA) and CHO with an optimized TOF of 32 h at 110 ºC. In 2017, Merna

Page 5 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
1
8b
ꢀ1
and coꢀworkers found that PPNCl promoted the ROAC of CHO and PA with a TOF of 23 h at
110 ºC. Recently, Zhao and coworkers revealed that the ROAC of PA and ethylene oxide (EO) was
efficiently initiated by 1,4ꢀbenzenedimethanol (BDM) and catalyzed by phosphazene superbases
18c
1
(tꢀBuP ). The new system highlighted a living nature with controlled molecular weights and very
narrow distributions (PDI < 1.1) by selfꢀbuffering mechanism; but somewhat low reactivity was
observed at 60 ºC in tetrahydrofuran solvent. Obviously, there are still lower catalytic activity and
lack of structural variation of small organic molecules compared to metal/ligand complexes. Also,
we have not still found that single organocatalyst could control regioselectivity of asymmetric
epoxides (such as propylene oxide) for the ROAC of anhydrides/epoxides in the literature.
Therefore, the development of highly active and simple catalysts for selective ROAC of anhydrides
and epoxides continues to be an inspiring objective.
Here, we anticipated that the combination of simple and proper Lewis acids with common
organic bases (Scheme 1c) could realize the ROAC of anhydrides with epoxides. For one, to avoid
the laborꢀconsuming synthesis of ligands and undesirable toxic metal contaminants like chromium
and cobalt, more environmentꢀfriendly ZnR
2
and organic bases were directly used to promote the
19aꢀc
copolymerization.
In the proceeding, Nꢀheterocyclic organobases showed relatively low
1
9d
cytotoxicity at a low concentration.
Removal of such organobase residues was also more
1
9e
convenient than metal/ligand combinations. For another, in contrast to small organic molecules
reported, cooperative Lewis pairs provided an attractive potential that the catalytic properties could
usually be finely tuned to compromise selectivity and reactivity. Besides, the new catalytic pairs
were developed based on the hypothesis that Lewis acid might activate epoxides for ringꢀopening
and Lewis base might act as nucleophile to initiate polymerization. The independent variability of
both components maybe allowed for a simultaneous increase of polymerization rates and selectivity
by simple means.
Results and Discussion
ROAC of PA and CHO. The ROAC of PA and CHO was firstly selected as the model reaction
with the molar ratio of 1:1:100:500 of LA, LB, PA and CHO in bulk. At 110 ºC, Zn(C
1:1) pair could efficiently catalyze the copolymerization with a moderate catalytic activity (run 1,
Table 1). A complete PA conversion was achieved within 1.5 h, yielding perfectly alternating
6 5 2
F ) /DMAP
(

Green Chemistry
Page 6 of 19
View Article Online
DOI: 10.1039/C7GC03261G
copolymer with ester linkage > 99%. No polyether segment was observed even in excess CHO (see
Figures S1ꢀS3 in Supplementary Information). Furthermore, homopolymerization of excess
epoxides did not occur even though the reaction time was extended to 9 h (Figure S4). While it was
i
noteworthy that Zn(C
6
F )
5 2
/PPh
3
, B(C
6
F
5
)
3
/DMAP, B(C
6
F
5
)
3
/PPh
3
and Al Bu
3
/DMAP Lewis pairs
were inefficient in the ROAC (runs 1ꢀ4, Table S1). A mass of ether linkages in the polymers showed
the inferior selectivity of these Lewis pairs, suggesting that steric effect and matched
acidityꢀbasicity of Lewis pairs were the decisive factors to produce alternating copolymers, which
7
agreed well with ringꢀopening copolymerization of COS/epoxides. With the efficient
Zn(C
6
F
5
)
2
/DMAP Lewis pair, PA monomer was consumed faster when changing Zn(C
6 5 2
F ) /DMAP
molar ratio into 1:2 (run 2, Table 1), whereas Zn(C
6 5 2
F )
/DMAP in 2:1 ratio generated
poly(esterꢀcoꢀether) with 73% ether linkage (run 3, Table 1). The results indicated that the molar
ratio of LA/LB greatly affects the chemoselectivity. To extend the scope of the Lewis pairs used in
the ROAC, some other organometallic zinc dicarbyl species and amines were further evaluated.
Obviously, the combination of DMAP with Zn(C
copolymerization without sacrificing any catalytic activity or structural selectivity (runs 4&5, Table
). However, replacing DMAP with DBU or MTBD led to a depressed catalytic efficiency and yet
6 5 2 2
H ) or ZnEt can also mediate the
1
perfect alternating degree under the same conditions (runs 6&7, Table 1). The results indicated that
the nature of Lewis bases had a stronger influence on the copolymerization rate than Lewis acids
under the new catalyst system. In addition, the copolymerization rate of PA and CHO depended
remarkably on solvent effect. As expected, PA conversion was lower in xylene compared to bulk
condition (run 8, Table 1). Dilution factors of CHO could account for this phenomenon, which has
11d,16a
been demonstrated to be common for many metal/ligand complexes.
copolymerization rate was much lower in a more polar solvent like 1,2ꢀdichlorobenzene (oDCB)
run 9, Table 1), indicating that solvent polarity considerably affected Lewis pairsꢀmediated the
ROAC of PA/CHO. We speculated a more polar medium would dissociate the interacting Lewis
While the
(
1e
pairs, favoring a slower propagation rate as compared to the experiment carried out in xylene.
Further kinetic experiments were performed to test chain initiation and the dependence of
copolymerization rates versus comonomers by pipetting aliquots during the copolymerization
(Figures S11&S12). In neat epoxide, the kinetic plots exhibited no induction period and a good
linearity about PA conversion, suggesting the ROAC expressly followed zeroꢀorder kinetics with

Page 7 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
respect to PA concentration for all of the Lewis pairs employed. Recently, Coates et al. also
deserved the same zero order dependence of the rate on norborneneꢀ2,3ꢀdicarboxylic anhydride
15c
(NA) by using (salph)AlCl/PPNCl.
This similar example indicated the ringꢀopening of
anhydrides was in a fast process. Obviously, LBs rather than LAs played a dominant role in the
catalytic activity according to the slope of kinetic plots.
a
Table 1. The ROAC of PA and CHO catalyzed by Lewis pairs.
e
Run Lewis pair
LA/LB/PA/CHO
t (h)
Conv.
TOF
Alternating degree Mn
，
M
n, GPC PDI
calc.
b
ꢀ1 c
d
e
e
(%)
(h )
(%)
>99
>99
27
(kDa)
17.0
11.4
15.3
11.8
11.7
18.9
11.8
12.2
10.3
23.6
42.3
12.3
(kDa)
6.1
1
2
3
4
5
6
7
8
9
1
1
1
Zn(C
Zn(C
Zn(C
Zn(C
6
6
6
6
F
F
F
5
5
5
)
)
)
2
2
2
/DMAP
/DMAP
/DMAP
1:1:100:500
1:2:100:500
2:1:100:500
1
69
93
30
96
95
77
48
99
84
96
86
>99
69
1.21
1.20
1.93
1.17
1.22
1.23
1.11
1.18
1.12
1.22
1.13
1.27
0.8
2.5
0.8
0.75
1.5
1.5
3
116
12
5.6
16.8
6.9
H
5
)
2
/DMAP 1:2:100:500
120
127
51
>99
>99
>99
>99
>99
>99
>99
>99
>99
ZnEt /DMAP
1:2:100:500
1:1:100:500
7.1
2
Zn(C
Zn(C
Zn(C
Zn(C
6
6
6
6
F
5
F
5
F
5
F
5
)
2
)
2
)
2
)
2
/DBU
5.9
/MTBD 1:1:100:500
32
4.7
f
/DMAP
/DMAP
1:2:100:100
1:2:100:100
1:2:200:500
1:2:400:500
1:2:100:500
33
7.7
g
3
28
3.0
0
1
2
Zn(C F ) /DMAP
1.5
3
128
115
ꢀ
13.5
17.9
10.5
6
6
6
5 2
Zn(C
Zn(C
F
5
5
)
2
2
/DMAP
/DMAP
h
F
)
9
a
Unless otherwise mentioned, the copolymerizations were carried out at 110 °C in neat epoxide with PA
b
1
c
d
concentration of 2 mol/L. PA conversion was determined by H NMR. TOF = turnover frequency. Ether
1
e
linkage was determined via H NMR by integrating ester and ether signals. M and PDI were determined by GPC
n
f
g
h
analysis. In xylene. In oDCB. PA was sublimed three times.
Under aforementioned conditions, moderate MWs and narrow molecular weight distributions
in the comonomer molar ratio of 100:500 was observed in the resulting poly(PAꢀaltꢀCHO).
n
Evolution of the number average molecular weights (M ) against PA conversion for the ROAC was
in a linear correlation with a narrow molecular weight distribution (PDI ~ 1.20) (Figure 1),
suggesting a controlled behavior of the copolymerization system. Gel permeation chromatography
(GPC) analysis revealed that data of molecular weights (MWs) obtained from crude reaction
mixture was in a good accordance with these obtained from precipitated polymers (Figure S14ꢀS16).

Green Chemistry
Page 8 of 19
View Article Online
DOI: 10.1039/C7GC03261G
The resultant alternating copolymers remained unimodal GPC traces at low PA conversion (Figure
S14). Nevertheless with PA conversion increasing, the samples obtained from a single
polymerization reaction showed bimodal GPC traces (Figure S15&S16). Besides, the polyesters had
lower MWs than calculated ones. With molar ratio of monomers increased to 200:500 and 400:500
(runs 10&11, Table 1), molecular weights of corresponding polymers rose to 13.5 and 17.9 kg/mol,
respectively. Actually, lower MWs and bimodal GPC traces implied the presence of adventitious
water or trace amounts of hydrolysed anhydride, which acted as a bifunctional chain transfer agent.
The bifunctional agent would be expected to propagate at the similar rate as chains initiated from
catalysts, leading to chain growth from both hydroxyl moieties (see Figure S43 and Scheme S1) and
9
c,11dꢀf, 13a,f,g,16a,18a,b
two distinct distributions of MWs differing by approximately a factor of 2.
Considering the presence of hydrolysed anhydride, PA was refined three times by sublimation. The
refined PA led to an appreciable increase of MWs (run 12, Table 1), suggesting that hydrolysed PA
acted as an efficient initiator of the copolymerization. When the copolymerization reaction was
further extended to 9 h, molecular weight distribution widen from 1.16 to 1.27 (Figure S17). The
result showed that intermolecular transesterification side reaction tended to happen after full
conversion of PA was reached (intermolecular transesterification was not observed, vide infra).
n
Figure 1. Plots of M and PDI versus PA conversion.
ROAC of different anhydrides and epoxides. After model experiments investigation of
PA/CHO, various comonomers were selected to explore the feasibility for the synthesis of
6 5 2
structurally diverse polyesters. The results suggested that Zn(C F ) /DMAP pair could effectively

Page 9 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
catalyze the ROAC of PA with a variety of epoxides for the synthesis of various semiaromatic
polyesters with wellꢀdefined structures (runs 1ꢀ5, Table 2). The type of epoxides dramatically
affected copolymerization rates. Propylene oxide (PO), one of the widest studies, had a similar
reactivity with CHO in this work (run 1, Table 2). While ECH exhibited an enhanced
ꢀ1
polymerization rate (TOF = 210 h ) to afford polyesters (run 2, Table 2). High reactivity of ECH
1
2d
could attribute to electronꢀwithdrawing chlorine atom. However, a lower molecular weight was
observed from poly(PAꢀaltꢀECH), in which chain transfer might be faster due to the stronger polar
ECH. Decreasing temperature from 110 ºC to 60 ºC resulted in a slower polymerization rate (run 4,
Table 2), but the TOF value was still eight times as high as the system of alcohol/phosphazene
18c
base. Another electronꢀpoor styrene oxide (SO) was also viable comonomers with a higher TOF
value than CHO and PO (run 5, Table 2).
a
Table 2. The ROAC of various anhydrides and epoxides catalyzed by Zn(C F ) /DMAP pair.
6
5 2
b
ꢀ1 c
b
d
d
Run Anhydride
Epoxide
PO
Solvent
bulk
t (h)
0.8
0.4
4
Conv. (%)
TOF (h )
Alternating degree (%)
M
n
(kDa)
PDI
1
2
3
4
5
6
7
8
9
1
PA
81
84
81
84
83
21
41
14
56
82
101
210
20
42
166
11
41
7
>99
>99
>99
>99
>99
33
7.4
2.1
5.6
2.6
3.8
1.9
5.3
1.25
1.17
1.21
1.23
1.28
1.03
1.05
PA
ECH
CHO
ECH
SO
bulk
e
f
PA
bulk
PA
bulk
2
PA
bulk
0.5
2
SA
CHO
CHO
CHO
CHO
CHO
bulk
SA
xylene
oDCB
bulk
1
>99
72
g
g
SA
2
ꢀ
ꢀ
CHA
CHA
4
14
21
75
5.4
9.8
1.07
1.09
0
xylene
4
96
a
Unless otherwise mentioned, copolymerizations were carried out at 110 °C; Zn(C
6
F
5
)
2
/DMAP/Anhydride/
b
1
Epoxide = 1:2:100:500 in bulk or 1:2:100:100 in solution. The conversion of anhydride was determined by H
c
d
e
f
g
NMR. TOF = turnover frequency. Determined by GPC analysis. At 80 °C. At 60 °C. Not defined.

Green Chemistry
Page 10 of 19
View Article Online
DOI: 10.1039/C7GC03261G
In general, ringꢀopening of asymmetric terminal epoxides (including PO, ECH and SO) could
occur either at the methylene or at the more sterically hindered methine, in which the resultant
polyesters would contain headꢀtoꢀtail (HT), headꢀtoꢀhead (HH) and tailꢀtoꢀtail (TT) junctions. When
ringꢀopening of epoxides occurred exclusively at one of these positions via a regioselective
12c,d
13
catalyst,
showed useful information for the analysis of regio(ir)regular insertion. As shown in Figure 2,
66.87 and 166.73 ppm of carbon chemical shifts were assigned to carbonyl signals from HT
regioregular polyesters could be fabricated. C NMR spectra of the resultant polyesters
1
linkages for poly(PAꢀaltꢀPO). Two separate resonance peaks at 167.01 and 166.60 ppm represented
carbonyl signals of HH and TT linkages, respectively. Clearly, a highly regioselective ROAC of PA
6 5 2
with PO could be achieved by using Zn(C F ) /DMAP (1:1) pair (Figure 2B). In order to verify
quantificationally the regioregularity of copolymers, enantiopure terminal epoxides (SꢀPO and SꢀSO)
were used for the ROAC. Subsequently, enantiopurity of diols resulting from the hydrolysis of the
11c,13h
resultant polyesters was determined by gas chromatography analysis.
The high enantiomeric
ratio, representing the percent HT junctions in the copolymers, certifying the regioselective ROAC
of asymmetric terminal epoxides (PO) with anhydrides (Figure S21). Indeed, this was the first
observation of regioselective insertion for the ROAC by using simple catalytic system. We
speculated the interacting Lewis pair could exclusively attack the less sterically crowded methylene
carbon of PO in chain initiation and propagation.
1
3
Figure 2. C NMR spectra of carbonyl regions of poly(PAꢀaltꢀPO): (A) regioirregular copolymer and (B)
regioregular copolymer catalyzed by Zn(C /DMAP (1:1).
6 5 2
F )

Page 11 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
Next, the ROAC of different anhydrides with CHO was also evaluated by using
Zn(C /DMAP pair. For SA/CHO copolymerization, bulk mode revealed a low TOF value and
6 5 2
F )
inferior alternating selectivity (run 6, Table 2). However, an unanticipated result suggested that
nonpolar xylene was more beneficial to the copolymerization (run 7, Table 2). In contrast, oDCB
solvent played a poorer role in the ROAC of SA and CHO (run 8, Table 2). Polar medium offered
rather low conversions and a mass of polyether segment compared with those in xylene. Inspired by
2
0
proton transfer reactions of weak carbonꢀacids with amine bases in polar solvent, we envisioned
that the αꢀH of cyclic anhydrides activated by carbonyl is rendered sufficiently activity to undergo
deprotonating with organic bases without the need for intervention by other catalysts. That is, the
αꢀH was easily extracted by the inherent organic amines to yield the undesired salts in
stoichiometric amounts in strong polar medium. The propensity of amines to react with common
anhydrides bearing αꢀH easily results in their inactivation, which shed light on the observation that
1
8b
SA was difficult to copolymerize with epoxides in polar medium by using organic bases. The
1
deprotonation was revealed by the H NMR spectra involving different anhydrides (MA and SA)
and amines. At 25 ºC, the observations that αꢀH signals of MA and SA disappeared or decreased
with equivalent amines (the αꢀH signal was not observed because of fast proton exchange between
3
MA and DMAP), exhibited a high reactivity for deprotonation in polar CDCl (Figure 3). A
8
contrastive experiment was performed to monitor αꢀH signals of MA in nonpolar tolueneꢀd (Figure
1
S29). The H NMR spectra revealed that nonpolar solvents could suppress the deprotonation
1
between αꢀH of anhydrides and amines in some degree. Importantly, as shown in Figure S30, the H
6 5 2 3
NMR spectra of Zn(C F ) /DMAP/MA in polar CDCl also showed the signals at 7.05 ppm
assigned to the αꢀH of MA. Obviously, the interaction of LAs with LBs impeded the deprotonation
side reaction as well. Intrigued by these results, we selected nonpolar solvents to dilute polar
epoxides. In this endeavor, the ROAC of anhydrides bearing αꢀH and epoxides can be achieved
under the facile conditions (Figure S32ꢀS40).

Green Chemistry
Page 12 of 19
View Article Online
DOI: 10.1039/C7GC03261G
Figure 3. Possible structures of the products for the proton transferof anhydrides bearing αꢀH and Lewis
bases in polar CDCl , (a) MA/DMAP, (b) SA/DBU.
3
Mechanism and microstructure analysis. To gain insight into the mechanism of the ROAC
1
catalyzed by Lewis pairs, we performed variable temperature (VT) H NMR experiments about well
6 5 2
reactive PA/CHO substrate. Firstly, as our previous report, Zn(C F ) /DMAP had a coordination to
6
generate tetrahedron complex at low temperature. With temperature elevating from 25 ºC to 110 ºC,
the interaction of the LA and the LB was impaired. Further, no obvious chemical shifts were
1
observed in VT H NMR spectra of Zn(C
6
F )
5 2
/PA, revealing an impossible pathway involving that
was very good at
Zn(C
6
F
5
)
2
facilitated PA (Figure S41). Meanwhile, considering that sole Zn(C
6
F
5
)
2
catalyzing ROP of CHO to form polyether, we clearly concluded that LA acted as epoxide activator
with the presence of LBs. During our study, two elegant publications about ringꢀopening
copolymerization of CO /epoxides and COS/epoxides promoted by BEt /LB similarly proved the
2
3
7
,1e
1
rationality that LA served as epoxide activator for ringꢀopening.
In addition, the VT H NMR
spectra of Zn(C F ) /DMAP/CHO in C D Cl in an NMR tube showed that the protons chemical
6
5 2
2
2
4
shifts at 3.60 and 3.79 ppm gradually appeared (Figure S42), strongly suggesting that CHO was
firstly activated and ringꢀopened by Lewis pairs. Then a zwitterion generated with an amine
initiator in one end and Zn(C
selectively reacted with an anhydride to form an ester unit and a carboxylate nucleophile (5.14 and
.17 ppm in Figure 4). In chain propagation cycle, Zn(C worked to activate epoxides. Alkoxide
and carboxylate intermediates alternately regenerated, enhancing one ester unit into growing chain
6 5 2
F ) in the other end (Scheme 2). Subsequently, the active alkoxide
4
6 5 2
F )

Page 13 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
end. In this cyclic period, it was considered that activating and ringꢀopening of epoxide to alkoxide
15c
anion was the rateꢀdetermining step (vide infra & Figure S13). Notably, different from cationic
6
mechanism of polyether formation using sole LAs, loose pair interaction of LA and LB suppressed
the possibility of alkoxide zwitterionic intermediate attacking epoxides, proven by the result that
Zn(C
6
F
5
)
2
/DMAP/CHO cannot gain any polymer at 110 °C even prolonging reaction time.
1
Figure 4. VT H NMR spectra of the mixture of Zn(C H ) /DMAP/PA/CHO in an NMR tube with the molar
6
5 2
2 2 4
ratio of 1:1:1:1 in C D Cl .
Scheme 2. Proposed mechanism schematic illustration of the ROAC of PA/CHO mediated by Lewis pairs.

Green Chemistry
Page 14 of 19
View Article Online
DOI: 10.1039/C7GC03261G
Figure 5. (a) Positive ion MALDIꢀTOF MS of poly(PAꢀaltꢀCHO) in a low PA conversion (18%, PA was
sublimed three times) precipitated in nꢀhexane without quenching. (b) A detail of blue grid in (a) with
calculated and exact masses. (c) The fitting relationship of M vs repeating unit.
n
Meanwhile, the obtained polyesters were analyzed by matrixꢀassisted laser
desorption/ionization timeꢀofꢀflight mass spectrometry (MALDIꢀTOF MS). Notably, anhydrous
nꢀhexane was carefully used to isolate and prepare copolymer samples from reaction mixture
without quenching in case that quaternary ammonium moieties of chain ends were substituted by
active species (water and acids). As shown in Figure 5, when PA was at a low conversion (18%),
only one distribution was observed with DMAP endꢀcapped chains. The great agreement of
11d
calculated mass with exact mass suggested that the polymer samples were ionized beforehand.
The distribution exactly consisted of n [CHO + PA], implying that the epoxide is the preferred
monomer to be initiated with Zn(C /DMAP and the ringꢀopening of epoxide was
6 5 2
F )
rateꢀdetermining step in chain propagation. In accordance with the possible mechanism we
proposed, DMAP reacted with epoxides rather than anhydrides to generate chain initiating
zwitterionic species. When PA conversion was further increased to 44%, two distinct distributions
could be observed (Figure S43). Each distribution showed an m/z interval of 246.26 among the
consecutive peaks corresponding to the addition of a [CHO + PA] repeating unit, which is in well
agreement with a fully alternating structure. The end group of the intense distribution (86.1%) is
ωꢀhydroxyl and DMAP. The low distribution (11.2%) is α,ωꢀdiꢀhydroxyl endꢀcapped, in which the

Page 15 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
calculated residue corresponds to the molecular weight of one phthalic acid. The observation
18b
confirmed that trace hydrolytic PA can be bifunctional chain initiating species (Scheme S1).
Finally, macrocyclic copolymers formed by intramolecular transesterification were not observed
from all of MALDI TOF MS (Figures 5 and S43). The possible reason was ascribed to a big rigid
backbone of poly(PAꢀaltꢀCHO).
Conclusions
In conclusion, we have introduced here a group of simple and lowꢀtoxic Lewis pairs for the
synthesis of polyesters by the way of ROAC. At 110 °C, TOFs of Lewis pairs used in this work
ꢀ1
were as high as 210 h in bulk, comparable to many metal/ligand complexes. The polarity of
solvents, molar ratio and the interaction mode of LA/LB had a great impact on copolymerization
rate and chemoselectivity. Furthermore, through optimization of copolymerization conditions,
deprotonation side reaction was suppressed by nonpolar solvents and Lewis pairs. Subsequently, an
array of structurally diverse polyesters, including semiaromatic and aliphatic polyesters, could be
efficiently synthesized by using the versatile Lewis pairs in a controlled manner. The resultant
polyesters showed high alternating degree (> 99%) and high regioregularity (for asymmetric
epoxides such as PO, ECH and SO). Besides, possible chain initiation and propagation mechanism
revealed that epoxides were firstly attacked by Lewis bases, and then carboxylates and alkoxides
alternately regenerated, resulting into perfect polyester structures. Therefore, a simple and
environmentally benign approach was developed for the preparation of polyesters with diverse
geometrical structures.
Supplementary Information
Characterizations, NMR spectra, additional GPC traces and statistical analysis (PDF)
Corresponding Author
*
*
Eꢀmail binwang@tju.edu.cn (B.W.).
Eꢀmail ysli@tju.edu.cn (Y. S. L.).
Notes
The authors declare no competing financial interests.

Green Chemistry
Page 16 of 19
View Article Online
DOI: 10.1039/C7GC03261G
Acknowledgements
The authors are grateful for financial support by the National Natural Science Foundation of the
People’s Republic of China (Grant No. 21234006).
Keywords: ringꢀopening alternating copolymerization • Lewis pair • polyester • regioselectivity •
deprotonation reaction
References
1
(a) C. K. Ober, S. Z. D. Cheng, P. T. Hammond, M. Muthukumar, E. Reichmanis, K. L. Wooley
and T. P. Lodge. Macromolecules, 2009, 42, 465–471; (b) W. N. Ottou, H. Sardon, D.
Mecerreyes, J. Vignolle and D. Taton, Prog. Polym. Sci., 2016, 56, 64–115; (c) S. Hua, J. Zhao,
G. Zhang, H. Schlaad, Prog. Polym. Sci., 2017, 57, 34–77; (d) B. Lin and R. M. Waymouth, J.
Am. Chem. Soc., 2017, 139, 1645−1652; (e) D. Zhang, S. K. Boopathi, N. Hadjichristidis, Y.
Gnanou and X. Feng, J. Am. Chem. Soc., 2016, 138, 11117−11120.
2
3
4
Some reviews and examples: (a) N. E. Kamber, W. Jeong, R. M. Waymouth, R. C. Pratt, B. G.
G. Lohmeijer and J. L. Hedrick, Chem. Rev., 2007, 107, 5813−5840; (b) M. Fevre, J. Pinaud, Y.
Gnanou, J. Vignolle and D. Taton, Chem. Soc. Rev., 2013, 42, 2142–2172; (c) M. Hong and E.
Y. X. Chen, Angew. Chem. Int. Ed., 2016, 55, 4188–4193.
Some recent reviews and examples: (a) C. Thomas and B. Bibal, Green Chem., 2014, 16,
1
8
687−1699; (b) X. Zhang, G. O. Jones, J. L. Hedrick and R. M. Waymouth, Nat. Chem., 2016,
, 1047−1053; (c) S. Xu, H. Sun, J. Liu, J. Xu, X. Pan, H. Dong, Y. Liu, Z. Li and K. Guo,
Polym. Chem., 2016, 7, 6843−6853.
(a) Y. Zhang, G. M. Miyake and E. Y. X Chen, Angew. Chem. Int. Ed., 2010, 49, 10158−10162;
(
b) Y. Zhang, G. M. Miyake, M. G. John, L. Falivene, L. Caporaso, L. Cavallo and E. Y. X.
Chen, Dalton Trans., 2012, 41, 9119−9134; (c) T. Xu and E. Y. X. Chen, J. Am. Chem. Soc.,
014, 136, 1774−1777; (d) Y. B. Jia, Y. B. Wang, W. M. Ren, T. Xu, J. Wang and X. B. Lu,
2
Macromolecules, 2014, 47, 1966−1972; (e) Y. B. Jia, W. M. Ren, S. J. Liu, T. Xu, Y. B. Wang
and X. B. Lu, ACS Macro Lett., 2014, 3, 896−899; (f) M. G. M. Knaus, M. M. Giuman, A.
Pöthig and B. Rieger, J. Am. Chem. Soc., 2016, 138, 7776−7781; (g) W. N. Ottou, E.
CondeꢀMendizabal, A. Pascual, A. ꢀL. Wirotius, D. Bourichon, J. Vignolle, F. Robert, Y.
Landais, J. ꢀM. Sotiropoulos, K. Miqueu and D. Taton, Macromolecules, 2017, 50, 762–774.
(a) E. PiedraꢀArroni, C. Ladavière, A. Amgoune and D. Bourissou, J. Am. Chem. Soc., 2013,
5
1
35, 13306−13309; (b) S. Naumann, P. B. V. Scholten, J. A. Wilson and A. P. Dove, J. Am.

Page 17 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
Chem. Soc., 2015, 137, 14439−14445; (c) S. Naumann and D. Wang, Macromolecules, 2016,
4
5
9, 8869−8878; (d) Q. Wang, W. Zhao, J. He, Y. Zhang, E. Y. X. Chen, Macromolecules, 2017,
0, 123−136.
6
7
X. Q. Li, B. Wang, H. Y. Ji and Y. S. Li, Catal. Sci. Technol., 2016, 6, 7763−7772.
J. L. Yang, H. L. Wu, Y. Li, X. H. Zhang and D. J. Darensbourg, Angew. Chem. Int. Ed., 2017,
56, 5774−5779.
8
(a) J. M. Longo, M. J. Sanford and G. W. Coates, Chem. Rev., 2016, 116, 15167−15197; (b) S.
Paul, Y. Zhu, C. Romain, R. Brooks, P. K. Saini and C. K. Williams, Chem. Commun., 2015, 51,
6459−6479; (c) R. Tong, Ind. Eng. Chem. Res., 2017, 56, 4207−4219; (d) M. Rabnawaz, I.
Wyman, R. Auras and S. Cheng, Green Chem., 2017, 19, 4737−4753.
9
(a) E. Mahmoud, D. A. Watson and R. F. Lobo, Green Chem., 2014, 16, 167–175; (b) M. N.
Rashed, S. M. A. H. Siddiki, M. A. Ali, S. K. Moromi, A. S. Touchy, K. Kon, T. Toyao and K.
Shimizu, Green Chem., 2017, 19, 3238−3242; (c) M. Winkler, C. Romain, M. A. R. Meier and
C. K. Williams, Green Chem., 2015, 17, 300−306.
1
0 (a) Z. T. Du, J. P. Ma, F. Wang, J. X. Liu and J. Xu, Green Chem., 2011, 13, 554–557; (b) N.
AlonsoꢀFagundez, M. L. Granados, R. Mariscal and M. Ojeda, ChemSusChem, 2012, 5, 1984–
1990.
1
1 (a) C. Robert, F. de Montigny and C. M. Thomas, Nat. Commun., 2011, 2, 586. (b) R. C. Jeske,
A. M. DiCiccio and G. W. Coates, J. Am. Chem. Soc., 2007, 129, 11330−11331; (c) J. M.
Longo, A. M. DiCiccio and G. W. Coates, J. Am. Chem. Soc., 2014, 136, 15897−15900. (d) S.
Huijser, E. Hosseini Nejad, R. Sablong, C. de Jong, C. E. Koning and R. Duchateau,
Macromolecules, 2011, 44, 1132–1139; (e) E. Hosseini Nejad, A. Paoniasari, C. E. Koning and
R. Duchateau, Polym. Chem., 2012, 3, 1308–1313; (f) E. Hosseini Nejad, A. Paoniasari, C. G.
van Melis, C. E. Koning and R. Duchateau, Macromolecules, 2013, 46, 631–637.
1
2 (a) R. A. Sheldon, Green Chem., 2014, 16, 950–963; (b) A. M. DiCiccio and Coates, G. W. J.
Am. Chem. Soc., 2011, 133, 10724−10727; (c) J. Liu, Y. Y. Bao, Y. Liu, W. M. Ren and X. B.
Lu, Polym. Chem., 2013, 4, 1439–1444. (d) A. M. DiCiccio, J. M. Longo, G. G.
RodríguezꢀCalero and G. W. Coates, J. Am. Chem. Soc., 2016, 138, 7107−7113.
1
3 (a) E. Hosseini Nejad, C. G. W. van Melis, T. J. Vermeer, C. E. Koning, R. Duchateau,
Macromolecules, 2012, 45, 1770−1776; (b) D. J. Darensbourg, R. R. Poland, C. Escobedo,
Macromolecules, 2012, 45, 2242−2248; (c) N. D. Harrold, Y. Li, M. H. Chisholm,
Macromolecules, 2013, 46, 692−698; (d) R. Baumgartner, Z. Song, Y. Zhang, J. Cheng, Polym.
Chem., 2015, 6, 3586–3590; (e) G. Si, L. Zhang, B. Han, Z. Duan, B. Li, J. Dong, X. Li, B. Liu,
Polym. Chem., 2015, 6, 6372–6377; (f) N. J. Van Zee, G. W. Coates, Angew. Chem. Int. Ed.,

Green Chemistry
Page 18 of 19
View Article Online
DOI: 10.1039/C7GC03261G
2015, 54, 2665−2668; (g) B. Han, L. Zhang, M. Yang, B. Liu, X. Dong, P. Theato,
Macromolecules, 2016, 49, 6232−6239; (h) J. Li, Y. Liu, W. M. Ren and X. B. Lu, J. Am. Chem.
Soc., 2016, 138, 11493−11496.
1
4 C. Robert, T. Ohkawara and K. Nozaki, Chem. Eur. J., 2014, 20, 4789–4795.
5 (a) N. J. Van Zee, M. J. Sanford and G. W. Coates, J. Am. Chem. Soc., 2016, 138, 2755−2761;
1
(b) M. J. Sanford, L. Peña Carrodeguas, N. J. Van Zee, A. W. Kleij and G. W. Coates,
Macromolecules, 2016, 49, 6394–6400; (c) M. E. Fieser, M. J. Sanford, L. A. Mitchell, C. R.
Dunbar, M. Mandal, N. J. Van Zee, D. M. Urness, C. J. Cramer, G. W. Coates and W. B.
Tolman, J. Am. Chem. Soc., 2017, 139, 15222ꢀ15231.
1
6 (a) P. K. Saini, C. Romain, Y. Zhu and C. K. Williams, Polym. Chem., 2014, 5, 6068–6075; (b)
Y. Zhu, C. Romain and C. K. Williams, J. Am. Chem. Soc., 2015, 137, 12179−12182; (c) J. A.
Garden, Saini, P. K.; Williams, C. K. J. Am. Chem. Soc., 2015, 137, 15078−15081; (d) C.
Romain, Y. Zhu, P. Dingwall, S. Paul, H. S. Rzepa, A. Buchard and C. K. Williams, J. Am.
Chem. Soc., 2016, 138, 4120−4131; (e) P. K. Saini, G. Fiorani, R. T. Mathers and C. K.
Williams, Chem. Eur. J., 2017, 23, 4260–4265.
1
7 L. Peña Carrodeguas, C. Martín, A. W. Kleij, Macromolecules, 2017, 50, 5337–5345.
8 (a) B. Han, L. Zhang, B. Liu, X. Dong, I. Kim, Z. Duan, P. Theato, Macromolecules, 2015, 48,
1
3
431−3437; (b) Z. Hošťálek, O. Trhlíková, Z. Walterová, T. Martinez, F. Peruch, H. Cramail, J.
Merna, Eur. Polym. J., 2017, 88, 433−447; (c) H. Li, J. Zhao, G. Zhang, ACS Macro Lett., 2017,
, 1094−1098.
9 (a) J. Hu, J. Ma, Q. Zhu, Q. Qian, H. Han, Q. Mei and B. Han, Green Chem., 2016, 18,
6
1
3
82−385; (b) A. P. Thankachan, K. S. Sindhu, K. K. Krishnan and G. Anilkumar, RSC Adv.,
015, 5, 32675−32678; (c) M. Schappacher, M. Le Hellaye, R. Bareille, M. C. Durrieu and S.
2
M. Guillaume, Macromol. Biosci., 2010, 10, 60–67; (d) A. Nachtergael, O. Coulembier, P.
Dubois, M. Helvenstein, P. Duez, B. Blankert and L. Mespouille, Biomacromolecules, 2015, 16,
507–514; (e) Y. Xia, J. Shen, H. Alamri, N. Hadjichristidis, J. Zhao, Y. Wang and G. Zhang,
Biomacromolecules, 2017, 18, 3233−3237.
20 (a) N. Mittal, K. M. Lippert, C. K. De, E. G. Klauber, T. J. Emge, P. R. Schreiner and D. Seidel,
J. Am. Chem. Soc., 2015, 137, 5748−5758; (b) A. Jarczewski and C. D. Hubbard, J. Mol. Struct.,
2003, 649, 287−307; (c) J. R. Gandler, O. L. Saunders and R. Barbosa, J. Org. Chem., 1997, 62,
4
677−4682; (d) X. Zhang, S. R. Yeh, S. Hong, M. Freccero, A. Albini, D. E. Falvey and P. S.
Mariano, J. Am. Chem. Soc., 1994, 116, 4211−4220.

Page 19 of 19
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03261G
Table of contents
Lewis Pairs for Ring-Opening Alternating Copolymerization of Cyclic Anhydrides and
Epoxides
HeꢀYuan Ji, Bin Wang, Li Pan, YueꢀSheng Li
Simple and lowꢀtoxic Lewis pairs were introduced for ringꢀopening alternating copolymerization of
anhydrides and epoxides to diversified polyesters.