Cobalt(II) phenoxy-imine complexes in radical polymerization of vinyl acetate: The interplay of catalytic chain transfer and controlled/living radical polymerization

Article abstract of DOI:10.1002/pola.29460

A series of cobalt(II) phenoxy-imine complexes (Co^II(FI)₂) have been synthesized to mediate the radical polymerization of vinyl acetate (VAc) and methyl acrylate (MA) to evaluate the influence of chelating atoms and configuration to the control of polymerization. The VAc polymerizations showed the properties of controlled/living radical polymerization (C/LRP) with complexes 1a and 3a, but the catalytic chain transfer (CCT) behaviors with complexes 2a, 1b, 2b, and 3b. The control of VAc polymerization mediated by complex 1a could be improved by decreasing the reaction temperature to approach the molecular weights that not only linearly increased with conversions but also matched the theoretical values and relatively narrow molecular weight distributions. The catalytic chain transfer polymerizations (CCTP) mediated by complexes 2a, 1b, 2b, and 3b were characterized by Mayo plots and the polymer chain end double bonds were observed by ¹H NMR spectra. The tendency toward C/LRP or CCTP in VAc polymerization mediated by Co^II(FI)₂ could be determined by the ligand structure. Cobalt complex coordinated by the ligand with more steric hindered and less electron-donating substituents favored the controlled/living radical polymerization. In contrast, the efficiency of CCT process could be enhanced by less steric hindered, more electron-donating ligands. The controlled/living radical polymerization of MA, however, could not be achieved by the mediation of these cobalt(II) phenoxy-imine complexes. Associated with the results of polymerization mediated by other cobalt complexes, this study implied that the configuration and spin state of cobalt complexes were more critical than the chelating atoms to the control behavior of radical polymerization.

Full text of DOI:10.1002/pola.29460

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
Cobalt(II) Phenoxy-imine Complexes in Radical Polymerization of Vinyl
acetate: The Interplay of Catalytic Chain Transfer and Controlled/Living
Radical Polymerization
1
†
1†
1
2
3
3
Yi-Hao Chen, Shih-Ji Chen, Jia-Qi Li, Zhenqiang Wu, Gene-Hsiang Lee, Yi-Hung Liu,
4
4
1
Wei-Ting Cheng, Chen-Yu Yeh, Chi-How Peng
Department of Chemistry and Frontier Research Center on Fundamental and Applied Sciences of Matters, National Tsing Hua
University, Hsinchu 30013, Taiwan
1
2
Beijing National Laboratory for Molecular Sciences, State Key Lab of Rare Earth Materials Chemistry and Applications, College of
Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
3
Instrumentation Center, National Taiwan University, Taipei 10617, Taiwan
4
Department of Chemistry and Research Center for Sustainable Energy and Nanotechnology, National Chung Hsing University,
Taichung 402, Taiwan
Correspondence to: C.-H. Peng (E-mail: chpeng@mx.nthu.edu.tw)
Received 20 June 2019; Revised 29 July 2019; accepted 31 July 2019
DOI: 10.1002/pola.29460
II
ABSTRACT:
A
series of cobalt(II) phenoxy-imine complexes
polymerization mediated by Co (FI)
2
could be determined by the
II
(
Co (FI)
2
) have been synthesized to mediate the radical polymeriza-
ligand structure. Cobalt complex coordinated by the ligand with
more steric hindered and less electron-donating substituents
favored the controlled/living radical polymerization. In contrast, the
efficiency of CCT process could be enhanced by less steric hin-
dered, more electron-donating ligands. The controlled/living radi-
cal polymerization of MA, however, could not be achieved by the
mediation of these cobalt(II) phenoxy-imine complexes. Associated
with the results of polymerization mediated by other cobalt com-
plexes, this study implied that the configuration and spin state of
cobalt complexes were more critical than the chelating atoms to
the control behavior of radical polymerization. © 2019 Wiley Peri-
odicals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2019
tion of vinyl acetate (VAc) and methyl acrylate (MA) to evaluate
the influence of chelating atoms and configuration to the control
of polymerization. The VAc polymerizations showed the properties
of controlled/living radical polymerization (C/LRP) with complexes
1a and 3a, but the catalytic chain transfer (CCT) behaviors with
complexes 2a, 1b, 2b, and 3b. The control of VAc polymerization
mediated by complex 1a could be improved by decreasing the
reaction temperature to approach the molecular weights that not
only linearly increased with conversions but also matched the the-
oretical values and relatively narrow molecular weight distribu-
tions. The catalytic chain transfer polymerizations (CCTP) mediated
by complexes 2a, 1b, 2b, and 3b were characterized by Mayo plots
1
and the polymer chain end double bonds were observed by
H
KEYWORDS: catalytic chain transfer; cobalt; controlled/living radical
NMR spectra. The tendency toward C/LRP or CCTP in VAc
polymerization; Phenoxy-imine; Vinyl acetate
1
3
INTRODUCTION Cobalt complexes have played an important
synthesize the sequential-controlled multiblock copolymers.
role in the radical polymerization of vinyl monomers. Since
Poli et al. expanded this area by developing a new cobalt
II
1975, cobalt porphyrin complex has been applied to accelerate
complex, Co (OPN) [Scheme 1(b)], to mediate the catalytic
2
1
4
the chain transfer reaction in methyl methacrylate polymeriza-
chain transfer polymerization (CCTP) of vinyl acetate (VAc).
1–3
tion, which was reported by Smirnov et al. and Gridnev. The
More details of recent progress in CCTP could be found in the
15
II
cobaloxime [Co (dmgH) , Scheme 1(a)] was then found to be a
2
review article published by Heuts and Smeets.
highly efficient catalytic chain transfer (CCT) agent in methyl
4–10
methacrylate radical polymerization
and was industrialized
In addition to catalytic CCTP, cobalt complexes have also shown
a remarkable capability to achieve the controlled/living radical
polymerization (C/LRP) of both conjugated vinyl monomers and
for the preparation of polymeric products with low-molecular
1
1,12
weight.
Recently, Engelis et al. used the cobaloxime to
Additional supporting information may be found in the online version of this article.
†
These authors contributed equally to this work.
©
2019 Wiley Periodicals, Inc.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
1

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
H
C/LRP or CCTP pathway depending on the ligand structure.
Controlled/living radical polymerization could be achieved by
using cobalt complexes with more steric hindered and less
electron-donating ligands. On the other hand, CCTP would be
observed when the cobalt mediators had less steric hindered and
more electron-donating ligands.
O
N
O
N
O
O
O
O
Co
Co
N
O
N
O
II
(b) Co (OPN)2
H
II
(
a) Co (dmgH)
O
O
O
O
2
EXPERIMENTAL SECTION
Co
Material
II
2,4-Di-tert-butylphenol (Alfa Aesar Ward Hill, Massachusetts,
MA, USA), 2-hydrxoy-5-methoxybenzaldehyde (Lancaster),
(
d) Co (acac)2
3
-tert-butyl-4-hydroxyanisole (Acros), VAc (Merck, 99%) were
N
degassed by freeze-pump-thaw cycle for three times before
use. Tributylamine (Alfa Aesar, 98%), tin(IV) chloride (Aldrich,
N
Co
N
N
N
O
N
O
1.0 M in heptane), paraformaldehyde (Alfa Aesar, 97%), cyclo-
Co
hexylamine (Alfa Aesar, 98 + %), cobalt(II) acetate tetrahydrate
(
4
Acros, 97%), sodium hydroxide (SHIMAKYU’S), diphenyl(2,-
,6-trimethyl-benzoyl)phosphine oxide (TPO, TCI, 98%), 2,2 -azo-
0
bisisobutyronitrile (AIBN, Showa) were used without any purifica-
tion. Toluene was dried by CaH before use.
II
II
(
c) Co (TMP)
(e) Co (salen*)
2
SCHEME 1 Examples of cobalt complexes used to mediate
II
II
Instruments
radical polymerization. (a) Co (dmgH)
2
(cobaloxime); (b) Co
II
All manipulations were carried out under a dry nitrogen atmo-
sphere using an MBraun glove box and Schlenk flask. NMR
spectra were recorded by Mercury 400 MHz and 500 MHz
(
(
(
OPN)
II)
2
(cobalt(II) 9-oxyphenalenone); (c) Co (TMP) (cobalt
II
tetramesitylporphyrin);
(d)
Co (acac)
2
(cobalt
II
0
II) acetylacetonate); (e) Co (salen*) (cobalt(II) [N,N -bis(3,5-di-tert-
1
butylsalicylidene)-1,2-cyclohexanediamine]).
Spectrometer. H NMR chemical shifts are given in ppm versus
residual protons in deuterated solvents as follows: δ 7.24 ppm
1
6–23
13
unconjugated vinyl monomers.
Cobalt(II) porphyrin com-
(CDCl3). C NMR chemical shifts are given in ppm versus
II
13
plex, Co (TMP) [Scheme 1(c)], was first reported to mediate the
residual C in solvents as follows: δ 77 ppm (CDCl ). NMR
3
controlled/living radical polymerization (C/LRP) of methyl
spectroscopy was used to identify the ligand structure and
monomer conversion. The UV–vis spectroscopy was measured
by a SHIMADZU UV-1800 instrument from 1100 to 190 nm.
The cyclic voltammetry (CV) was measured by three electrode
2
4
acrylate (MA) by Wayland et al. Another milestone was attrib-
uted to Jérôme et al. who used cobalt(II) acetylacetonate
II
25
[
Co (acac)
2
, Scheme 1(d)] to control the VAc polymerization.
Afterward, several cobalt complexes were then applied to medi-
potentiostat (CH Instruments, Model 750A) in TBAPF /CH Cl
6 2 2
2
6–29
ate controlled/living radical polymerization.
However, the
(0.1 M) using a Pt wire as the working electrode. The potentials
were measured under scan rate = 0.1 V s− at 298 K with fer-
rocene as an internal standard. Polymers were characterized by
a gel permeation chromatography (GPC) system equipped with
three Shodex columns (Shodex KF-802, Shodex KF-803, and
1
ligand effect has not been systematically studied. Particularly,
II
II
2
the comparison of Co (TMP) and Co (acac) is one of interest
to polymer scientists because they have totally different coordi-
nating atoms, configuration, and spin state. Although the use of
ꢀ
II
Shodex KF-805) using THF eluent at 30 C and at a flow rate of
1
Co (salen*), a cobalt complex with square planar O,N,N,O-
.0 mL min− . The signal was collected by DIONEX Shodex RI-101
1
coordinated environment [Scheme 1(e)], as a mediator in
3
0–33
II
refractometer (RI) detector. The molecular weight was calibrated
with narrow linear poly(styrene) Shodex standard (SM-105)
C/LRP
provided a reference between Co (TMP) and
II
Co (acac) from the aspect of coordinating atoms, the influence
2
2
−1
ranging in molecular weight from 1.20 × 10 g mol –2.61 ×
of configuration, and spin state of the cobalt complexes to the
C/LRP is still ambiguous.
6
−1
1
0 g mol . Molecular weights and the M /M values were calcu-
w n
lated using the DIONEX chromeleon software.
II
Herein, a series of cobalt(II) phenoxy-imine complexes (Co (FI)
2
,
II
Scheme 2), which have same chelating atoms with Co (salen*),
General Procedure for Polymerization of VAc and MA
II
II
but identical configuration and spin state with Co (acac) , were
Co (FI) (0.07 mmol) and AIBN (0.127 g, 0.77 mmol) were
2
2
selected to mediate the radical polymerization of VAc and MA for
the investigation of ligand effect. The comparison of C/LRPs
mediated by Co (TMP), Co (salen*), Co (FI) , and Co (acac)
2 2
revealed that all four kinds of cobalt complexes can control
the radical polymerization of VAc but only the cobalt complexes
with low spin, square planar configuration can control the
MA polymerization. Besides, cobalt(II) phenoxy-imine complexes
were able to mediate the radical polymerization of VAc via
mixed in a 50 mL Schlenk flask and degassed by three
vacuum/nitrogen cycles. Degassed VAc (5 mL, 54.24 mmol)
was then injected into the Schlenk flask. The red mixture was
II
II
II
II
ꢀ
heated to 60 C under stir. The monomer conversion was mea-
1
sured by H NMR. Molecular weight of polymer was measured
by GPC. The polymerization of VAc at difference temperature
and MA polymerization were performed under the same
procedure.
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
SCHEME 2 Structures of cobalt(II) phenoxy-imine complexes investigated in this study.
Photopolymerization of VAc
before evaporation under reduced pressure. The crude product
was purified by column chromatography (silica gel,
hexane/EtOAc = 9/1). Yield of aldehyde 1 (3,5-di-tert-butyl-
Photopolymerizations were carried out in a black box. The
length, width, and height of the box were 26, 26, and 35 cm,
respectively. Four UV lamps (PHILIPS, TL, 8 W, BLB, Poland)
were used as the 365 nm light source. The sample was placed
at an approximate distance of 12 cm to the lamps.
1
2-hdroxybenzaldehyde) was 75.3%, yellow solid. H NMR
(400 MHz, CDCl ): δ (ppm) 11.62 (s, 1H, ─OH), 9.85 (s, 1H,
3
─CHO), 7.57 (d, J = 1.14 Hz, 1H, Ar-H), 7.33 (d, J = 1.24 Hz, 1H,
1
3
Ar-H), 1.42 (s, 9H, ─C(CH ) ), 1.31 (s, 9H, ─C(CH ) ). C NMR
3
3
3 3
(
1
100 MHz, CDCl
31.69, 127.67, 119.86, 35.09, 34.32, 31.41, 29.37. Yield of
3
): δ (ppm) 197.00, 158.86, 141.42, 137.39,
Chain Extension of VAc Polymerization Using Macro-
Initiator Co-PVAc
aldehyde 3 (3-(tert-butyl)-2-hdroxy-5-methoxybenzaldehyde)
was 57.1%, yellow solid. H NMR (400 MHz, CDCl ): δ (ppm)
11.51 (s, 1H, ─OH), 9.79 (s, 1H, -CHO), 7.15 (d, J = 1.54 Hz, 1H,
Ar-H), 6.78 (d, J = 1.54 Hz, 1H, Ar-H), 3.78 (s, 3H, ─OCH ), 1.39
Using a produce similar to general VAc polymerization.
1
1
a
III
3
(
[
FI )
2
Co -PVAc was synthesized under the condition of [1a]
0
/
ꢀ
TPO]
0
/[VAc]
0
= 1/5/500 at 24 C in bulk by light initiated
3
polymerization. After the desired conversion was obtained,
residual VAc was removed under vacuum. Then, fresh VAc was
added into the mixture, and chain extension of VAc polymeriza-
1
3
(
1
s, 9H, ─C(CH
55.80, 151.72, 139.76, 123.58, 119.58, 111.48, 55.56,
34.93, 29.09.
3 3 3
) ). C NMR (100 MHz, CDCl ): δ (ppm) 196.20,
ꢀ
tion was performed at 24 C under UV light.
General Procedure for Preparing PVAc Macro-Monomer
Synthesis of Phenoxy-Imine (1a-3b)
Phenoxy-imine compounds were synthesized following the
modified method based on literature. Aldehyde (4.2 mmol)
and amine (4.2 mmol) was dissolved in methanol (10.0 mL)
and refluxed for 6 h under nitrogen atmosphere. The mixture
was cooled to room temperature and the precipitation of yel-
low solid was observed. The product was filtrated and washed
by cold methanol for three times and then dried under high
vacuum. Yield of phenoxy-imine 1a ((E)-2,4-di-tert-butyl-
II
PVAc macro-monomer was prepared with Co (FI)
2
as media-
II
tor under the condition of [Co ] /[AIBN] /[VAc] = 1/10/700 at
34
0
0
0
ꢀ
6
0 C in bulk. The reaction was stopped by removing residual
VAc under vacuum.
Synthesis of Aldehydes 1 and 3
Aldehydes were synthesized following the modified method
3
4
based on literature. Phenol (0.05 mol) and tributylamine
5 mL, 0.02 mol) were mixed with toluene (15 mL) in a 50 mL
(
6-((cyclohexylimino)methyl)phenol) was 70.3%, yellow solid.
1
Schlenk flask under nitrogen. Tin(IV) chloride (1.0 M in hep-
tane) was added to the solution dropwise under nitrogen atmo-
sphere then the mixture reacted at room temperature for
H NMR (400 MHz, CDCl ): δ (ppm) 8.37 (s, 1H, ─HC N), 7.36
3
(s, 1H, Ar-H), 7.12 (s, 1H, Ar-H,), 3.14–3.24 (m, 1H, ─C NCH),
1.79 (d, J = 4.7 Hz, 4H, Cy-H), 1.50–1.65 (m, 6H, Cy-H), 1.42
1
3
1
5 min. Then, paraformaldehyde (2.078 g, 66.5 mmol) was
3 3 3 3
(s, 9H, ─C(CH ) ), 1.28 (s, 9H, ─C(CH ) ). C NMR (100 MHz,
ꢀ
added into solution and the solution was heated to 100 C for
another 10 h. The reaction was stopped by the addition of
CDCl ): δ (ppm) 163.25, 158.24, 139.76, 136.60, 126.51,
3
125.63, 117.96, 67.63, 35.01, 34.42, 34.09, 31.51, 29.45, 25.59,
1
00 ml water. The desired compound was then extracted by
24.48. Yield of phenoxy-imine 2a ((E)-2-((cyclohexylimino)
methyl)-4-methoxyphenol) was 62.4%, yellow solid. H NMR
1
diethyl ether (30 mL) from the solution. The organic phase was
separated, washed by water (200 mL), and dried over MgSO4
(400 MHz, CDCl ): δ (ppm) 8.30 (s, 1H, ─HC N), 6.85–6.90
3
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
3

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
(
─
─
CDCl
m, 2H, Ar-H), 6.74 (dd, J = 1.2, 0.4 Hz, 1H, Ar-H), 3.75 (s, 3H,
absorption wavelength and molar absorption coefficient were
ε307 nm = 7300 L cm− mol , ε395 nm = 8700 L cm mol
.
1
−1
−1
−1
OCH ), 3.16–3.25 (m, 1H, ─C NCH), 1.79 (d, J = 4.6 Hz, 4H,
3
1
3
Cy─H), 1.25–1.65 (m, 6H, ─Cy─H). C NMR (100 MHz,
): δ (ppm) 161.62, 155.13, 151.60, 118.67, 118.43,
17.47, 114.56, 67.64, 55.93, 55.91, 34.34, 25.58, 24.41. Yield
Yield of complex 3a was 89.5%, orange solid. Observed value in
mass spectrum (FAB, m/z) was 635, matched well with calc.
3
1
exact mass C36
52CoN : N 4.41 (4.43), C 68.01 (68.17), H 8.24 (8.38) %.
The maximum absorption wavelength and molar absorption
2 4
H52CoN O 635.76. Anal. Calc. (found) for
of phenoxy-imine 3a ((E)-2-(tert-butyl)-6-((cyclohexylimino)
methyl)-4-methoxyphenol) was 89.5%, yellow solid. H NMR
36
C H
2 4
O
1
= 8200 L cm− mol , ε400
1
−1
=
nm
(
400 MHz, CDCl ): δ (ppm) 8.31 (s, 1H, ─HC N), 6.94 (d,
coefficient were ε
3
308 nm
−1
10,300 L cm− mol .Yield of complex 1b was 87.3%, red solid.
1
J = 1.5 Hz, 1H, Ar-H), 6.59 (d, J = 1.4 Hz, 1H, Ar-H), 3.75 (s, 3H,
─
OCH ), 3.15–3.24 (m, 1H, ─C NCH), 1.50–1.90 (m, 10H,
3
Observed value in mass spectrum (FAB, m/z) was 735, matched
1
3
─Cy─H), 1.41 (s, 9H, ─C(CH
3 3
)
). C NMR (100 MHz, CDCl
3
): δ
well with calc. Exact mass C44
(found) for C44 56CoN : N 3.81 (3.51), C 71.82 (66.76), H
7.67 (7.12) %. The maximum absorption wavelength and molar
2 4
H56CoN O 735.88. Anal. Calc.
(
ppm) 162.46, 154.80, 150.82, 138.70, 117.95, 117.61, 111.24,
H
2 4
O
67.74, 55.82, 55.81, 35.03, 34.47, 29.36, 25.65, 24.57. Yield
absorption coefficient ε305 nm = 25,400 L cm− mol , ε403
1
−1
of phenoxy-imine 1b ((E)-2,4-di-tert-butyl-6-((phenylimino)
methyl)phenol) was 87.3%, yellow solid. H NMR (400 MHz,
1
−1
−1
= 14,300 L cm mol . Yield of complex 2b was 70.0%, red-
nm
CDCl ): δ (ppm) 8.65 (s, 1H, ─HC N), 7.42 (d, J = 1.2 Hz, 1H,
Ar-H), 7.25–7.35 (m, 3H, Ar-H), 6.89 (d, J = 4.3 Hz, 1H, Ar-H),
purple solid. Observed value in mass spectrum (FAB, m/z) was
572, matched well with calc. exact mass C30 28CoN 571.50.
Anal. Calc. (found) for C30 28CoN : N 4.90 (4.90), C 63.05
(62.99), H 4.94 (4.97) %. The maximum absorption wavelength
3
H
2 6
O
3
.81 (s, 3H, ─OCH
3
), 1.45 (s, 9H, ─C(CH
3
)
3
), 1.30 (s, 9H, ─C
H
2 6
O
1
3
(
CH
3 3
) ). C NMR (100 MHz, CDCl ): δ (ppm) 161.44, 158.29,
3
−
1
−1
1
1
57.82, 141.41, 140.23, 136.63, 127.35, 126.39, 122.06, 118.30,
14.39, 55.48, 35.16, 34.25, 31.59, 29.54. Yield of phenoxy-
and molar absorption coefficient ε310 nm = 27,800 L cm mol ,
ε424 nm = 19,800 L cm− mol . Yield of complex 3b was 73.6%,
deep-red powder. Observed value in mass spectrum (FAB, m/z)
was 683, matched well with calc. exact mass C H CoN O
2 6
1
−1
imine 2b ((E)-2-((cyclohexylimino)methyl)-4-methoxyphenol)
1
was 90.3%, yellow solid. H NMR (400 MHz, CDCl ): δ (ppm)
3
38 44
8
2
─
.56 (s, 1H, ─HC N), 7.24–7.30 (m, 2H, Ar-H), 6.95 (d, J = 0.8 Hz,
2 6
683.71. Anal. Calc. (found) for C38H44CoN O : N 4.10 (4.08), C
66.76 (66.03), H 6.49 (6.46) %. The maximum absorption wave-
H, Ar-H), 6.90–6.95 (m, 2H, Ar-H,), 6.89 (s, 1H, Ar-H), 3.83 (s, 3H,
13
OCH
3
), 3.79 (s, 3H, ─OCH
3
). C NMR (100 MHz, CDCl
3
): δ (ppm)
length and molar absorption coefficient ε303 nm = 23400
−1
−1
−1
−1
159.74, 158.53, 154.87, 151.90, 141.04, 122.09, 119.61, 118.79,
L cm mol , ε424 nm = 12700 L cm mol .
1
17.66, 114.95, 114.37, 55.83, 55.44. Yield of phenoxy-imine 3b
II
The structures of Co (FI) were confirmed by single crystal X-
(
(E)-2-(tert-butyl)-4-methoxy-6-((phenylimino)methyl)phe-
2
1
ray crystallography. The information in detail was described
in Supporting Information (Tables S21 to S48).
nol) was 63.5%, orange solid. H NMR (400 MHz, CDCl
3
): δ
ppm) 8.57 (s, 1H, ─HC N), 7.27 (d, J = 4.4 Hz, 2H, Ar-H), 7.00
d, J = 1.6 Hz, 1H, Ar-H), 6.92 (d, J = 4.4 Hz, 2H, Ar-H), 6.72
s, 1H, Ar-H), 3.82 (s, 3H, -OCH ), 3.79 (s, 3H, ─OCH ), 1.44
s, 9H, ─C(CH ) ). C NMR (100 MHz, CDCl ): δ (ppm) 160.65,
(
(
(
(
3
3
RESULTS AND DISCUSSION
1
3
3
3
3
1
58.41, 154.68, 151.17, 141.13, 138.95, 122.07, 118.68, 118.36,
Synthesis and Characterization of Cobalt(II) Phenoxy-
Imine Complexes
114.40, 111.61, 55.75, 55.47, 35.08, 29.33.
Cobalt(II) phenoxy-imine complexes were obtained from the
reactions of cobalt(II) acetate and phenoxy-imine ligands which
were converted from phenoxy-aldehydes that synthesized by
Synthesis of Cobalt(II) Phenoxy-Imine Complexes
34,35
Cobalt(II) phenoxy-imine complexes were synthesized following
phenol derivatives and formaldehyde (Scheme 3).
The
35
the modified method based on the literature. Phenoxy-imine
1.9 mmol), NaOH (0.9 mmol), and methanol (10 mL), which
were mixed in a 50 mL Schlenk flask. Then cobalt(II) acetate
1.1 mmol) in MeOH solution was added dropwise to the mix-
cobalt(II) complexes were mainly orange-yellowish powder with
two major UV–vis absorption bands at the regions of
300–310 nm and 375–424 nm (Supporting Information Fig. S1)
possibly corresponding to n–π* transitions of the imine groups
and ligand-to-metal charge-transfer (LMCT) transition from sali-
(
(
ꢀ
ꢀ
ture at 40 C. The solution was subsequently heated to 60 C
35–37
and reacted for 6 h. The precipitate was filtered and washed by
cylic oxygen to cobalt,
respectively. The redox potentials
methanol (10 mL) for three times, and then dried under high
(E1/2) of cobalt(II) phenoxy-imine complexes have also been
measured. The values were found in the range of 0.38–0.25 V
II
35
vacuum to obtain Co (FI)
2
complexes. Yield of complex 1a was
9
2.8%, yellow solid. Observed value in mass spectrum (FAB, m/
z) was 687, matched well with calc. exact mass C H CoN O
2 2
with the order of 1b > 1a > 3b > 2b > 3a > 2a (Supporting
Information Table S2). The structures of cobalt(II) phenoxy-
4
2
64
6
87.92. Anal. Calc. (found) for C H CoN O : N 4.07 (4.09), C
imine complexes were characterized by X-ray crystallography.
4
2
64
2 2
II
73.33 (73.08), H 9.38 (9.42) %. The maximum absorption
The crystals of Co (FI) were obtained from bilayer diffusion of
2
wavelength and molar absorption coefficient were ε300 nm
1
=
dichloromethane solution with cobalt(II) phenoxy-imine com-
plexes and methanol. According to the Oak Ridge Thermal Ellip-
soid Plot (ORTEP) plots (Supporting Information Fig. S3), except
complex 3b, other five complexes have a cobalt center coordi-
nated by two phenoxy-imine ligands with a tetrahedral configu-
ration. And the structure of complex 1b is consistent with
1,000 L cm− mol , ε375 nm = 10,100 L cm mol . Yield of
1
−1
−1
−1
complex 2a was 62.4%, red solid. Observed value in mass spec-
trum (FAB, m/z) was 524, matched well with calc. exact mass
C H CoN O 523.54. Anal. Calc. (found) for C H CoN O : N
28
36
2
4
28 36
2 4
5.35 (5.35), C 64.24 (64.30), H 6.93 (6.99) %. The maximum
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
SCHEME 3 Synthetic process with the yield of cobalt phenoxy-imine complexes.
3
8
published result. The bond lengths of Co─N and Co─O are
both quite similar in all five complexes as 1.99–2.00 Å and
1
Fig. S10b), indicating a spin-only tetrahedral cobalt(II) com-
39,40
plex.
The high-spin electron configuration of complex 1a
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
.89–1.91 Å, respectively. Although complex 3b showed an octa-
was demonstrated by S = 3/2 calculated from μeff = g SðS + 1Þ
associated with g = 2.0010 and μeff = 4.21.
hedral configuration with the coordination of three phenoxy-
imine ligands in X-ray crystallography, the mass spectrum
showed the largest m/z signal at 683 (molecular weight of one
cobalt and two ligands, Supporting Information Fig. S9), indicat-
ing that the complex 3b should be more favorable to two ligands
coordinated configuration during radical polymerization, similar
with other five complexes. The summary of crystallographic data
is shown in the Supporting Information Table S3, and the com-
plete results of X-ray crystallography and mass spectra of
Polymerization of VAc Mediated by Cobalt(II) Phenoxy-
Imine Complexes: Features of Controlled/Living Radical
Polymerization
The polymerization of VAc was mediated by cobalt(II)
II
phenoxy-imine complexes under the condition of [Co ] /
0
ꢀ
[AIBN]
0
/[VAc]
0
= 1/10/700 in bulk at 60 C and the results
cobalt(II) phenoxy-imine complexes are described in the
are shown in Table 1. Induction periods in the range of
15–60 min followed by the linearly increased first-order
kinetic plots were observed [Fig. 1(a)]. During the induction
period, cobalt(II) species were converted to organo-cobalt(III)
II
Supporting Information. The spin state of Co (FI)
2
was recog-
nized by the measurements of electron paramagnetic resonance
EPR) and superconducting quantum interference device
SQUID) using complex 1a as the representative complex. The
(
(
4
1,42
complexes through the reaction with radicals.
The first-
EPR spectrum of complex 1a was recorded under the condition
of 77 K in toluene to give a g value as 2.0010 (Supporting Infor-
mation Fig. S10a). The eight peaks observed in the spectrum
matched the hyperfine coupling of cobalt(II) complex. The effec-
order kinetic plots implied that the radical concentration was
4
3,44
constant,
controlled/living radical polymerization. However, the plots of
n
molecular weight and polydispersity index (PDI = M /M )
which is the feature of cobalt-mediated
w
tive magnetic moment (μ ) of complex 1a in the solid state was
measured by a SQUID magnetometer as a relatively constant
value of 4.21 μB from 50 to 300 K (Supporting Information
versus monomer conversion [Fig. 1(b)] showed that only poly-
merizations mediated by complexes 1a and 3a demonstrated
the molecular weight increased properly with conversion,
eff
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
5

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
II
ꢀ
II
TABLE 1 Polymerization of VAc Mediated by Co (FI)
AIBN] = 0.155 M
2
in Bulk at 60 C with the Condition of [Co ]
0
/[AIBN]
0
/[VAc]
0
= 1/10/700,
PDI
[
0
a
b
3
c
3
Mediator
Time (min)
25
Conv. (%)
M
n,exp (10 g/mol)
M
n,th (10 g/mol)
1
2
3
1
2
3
a
a
a
b
b
b
3.1
8.3
92.8
115.0
135.0
54.3
54.3
53.7
84.9
85.5
88.0
61.4
67.5
71.1
17.0
18.1
18.6
37.5
38.1
33.4
1.8
1.25
1.24
1.27
1.55
1.55
1.57
1.70
1.72
1.70
1.54
1.52
1.47
1.86
1.84
1.83
1.74
1.75
1.70
3
3
0
5
4.8
10.8
15.8
18.0
225
18.4
26.3
29.8
37.4
24.7
28.0
32.4
38.5
42.6
44.4
14.5
21.3
23.7
20.9
27.8
30.2
105
1
1
20
50
80
00
20
14.9
16.9
19.6
23.5
25.9
26.5
8.7
1
1
240
3
3
00
60
360
4
4
20
80
12.8
14.3
12.7
16.9
18.1
360
4
4
20
80
a
b
1
c
Conversion was detected by H NMR.
was determined by GPC with polystyrene as standard.
Mn,th = Mw(mediator) + Mw(VAc) × 700 × Conv.
M
n
(a)
(b)
1
1
2
2
3a
3
a
b
a
b
2.5
2.0
1.5
1
0
0
0
0
.2
.9
.6
.3
.0
1
.0
1
2.0
b
Mn, th
1
a
9.0
1b
2
a
6
3
0
.0
.0
.0
2b
3
a
3b
0
10000
20000
30000
0
0.2 0.4 0.6 0.8
Conversion
Time (sec)
(c)
2
0 min
5 min
0 min
5 min
(d)
4
5
6
7
8
9
1
0 min
0 min
2
3
3
0 min
0 min
0 min
0 min
00 min
1
8
20
22
24
18
20
22
24
26
Elution Time (min)
Elution time (mine)
II
ꢀ
II
FIGURE 1 VAc polymerization mediated by Co (FI)
2
at 60 C in bulk under the condition of [Co ]
0
/[AIBN]
0
/[VAc]
0
= 1/10/700 with
−
4
−5
[
0
AIBN] = 0.155 M. (a) First-order kinetic plots. The slopes in the kinetic plots are 2.06 × 10 (■ complex 1a), 2.07 × 10 (□ complex
b), 5.70 × 10− (● complex 2a), 4.70 × 10 (○ complex 2b), 9.65 × 10 (▲ complex 3a), and 1.33 × 10−5 (4 complex 3b). (b) The
5
−5
−5
1
n
plots of M and PDI versus conversion. The GPC traces of VAc polymerization mediated by (c) complex 1a and (d) complex 3a. [Color
figure can be viewed at wileyonlinelibrary.com]
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
(a)
(b)
0
0
0
0
0
0
.5
.4
.3
.2
.1
.0
2
1
1
.0
.5
.0
2
5
1
0
2
1
4.0
6.0
M
2
n, th
8
0
.0
.0
5
10
0
4000 8000 12000
Time (sec)
0
0.1
0.2
0.3
0.4
Conversion
ꢀ
FIGURE 2 VAc polymerization mediated by complex 1a at 60 C under the condition of [1a]
0
/[AIBN]
0
/[VAc]
0
= 1/X/700. (▲) X = 2, (■)
X = 5, and (●) X = 10. (a) First-order kinetic plots. The slopes in kinetic plot are (▲) 4.67 × 10− , (■) 1.66 × 10 , and (●) 2.06 × 10
5
−4
−4
.
(
b) The plots of M and PDI versus conversion. [Color figure can be viewed at wileyonlinelibrary.com]
n
while the molecular weights of polymers in other polymeriza-
tions stayed constant. Comparing the shift of GPC traces of
VAc polymerization mediated by complexes 1a and 3a [Fig. 1
from the lower radical concentration due to the slower dissocia-
tion of AIBN at declined temperature [Fig. 3(a)]. More impor-
tantly, the molecular weight deviation decreased significantly
with declined temperature [Fig. 3(b)], which encouraged us to
perform the polymerization at even lower temperature. How-
ever, due to the low dissociation rate of AIBN at the desired
temperature, photo-initiator of TPO was selected for following
polymerization. VAc polymerization was then initiated by
(
c,d)], complex 1a was proposed to be the best mediator for
C/LRP of VAc among all investigated cobalt complexes. How-
ever, the significant molecular weight deviation from theoreti-
cal value motivated us to further optimize the reaction
condition and study the mechanism.
3
65 nm UV irradiation from four light bulbs (8 W for each)
The effect of AIBN concentration in VAc polymerization medi-
ated by complex 1a was first evaluated. Decreasing the equiva-
lent of AIBN, which lowered the radical concentration, caused
only the longer induction period and slower polymerization
rate observed by the first-order kinetic plots [Fig. 2(a)] but did
not lower the PDI value nor improved the molecular weight
deviation [Fig. 2(b)]. However, the slower polymerization
allowed us to observe two periods of propagation in the plots
under the condition of [1a]
0
/[TPO]
0
/[VAc]
0
= 1/0.8/500 at
ꢀ
37, 24, and 0 C in bulk. As expected, the temperature effect to
the length of induction period was similar with the polymeriza-
tion initiated by AIBN [Fig. 4(a)] and the molecular weight devi-
ation kept decreasing with lowered temperature [Fig. 4(b)].
However, because the molecular weight deviations observed in
VAc polymerization at 24 and 0 C were similar, the polymeriza-
tion at 24 C was chosen as the optimized condition due to the
ꢀ
ꢀ
n
of M and PDI versus conversion. With two equivalent of AIBN,
the first period of propagation showed a faster chain growth
and decline of PDI before 10% monomer conversion. Then the
consideration of polymerization rate. With the adjustment of
TPO equivalent, the VAc polymerization mediated by complex
1a at 24 C matched the features of C/LRP when the ratio of
ꢀ
values of M and PDI obtain during the second period became
[TPO] /[1a] equaled to 2 (Fig. 5). The monomer conversion
n
0
0
similar to those in polymerization with 5 and 10 equivalent of
AIBN. These two periods of propagation should be attributed
to the increase of effective amount of cobalt mediators because
achieved 63% in 77 min with linearly increased molecular
weights that matched well with the theoretical values, relatively
narrow PDIs between 1.44 and 1.75, and smoothly shifted GPC
traces (Supporting Information Fig. S15b), which demonstrated
that complex 1a was a good mediator for C/LRP of VAc at room
temperature. The results of VAc polymerization mediated by 1a
using different initiators and under different temperature were
summarized in table 2.
p
the k ×[radical] was relatively constant before and after 10%
conversion as demonstrated by the first order kinetic plots.
Since the change of AIBN did not significantly affect the molec-
ular weight and polydispersity of the poly(vinyl acetate),
1
a
III
indicating that the concentration of (FI )
2
Co -PVAc is inde-
pendent to the equivalent of radical and thus the polymeriza-
tion was not controlled by Co /Co equilibrium. Therefore,
II
III
High chain-end fidelity of polymer is another significant feature
of C/LRP. A successful chain extension of C/LRP mediated by
cobalt complex indicated that the chain-end functional group of
complex 1a was proposed to mediate the VAc polymerization
2
9
via the mechanism of degenerative transfer (DT).
1
a
III
polymer was cobalt complex. Here, macro-initiator (FI ) Co -
2
3
In addition to the concentration of AIBN, VAc polymerization
n
PVAc (M = 5.8 × 10 g/mol, PDI = 1.43) was obtained from
mediated by complex 1a was performed under 60, 50, and
C/LRP mediated by complex 1a under the condition of [1a]
0
/
ꢀ
ꢀ
40 C, respectively, to evaluate the temperature effect. The lon-
[TPO]
0
/[VAc]
0
= 1/5/500 at 24 C in bulk. Then chain exten-
1
a
III
ger induction period and slower polymerization rate resulted
2
sion was performed by the mediation of (FI ) Co -PVAc with
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
7

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
(a)
(b) 2.0
0.5
0.4
0.3
0.2
0.1
0
1
.5
6
5
4
0 °C
0 °C
0 °C
1.0
0.0
M
n, th
2
1
1
60 °C
50 °C
4
0 °C
6.0
2.0
8.0
4
.0
.0
0
10000 20000 30000
Time (sec)
0
0
0.1
0.2
0.3
0.4
Conversion
ꢀ
0 0 0
/[VAc] = 1/2/700 at 60 C, (■) [1a] /
FIGURE 3 VAc polymerization mediated by complex 1a under the condition of (▲) [1a]
0
/[AIBN]
ꢀ
0 0 0 0 0
AIBN] /[VAc] = 1/12/700 at 50 C, and (●) [1a] /[AIBN] /[VAc] = 1/70/700 at 40 C. (a) First-order kinetic plots. The slopes in kinetic
ꢀ
[
−5
−5
−5
n
plots are (▲) 4.67 × 10 , (■) 3.72 × 10 , and (●) 2.77 × 10 . (b) The plots of M and PDI versus conversion. [Color figure can be
viewed at wileyonlinelibrary.com]
(
a)
(b)
0
0
0
0
.8
.6
.4
.2
0
3
2
0
7 °C
4 °C
°C
2.0
1.0
M
n, th
3
2
1
0
.0
.0
.0
.0
37 °C
2
0
4 °C
°C
0
10000 20000 30000
Time (sec)
0
0.1 0.2 0.3 0.4 0.5
Conversion
FIGURE 4 VAc polymerization mediated by complex 1a with TPO as initiator under 365 nm irradiation with 4 bulbs of light (8 W for
ꢀ ꢀ ꢀ
each) under the condition of [1a] /[TPO] /[VAc] = 1/0.8/500 at (▲) 37 C, (■) 24 C, and (●) 0 C. (a) First-order kinetic plots. The
0 0 0
slopes in kinetic plots are (▲) 7.65 × 10− , (■) 6.92 × 10 , and (●) 4.38 × 10 . (b) The plots of M
5
−5
−5
and PDI versus conversion. [Color
n
figure can be viewed at wileyonlinelibrary.com]
(
b) 2.0
(
a)
1
0
0
0
0
.2
.9
.6
.3
.0
0
0
2
.6
1.5
.8
1
.0
.0
M
n, th
4
2
0.8
3.0
0.6
2
.0
.0
1
0.0
0
5000
10000 15000
0
0.2
0.4
0.6
Time (sec)
Conversion
ꢀ
FIGURE 5 VAc polymerization mediated by complex 1a with TPO as initiator at 24 C under the condition [1a]
0
/[TPO]
0
/[VAc]
0
4
= 1/
−
X/700. (▲) X = 2, (■) X = 0.8, and (●) X = 0.6. (a) First-order kinetic plots. The slopes in kinetic plots are (▲) 3.03 × 10 , (■)
−
5
−5
6.92 × 10
,
and (●) 5.02 × 10
.
(b) The plots of
M
n
and PDI versus conversion. [Color figure can be viewed at
wileyonlinelibrary.com]
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
TABLE 2 Data of VAc Polymerization Mediated by Complex 1a in Bulk Under Different Conditions
ꢀ
a
b
3
c
3
b
[
1a]
/2/700
/5/700
0
/[I]
0
/[VAc]
0
Temp. ( C)
Initiator (I)
Time (min)
Conv. (%)
M
n,exp (10 g/mol)
M
n,th (10 g/mol)
PDI
1
1
1
1
1
1
1
1
1
1
60
60
60
50
40
37
24
0
AIBN
AIBN
AIBN
AIBN
AIBN
200
60
34
35
18
36
35
39
24
47
63
25
189.6
244.3
135.0
84.3
53.7
25.2
10.8
20.5
29.7
11.7
20.5
21.1
10.8
21.9
21.1
16.9
10.1
20.2
27.1
10.8
1.37
1.33
1.27
1.65
1.69
1.53
1.51
2.06
1.64
1.55
/10/700
/12/700
/70/700
35
330
420
140
140
450
77
d
/0.8/500
/0.8/500
/0.8/500
/2/500
TPO
d
TPO
d
TPO
d
24
24
TPO
d
/0.6/500
TPO
210
a
b
c
1
d
Conversion was detected by H NMR.
With TPO as an initiator under 365 nm irradiation with four bulbs of
light (8 W for each).
n
M was determined by GPC with polystyrene as standard.
n,th = Mw(mediator) + Mw(VAc) × ([VAc] /[Co ] ) × conv.
0 0
II
M
III
the ratio of [Co ] /[VAc] = 1/700, generating polymer with
transfer constant that is corresponding to the efficiency of
0
0
3
4
molecular weight increased from 5.8 × 10 to 8.1 × 10 g/mol
Fig. 6). Symmetrically monomodal GPC trace was smoothly
shifted to higher molecular weight, which demonstrated the
chain-end functional group of polymer was cobalt phenoxy-
imine complex (Table 2).
chain transfer agent.
ꢀ
ꢁ
(
1
1
½Coꢁ
½Mꢁ
=
+ CS
ð1Þ
DPn DP0
The values of Cs for complexes 2a, 1b, 2b, and 3b were
obtained from the slope of Mayo plots with the VAc polymeri-
II
zation performed under the condition of [Co ] /[AIBN] /
0
0
ꢀ
[
VAc] = x/100/7000 at 60 C in bulk (Fig. 7 and Table 3).
0
Polymerization of VAc Mediated by Cobalt(II) Phenoxy-
Imine Complexes: Characterization of Catalytic Chain
Transfer Polymerization
The polymerization was stopped at monomer conversion
below 15% to avoid the interferences from increased viscosity
or other side reactions. Given by the slope in Mayo plots, com-
The molecular weight of poly(vinyl acetate) stayed at a rela-
tively low and constant value during the polymerization medi-
ated by complexes 2a, 1b, 2b, and 3b, which matched the
plex 2b showed the highest C as 5.41, followed by complexes
s
3
b, 1b, and then 2a (Table 3). This order matched well with
1
2
the trend that complex with more electron donating and less
steric hindered environment would have higher activity
in CCTP.
feature of CCTP. CCTP is generally used to prepare the poly-
mers with low-molecular weight and chain end double bond.
The correlation between the degree of polymerization (DPn)
and concentration of chain transfer agent ([Co]) was
Besides the suppression of molecular weight, generation of
polymer chains with chain end double bond that could be used
as “macro-monomers” for the synthesis of macromolecules with
4
5,46
described by Mayo equation (eq 1),
in which the DP is
0
the degree of polymerization without chain transfer agent,
[
M] is the concentration of monomer, and C
s
is the chain
12
varied structures is another important feature of CCTP. Thus,
the poly(vinyl acetate) obtained from CCTP mediated by com-
plex 2b, which showed the highest C , was used as monomer
s
for free-radical polymerization initiated by AIBN with the con-
ꢀ
dition of [AIBN] = 0.039 M at 60 C in benzene [Fig. 8(a)]. The
0
GPC traces demonstrated that the PVAc can be further poly-
merized to form the product with multimodal GPC trace. The
chain end functionality was preliminarily estimated by the area
of GPC peaks as 65% and further recognized by the integration
1
1
of H NMR spectrum as 83% [Fig. 8(b)]. The H NMR signals of
terminal double bond on PVAc were confirmed by chemical
shifts at 4.56, 5.54, and 6.08 ppm by using the reaction of dou-
1
4
ble bond and thiophenol reported by Poli et al.
To recognize the performance of complex 2b in CCTP of VAc,
1
a
III
FIGURE 6 The GPC traces of (FI
)
2
Co -PVAc and the chain
the Mayo plot with complex 2b was compared with
1
a
III
II
extended PVAc synthesized via the mediation of (FI
)
2
Co -
Co (dmgBF
)
(H
2
O)
(Fig. 9) which was patented as a highly
2
2
2
ꢀ
III
PVAc at 24 C in bulk, [Co ]
0
/[VAc]
0
= 1/700. [Color figure can be
efficient CCT agent for the polymerization of methyl methacry-
viewed at wileyonlinelibrary.com]
late and also reported to be able to mediate the CCTP of
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
9

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
(
a) 2a - Conv. = 13%
10
20
(b) 1b - Conv. = 7%
5
4
3
2
1
0
6
4
2
0
5
1
3
5
5
0
10
15
20
y = 0.74 x + 0.61
y = 1.66 x + 0.52
0.0
1.5
3.0
4.5
0.0
1.0
2.0
3.0
II
-3
II
-3
[
Co ]/[VAc] (×10 )
[Co ]/[VAc] (×10 )
(
c) 2b - Conv. = 5%
(d) 3b - Conv. = 8%
12
5
1
1
2
1
6
5
0
5
0
1
1
2
0
1
2
8
4
0
8
4
0
5
0
y = 5.41 x - 0.46
y = 2.56 x + 1.44
0.0
1.0
2.0
3.0
0
.0
1.0
2.0
3.0
II
-3
II
-3
[
Co ]/[VAc] (×10 )
[Co ]/[VAc] (×10 )
II
FIGURE 7 Mayo plots of VAc polymerization mediated by complexes (a) 2a, (b) 1b, (c) 2b, (d) 3b under the condition of [Co ]
0
/[AIBN]
0
/
ꢀ
[
VAc]
0
= x/100/7000 at 60 C in bulk. The monomer conversion was (a) 13%, (b) 7%, (c) 5%, (d) 8%. [Color figure can be viewed at
wileyonlinelibrary.com]
4
8
to solvent is also significant in VAc polymerization and thus
the addition of solvent was expected to further suppress the
molecular weight of polymeric product. Toluene was selected
as the solvent due to a relatively high chain transfer constant
in the literature. Given by the Mayo plots (Fig. 10), the addi-
tion of toluene indeed further lowered the molecular weight
of poly(vinyl acetate), which could be observed from either
TABLE
3 Chain Transfer Constant for VAc Polymerization
Mediated by Different Complexes
a
s
Complex
C
4
9
2a
1b
2b
2b
3b
0.74(07)
1.66(07)
5.41(47)
4.70(50)
2.55(16)
2.23(15)
directly the value of molecular weight (from 22,000 to
b
II
1
1,800 g/mol when [Co ]
0
/[VAc]
0
= 10/14000 = 0.71) or the
intercept of Mayo plots (1/DP0) that changed from
Co(dmgBF
2
)
2
(H
2
O)
2
−3
−3
−
0.46 × 10 to 4.38 × 10 . However, the slopes of Mayo
a
plots with and without toluene were similar (5.41 and 4.70,
respectively, Table 3), indicating the efficiency of chain trans-
fer for complex 2b was almost independent to solvent.
In bulk.
In toluene.
b
1
5,46,47
II
VAc.
same condition was 2.24, which was lower than the value of
.41 for complex 2b, suggesting that complex 2b was a better
The value of C for Co (dmgBF ) (H O) under the
s 2 2 2 2
Polymerization of MA Mediated by Cobalt(II) Phenoxy-
Imine Complexes
5
II
CCT agent than Co (dmgBF
2
)
2
(H
2
O)
2
in VAc polymerization.
Cobalt(II) phenoxy-imine complexes were then investigated for
II
The solvent effect has been evaluated because chain transfer
C/LRP of MA under the condition of [Co ] /[AIBN] /[MA]
=
0
0
0
FIGURE 8 (a) The GPC traces of PVAc macro-monomer and the polymeric product of free-radical polymerization of PVAc macro-
ꢀ
monomer at 60 C in benzene, [AIBN]
0
= 0.039 M. The macro-monomer was prepared from VAc polymerization mediated by complex
II
ꢀ
1
2
b under the condition of [Co ]
= 6700 g/mol prepared under the condition of [2b]
0
/[AIBN]
0
/[VAc]
0
= 1/10/700 at 60 C in bulk. (b) H NMR spectrum of PVAc macro-monomer with
ꢀ
M
n
0
/[AIBN]
0
/[VAc]
0
= 5.5/10/700 at 60 C in bulk before (lower) and after (upper)
quenched by thiophenol. [Color figure can be viewed at wileyonlinelibrary.com]
1
0
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
II
Co (dmgBF ) (H O) - Conv. = 11%
TABLE
4
Summary of Cobalt Complexes with Varied
2
2
2
2
4
3
2
1
0
.0
Configuration and Chelating Atoms, as Well as the C/LRP They
can Mediate
3
6
1
1
.0
.0
.0
.0
0
5
Complex
Configuration
Chelating atoms
Monomer
II
Co (TMP)
Square planar
Square planar
Tetrahedral
N,N,N,N
N,N,O,O
N,N,O,O
O,O,O,O
VAc, MA
VAc, MA
VAc
y = 2.24 x + 0.29
II
Co (salen*)
II
0
.0
0.5
1.0
1.5
-3
2.0
Co (FI)
2
II
II
[
Co ]/[VAc] (×10 )
Co (acac)
Tetrahedral
VAc
2
FIGURE
Co(dmgBF
VAc]
9 Mayo plot of VAc polymerization mediated by
II
2
)
2
(H
2
O)
2
under the condition of [Co ]
0
/[AIBN]
0
/
and polydispersities between 1.24 and 1.58, the cobalt(II)
phenoxy-imine complexes actually could not mediate the
controlled/living radical polymerization of MA.
ꢀ
[
0
= x/100/140,00 at 60 C in bulk. The monomer conversion
was 11%. [Color figure can be viewed at wileyonlinelibrary.com]
The results of VAc and MA polymerization mediated by
cobalt(II) phenoxy-imine complexes and other cobalt(II) com-
plexes are summarized in Table 4 to demonstrate the influence
of coordinating atoms and configuration. All four kinds of
cobalt complexes can control the polymerization of VAc but
2
b in toluene - Conv. = 11%
1
6
2
8
4
0
1
1
2
3
0
5
0
0
1
II
II
only Co (TMP) and Co (salen*), which have a square planar
II
y = 4.70 x + 4.38
configuration, can mediate the C/LRP of MA. Even the Co (FI)
2
II
has the same chelating atoms with Co (salen*), still could not
control the MA polymerization, suggesting that configuration
and thus spin state were more critical than chelating atoms for
the controlled/living radical polymerization of MA and perhaps
a broader scope of vinyl monomers.
0
.0 0.5 1.0 1.5 2.0 2.5
II
-3
[
Co ]/[VAc] (×10 )
FIGURE 10 Mayo plot of VAc polymerization mediated by complex
b under the condition of [Co ]
0 C in toluene with [VAc]
II
2
6
0
/[AIBN]
0
/[VAc]
0
= x/200/14000 at
ꢀ
0
= 3.60 M. The monomer conversion
was 11%. [Color figure can be viewed at wileyonlinelibrary.com]
Mechanism
Cobalt(II) phenoxy-imine complexes have been demonstrated
to be capable of mediating controlled/living radical polymeriza-
tion and CCTP of VAc depending on the ligand structure, imply-
ing these two polymerizations are correlated and could be
competing reactions. The mechanism of cobalt-mediated radical
polymerization includes C/LRP and CCTP is thus proposed as
Scheme 4. In most cases, both C/LRP and CCTP mediated by
cobalt complexes start from the reaction of cobalt(II) com-
ꢀ
1
/1/500 in benzene 60 C with [MA]
0
= 3.6 M (Supporting
Information Table S20). All polymerizations showed an induc-
tion period followed by the linearly increased first-order kinetic
plots [Fig. 11(a)]. However, the molecular weight of poly(methyl
acrylate) approached 2.5–3.7 × 10 g/mol [Fig. 11(b)] at the
beginning of polymerization and then stayed constant. Associ-
ated with unshifted GPC traces (Supporting Information Fig. S17)
5
7
,33,50
plexes and radicals to form the organo-cobalt(III) species,
(a)
(
b) 2.0
0.4
0.3
0.2
0.1
0.0
1
2
3
1
2
3
a
a
a
b
b
b
1.5
1.0
40.0
30.0
20.0
10.0
0.0
1
1
a
b
2a
2b
3a
3b
0
2000
4000
6000
0
0.1
0.2
0.3
Time (sec)
Conversion
II
ꢀ
II
FIGURE 11 MA polymerization mediated by Co (FI)
2
at 60 C in benzene under the condition of [Co ]
0
/[AIBN]
0
/[MA]
0
= 1/1/500,
−
4
−5
−4
[
1
0
MA] = 3.6 M. (a) First-order kinetic plots. The slope in kinetic plot are 1.00 × 10 (■ 1a), 2.96 × 10 (□ 1b), 2.86 × 10 (● 2a),
.72 × 10− (○ 2b), 8.00 × 10 (▲ 3a), and 1.24 × 10 (4 3b). (b) The plots of M
4
−5
−4
n
and PDI versus conversion. [Color figure can be
viewed at wileyonlinelibrary.com]
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
11

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 4 Correlation of RT, DT, and CCT mechanisms in cobalt-mediated radical polymerization.
1
7,25,30,31,40
resulting in an induction period.
C/LRPs directly ini-
be rationalized that the change of coordination environment of
cobalt(II) phenoxy-imine complexes by modifying the ligands
could shift their tendency of polymerization pathways.
tiated by organo-cobalt(III) species, which was synthesized
from cobalt(II) complex in advance, do not have induction
4
2,51,52
period.
If the Co─C bond dissociation energy of organo-
cobalt(III) species is in a proper range that cobalt(II) complexes
and radicals can be regenerated via bond homolysis to initiate
the polymerization without addition of external radicals, the
mechanism of C/LRP could be rationalized as reversible termi-
nation (RT) [Scheme 4(a), RT]. In this case, the radical concen-
tration is mainly dominated by the equilibrium of cobalt(II)
CONCLUSIONS
A series of cobalt(II) phenoxy-imine complexes have been pre-
pared and applied to mediate the radical polymerization of
VAc and MA to study the influence of coordination configura-
tion and chelating atoms to the control of radical polymeriza-
tion. Complex 1a showed the best control behavior of VAc
polymerization supported by the living characters of molecu-
lar weights not only increasing linearly with conversion but
also matching with the theoretical values and moderate
molecular weight distributions at room temperature. Com-
and organo-cobalt(III) species ([P•]
=
[Co(III)-P]/([Co
(
II)] × K )), which can be another characteristic to recognize
eq
RT mechanism. However, when the Co─C bond dissociation
energy is too high for organo-cobalt(III) species to release the
radicals, the initiation of polymerization requires an external
radical source such as AIBN or TPO and the radical concentra-
tion is mainly determined by the concentration of external radi-
II
II
pared to the published results using Co (TMP), Co (salen*),
II
and Co (acac) , coordination configuration is more critical to
2
1
/2
cal source ([P•] = (k [initiator]/2k ) ). This mechanism is
the mediation of controlled/living radical polymerization than
chelating atoms, demonstrated by that cobalt complexes with
square planar structure like Co (TMP) and Co (salen*) can
i
t
known as DT [Scheme 4(b)], in which the propagating radicals
rapidly exchange with the organic radicals on organo-cobalt
II
II
(
III) species to provide an equal chance for each polymer chain
mediate C/LRP of both unconjugated monomer of VAc and
II
II
to grow. Both RT and DT can achieve the C/LRP and are distin-
guished by the source and concentration of radicals.
Another pathway is known as CCT [Scheme 4(c))] in which the
metal center of organo-cobalt(III) species abstracts the hydro-
gen from the organic group to form the cobalt hydride and the
conjugated monomer of MA, but Co (FI) and Co (acac) can
2
2
2
9,30
II
only control VAc polymerization, even Co (salen*) and
Co (FI) have the same chelating atoms. In addition to
controlled/living radical polymerization, Co (FI) also could
II
2
II
2
mediate the CCTP of VAc. Complex 2b showed the highest
activity in CCTP of VAc, even higher than the famous CCT
agent of Co(dmgBF ) (H O) , supported by the larger slope in
7
,50
corresponding vinyl species.
Although this reaction was pro-
posed to be an intramolecular process in our study due to the
occurrence of induction period, implying that the formation of
organo-cobalt(III) species is prior to the polymerization, the
possibility of coexistence of β-H transfer via intermolecular
reaction between cobalt(II) complex and radicals cannot be
2
2
2
2
the Mayo plots. Given by the polymerization of VAc mediated
by these six cobalt(II) phenoxy-imine complexes, the charac-
teristics of controlled/living radical polymerization become
more significant when the ligand has more bulky and less
electron-donating substituents. Reversely, the features of
CCTP would be more obvious. The different performance of
cobalt(II) phenoxy-imine complexes in VAc polymerization
could be rationalized by the proposed mechanism correlating
the pathways of RT and DT in C/LRP and the reaction of CCT
5
3–55
excluded.
The cobalt hydride is highly active and can react
with monomers or other vinyl compounds to generate organo-
cobalt(III) species. Because the long propagating radicals (P•)
are replaced by small radicals (m•), and polymers with double
bond chain end could be formed by β-H transfer, this process
would suppress the molecular weight of polymer and generate
macro-monomer for further polymerization. Therefore, it could
II
in CCTP and thus illustrates the potential of Co (FI) in the
2
application of combining C/LRP and CCTP techniques.
1
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
ACKNOWLEDGMENT
25 A. Debuigne, J. R. Caille, R. Jerome, Angew. Chem. Int. Ed.
2
005, 44, 1101.
We thank the Ministry of Science and Technology, Taiwan,
2
6 K. S. S. Kumar, Y. Gnanou, Y. Champouret, J.-C. Daran,
(
MOST 107-2113-M-007-014) for support of this research. We
R. Poli, Chem. A Eur. J. 2009, 15, 4874.
7 Y.-C. Lin, Y.-L. Hsieh, Y.-D. Lin, C.-H. Peng, Macromolecules
2014, 47, 7362.
8 Y. K. Redjel, L. Thevenin, J.-C. Daran, M. Benslimane, R. Poli,
C. Fliedel, Polyhedron 2019, 158, 83.
9 Y. H. Chen, H. H. Lu, J. Q. Li, C. H. Peng, Reversible Deactiva-
also thank Chuan Huang and Shan-Cheng Tang for their help
to the experiments.
2
2
REFERENCES AND NOTES
2
1
B. R. Smirnov, V. D. Plotnikov, B. V. Ozerkovskii,
tion Radical Polymerization: Mechanisms and Synthetic Method-
ologies; ACS Symposium Series 1284; American Chemical
Society: Washington, DC, 2018, p. 335.
V. P. Roshchupkin, N. S. Enikolopyan, Polym. Sci. U.S.S.R 1981,
23, 2807.
2
N. S. Enikolopyan, B. R. Smirnov, G. V. Ponomarev,
3
0 C. M. Liao, C. C. Hsu, F. S. Wang, B. B. Wayland, C. H. Peng,
I. M. Belgovskii, J. Polym. Sci. A-Polym. Chem. 1981, 19, 879.
Polym. Chem. 2013, 4, 3098.
3
A. Gridnev, J. Polym. Sci. A-Polym. Chem. 2000, 38, 1753.
D. J. Forster, J. P. A. Heuts, F. P. Lucien, T. P. Davis, Macro-
3
1 C. S. Hsu, T. Y. Yang, C. H. Peng, Polym. Chem. 2014, 5, 3867.
2 Y. Zhao, M. Yu, S. Zhang, Z. Wu, Y. Liu, C.-H. Peng, X. Fu,
4
3
molecules 1999, 32, 5514.
J. P. A. Heuts, D. Kukulj, D. J. Forster, T. P. Davis, Macromole-
cules 1998, 31, 2894.
K. G. Suddaby, D. R. Maloney, D. M. Haddleton, Macromole-
cules 1997, 30, 702.
J. P. A. Heuts, D. J. Forster, T. P. Davis, B. Yamada,
H. Yamazoe, M. Azukizawa, Macromolecules 1999, 32, 2511.
Chem. Sci. 2015, 6, 2979.
3 Y. Zhao, S. Zhang, Z. Wu, X. Liu, X. Zhao, C.-H. Peng, X. Fu,
Macromolecules 2015, 48, 5132.
5
3
6
3
4 G. W. Theaker, C. Morton, P. Scott, Dalton Trans. 2008, 6883.
5 V. T. Kasumov, M. Özdemir, E. Tas, Ö. Sahin, Transit. Met
3
7
Chem. 2005, 30, 191.
6 L. Sacconi, M. Ciampolini, F. Maggio, F. P. Cavasino, J. Am.
Chem. Soc. 1962, 84, 3246.
7 M. A. Hitchman, Inorg. Chem. 1977, 16, 1985.
3
8
2
S. C. J. Pierik, B. Smeets, A. M. van Herk, Macromolecules
003, 36, 9271.
3
9
M. D. Haddleton, D. R. Maloney, K. G. Suddaby, Macromole-
cules 1996, 29, 481.
K. G. Suddaby, D. M. Haddleton, J. J. Hastings,
S. N. Richards, J. P. O’Donnell, Macromolecules 1996, 29, 8083.
1 W. Wang, P. A. Stenson, A. Marin-Becerra, J. McMaster,
M. Schröder, D. J. Irvine, D. Freeman, S. M. Howdle, Macromol-
ecules 2004, 37, 6667.
38 B. Chakraborty, S. Banerjee, J. Coord. Chem. 2013, 66, 3619.
1
0
39 B. S. Jaynes, L. H. Doerrer, L. Shuncheng, J. Stephen, Inorg.
Chem. 1995, 34, 5735.
1
40 S. A. Carabineiro, L. C. Silva, P. T. Gomes, L. C. J. Pereira,
L. F. Veiros, S. I. Pascu, M. T. Duarte, S. Namorado,
R. T. Henriques, Inorg. Chem. 2007, 46, 6880.
1
2 A. A. Gridnev, S. D. Ittel, Chem. Rev. 2001, 101, 3611.
41 C.-H. Peng, J. Scricco, S. Li, M. Fryd, B. B. Wayland, Macro-
molecules 2008, 41, 2368.
1
3 N. G. Engelis, A. Anastasaki, R. Whitfield, G. R. Jones,
E. Liarou, V. Nikolaou, G. Nurumbetov, D. M. Haddleton, Macro-
molecules 2018, 51, 336.
42 A. Debuigne, Y. Champouret, R. Jérôme, R. Poli,
C. Detrembleur, Chem. A Eur. J. 2008, 14, 4046.
1
4 E. V. Bellan, L. Thevenin, F. Gayet, C. Fliedel, R. Poli, ACS
43 G. Moad, E. Rizzardo, D. H. Solomon, B. ALJ, Polym. Bull.
1992, 29, 647.
Macro. Lett. 2017, 6, 959.
1
5 J. P. A. Heuts, N. M. B. Smeets, Polym. Chem. 2011, 2, 2407.
6 S. Banerjee, E. V. Bellan, F. Gayet, A. Debuigne, C. Detrembleur,
44 S. Li, B. d. Bruin, C.-H. Peng, M. Fryd, B. B. Wayland, J. Am.
Chem. Soc. 2008, 130, 13373.
1
R. Poli, B. Améduri, V. Ladmiral, Polymers 2017, 9, 702.
7 A. Debuigne, J. Warnant, R. Jérôme, I. Voets, A. de Keizer,
M. A. Cohen Stuart, C. Detrembleur, Macromolecules 2008, 41,
353.
8 P. G. Falireas, V. Ladmiral, A. Debuigne, C. Detrembleur,
R. Poli, B. Ameduri, Macromolecules 2019, 52, 1266.
45 G. Odian, Principles of Polymecrization; Wiely: New York, NY,
1
981.
1
46 A. Rudin, The Elements of Polymer Science and Engineering;
Academic Press: New York, NY, 1982.
2
1
47 A. H. Janowicz, U.S. Patent, 4,694,054, 1991.
4
8 S. Salehpour, M. A. Dubé, Green Chem. 2008, 10, 321.
9 F. R. Mayo, J. Am. Chem. Soc. 1943, 65, 2324.
1
9 C. Fliedel, R. Poli, J. Organomet. Chem. 2019, 880, 241.
J. Demarteau, B. Améduri, V. Ladmiral, M. A. Mees,
4
20
5
1
0 J. P. A. Heuts, D. J. Forster, T. P. Davis, Macromolecules
999, 32, 3907.
R. Hoogenboom, A. Debuigne, C. Detrembleur, Macromolecules
2017, 50, 3750.
5
1 A. Debuigne, J.-R. Caille, N. Willet, R. Jérôme, Macromole-
21
S. Yamago, B. Ray, K. Iida, J.-i. Yoshida, T. Tada,
cules 2005, 38, 9488.
2 M. Hurtgen, A. Debuigne, C. Jérôme, C. Detrembleur, Macro-
molecules 2010, 43, 886.
3 S. Slavin, K. McEwan, D. M. Haddleton, In Polymer Science;
K. Yoshizawa, Y. Kwak, A. Goto, T. Fukuda, J. Am. Chem. Soc.
5
2004, 126, 13908.
22
A. Kermagoret, Y. Nakamura, M. Bourguignon,
5
C. Detrembleur, C. Jérôme, S. Yamago, A. Debuigne, ACS
Macro Lett. 2014, 3, 114.
A. Comprehensive Reference, K. Matyjaszewski, M. Möller, Eds.;
Elsevier BV: Amsterdam, 2012; Vol. 3, p. 249.
23 J. J. Lessard, L. F. Garcia, C. P. Easterling, M. B. Sims,
5
4 A. Debuigne, R. Poli, C. Jérôme, R. Jérôme, C. Detrembleur,
Prog. Polym. Sci. 2009, 34, 211.
5 G. E. Roberts, C. Barner-Kowollik, T. P. Davis, J. P. A. Heuts,
Macromolecules 2003, 36, 1054.
K. C. Bentz, S. Arencibia, D. A. Savin, B. S. Sumerlin, Macromol-
ecules 2019, 52, 2105.
5
24 B. B. Wayland, G. Poszmik, S. L. Mukerjee, M. Fryd, J. Am.
Chem. Soc. 1994, 116, 7943.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
13