Switchable Polymerization Triggered by Fast and Quantitative Insertion of Carbon Monoxide into Cobalt–Oxygen Bonds

Article abstract of DOI:10.1002/anie.201914216

A strategy that uses carbon monoxide (CO) as a molecular trigger to switch the polymerization mechanism of a cobalt Salen complex [salen=(R,R)-N,N′-bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexanediamine] from ring-opening copolymerization (ROCOP) of ep

Full text of DOI:10.1002/anie.201914216

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Switchable Polymerization Triggered by Fast and Quantitative
Insertion of Carbon Monoxide into Cobalt-Oxygen Bonds
Authors: Yong Wang, Yajun Zhao, Shuaishuai Zhu, Xingping Zhou,
Jing Xu, Xiaolin Xie, and Rinaldo Poli
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.201914216
Angew. Chem. 10.1002/ange.201914216
Link to VoR: http://dx.doi.org/10.1002/anie.201914216
http://dx.doi.org/10.1002/ange.201914216

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
Switchable Polymerization Triggered by Fast and Quantitative
Insertion of Carbon Monoxide into Cobalt-Oxygen Bonds
Yong Wang,*[a] Yajun Zhao,[a] Shuaishuai Zhu,[a] Xingping Zhou,*[a] Jing Xu,[b] Xiaolin Xie[a] and Rinaldo
Poli*[c]
Abstract: A strategy that uses carbon monoxide (CO) as a molecular
The
ring-opening
copolymerization
(ROCOP)
of
trigger to switch the polymerization mechanism of a cobalt Salen
epoxides/anhydrides has emerged as a burgeoning method to
produce polyesters.5 The wide availability of renewable
monomers and the opportunity for post-polymerization
modification open access to a myriad of new materials with unique
properties that are not accessible through traditional
condensation reactions.6 Cobalt Salcy [bis(salicylaldimine)]
complexes featuring nucleophilic axial groups [(Salcy)CoIIIX; X =
halides, carboxyl, etc.] are among the most investigated catalysts
for the copolymerization, wherein the alternative insertion of
anhydrides and epoxides is directly correlated to the reversible
formation and dissociation of the Co-O bonds (Scheme 1A).7
Relative to the polar Co-O bonds, Co-C bonds tend to break
homolytically to generate carbon-based radicals as is known to
occur, for instance, in the biological action of vitamin B12.8 This
process may be promoted by thermal treatment or by light
irradiation.9 Recently, cobalt complexes have been used to
ensure the well-controlled polymerization of a wide variety of the
challenging nonconjugated monomers through the strategy
known as organometallic mediated controlled radical
polymerization (OMRP) (Scheme 1B).10 Especially, Fu and Peng
et al. demonstrated that an alkoxycarbonyl cobalt Salen complex
(Salen is a specific Salcyl ligand with a trans-cyclohex-diyl tether,
Scheme 2) could serve as the photoinitiator as well as the
mediator for visible light controlled OMRP.11
In order to expand the application scopes of biodegradable
polymers resulted from ROCOP, intensive research efforts have
been directed to the rational design and synthesis of block
copolymers by combining ROCOP and other living/controlled
polymerization techniques.12 We have reported the one-step
synthesis of well-defined CO2-based block copolymers via
concurrent ROCOP and RAFT polymerization using a bifunctional
chain transfer agent in 2018, however, only conjugated
monomers such as methyl methacrylate (MMA) and styrene (St)
were involved.12b In a recent contribution, we have illustrated a
complex [Salen
= (R,R)-N,N′-bis(3,5-di-tertbutylsalicylidene)-1,2-
cyclohexanediamine] from ring-opening copolymerization (ROCOP)
of epoxides/anhydrides to organometallic mediated controlled radical
polymerization (OMRP) of acrylates is described. The key
phenomenon is a rapid and quantitative insertion of CO into the Co-O
bond, allowing for in situ transformation of the ROCOP active species
(Salen)CoIII-OR into the OMRP photoinitiator (Salen)CoIII-CO2R. The
proposed mechanism, which involves CO coordination to (Salen)CoIII-
OR and subsequent intramolecular rearrangement via migratory
insertion has been rationalized by DFT calculations. Regulated by
both CO and visible light, on-demand sequence control can be
achieved for the one-pot synthesis of polyester-b-polyacrylate diblock
copolymers (Đ < 1.15).
Introduction
The fundamental properties and functionalities of polymers are
directly correlated to their primary microstructures. Whereas
enzymes display exceptional efficiency for the synthesis of
precisely programmed biopolymers within a living cell, it is a major
challenge to regulate polymer sequence and architecture from
mixture of monomers for modern polymer chemistry.1 An
attractive solution may lay in the rapidly evolving field of
switchable catalysis, whereby the reactivity of a metal-based
polymerization catalyst is altered in response to an external
stimulus.2 Typically, the switch involves a change of the active
species, which is achieved either via in situ modification of the
metal coordination sphere through
a stimulus-responsive
functional group,3 or by a metal oxidation state change via a redox
process4. In this regard, current achievements in switchable
polymerization have mostly been limited to catalysts that iterate
between mechanistically similar polymerization reactions.
mechanistic switch involving ROCOP and OMRP with
a
(Salen)Co system as catalyst/mediator for both mechanisms. The
switch operated in the OMRP-to-ROCOP direction via a O2-
triggered modification of the polymer chain end from Co-R to Co-
OR and was implemented for the development of diblock
copolymers with a poly(vinyl acetate) block and a polycarbonate
block (by ROCOP of CO2/epoxides).2d Herein, we report a switch
of polymerization mechanism in the opposite direction, from
ROCOP of epoxides/anhydrides to OMRP, using carbon
monoxide (CO) as a molecular trigger and the (Salen)CoIIICl
complex as the mediator (Scheme 2). Central to this protocol is
the direct and quantitative insertion of CO into Co-O bonds, which
enables an in situ transformation of the ROCOP active species
(Salen)CoIII-OR into the OMRP photoinitiator (Salen)CoIII-CO2R.
Although CO insertions into metal-carbon single and multiple
bonds are implicated in many stoichiometric and catalytic
transformations of great significance in both industrial and acade-
[a]
Dr. Y. Wang,Y. J. Zhao, S. S. Zhu, Prof. Dr. X. P. Zhou, Prof. Dr. X.
L. Xie.
School of Chemistry and Chemical Engineering
Huazhong University of Science and Technology
Wuhan 430074, P. R. China
E-mail: tcwy@mail.hust.edu.cn; xpzhou@mail.hust.edu.cn
Prof. Dr. J. Xu.
College of Chemistry and Material Science
Shandong Agricultural University
Taian 271018, P. R. China
Prof. Dr. R. Poli.
Laboratoire de Chimie de Coordination (LCC-CNRS)
Université de Toulouse, UPS, INPT
205, route de Narbonne, 31077 Toulouse (France)
E-mail: rinaldo.poli@lcc-toulouse.fr
[b]
[c]
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
Scheme 1. Scheme 1 Chain propagation processes: cobalt(III) complex mediated ROCOP of anhydrides/epoxides (A); cobalt complex mediated controlled/ living
radical polymerization of vinyl monomers (B).
carbonyl carbon atom is quite positive charged.16 Under the
assumption of a facile dissociation of (Salen)CoIII-OR into
[(Salen)CoIII]+ and the highly nucleophilic –OR (Scheme 3A1),17 it
is possible to envision that (Salen)CoIII-OR may transform into
(Salen)CoIII-CO2R in the presence of CO. Alternatively, CO may
directly coordinate to (Salen)CoIII-OR and the resulting
[(Salen)CoIII(CO)]OR adduct may intramolecularly rearrange via
insertion (Scheme 3A2). Yet another possibility is ionic
dissociation after CO coordination (Scheme 3A3). All these
mechanisms were previously shown to occur for other systems,
depending on the nature of the complex and on the conditions.18
It is worth mentioning that the methoxycarbonyl cobalt complex
(Salen)CoIII-CO2Me used by Fu and Peng was synthesized by
sequential oxidation of (Salen)CoII with oxone in methanol and
carbonylation process with CO, which is indicative of the
formation of a (Salen)CoIII-OMe intermediate.11
Scheme 2. CO-triggered switch from cobalt mediated ROCOP to OMRP for the
one-pot synthesis of polyester-b-polyacrylate.
mic fields,13 those into a metal-oxygen bond are less common.14
Most are related to the Group 10 metals and all concern metals in
oxidation states up to II. In those systems, the metal-alkoxide
bonds are weakened by filled-filled orbital interactions between
the O lone pairs and the metal d orbitals. This process is
unprecedented, to the best of our knowledge, for cobalt and for a
metal complex in an oxidation state as high as III.
Scheme 3. Mechanistic aspects concerning the CO insertion into Co-O bonds
to form alkyloxy cobalt Salen complexes.
Results and Discussion
To verify this assumption, the stoichiometric ring-opening of
propylene oxide (PO) operated by (Salen)CoIIICl (complex 1) was
CO-insertion study. In our primary effort to accomplish a switch
from ROCOP to OMRP, we turned for inspiration to the work of
Oertling and Kadish, which described the formation of
[(Porphyrin)CoIII(CO)]+ via binding of CO with [(Porphyrin)Co]+ at
room temperature.15 The weak metal dπ→ligand π* backbonding
revealed by Raman studies of the resultant [(Porphyrin)CoIII(CO)]+
suggests that CO is largely σ-bonded to the metal and that the
carried out in ampule at room temperature, aiming to obtain the
moisture sensitive (Salen)Co-alkoxo intermediates (2a and 2b)
(Scheme 3B).19 The reaction mixture was then allowed to stir in
the dark under 1 atm CO for 12 h before removing the excess PO
by evaporation under vacuum. The electrospray ionization mass
spectrum (ESI-MS) of the resultant dark green powder displayed
only
one
series
of
signals
corresponding
to
This article is protected by copyright. All rights reserved.

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
[(Salen)CoIIICl+PO+CO]+ (see Figure S1 in the Supporting
Information), suggesting that the (Salen)cobalt-alkoxo complexes
were completely converted to the targeted alkoxycarbonyl cobalt
Salen complexes (3a and 3b). This green powder was also
analyzed by 1H and 13C NMR (see Figures S2 and S3). By
comparison of these spectra with those of pure 3a that was
obtained following the protocol described in literature (see Figures
S4 and S5),11 the 3a/3b ratio could be determined as 77:23,
indicative of the preferential attack on the sterically unhindered
epoxide ring carbon atom during the ring opening reaction. The
direct treatment of 1 with PO and 1 atm of CO also generated the
alkoxycarbonyl cobalt Salen complexes, suggesting that CO does
not hinder the PO ring opening on the cobalt Salen complex. The
3a/3b mixture initiated a well-controlled OMRP of methyl acrylate
(MA) after an induction period of about 1 h under a light intensity
of 10 mW·cm-2. MALDI-TOF-MS spectra of the resultant
poly(methyl acrylate) (PMA) exhibited two series of peaks in
accordance with (Cl + PO + CO + MAn + OH)Na+ and (Cl + PO +
CO + MAn + OH)K+, respetively (Figure S7). This suggests that
the alkoxycarbonyl cobalt Salen complexes undergo Co–acyl
bond homolysis and generate alkoxycarbonyl radicals able to
initiate the polymerization, which is in full agreement with the work
by Fu and Peng.11 Further experiments involving the
epoxides/anhydrides ROCOP followed by the CO addition were
performed to test whether each of the ROCOP active species
could be effectively carbonylated (Scheme 3C). In a first
experiment, the ROCOP of PO and phthalic anhydride (PA)
catalyzed by 1 was conducted in the presence of a large PO
excess ([PO]/[PA]/[1]=1000/100/1), ensuring the generation of the
[(Salen)CoIII-O-PPE] chain end (Scheme 3C1). CO was then
added after a specific polymerization time. As revealed by NMR
spectroscopy, the PA consumption stopped after the CO addition
(see Figures S8 and S9). This suggests that CO quickly
terminates the copolymerization reaction. The final product had a
dark green color, consistent with the formation of a carbonyl
cobalt Salen complex. The 1H NMR and GPC analyses confirmed
that the product is poly(propylene phthalate) (PPE) with complete
alternation of PO and PA monomers and with a number average
molecular weight (Mn) of 4.3 kg/mol and dispersity of 1.10 (see
Figures S10 and S11). The ESI-MS analysis showed two series
of signals in accordance with [(Salen)CoIIICl + PO + CO + (PO +
PA)n]Na+ and [(Salen)CoIIICl + PO + CO + (PO + PA)n]K+,
respectively (see Figure S12), indicative of the full conversion of
the alkyoxy active species [(Salen)CoIII-O-PPE] into (Salen)CoIII-
CO2-PPE. Furthermore, an isotope labeling experiment using
13CO was conducted and the resultant polymeric product was
confirmed as (Salen)CoIII-C13O2-PPE according to the ESI-MS
analysis (Figure S14), which further elucidates the mechanistic
aspects related to the CO insertion process. This material was
able to photoinitiate the OMRP of MA (vide infra). The
comparative ROCOP experiments with excess PA (Scheme 3C2),
aiming at the selective formation of a carboxyl chain end
[(Salen)CoIII-O2C-PPE] after full PO conversion, was executed in
toluene ([PO]/[PA]/[1]/[toluene]=20/100/1/200). As for the
previous experiment, CO was added at the end of the
polymerization, but a white product was obtained in this case (see
Figures S15 and S16). The ESI-MS analysis of the product
exhibited two series of signals, which could be assigned to [Cl +
PO + (PO + PA)n + H]Na+ and [Cl + PO + (PO + PA)n + H]K+,
respectively (see Figure S17). Moreover, this material was not a
suitable photoinitiator for the OMRP of MA. Therefore, we
envision that CO can only insert into the Co-O bond of
(Salen)CoIII-O-PPE and not into that of (Salen)CoIII-O2C-PPE. In
this regard, we propose that (Salen)CoIII-O2C-PPE can
stoichiometrically ring-open PO in the presence of CO to generate
(Salen)CoIII-CO2-PPE, thus enabling the full conversion of the
ROCOP active species into the carboalkoxo cobalt Salen
complexes (Scheme 3C).
DFT Calculations.The remarkable and unprecedented experi-
mental finding of a CO insertion into a CoIII-alkoxide species to
yield a CoIII-carboalkoxy species was further investigated by DFT
calculations in order to confirm its feasibility and to assess which
of the three envisaged mechanisms in Scheme 3A would be
preferred. Since two of the three mechanisms involve a charge
separation (Scheme 3, A1 and A3), the calculations included a
polarizable continuum model in propylene oxide, which was used
in large excess, as solvent (ε = 16.0) and a correction for
dispersion forces was also included.20 The calculations were
carried out with the DFT approach at the full QM level, using
slightly simplified ligands for the sake of computational efficiency.
The tBu substituents on the Salen ligand were replaced with Me
groups to yield the Salen* model and the polymer chain beyond
the cobalt-bonded PO unit in the polymeric –O-PPE alkoxide
ligand was replaced by an H atom to yield the O-CH(Me)CH2OH
alkoxide model. For other computational details, see the SI. The
salient results are summarized in Figure 1. The ionic dissociation
of the alkoxide to yield the [Co(Salen*)]+/-OCHMeCH2OH ion pair
is energetically very costly, even in the dielectric solvent (56.8
kcal/mol). Before solvation correction, the corresponding
ΔG298K,1M is 142.4 kcal/mol. Therefore, this pathway appears
unfeasible. On the other hand, CO addition to the
[Co(Salen*)(CO)(OCHMeCH2OH)] intermediate is relative facile.
Two isomers may be obtained, the most stable of which has a
trans configuration (ΔG298K,PO,1M = -7.6 kcal/mol). This isomer
does not have an appropriate ligand arrangement for the
migratory insertion step of pathway A2, but could lead to the ionic
dissociation of pathway A3. This, however, is once again
energetically too costly (ΔG298K,PO,1M = 47.4 kcal/mol). The
alternative [Co(Salen*)(CO)(OCHMeCH2OH)] isomer with a cis
relative arrangement of the CO and alkoxide ligands is slightly
endoergic but kinetically accessible (1.4 kcal/mol), with a slightly
exoenergetic bond formation (-4.2 kcal/mol) and
a mild
stabilization by solvation (-4.1 kcal/mol). The subsequent CO
insertion is very facile, with a barrier of only 4.8 kcal/mol from the
carbonyl adduct or 3.4 kcal/mol from the separated reagents, and
is very exothermic. Thus, the transformation of the alkoxide into
the carboalkoxyl group by CO addition is predicted to be
irreversible. Calculations were also carried out to assess the
This article is protected by copyright. All rights reserved.

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
56.8
+
39.8
A1
+
A3
-
Ionic
dissociation
-
4.8
1.4
(cis)
-7.6
0.0
[Co(Salen*)]+ -O-CHMeCH2OH
[Co(CO)(Salen*)]+ -O-CHMeCH2OH
(trans)
[Co(Salen*)(O-CHMeCH2OH)]
[Co(Salen*)(CO)(O-CHMeCH2OH)]
-29.0
Ionic
dissociation
[Co(Salen*)(COO-CHMeCH2OH)]
Insertion
A2
+CO
Figure 1. Free energy profile (ΔG298K,1M) in PO (ε = 16.0) for the reaction between [Co(Salen*)(OCHMeCH2OH)] and CO to yield [Co(Salen*)(COOCHMeCH2OH)].
homolytic bond strength of the CoIII-E (E = O or C) bond in the
various complexes involved in this work. As a reference, the bond
suggesting that the polymerization does not require constant
irradiation with light, but at the same time the polymerization rate
scales with the light intensity (Figures S18-S21). It is notable that
the induction periods under light density of 5 mW·cm-2 and 10
mW·cm-2 are around 2 h and 1 h, respectively, indicating that
stronger light density is favorable for shortening the induction
periods of the OMRP mediated by the in-situ generated
alkoxycarbonyl cobalt Salen complexes via CO insertion into Co-
O bonds.
dissociation
CH(CH3)CO2CH3] in the PO medium at 298 K was calculated as
19.
kcal/mol. The CoIII-O bond in [(Salen*)CoIII-
free
energy
(BDFE)
of
[(Salen*)CoIII-
7
OCH(CH3)CH2OH] is homolytically stronger (28.1 kcal/mol),
consistent with the observation that the [(Salen)CoIII-O-PPE] does
not initiate the OMRP of MA. Finally, the CoIII-C bond in
[(Salen*)CoIII-CO2CH (CH3)CH2OH] is even stronger (36.7
kcal/mol). This suggests that [(Salen)CoIII-CO2-PEE] produced by
CO insertion will not spontaneously initiate the radical
polymerization of MA under thermal activation.
CO-triggered switchable polymerization from ROCOP to
OMRP. Encouraged by these results, several one-pot
terpolymerizations of epoxides, anhydrides and acrylates were
Light initiated OMRP of MA mediated by (Salen)CoIII-CO2-PPE. conducted in ampules at room temperature (Table 1). The
The (Salen)CoIII-CO2-PPE product obtained from the procedure
of Scheme 3C2 (Mn = 5.3 kDa, Ð = 1.15) was explored as the
photoinitiator for the OMRP of MA under periodic light on–off
process (Figure 2). Figure 2a reveals that visible light could give
ON/OFF control over the living chain growth of PPE-b-PMA.
During the whole procedure, the polymerization was well-
controlled, as indicated by the kinetic study (Figures 2b and 2c)
and by the excellent match between experimental and observed
molar masses. The OMRP of MA was observed to proceed with
linear first-order kinetics with an induction period and an apparent
polymerization rate of 5.4×10-2·h-1, from which the radical
concentration was estimated as ~1.11×10-9 M. The low
concentration of propagating radicals largely minimized
undesirable side reactions leading to dead polymers and ensured
the living character. This was expected on the basis of the
previous MA polymerization study with control by (Salen)CoIII-
CO2-Me.11 Moreover, the GPC curves of the polymeric products
displayed monomodal and narrow distributions as well as a
continuous molecular weight increase (Figure 2d), in agreement
with the formation of PPE-b-PMA. In order to investigate the effect
of light on the OMRP of MA after the induction period,
comparative experiments have been conducted (i) for 4 h under a
light intensity of 10 mW·cm-2, (ii) for 1 h under a light intensity of
10 mW·cm-2 + 3 h under a light intensity of 5 mW·cm-2, (iii) for 1 h
under a light intensity of 10 mW·cm-2 + 3 h in the dark and (iv)
only for 1 h under a light intensity of 10 mW·cm-2. The conversions
of MA were 28.8%, 17.6%, 14.3% and 5.4%, respectively,
reactions comprised three stages: 1) ROCOP of the
epoxides/anhydrides mixture; 2) chain-end modification by
addition of CO; 3) visible light initiated OMRP of acrylates. The
ampules were kept in the dark (wrapped by aluminum foil) during
steps 1 and 2. The reactions were monitored by 1H NMR analysis
of aliquots taken at regular intervals. At first, PO, PA and MA were
used as the model substrates (Table 1, entries 1-5). Given that
CO and visible light could provide temporal control over the
ROCOP/OMRP switch and the OMRP process, respectively, it is
possible to precisely manipulate the incorporation of monomers
into the resulting polymer chains. Taking entry 1 for example,
ROCOP and OMRP occurred during the first stage and the third
stage, respectively, as revealed by the 1H NMR spectra (Figures
S22-23). According to the NMR characterization of the isolated
products, both PPE with a completely alternating PO/PA structure
and PMA formed in the one-pot procedure. However, no signals
corresponding to the junction units could be detected (see Figures
S24-S26). Both the GPC curves of products of the first stage and
the third stage exhibited monomodal and narrow distribution
(Figure S27). Moreover, the theoretical molecular weights
calculated from the monomer conversions were all in good
agreement with the values obtained from the GPC curves. It is
notable that polyester/polyacrylate ratio measured by NMR
(1:2.26) agrees well with that calculated from GPC analysis
(1:2.29). To unambiguously verify the block structure, the
polymers were analyzed by DOSY NMR spectroscopy (Table 1,
entry 1). The characteristic signals of both PPE and PMA blocks
This article is protected by copyright. All rights reserved.

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
shared the same diffusion coefficient, whereas a blend of the
PPE/PMA constituents, which were expressly synthesized by
PO/PA ROCOP and by MA OMRP with approximately the same
chain length ratio as the diblock copolymer, displayed two
diffusion coefficients (see Figures S28 and S29). Furthermore,
the quantitative insertion of CO into Co-O bonds also occurred
when the CO insertion process was reduced from 12 h to 30 min
(Table 1, entry 5). It is notable that, after the final OMRP stage,
the PPE-b-PMA product has a (Salen)Co chain-end function
linked via a Co-C bond, allowing postpolymerization modification
via treatment with radical scavengers such as thiols or nitroxides.
By direct addition of 2,2,6,6-tetramethylpiperidine-1-oxyl
(TEMPO) into the reaction mixture after the third stage of the
reaction (Table 1, entry 2), we successfully synthesized PPE-b-
PMA macromolecules end-capped with TEMPO (Scheme 4), as
confirmed by 1H NMR spectroscopy (see Figure S30), yielding an
isolated polymer in the form of a white powder.
Figure 2. Visible light initiated OMRP of MA mediated by (Salen)CoIII-CO2-PPE (Mn = 5.3 kDa, Ð = 1.15) at room temperature; a household white LED lamp was
used as the light source (the light intensity was 10 mW·cm-2 at the sample position). (a) Investigation of the “ON/OFF” switching regulated by light. (b) Evolution of
number-average molecular weight and polydispersity (Mw/Mn) with the MA conversion; the straight line represents the theoretical mass weight evolution. (c) Kinetic
plot for the OMRP of MA. (d) Evolution of the GPC traces.
Table 1. One-pot synthesis of block copolymers via CO triggered switchable polymerization from the ROCOP of epoxides/anhydrides to the OMRP of acrylates.a
ROCOP
Conv. (%)b
OMRP
Conv. (%)b
c
d
Polyester
Mn,theo
Mn,GPC
Entry
Monomer
Ðd
/Polyacrylateb
(kg/mol)
(kg/mol)
Time (h)
Time (min)
(anhydride)
(acrylate)
1
2e
3
[PO]/[PA]/[MA]
[PO]/[PA]/[MA]
[PO]/[PA]/[MA]
[PO]/[PA]/[MA]
[PO]/[PA]/[MA]
[PO]/[PA]/[nBA]
[EO]/[PA]/[MA]
[EO]/[CHA]/[MA]
[EO]/[THPA]/[MA]
[PO]/[NBA]/[MA]
[PO]/[NBA]/[nBA]
48
60
72
72
96
48
24
12
25
65
60
31
40
46
46
58
31
73
23
70
35
29
60
10
50
90
50
20
60
60
50
40
25
17
1
1/2.26
1/0.10
1/8.73
1/15.7
1/1
21.0
9.1
20.2
8.2
1.09
1.13
1.25
1.23
1.13
1.11
1.11
1.16
1.16
1.15
1.14
35
63
16
4
39.5
63.5
25.8
11.5
21.7
33.0
29.2
14.6
9.0
36.7
59.6
24.6
9.2
4
5f
6
1/2.20
1/1.26
1/4.28
1/0.75
1/0.60
1/0.30
7
9
19.8
30.4
27.3
12.7
7.7
8
33
18
8
9
10
11
2
aConditions: room temperature; [epoxide]/[anhydride]/[acrylate]/[Complex 1] = 1000/100/1000/1; CO insertion processes run for 12 h; light source for OMRP:
c
household white LED lamp (10 mWcm-2 at the sample position). bDetermined by 1H NMR spectroscopy. Mn,theo = [Mw(epoxide) + Mw(anhydride)] × 100 ×
Conv.(anhydride) + Mw(acrylate) × 1000 × Conv.(acrylate). dDetermined by GPC analysis in THF, calibrated against polystyrene standards. eTerminated with
TEMPO. fThe CO insertion process runs for 30 min.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
Typical procedure for visible light initiated OMRP of MA mediated by
(Salen)CoIII-CO2-PPE. In an argon filled glove box, (Salen)CoIII-CO2-PPE
(Mn = 5.3 kDa, Ð = 1.15), MA were charged into a 25 mL ampoule equipped
with a magnetic stir bar, [(Salen)CoIII-CO2-PPE]/[MA]=1/1000, [M]0=1.0 M
solvent: CH2Cl2. The ampoule was then taken out of the glove box and
allowed to stirred at 25 oC under irradiation of a household white LED lamp
with light intensity of 10 mW·cm-2. A small aliquot of the polymerization
mixture was taken out for 1H NMR spectroscopy and the remained crude
mixture was precipitated in cold n-hexane.
The monomer scope was subsequently expanded using ethylene
oxide (EO), cyclohexene anhydride (CHA), cis-5-norbornene-
endo-2,3-dicarboxylic
anhydride
(NBA),
cis-1,2,3,6-
tetrahydrophthalic anhydride (THPA) and n-butyl acrylate (nBA)
(Table 1, entries 6-10). In all cases, the targeted polyester-b-
polyacrylates were successfully obtained with narrow distributions
(Ð < 1.17) (see Figures S32-S62).
Typical procedure for CO triggered switchable polymerization from
ROCOP to OMRP. In an argon filled glove box, (Salen)CoIIICl complex
(63.9 mg, 0.1 mmol), PA (1.48 g, 10 mmol), MA (9.0 mL, 100 mmol) and
PO (7.0 mL, 100 mmol) were charged into a 25 mL ampoule, which was
wrapped in aluminum foil and equipped with a magnetic stir bar. The
ampoule was then taken out of the glove box and allowed to stirred at 25
oC. After the allotted reaction time, it was degassed and pressurized with
1 atm CO and allowed to stir at 25 oC for 12 h. Next, the aluminum foil was
uncovered and the ampule was degassed and pressurized with 1 atm
argon and allowed to stir at ambient temperature under irradiation of a
household white LED lamp with light intensity of 10 mW·cm-2. A small
aliquot of the polymerization mixture was taken out for 1H NMR
spectroscopy and the remained crude mixture was precipitated in cold n-
hexane.
Scheme 4. Postpolymerization modification of (Salen)Co terminated diblock
PPE-b-PMA via treatment with TEMPO.
Conclusion
Acknowledgements
In summary, the first example of a switch from the ring-opening
copolymerization (ROCOP) of epoxides/anhydrides to a light-
YW, YJZ and XPZ thank the Financial support from National Key
R&D Plan (No.2016YFB0302400), NSFC (No.21604027) of china,
the Fundamental Research Funds for the Central Universities
(HUST:2019JYCXJJ015) as well as the analytical and testing
assistance from the Analysis and Testing Center of HUST. RP
thanks the CNRS (Centre National de la Recherche Scientifique)
for support and the CALMIP mesocenter of the University of
Toulouse for the allocation of computational resources.
initiated
organometallic
mediated
controlled
radical
polymerization (OMRP) of vinyl monomers has been described,
using a (Salen)Co system as a common catalyst/moderating
agent and CO as a molecular trigger. Central to this protocol is
the direct insertion of CO into the Co-O bond of the ROCOP active
species (Salen)CoIII-OR to afford (Salen)CoIII-CO2R, which acts
as successful OMRP photoinitiator for the OMRP of MA.
Mechanistic aspects concerning the quantitative transformation of
the Co-O bond into a Co-CO2 bond have been reasonably
proposed, as confirmed by experimental results and supported by
DFT calculations. This constitutes the first example of CO
insertion into a metal-alkoxide bond for cobalt and for any metal
in the oxidation state III. Regulated by both CO and light, on-
demand monomer incorporation has been demonstrated for the
one-pot terpolymerization of epoxides, anhydrides and acrylates,
allowing the elaboration of polyester-b-polyacrylate diblock
copolymers with precisely programmed sequence, relative block
lengths and chain-end functionality.
Conflict of interest
The authors declare no conflict of interest
Keywords: Ring-opening copolymerization • organometallic
mediated radical polymerization • switch catalysis • block
copolymer • polypropylene carbonate
References
Experimental Section
(1)
Fredrickson, G. H. Science 2012, 336, 434-440.
(2) a) Eisenreich, F.; Kathan, M.; Dallmann, A.; Ihrig, S. P.; Schwaar, T.;
Bates, F. S.; Hillmyer, M. A.; Lodge, T. P.; Bates, C. M.; Delaney, K. T.;
Typical procedure for synthesis of (Salen)CoIII-CO2-PPE [PPE=poly-
(propylene phthalate)]. In an argon filled glove box, (Salen)CoIIICl
complex (63.9 mg, 0.1 mmol), PA (1.48 g, 10 mmol) and PO (7.0 mL, 100
mmol) were charged into a 25 mL ampoule, which was wrapped in
aluminum foil and equipped with a magnetic stir bar. The ampoule was
taken out of the glove box and allowed to stirred at 25 oC for 24 h before it
was degassed and pressurized with 1 atm CO. Then, the ampoule was
allowed to stir at 25 oC for 12 h. A small aliquot of the polymerization
mixture was taken out for 1H NMR spectroscopy and the remained crude
mixture was precipitated in cold n-hexane.
Schmidt, B. M.; Hecht, S. Nat. Catal. 2018, 1, 516-522. b) Chen, C. Nat. Rev.
Chem. 2018, 2, 6-14. c) Teator, A. J.; Lastovickova, D. N.; Bielawski, C. W.
Chem. Rev. 2016, 116, 1969-1992. d) Zhao, Y. J.; Wang, Y.; Zhou, X. P.; Xue,
Z. G.; Wang, X. H.; Xie, X. L.; Poli, R. Angew. Chem. Int. Ed. 2019, 58, 14311-
14318; Angew. Chem. 2019, 131, 14449-14456. e) Zhu, S. S., Zhao, Y. J., Ni,
M. L., Xu, J., Zhou, X. P., Liao, Y. G., Wang, Y., Xie, X. L. ACS Macro Lett. 2020,
9, 204−209. f) Zhu, S. S., Wang, Y., Ding, W. Z., Zhou, X. P., Liao, Y. G., Xie, X.
L. Polym. Chem. 2020, 10.1039/c9py01508f.
(3)
a) Wang, X.; Thevenon, A.; Brosmer, J. L.; Yu, I.; Khan, S. I.;
Mehrkhodavandi, P.; Diaconescu, P. L. J. Am. Chem. Soc. 2014, 136, 11264-
11267. b) Broderick, E. M.; Guo, N.; Vogel, C. S.; Xu, C.; Sutter, J. R.; Miller, J.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
T.; Meyer, K.; Mehrkhodavandi, P.; Diaconescu, P. L. J. Am. Chem. Soc. 2011,
133, 9278-9281. c) Wei, J.; Diaconescu, P. L. Accounts Chem. Res. 2019, 52,
415-424. d) Chen, C. ACS Catal. 2018, 8, 5506-5514.
(4)
a) Qi, M.; Dong, Q.; Wang, D.; Byers, J. A. J. Am. Chem. Soc. 2018, 140,
5686-5690. b) Biernesser, A. B.; Delle Chiaie, K. R.; Curley, J. B.; Byers, J. A.
Angew. Chem. Int. Ed. 2016, 55, 5251-5254; Angew. Chem. 2016, 128, 5337-
5340. c) Anderson, W. C.; Rhinehart, J. L.; Tennyson, A. G.; Long, B. K. J. Am.
Chem. Soc. 2016, 138, 774-777.
(5)
a) Longo, J. M.; Sanford, M. J.; Coates, G. W. Chem. Rev. 2016, 116,
15167-15197. b) Fieser, M. E.; Sanford, M. J.; Mitchell, L. A.; Dunbar, C. R.;
Mandal, M.; Van Zee, N. J.; Urness, D. M.; Cramer, C. J.; Coates, G. W.; Tolman,
W. B. J. Am. Chem. Soc. 2017, 139, 15222-15231.
(6)
a) Van Zee, N. J.; Coates, G. W. Angew. Chem. Int. Ed. 2015, 54, 2665-
2668; Angew. Chem. 2015, 127, 2703-2706. b) Chen, Y.; Wilson, J. A.; Petersen,
S. R.; Luong, D.; Sallam, S.; Mao, J.; Wesdemiotis, C.; Becker, M. L. Angew.
Chem. Int. Ed. 2018, 57, 12759-12764; Angew. Chem. 2018, 130, 12941-12946.
(7)
Am. Chem. Soc. 2016, 138, 7107-7113.
(8) a) Abeles, R. H.; Dolphin, D. Accounts Chem. Res. 1976, 9, 114-120. b)
Halpern, J. Science 1985, 227, 869.
(9) Demarteau, J.; Debuigne, A.; Detrembleur, C. Chem. Rev. 2019, 119,
DiCiccio, A. M.; Longo, J. M.; Rodríguez-Calero, G. G.; Coates, G. W. J.
6906-6955.
(10) a) Debuigne, A.; Poli, R.; Jérôme, C.; Jérôme, R.; Detrembleur, C. Prog.
Polym. Sci. 2009, 34, 211-239. b) Banerjee, S.; Ladmiral, V.; Debuigne, A.;
Detrembleur, C.; Poli, R.; Améduri, B. Angew. Chem. Int. Ed. 2018, 57, 2934-
2937; Angew. Chem. 2018, 130, 2984-2987. c) Debuigne, A.; Caille, J.-R.;
Jérôme, R. Angew. Chem. Int. Ed. 2005, 44, 1101-1104; Angew. Chem. 2005,
117, 1125-1128. d) Poli, R. Angew. Chem. Int. Ed. 2006, 45, 5058-5070; Angew.
Chem. 2006, 118, 5180–5192. e) Poli, R. Polymer Science: A Comprehensive
Reference, Vol. 3, Elsevier BV, Amsterdam 2012, 351-357.
(11) Zhao, Y.; Yu, M.; Zhang, S.; Wu, Z.; Liu, Y.; Peng, C. H.; Fu, X. Chem.
Sci. 2015, 6, 2979-2988.
(12) a) Li, Y.; Zhang, Y. Y.; Hu, L. F.; Zhang, X. H.; Du, B. Y.; Xu, J. T. Prog.
Polym. Sci. 2018, 82, 120-157. b) Wang, Y.; Zhao, Y. J.; Ye, Y. S.; Peng, H. Y.;
Zhou, X. P.; Xie, X. L.; Wang, X. H.; Wang, F. S.; Angew. Chem. Int. Ed. 2018,
57, 3593– 3597; Angew. Chem. 2018, 130, 3655– 3659. c) Darensbourg, D.J.,
Green Chem. 2019, 21, 2214-2223. d) Zhou, H. J.; Yang, G. W.; Zhang, Y. Y.;
Xu, Z. K.; Wu, G. P. ACS Nano 2018, 12, 11471-11480.
(13) a) Peng, J. B.; Wu, F. P.; Wu, X. F. Chem. Rev. 2019, 119, 2090-2127. b)
Drance, M. J.; Sears, J. D.; Mrse, A. M.; Moore, C. E.; Rheingold, A. L.; Neidig,
M. L.; Figueroa, J. S. Science 2019, 363, 1203-1205. c) Wu, X. F.; Fang, X.; Wu,
L.; Jackstell, R.; Neumann, H.; Beller, M. Accounts Chem.Res. 2014, 47, 1041-
1053.
(14) a) Bennett, M. A.; Yoshida, T. J. Am. Chem. Soc. 1978, 100, 1750-1759.
b) Rees, W. M.; Churchill, M. R.; Fettinger, J. C.; Atwood, J. D. Organometallics
1985, 4, 2179-2185. c) Kim, Y. J.; Osakada, K.; Takenaka, A.; Yamamoto, A. J.
Am. Chem. Soc. 1990, 112, 1096-1104. d) Dockter, D. W.; Fanwick, P. E.;
Kubiak, C. P. J. Am. Chem. Soc. 1996, 118, 4846-4852. e) Kapteijn, G. M.;
Dervisi, A.; Verhoef, M. J.; Frederik, M. A.; van den Broek, H.; Grove, D. M.; van
Koten, G. J. Organomet. Chem. 1996, 517, 123-131.
(15) a) Schmidt, E.; Zhang, H.; Chang, C. K.; Babcock, G. T.; Oertling, W. A.
J. Am. Chem. Soc. 1996, 118, 2954-2961. b) Kadish, X. H. M. a. K. M. Inorg.
Chem. 1989, 28, 3743-3747.
(16) Willner, H.; Aubke, F. Angew. Chem. Int. Ed. 1997, 36, 2402-2425;
Angew. Chem. 1997, 109, 2506.
(17) a) Cohen, C. T.; Chu, T.; Coates, G. W. J. Am. Chem. Soc. 2005, 127,
10869-10878. b) Ren, W. M.; Liu, Z. W.; Wen, Y. Q.; Zhang, R.; Lu, X. B. J. Am.
Chem. Soc. 2009, 131, 11509-11518.
(18) a) Macgregor, S. A.; Neave, G. W. Organometallics 2003, 22, 4547-4556.
b) Macgregor, S. A.; Neave, G. W. Organometallics 2004, 23, 891-899. c) Bi, S.;
Zhao, J.; Fan, W. E. I.; Li, P. J. Theor. Comput. Chem. 2012, 11, 1-17.
(19) Robert, C.; Schmid, T. E.; Richard, V.; Haquette, P.; Raman, S. K.; Rager,
M.-N.; Gauvin, R. M.; Morin, Y.; Trivelli, X.; Guérineau, V.; del Rosal, I.; Maron,
L.; Thomas, C. M. J. Am. Chem. Soc. 2017, 139, 6217-6225.
(20) Smyth, C. P. Dielectric Behavior and Structure, Mc. Graw-Hill 1955, 79.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914216
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents (Please choose one layout)
RESEARCH ARTICLE
Yong Wang,*,† Yajun Zhao,† Shuaishuai
Zhu,† Xingping Zhou,*,† Jing Xu,‡ Xiaolin
Xie† and Rinaldo Poli*,§
Page No. – Page No.
Switchable Polymerization Triggered
by Fast and Quantitative Insertion of
Carbon Monoxide into Cobalt-Oxygen
Bonds
This article is protected by copyright. All rights reserved.