Controlled synthesis of photosensitive graft copolymers with high azobenzene-chromophore loading densities in the main and side chains by combining ATRP and ADMET polymerization

Article abstract of DOI:10.1016/j.reactfunctpolym.2015.05.003

Photosensitive graft copolymers containing azobenzene chromophores in the main and side chains were successfully synthesized by combining living atom transfer radical polymerization (ATRP) and acyclic diene metathesis (ADMET) chemistry based on the results of proper heterodifunctional inimer and azobenzene monomer design. The precise copolymer architectural features were manipulated by combining the macromonomer technique and the macroinitiator method. The as-prepared copolymer containing azobenzene chromophores in the main and side chains showed unique reversible isomerization processes, suggesting that the photoisomerization of azobenzene chromophores occured mainly in one of the two types of azobenzene groups in the main or side chains with similar probabilities due to their main and side-on structure.

Full text of DOI:10.1016/j.reactfunctpolym.2015.05.003

Reactive & Functional Polymers 91–92 (2015) 85–92
Contents lists available at ScienceDirect
Reactive & Functional Polymers
journal homepage: www.elsevier.com/locate/react
Controlled synthesis of photosensitive graft copolymers with high
azobenzene-chromophore loading densities in the main and side chains
by combining ATRP and ADMET polymerization
Liang Ding a, , Juan Li , Chengshuang Wang , Ling Lin
⇑
a
a
a,b
a
School of Materials Engineering, Yancheng Institute of Technology, Yancheng 224051, China
Key Laboratory of Eco-Textile, Ministry of Education, Jiangnan University, Wuxi 214122, China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Photosensitive graft copolymers containing azobenzene chromophores in the main and side chains were
successfully synthesized by combining living atom transfer radical polymerization (ATRP) and acyclic
diene metathesis (ADMET) chemistry based on the results of proper heterodifunctional inimer and
azobenzene monomer design. The precise copolymer architectural features were manipulated by
combining the macromonomer technique and the macroinitiator method. The as-prepared copolymer
containing azobenzene chromophores in the main and side chains showed unique reversible isomeriza-
tion processes, suggesting that the photoisomerization of azobenzene chromophores occured mainly in
one of the two types of azobenzene groups in the main or side chains with similar probabilities due to
their main and side-on structure.
Received 25 March 2015
Received in revised form 2 May 2015
Accepted 7 May 2015
Available online 14 May 2015
Keywords:
Graft copolymer
Azobenzene
Main–side chains
ATRP
Ó 2015 Elsevier B.V. All rights reserved.
ADMET polymerization
1
. Introduction
Polymers containing azobenzene chromophores have received
development of new main and side chain azo-functionalized poly-
mers is important.
Graft copolymers are regularly branched macromolecules con-
sisting of a backbone, and many side chains [11]. They offer unique
material properties that can be tailored by changing the polymer
backbone and the graft chains [12]. The design of controlled graft
copolymers through various methods has become the focus of
interesting research in which the physical and chemical properties
can be purposely altered by combining different polymer compo-
nents in an ordered fashion [13]. Generally, well-defined graft
copolymers with various architectures have been synthesized
using multiple controlled/living polymerization methods [14].
Among these diverse polymerization strategies, metathesis poly-
merization and radical polymerization are particularly suitable
for the preparation of graft copolymers due to the controlled
molecular weight, relatively narrow molecular weight distribution,
and desired topology of the resulting polymers [15–19].
significant attention because of their two unique properties: pho-
toinduced reversible trans–cis isomerization and nonlinear optical
properties [1,2]. Generally, azobenzene-containing polymers (azo
polymers) can be divided into main-chain and side-chain types
according to the locations of the azo groups in the polymers
[
3,4]. These two polymer types have some distinct differences in
their thermal stabilities [5] and chain anisotropies [6]. Polymers
containing azo group in the main and side chains can be easily pre-
pared by either random copolymerization of azo monomers with-
out reactive groups with another monomer with a reactive group
[
7,8] or direct living polymerization of azo monomers with reactive
groups [9,10]. However, most investigations have focused on the
incorporation of azo groups into either the main chains or the side
chains, and there are few reports on polymers with azo chro-
mophores in both the main and side chains, most likely due to
the rather limited synthetic methods. Based on the above state-
ments, it can be concluded that, despite the fact that the synthesis
of main and side chain, azo polymers remains a challenge, they
have a promising future in many applications. Therefore, the
Olefin metathesis-based step-growth polymerization, called
acyclic diene metathesis (ADMET) polymerization, is performed
with
a,x-dienes and proceeds via the release of ethylene as a
condensate [20,21]. It is well known that ADMET polymerization
successfully yields unsaturated, linear, and high molecular weight
polymers possessing various functional groups, both within
and pendant to them [22]. This approach has led to a wide range
of unsaturated and saturated polymers bearing different
⇑
Corresponding author. Tel.: +86 515 88298872.
E-mail address: dl1984911@ycit.edu.cn (L. Ding).
http://dx.doi.org/10.1016/j.reactfunctpolym.2015.05.003
381-5148/Ó 2015 Elsevier B.V. All rights reserved.
1

8
6
L. Ding et al. / Reactive & Functional Polymers 91–92 (2015) 85–92
functionalities with the possibility of modifying the polyolefin
architectures by placing precise functional groups throughout the
backbone [23]. Wagener explored this polymerization technique
and prepared a new type of graft copolymer by combining atom
transfer radical polymerization (ATRP) and ADMET polymerization
a
,
x
-diene monomer (M2) was synthesized as reported previously
by our group [25c].
2.2. Characterization
[
24]. These methods provide the possibility of preparing
ATR–IR spectra were recorded on a Bruker Tensor 37 in the
azo-containing main–side chain graft copolymers. However, to
the best of our knowledge, up to now, only three classes of
metathesis azo-polymer based on more defined polymeric
structures (linear, hyperbranched and long-chain highly branched
polymers) have been reported by our group [25].
Consequently, the goal in this research is to exploit the ADMET
polymerization scheme for the preparation of densely grafted
molecular brushes with azo units in the main and side chains.
ꢁ1
region of 4000–400 cm using attenuated total reflection (ATR).
UV–vis absorption spectra were measured on a Cary 60 spectrom-
eter. UV irradiation was performed using an 8-W UV lamp with a
wavelength of 365 nm. Irradiation by visible light was performed
using a 23-W Philips day light bulb (>400 nm). Elemental analysis
(
EA) was conducted using an Elementar Vario EL. Gas chromatog-
raphy (GC) was measured using an Agilent 6890 series GC system
instrument equipped with a flame ionization detector and a capil-
lary column (HP–5. 0.25 mm ꢂ 30 m), using decane as an internal
standard; Tinj 280 °C, Tdetec 280 °C, Tinit 50 °C (10 °C/min), carrier
gas: N . High–resolution mass spectrometry (HRMS) data were
2
recorded on a Waters GCT Premier mass spectrometer in electron
ionization mode. LC/MS measurements were performed using an
According to ‘‘Route 1’’, the azo-functionalized
4
a,x-diene inimer
first initiated ATRP of the as-prepared azo monomer M1 using
CuBr/PMDETA as a catalyst and ligand, yielding a macromonomer
in the subsequent ADMET polymerization. The second route relies
upon ADMET polymerization of inimer 4 catalyzed by a Grubbs
third generation catalyst (Ru-III) to produce a macroinitiator fol-
lowed by initiating ATRP of M1 (Scheme 1).
Agilent technology 1200 series at 45 °C using H
the mobile phase on a 50 m ꢂ 4.6 mm ꢂ 3.5 m diffused column.
H (500 MHz) and C (125 MHz) NMR spectra were recorded
using tetramethylsilane as an internal standard in CDCl on a
2 3
O or CH CN as
l
1
13
3
2
. Experimental
Bruker DPX spectrometer. Multiangle laser light scattering–gel
permeation chromatography (MALLS–GPC) was employed to eval-
uate the molecular weights of the synthesized samples. The
MALLS–GPC system consisted of a Waters 2690D Alliance liquid
chromatography system, a Wyatt Optilab DSP differential refrac-
tometer detector (690 nm, Wyatt Technology), and a Wyatt
DAWN EOS MALLS detector (30 mW GaAs linearly polarized laser,
690 nm). Two chromatographic columns (Styragel HR3, HR4) with
a precolumn were used in series. THF was used as the mobile phase
2.1. Materials
Methacrylic
acid
(98%),
(99%),
10-undecenoic
acid
(99%),
(90%),
2
2
-bromoethanol
3-mercapto-1,2-propanediol
-bromoisobutyryl bromide (97%), dichloro[1,3-bis(2,4,6-trimethyl
phenyl)-2-imidazolidinylidene](benzylidene)bis(3-bromopyridine)
ruthenium (II) (Grubbs third generation catalyst, Ru-III), 1-[3-(di
methylamino)propyl]-3-ethylcarbodiimide hydrochloride (EDCIꢀ-
at a flow rate of 1.0 mL/min at 30 °C. A 100-
lL sample of a
0
00
00
HCl, >99%), and N,N,N ,N ,N -pentamethyldiethylenetriamine
3.0 mg/mL solution, which was filtered through
a
0.2-
lm
(
(
PMDETA, 98%) were purchased from Energy Chemical
Shanghai) and used as received, without purification. The solvents
Whatman filter prior to use, was injected for all measurements.
The increase in the refractive index (dn/dc) of the polymer samples
in THF was determined using a Wyatt Optilab DSP differential
refractometer at 690 nm. The data analysis was performed using
Astra software (ver. 4.90.04, Wyatt Technology).
were distilled over drying agents under nitrogen prior to use.
Copper(I) bromide (CuBr, Alfa Aesar, 97%) was purified by washing
with glacial acetic acid, followed by ethanol and ethyl ether, and
then dried under vacuum. Triethylamine (Et
freshly distilled and dried. The azobenzene functional asymmetric
3
N) and pyridine were
Polymerizations were performed in Schlenk tubes using a nitro-
gen flow to drive off the ethylene condensate during ADMET.
Scheme 1. Synthetic routes to azobenzene functionalized graft copolymers via commutative combination of ATRP and ADMET polymerization.

L. Ding et al. / Reactive & Functional Polymers 91–92 (2015) 85–92
87
2.3. Synthesis of methacrylic acid 2-bromo-ethyl ester (1)
2.6. Synthesis of the hydroxyl bromoisobutyryloxy azobenzene
compound (3)
Methacrylic acid (4.30 g, 50 mmol), 2-bromoethanol (9.38 g,
5 mmol), and DMAP (0.75 g, 6 mmol) were added to a Schlenk
7
Compound 2 (5.05 g, 8 mmol) was dissolved in 30 mL of THF,
and dry Et N (2.42 g, 24 mmol) was added under nitrogen. The
flask and dissolved in 100 mL of dried CH
2
Cl
2
under a nitrogen
3
atmosphere. Then, EDCI (11.48 g, 60 mmol) was added at 0 °C with
rapid stirring. The reaction mixture was allowed to warm to room
temperature and stirred for an additional amount of time. After
reaction mixture was cooled to 0 °C before 2-bromoisobutyryl
bromide (0.9 mL, 7.2 mmol) was added dropwise to this solution
via syringe. The reaction mixture was then warmed to room
temperature and stirred for another 24 h. During this time, a
white precipitate formed. The precipitate was filtered off, and
the filtrate was subsequently concentrated and precipitated in
excess deionized water. The crude product precipitated and
was further purified using silica gel chromatography eluted with
3
days, the resulting solution was subsequently washed with 1 M
HCl, saturated NaHCO aq., and deionized water. The organic layer
was separated and dried over anhydrous Na SO . Then, it was
evaporated to afford a yellowish liquid (8.36 g, 86.6% yield).
3
2
4
1
H NMR (500 MHz, CDCl
3
, d): 6.51–6.43 (m, 1H, CH@CCH
.97–5.82 (m, 1H, CH@CCH COO), 4.42–4.17 (m, 2H, CH CH
.61–4.30 (m, 2H, CH CH Br), 2.09–1.73 (m, 3H, CH@CCH COO).
C NMR (125 MHz, CDCl , d): 165.6, 139.1, 123.3, 71.7, 33.1,
8.5. GC: single peak was observed. EI/HRMS: Calcd. for
Br: 193.0427; found: 193.0326. Anal. calcd. for C Br:
3
COO),
5
3
3
2
2
Br),
methylene chloride/methanol (50/1, R
f
= 0.5) to afford a red solid
(4.41 g, 70.6% yield). H NMR (500 MHz, CDCl , d): 7.98–7.89 (m,
4H, o-ArHAN@NAArH), 7.02–6.87 (m, 4H, m-ArHAN@NAArH),
5.84–5.77 (d, 1H, CH @CH), 5.03–4.96 (m, 2H, CH @CH), 4.55–
4.47 (m, 2H, CH CH OCO), 4.24–4.02 (m, 3H, CH CH OCO +
CHOCOCH OH), 3.72–3.59 (m, 2H, CHOCOCH OH), 2.77–2.53
m, 6H, CH CH SCH ), 2.22–2.10 (m, 2H, OCOCH ), 2.06–1.93
m, 8H, CH @CHCH + OCOCCH BrCH ), 1.79–1.62 (m, 2H,
CH ), 1.42–1.11 (m, 12H, CH
, d): 176.2, 172.1, 161.7, 144.2, 140.3, 123.6, 114.5, 80.6,
2.2, 67.7, 66.0, 52.1, 35.5, 33.7, 29.8, 27.9, 25.3, 24.1. LC: single
peak was observed. EI/MS: Calcd. for C37 SBr: 779.77;
found: 779.79. Anal. calcd. for C37 SBr: C 56.99, H 6.59,
O 18.47; Found: C 57.01, H 6.56, O 18.47.
1
2
2
3
3
1
3
3
1
C
2
2
H
6 9
O
2
6
H
9
O
2
2
2
2
2
C 37.33, H 4.70, O 16.58; Found: C 37.46, H 4.45, O 16.57.
2
2
(
(
2
2
2
2
2
2
3
3
2.4. Synthesis of 4-(2-ethyoxy methacrylate) azobenzene (M1)
1
3
OCOCH
CDCl
2
2
2
).
C NMR (125 MHz,
3
4
-Hydroxy azobenzene (5.94 g, 30 mmol), potassium carbon-
7
ate (16.56 g, 120 mmol), and 90 mL of DMF were charged into
a 250-mL Schlenk flask. The reaction mixture was heated at
51 9 2
H O N
51 9 2
H O N
8
0 °C for 6 h under nitrogen, thus allowing the potassium salt
to form. A solution of compound 1 (6.95 g, 36 mmol) in 30 mL
of DMF was then added dropwise to the above mixture. After
2
.7. Synthesis of the azobenzene-containing a,x-diene inimer (4)
2
1
4 h of stirring at 50 °C, the reaction mixture was poured into
L of deionized water and the crude product precipitated.
Compound 3 (3.90 g, 5 mmol), 10-undecenoic acid (1.11 g,
mmol), DMAP (0.08 g, 0.6 mmol), CH Cl (22 mL), and THF
This product was further purified by recrystallization from
6
2
2
1
ethanol to give a red crystal (7.55 g, 81.2% yield). H NMR
(
8 mL) were charged into a 100-mL round-bottom flask equipped
(
500 MHz, CDCl
3
,
d): 8.08–7.96 (m, 4H, o-ArHAN@NAArH),
.56–7.49 (m, 3H, m-ArHAN@NAArH + p-ArHAN@NAArH),
.02–6.93 (d, 2H, m-ArHAN@NAArH), 6.22–6.17 (m, 1H,
COO), 5.87–5.81 (m, 1H, CH@CCH COO), 4.62–4.49
2H, CH @CCH COOCH CH ), 4.36–4.27 (m, 2H,
@CCH COOCH CH ), 1.95–1.89 (m, 3H, CH@CCH COO).
NMR (125 MHz, CDCl , d): 165.8, 161.3, 153.5, 144.7, 136.9,
31.0, 128.2, 123.3, 122.8, 115.1, 73.5, 67.6, 18.2. LC: single
peak was observed. EI/MS: Calcd. for 310.34;
found: 310.32. Anal. calcd. for C18 : C 69.64, H 5.85,
O 15.45; Found: C 69.66, H 5.85, O 15.47.
with a magnetic stirrer under a nitrogen atmosphere, and the
mixture was stirred at 0 °C for 15 min. EDCI (1.12 g, 6 mmol)
was then added to the former solution, which was stirred for
7
7
CH@CCH
m,
CH
3
3
3
days after the solution warmed to room temperature. The
resulting solution was washed three times with deionized water
3 ꢂ 80 mL), and the organic layer was dried over anhydrous
Na SO . The solvent was then evaporated and the crude product
was purified by silica gel chromatography eluted with methylene
chloride/petroleum ether (10/1, R = 0.6) to give a dark red solid
3.58 g, 75.7% yield). H NMR (500 MHz, CDCl , d): 8.04–7.76 (m,
H, o-ArHAN@NAArH), 7.05–6.83 (m, 4H, m-ArHAN@NAArH),
.85–5.72 (d, 1H, CH @CH), 5.02–4.94 (m, 2H, CH @CH),
CHOCOCH OCO), 4.53–4.46 (m, 2H, CH CH OCO),
.37–4.07 (m, 4H, CH CH OCO + CH CHOCOCH OCO), 2.73–2.50
(m, 6H, CH CH SCH ), 2.32–2.11 (m, 4H, OCOCH ), 2.08–1.79
(m, 10H, CH @CHCH + OCOCCH BrCH ), 1.71–1.52 (m, 4H,
OCOCH CH ), 1.50–1.16 (m, 24H, CH
CDCl , d): 175.6, 172.0, 161.4, 145.0, 139.7, 122.9, 115.2, 77.1,
72.8, 69.3, 66.4, 53.2, 36.5, 33.6, 29.9, 28.5, 26.8, 25.2. LC: single
peak was observed. EI/MS: Calcd. for C48 SBr: 946.02;
found: 945.99. Anal. calcd. for C48 SBr: C 60.94, H 7.35,
O 16.91; Found: C 60.91, H 7.36, O 16.87.
(
2
3
2
2
1
3
2
3
2
2
3
C
(
3
2
4
1
18 18 3 2
C H O N :
f
18 3 2
H O N
1
(
4
5
3
2
2
2.5. Synthesis of the dihydroxyl azobenzene compound (2)
4.90–4.72 (CH
4
2
2
2
2
2
2
2
2
M2 (5.23 g, 10 mmol), 3-mercapto-1,2-propanediol (1.62 g,
5 mmol), Et N (2.02 g, 20 mmol) and THF (50 mL) were charged
2
2
2
2
1
3
2
2
3
3
1
3
into a round-bottom flask equipped with a magnetic stirrer under
nitrogen atmosphere, and the mixture was stirred overnight at
room temperature. The resulting solution was concentrated and
precipitated into excess deionized water. The product precipitated
2
2
2
).
C NMR (125 MHz,
3
69 10 2
H O N
H O N
69 10 2
1
and dried under vacuum to give a red solid (5.62 g, 89.1% yield). H
NMR (500 MHz, CDCl
3
, d): 7.97–7.86 (m, 4H, o-ArHAN@NAArH),
7
.01–6.89 (m, 4H, m-ArHAN@NAArH), 5.82–5.76 (d, 1H,
CH
2
@CH), 5.06–4.95 (m, 2H, CH
CH OCO), 4.26–4.03 (m, 2H, CH
OH), 2.73–2.50 (m, 6H, CH
), 2.04–1.98 (m, 2H, CH @CHCH
CH ), 1.47–1.06 (m, 12H, CH
, d): 171.8, 161.3, 144.7, 139.6, 123.3, 115.2, 76.5, 71.6,
8.1, 66.7, 36.8, 34.2, 29.4, 27.6, 23.2. LC: single peak was observed.
EI/MS: Calcd. for C33 S: 630.79; found: 630.74. Anal. calcd.
for C33 S: C 62.83, H 7.35, O 20.29; Found: C 62.80 H 7.35, O
0.27.
2
@CH), 4.51–4.42 (m, 2H,
CH OCO), 3.61–3.52 (m, 3H,
CH SCH ), 2.20–2.13 (m, 2H,
), 1.90–1.69 (m, 2H,
2.8. ATRP for preparing macromonomer (First Step in Route 1)
CH
2
2
2
2
CHOHCH
OCOCH
OCOCH
CDCl
2
2
2
2
A mixture of inimer (945 mg, 1 mmol), M1 (4.65 g, 15 mmol),
2
2
2
CuBr (288 mg, 2 mmol), PMDETA (420
lL, 2 mmol), 10 mL of THF,
13
2
2
2
).
C
NMR (125 MHz,
and 2 mL of CH OH was degassed by three freeze–vacuum–thaw
3
3
cycles and then heated at 60 °C for 12 h under vacuum. The mix-
ture was diluted in 20 mL of THF followed by passing through a
column of basic alumina. The purified polymer was precipitated
in excess methanol and then dried under vacuum for 24 h to give
the macromonomer with a monomer conversion of 92%. 1H NMR
6
46 8 2
H O N
46 8 2
H O N
2

8
8
L. Ding et al. / Reactive & Functional Polymers 91–92 (2015) 85–92
Scheme 2. General protocols for the preparation of the azobenzene functional inimer for ATRP.
(
(
500 MHz, CDCl
3
, d): 7.95–7.67 (m, o-ArHAN@NAArH), 7.58–7.29
CH
CH
2
CH
2
SCH
2
+ OCOCH
+ CH CBrCH
= 26,700, M
= 2.08 for 48 h;
= 127,800, M /M = 1.92 for 96 h; NMR: M
= 84,200 for 72 h.
2
), 2.22–2.10 (m, OCOCH
), 1.77–1.02 (m, CH
/M = 2.23 for 24 h; M
= 89,300,
2
), 2.07–1.81 (m,
+ OCOCCH CH ).
= 50,500,
m, p-ArHAN@NAArH), 7.01–6.78 (m, m-ArHAN@NAArH), 5.89–
2
@CHCH
2
2
3
2
3
3
5
.75 (d, CH
.52–4.43 (m, CH
CHOCOCH OCO), 2.76–2.44 (m, CH
.77 (m, CH @CHCH + CH CBrCH ), 1.71–1.52 (m, CH
OCOCCH CH ). MALLS–GPC: = 5300, /M = 1.07; NMR:
= 5200.
2
@CH), 5.03–4.79 (m, CH
CH OCO), 4.27–4.02 (m, CH
CH SCH + OCOCH
2
@CH + CH
2
CHOCOCH
CH
), 2.12–
2
OCO),
OCO +
MALLSꢁGPC: M
n
w
n
n
4
CH
1
2
2
2
2
M
M
M
w
/M
n
M
n
M
w
/M
n
= 1.89 for 72 h;
2
2
2
2
2
2
n
w
n
n
= 23,900 for 24 h;
2
2
2
3
2
+
n
3
3
M
n
M
w
n
M
n
2.10. General ADMET polymerization procedure for the synthesis of the
macroinitiator (First Step in Route 2)
2
.9. Typical ADMET polymerization procedure for the synthesis of the
In a nitrogen–filled Schlenk tube, a solution of Ru-III (0.5 mol%
to inimer 4) in toluene was degassed with three freeze–vacuum–
thaw cycles and then added to a degassed (with the same proce-
dure as above) solution of inimer 4 (0.5 mmol, 473 mg) in toluene.
After stirring at 60 °C for 24 h, the polymerization was quenched
by adding ethyl vinyl ether with stirring for 30 min. Then, the reac-
tion mixture was precipitated in excess methanol, and the precip-
azo-containing main–side chains graft copolymer (Second Step in
Route 1)
In a nitrogen-filled Schlenk tube, a solution of Ru-III (0.5 mol%
to macromonomer) in 0.2 mL of toluene degassed with three
freeze–vacuum–thaw cycles was added to a solution of the macro-
monomer (0.5 mmol) in 0.8 mL of toluene degassed with the same
procedure. After the reaction mixture was stirred at 60 °C for
itate was isolated by filtration and dried under vacuum for 24 h to
1
give the macroinitiator at high yields. H NMR (500 MHz, CDCl
3
, d):
2
4ꢁ96 h, the polymerization was quenched by adding ethyl vinyl
8
.03–7.83 (m, o-ArHAN@NAArH), 7.22–6.90 (m, m-ArHAN@NA
ArH), 5.47–5.21 (d, CH@CH), 4.92–4.83 (s, CH CHOCOCH OCO),
.61–4.43 (m, CH CH OCO), 4.28–3.96 (m, CH CH OCO +
ether with stirring for 30 min. The solution was precipitated in
excess methanol, and the precipitate was isolated by filtration
and dried under vacuum for 24 h to give the resulting graft copoly-
2
2
4
2
2
2
2
1
CH CHOCOCH OCO), 2.77–2.42 (m, CH CH SCH + OCOCH ), 2.25–
mer as a dark red semisolid to solid. H NMR (500 MHz, CDCl
.05–7.78 (m, o-ArHAN@NAArH), 7.61–7.28 (m, p-ArHAN@NA
ArH), 7.21–6.96 (m, m-ArHAN@NAArH), 5.49–5.24 (d, CH@CH),
3
, d):
2
2
2
2
2
2
2
1
.11 (m, OCOCH
.76–1.20 (m, CH
= 17,500.
2
), 2.06–1.78 (m, CH
2
@CHCH
= 18,300, M /M = 2.21;
2
3 3
+ OCOCCH CH ),
8
2
). MALLSꢁGPC:
M
n
w n
NMR: M
n
5
4
.00–4.89 (s, CH
.31–3.99 (m, CH
2
CHOCOCH
2
OCO), 4.68–4.47 (m, CH
2 2
CH OCO),
2
CH OCO + CH
2
2
CHOCOCH OCO), 2.73–2.46 (m,
2
2.11. Representative procedure for the preparation of the azo-
containing main–side chains graft copolymer via ATRP (Second Step in
Route 2)
The macroinitiator produced by ADMET polymerization
(
(
0.02 mmol) in 1 mL of THF, M1 (93 mg, 0.3 mmol), CuBr
14.4 mg, 0.1 mmol) and PMDETA (21 L, 0.1 mmol) in 1 mL of
OH were charged into a Schlenk tube.
l
THF and 0.5 mL of CH
3
After degassing of the mixture by three freeze–vacuum–thaw
cycles, the tube was sealed under vacuum and then heated at
6
0 °C for 24 h. The reaction was quenched by removal of the heat
and exposure to air. The mixture was diluted in 15 mL of THF
and then passed through a short column of basic alumina to
remove the copper residues. The purified copolymer was precipi-
tated in petroleum ether twice, and then dried under vacuum for
2
4 h to give the resulting graft copolymer as a dark red solid with
1
a monomer conversion of 20%. H NMR (500 MHz, CDCl
3
, d): 8.05–
.78 (m, o-ArHAN@NAArH), 7.61–7.28 (m, p-ArHAN@NAArH),
.21–6.96 (m, m-ArHAN@NAArH), 5.49–5.24 (d, CH@CH),
.00–4.89 (s, CH CHOCOCH OCO), 4.68–4.47 (m, CH CH OCO),
.31–3.99 (m, CH CH OCO + CH CHOCOCH OCO), 2.73–2.46 (m,
CH SCH + OCOCH ), 2.22–2.10 (m, OCOCH ), 2.07–1.81 (m,
7
7
5
4
2
2
2
2
2
2
2
2
Fig. 1. 1H NMR spectra for (a) ATRP monomer M1 and (b) inimer 4.
CH
2
2
2
2
2

L. Ding et al. / Reactive & Functional Polymers 91–92 (2015) 85–92
89
Fig. 2. ATR–IR spectra for (a) compound 2 and (b) inimer 4.
CH
2
@CHCH
2
+ CH
2
CBrCH
3
), 1.77–1.02 (m, CH
2
+ OCOCCH
3 3
CH ).
MALLSꢁGPC: M
n
= 35,500, M
w
/M = 2.37; NMR: M
n
n
= 36,800.
3
. Results and discussion
.1. Preparation of the ATRP monomer and inimer
The synthesis of 4-(2-ethyoxy methacrylate) azobenzene (M1),
3
which is depicted in Scheme 2, involved the Williamson coupling
reaction of the as-prepared methacrylic acid 2-bromo-ethyl ester
(
1) with 4-hydroxy azobenzene in high yield. 1H NMR (Fig. 1a),
LC/MS, and elemental analysis confirmed the structure and purity
of the ATRP monomer, M1. In fact, the key point of the studied syn-
thetic routes relies on the use of inimer 4, which has mixed func-
tionality: It contains two polymerizable terminal alkenes for
Fig. 3. 1H NMR spectra for the (a) macromonomer, (b) macroinitiator, and (c) graft
copolymer.
ADMET polymerization and one
a-bromoisobutyrate group to ini-
tiate ATRP and was designed and prepared using a three-step reac-
tion (Scheme 2). The effective formation and purity of 4 were
confirmed by ATR–IR (Fig. 2b) and H NMR spectroscopies (Fig. 1b).
1
was found to be close to Mn,NMR and the theoretical molecular
weight (Mn,theo = [M1]/[4] ꢂ yield% ꢂ M(M1) + M(2) = 5200).
The diene-containing macromonomer was then converted into a
graft copolymer via ADMET polymerization. The metathesis condi-
tions of the macromonomer required high temperatures and
extended reaction times [27]. The azo-containing macromonomer
was dissolved in toluene, heated to 60 °C, and placed under a slight
argon purge to remove ethylene and drive the equilibrium of the
3.2. Macromonomer synthesis and ADMET polymerization
According to the ‘‘Route 1’’ approach, the macromonomer tech-
nique, which was used to synthesize the graft copolymers, forms
the basis for our linking ATRP and ADMET polymerization for
well-defined graft copolymers possessing narrow molecular
weight distributions [24,26]. The controlled radical polymerization
of M1 was first initiated by using inimer 4 as the initiator and
1
polycondensation to completion. H NMR technique was used to
3
CuBr/PMDETA as the catalyst/ligand at 60 °C in THF/CH OH. To
obtain a macromonomer with a relatively low molecular weight,
we initially used a low monomer M1 to inimer 4 ratio of 15:1.
The product was characterized using 1H NMR spectroscopy as
shown in Fig. 3a. The structure of the macromonomer was verified
through the characteristic resonances of the terminal double bonds
at 5.78 (H
nyl proton, methylene proton and methyl proton on each M1 unit
at 7.35ꢁ7.56 ðH Þ and 2.12ꢁ1.83 ppm (H , H ). By comparing the
integration area ratio of phenyl proton from the M1 unit at
r s
) and 4.96 ppm (H ) as well as the resonances of the phe-
0
h
x
c
7
5
4
.35ꢁ7.56 ðH ; 3HÞ to that of the terminal alkene proton at
0
h
.78 ppm (H
r
, 2H) and including the molecular weight of inimer
1
, the H NMR number-average molecular weight ðMn;NMR
¼
½ðS
0
=3Þ=ðS =2Þꢃ ꢂ MðM1Þ þ Mð4Þ ¼ 4900Þ was determined. The living
r
h
free-radical conditions of ATRP can be used to generate monodis-
perse macromonomers before metathesis condensation. The
MALLS trace of the macromonomer (Fig. 4) displayed monomodal
peaks and a narrow molecular weight distribution of 1.12 with a
relatively moderate molecular weight (Mn,MALLS) of 5100, which
Fig. 4. Representative MALLS–GPC traces for the macromonomer and graft
copolymer.

9
0
L. Ding et al. / Reactive & Functional Polymers 91–92 (2015) 85–92
resonance signal of the internal olefin at 5.43ꢁ5.36 ppm was
observed. Additionally, all of the other peaks were easily assigned
for the macromonomer as well as for the graft copolymer. Then,
the number-average molecular weight of the graft copolymer was
determined by comparing the integration area ratio of the newly
formed internal olefin (HACH@CHA) to that of the terminal alkene
(
M
M
(
H
r
):
M
n,NMR = n ꢂ M(macromonomer) ꢁ (n ꢁ 1) ꢂ M(ethylene) = n ꢂ
(macromonomer) ꢁ n ꢂ M(ethylene) + M(ethylene) = n ꢂ (M(macromonomer) ꢁ
(ethylene)) + M(ethylene) = [(S(ACH@CHA)/2)/(S
r
/2)] ꢂ [M(macromonomer)
4900) ꢁ M(ethylene) (28)] + M(ethylene) (28).
The MALLS curves in Fig. 4 also showed the variation in the
molecular weights and molecular weight distributions of the graft
copolymers prepared from the macromonomer. We observed that
the molecular weights of the resulting copolymers shifted to
higher molecular weight positions as the polymerization time
increased. The starting macromonomer residue was not observed,
indicating that the macromonomer was completely consumed
after 24 h, which could be considered positive evidence for the for-
mation of the graft copolymers (Table 1). However, the short life-
time of the catalyst in the reaction solution limited the
effectiveness of prolonging the polymerization time for more than
Fig. 5. Representative MALLS–GPC traces for the macroinitiator and graft
copolymer.
determine completion of the reaction (Fig. 3c). After extensive reac-
1ꢁ2 days, so a fresh batch of 0.5 mol equiv of the catalyst was
tion times, the resonances of
a
,
x
-diene in macromonomer almost
), whereas the new
added to the reaction vessel after 48 h of polymerization [28]. As
a result, further growth in the molecular weight of the graft
disappeared at 5.78 (H ) and 4.95 ppm (H
r
s
Fig. 6. UV–vis spectral changes as a function of time for a main–chain azo-polymer (macroinitiator) solution in THF with a concentration of 10 mg/mL. (a) Absorbance
changes with UV light irradiation (365 nm) and (b) upon irradiating the solution at the photostationary state with visible light.
Table 1
Molecular characteristics of macroinitiator, macromonomer, and graft copolymers as-obtained through ‘‘Route 1’’ and ‘‘Route 2’’ synthetic process.
Yield (%)a
M
n,MALLS
b
M
/M b
M
n,NMR
DPn,backbone
DPn,graft
Polymer
t (h)
Synthetic route
w
n
Macromonomer
Macroinitiator
Graft copolymers
12
24
24
ATRP
ADMET
Route 1
92
91
89
87
92
91
20
5100
1.07
2.21
2.23
2.08
1.89
1.92
2.37
4900c
17,500
23,900
Nd
84,200
nd
Ndg
19
5
10
18
25
19
13
Nd
13
13
13
13
3
d
e
18,300
26,700
50,500
89,300
127,800
35,500
4
7
9
2
8
2
6
4
e
f
Route 2
36,800
a
b
c
d
e
f
Obtained gravimetrically from the dried polymer.
Number average molecular weight (M
n
) and molecular weight distribution (M
w
/M
0
n
) were obtained from multiangle laser light scattering analysis.
=3Þ=ðS =2Þ, M(M1) = 310 and M(inimer) = 946.
/4)/(S /2), M(inimer) = 946, and M(ethylene) = 28.
Obtained by 1H NMR spectroscopy: Mn,NMR = m ꢂ M(M1) + M(inimer), where m ¼ ðSh
r
1
Obtained by H NMR spectroscopy: Mn,NMR = n ꢂ M(inimer) ꢁ (n ꢁ 1) ꢂ M(ethylene), where n = (S
g
r
Obtained by 1H NMR spectroscopy: Mn,NMR = n ꢂ M(Macromonomer) ꢁ (n ꢁ 1) ꢂ M(ethylene), where n = (SACH@CHA/2)/(S
/2), M(Macromonomer) = 4900, and M(ethylene) = 28.
r
Obtained by 1H NMR spectroscopy: Mn,NMR = m ꢂ M(M1) ꢂ n + M(Macroinitiator), where m ¼ ðSh
=3Þ=ðSACH@CHA=2Þ, M(M1) = 310, n = 19, and M(Macroinitiator) = 17,500.
0
g
Not determined.

L. Ding et al. / Reactive & Functional Polymers 91–92 (2015) 85–92
91
Fig. 7. (a) The photoisomerization of the main–side chain graft azo-copolymer under UV light irradiation (365 nm) and (b) the absorbance of the polymer at 365 nm as a
function of irradiation time.
copolymer was detected by MALLS for 72 h and 96 h, as shown in
Fig. 4, which was in excellent agreement with the calculated value
that the initiation did not occur quantitatively, i.e., ATRP had lost
some efficiency in this system.
of Mn,NMR
.
3.4. Photoresponsive behavior in solution
3.3. Macroinitiator synthesis and ATRP
Polymers containing azo-type chromophores have cis and trans
According to ‘‘Route 2’’, the macroinitiator route was also used
isomers. Generally, the cis form of the azo molecule is thermody-
namically less stable than the trans form. Therefore, when there
is a lack of visible light, the cis form, which is obtained by photoi-
somerization, will change into the trans form. We first investigated
the photoresponsive properties of the representative main-chain
azo-polymer in solution. As illustrated by the UV–vis spectra in
Fig. 6, the macroinitiator THF solution was exposed to varying con-
ditions of UV or visible light irradiation, which underwent the trans
to cis photoisomerization (Fig. 6a) and the cis to trans
back-isomerization process (Fig. 6b). Then, we characterized the
to synthesize graft copolymers using a multifunctional precursor in
which branch growth occurred from multiple initiation sites [24].
The development of the macroinitiator offered several advantages
over the macromonomer method. Therefore, ADMET polymeriza-
tion of inimer 4 was performed first, yielding a higher molecular
weight polymer. Then, ADMET polycondensation resulted in
macroinitiator for ATRP. However, well-defined graft copolymer
structures are typically impossible to synthesize by this technique
because it is difficult to initiate every functional site on the
macroinitiator. The conversion of the azo-containing
a
,
x
-diene to
the internal olefin was closely monitored by H NMR spectroscopy
Fig. 3b). As previously noted with the macromonomer, ADMET
UV isomerization efficiency of the azo-polymer by calculating the
1
trans–cis
ratio of A
0
/A , which denoted the absorbance at the photosta-
(
tionary state of the UV-induced trans to cis isomerization process
[29]. A is the absorbance maximum of the sample after 24 h in
polymerization occurred. The successful polymerization was indi-
cated by reductions in the intensities of the peaks at 5.78 and
0
the dark. For the main-chain azo-polymer, the photoisomerization
efficiency was 0.88, which indicated that most of the azo groups
underwent the isomerization process.
In this case, there are two types of azo groups in the main and
side chains of the photosensitive graft copolymer; thus, the iso-
merization process of the chromophore will be very complicated
[30,31]. UV–vis spectroscopy was used to characterize the compli-
4
.95 ppm and an increase in the peak at 5.43ꢁ5.36 ppm.
Analogously, the number-average molecular weight of the
macroinitiator could be determined using the end group analysis
process (Mn,NMR = n ꢂ M(4) ꢁ (n ꢁ 1) ꢂ M(ethylene), n = (S
The MALLS trace in Fig. 5 shows a relatively moderate molecular
weight distribution (M /M = 2.21) and high molecular weight
n,MALLS = 17,500), which was in agreement with Mn,NMR (Table 1).
The obtained azo-functionalized ADMET polymer bearing many
g r
/4)/(S /2)).
w
n
(M
cated photoisomerization process of the azo moieties. It showed a
⁄
strong absorption at 364 nm, which is attributed to the
p ? p
initiating sites could be employed as a macroinitiator in subse-
quent ATRP of M1 to synthesize graft copolymers with azo side
chains. A homogeneous reaction mixture was retained throughout
the entire reaction process, even when the viscosity of the mixture
increased as the polymerization progressed. The combination of
ADMET polymerization and ATRP following the route depicted in
transition in the trans isomer, and a wide absorption peak at
⁄
456 nm, which is attributed to the n ?
p
transition in the cis iso-
mer. We also observed a shoulder peak at 495 nm, which might be
due to the absorbance of cis–trans, trans–cis or cis–cis isomers.
With increasing irradiation time, the absorbance of these two
peaks increased at similar rates (Fig. 7a). This result indicated that
cis isomers increasingly appeared. However, when the main or side
azo groups isomerized, the quantity of cis and trans isomers should
be equal in the cis–trans and trans–cis isomers. The shoulder peak
at 495 nm must be due to the absorption of cis–trans and/or trans–
cis isomers but not cis–cis isomers. If this peak originated from cis–
cis isomers, the absorbance at 364 nm should disappear because
there are no single trans azo groups remaining. After 60 s of irradi-
ation, the shoulder peak disappeared and merged into a broad peak
between 450 and 525 nm, indicating complete photochemical
‘
(
‘Route 2’’ yielded a graft copolymer, as shown by MALLSꢁGPC
Fig. 5), where a clear-cut shift in the copolymer trace toward
lower retention volumes (with respect to the macroinitiator) was
observed while the molecular weight distribution became slightly
more broad (Table 1). The growth of M1 graft chains was also indi-
1
cated by H NMR, and the molecular weight agreed with the MALLS
data. However, these data correlated with a M1 chain of approxi-
mately 3 repeat units per initiation site. These results did not agree
with the calculated values of 15 repeat units of M1 and indicated

9
2
L. Ding et al. / Reactive & Functional Polymers 91–92 (2015) 85–92
conversion from the trans–trans isomer configuration to the cis–cis
isomer configuration and the existence of a photostationary state.
Furthermore, Fig. 7b shows that the photoisomerization rates of
the azo functional macroinitiator (main chain) and graft copolymer
[6] C. Ohm, M. Brehmer, R. Zentel, Adv. Mater. 22 (2010) 3366–3387.
[
7] (a) Y.N. He, X.G. Wang, Q.X. Zhou, Polymer 43 (2002) 7325–7333;
b) P.C. Che, Y.N. He, X.G. Wang, Macromolecules 38 (2005) 8657–8663.
(
[
8] (a) X.Q. Xue, J. Zhu, Z.B. Zhang, N.C. Zhou, X.L. Zhu, React. Funct. Polym. 70
(2010) 456–462;
(
b) Y.N. Zhang, X. Zhu, N.C. Zhou, X.R. Chen, W. Zhang, Y.G. Yang, X.L. Zhu,
(
main–side chains) showed the same trend as the photoisomeriza-
Chem. Asian J. 7 (2012) 2217–2221.
tion percentages. The macroinitiator gave a higher isomerization
rate, while the graft copolymer gave a lower one. This result sug-
gested that one type of azo group had a faster photoisomerization
rate than the two types under UV irradiation, which agreed with
the possible photoisomerization mechanism of main–side-on azo
chromophores.
[
9] T. Yoshino, M. Kondo, J. Mamiya, M. Kinoshita, Y. Yu, T. Ikeda, Adv. Mater. 22
(2010) 1361–1363.
[
[
10] X.Y. Jiang, X.B. Chen, X.G. Yue, J.J. Zhang, S.W. Guan, H.B. Zhang, W.Y. Zhang,
Q.D. Chen, React. Funct. Polym. 70 (2010) 616–621.
11] B. Yameen, A. Farrukh, Chem. Asian J. 8 (2013) 1736–1753.
[12] S.S. Sheiko, M. Moller, Chem. Rev. 101 (2001) 4099–4124.
[
13] S.S. Sheiko, B.S. Sumerlin, K. Matyjaszewski, Prog. Polym. Sci. 33 (2008) 759–
85.
14] H. Jiang, F.J. Xu, Chem. Soc. Rev. 42 (2013) 3394–3426.
7
[
4
. Conclusion
[15] W.V. Camp, V. Germonpré, L. Mespouille, P. Dubois, E.J. Goethals, F.E.D. Prez,
React. Funct. Polym. 67 (2007) 1168–1180.
[
16] (a) C. Cheng, E. Khoshdel, K.L. Wooley, Macromolecules 40 (2007) 2289–2292;
b) M.R. Xie, J.Y. Dang, H.J. Han, W.Z. Wang, J.W. Liu, X.H. He, Y.Q. Zhang,
Well-defined graft copolymers containing azo units in all of the
(
main and side chains were successfully synthesized via the combi-
nation of living ATRP and ADMET chemistry. Compared with typi-
cal macroinitiator syntheses of graft copolymers (Route 2), the
macromonomer synthetic approach (Route 1) allowed for precise
control of the architecture. Upon irradiation with UV or visible
light, the photoisomerization process of the azo chromophores
were monitored by the UV–vis spectra. We found that the isomer-
ization of the azo groups in the main and side chains occurred
mainly on one type of azo group, and the main–side-on structure
resulted in the two types of azo groups having similar isomeriza-
tion probabilities. These results suggested that the azo chro-
mophore percentage was decisive in the photoisomerization rate
of the chromophores and the main–side-on structure of the azo
chromophores favored the photoisomerization.
Macromolecules 41 (2008) 9004–9010.
17] H.C. Shi, D. Shi, L.G. Yin, S.F. Luan, J. Zhao, J.H. Yin, React. Funct. Polym. 70
[
[
[
[
(
2010) 449–455.
18] (a) P. Seven, M. Co sß kun, K. Demirelli, React. Funct. Polym. 68 (2008) 922–930;
(b) M. Coskun, P. Seven, React. Funct. Polym. 71 (2011) 395–401.
19] L. Ding, J. Qiu, J. Wei, Z.S. Zhu, Macromol. Rapid Commun. 35 (2014) 1509–
1515.
20] K.L. Opper, K.B. Wagener, J. Polym. Sci., Part A: Polym. Chem. 49 (2011) 821–
831.
21] P. Atallah, K.B. Wagener, M.D. Schulz, Macromolecules 46 (2013) 4735–4741.
22] M.D. Schulz, R.R. Ford, K.B. Wagener, Polym. Chem. 4 (2013) 3656–3658.
23] H. Mutlu, L. Montero de Espinosa, M.A.R. Meier, Chem. Soc. Rev. 40 (2011)
1404–1445.
[
[
[
[24] (a) P.M. O’Donnell, K.B. Wagener, J. Polym. Sci., Part A: Polym. Chem. 41 (2003)
816–2827;
b) D. Markova, K.L. Opper, M. Wagner, M. Klapper, K.B. Wagener, K. Müllen,
2
(
Polym. Chem. 4 (2013) 1351–1363.
[25] (a) L. Ding, L.Y. Zhang, H.J. Han, W. Huang, C.M. Song, M.R. Xie, Y.Q. Zhang,
Macromolecules 42 (2009) 5036–5042;
(
3
(
b) L. Ding, G.D. Yang, M.R. Xie, D.Y. Gao, J.H. Yu, Y.Q. Zhang, Polymer 53 (2012)
33–341;
c) L. Ding, M.Y. Xu, J.J. Wang, Y. Liao, J. Qiu, Polymer 55 (2014) 1681–1687.
Acknowledgments
[
[
26] (a) K. Matyjaszewski, M. Teodorescu, P.J. Miller, M.L. Peterson, J. Polym. Sci.,
Part A: Polym. Chem. 38 (2000) 2440–2448;
The authors thank the National Natural Science Foundation of
China (No. 21304079) and the Initial Scientific Research
Foundation of Yancheng Institute of Technology (KJC2014002) for
financial support of this research.
(
b) H. Shinoda, P.J. Miller, K. Matyjaszewski, Macromolecules 34 (2001) 3186–
3194;
c) H. Shinoda, K. Matyjaszewski, Macromolecules 34 (2001) 6243–6248.
27] J.E. Schwendeman, A.C. Church, K.B. Wagener, Adv. Synth. Catal. 344 (2002)
97–613.
[28] I.A. Gorodetskaya, T.-L. Choi, R.H. Grubbs, J. Am. Chem. Soc. 129 (2007) 12672–
2673.
(
5
References
1
[
[
29] X.Q. Shen, H.W. Liu, Y.S. Li, S.Y. Liu, Macromolecules 41 (2008) 2421–2425.
30] M. Jin, R. Lu, Q.X. Yang, C.Y. Bao, R. Sheng, T.H. Xu, Y.Y. Zhao, J. Polym. Sci., Part
A: Polym. Chem. 45 (2007) 3460–3472.
31] J.J. Zhang, H.B. Zhang, X.B. Chen, J.H. Pang, Y.X. Zhang, Y.P. Wang, Q.D. Chen,
S.H. Pei, W.X. Peng, Z.H. Jiang, React. Funct. Polym. 71 (2011) 553–560.
[
[
1] A. Natansohn, P. Rochon, Chem. Rev. 102 (2002) 4139–4176.
2] (a) A. Shishido, Polym. J. 42 (2010) 525–533;
(
b) A. Priimagi, C.J. Barrett, A. Shishido, J. Mater. Chem. C 2 (2014) 7155–7162.
[
[
[
[
3] T. Seki, S. Nagano, M. Hara, Polymer 54 (2013) 6053–6266.
4] H.F. Yu, J. Mater. Chem. C 2 (2014) 3047–3054.
5] X.Q. Xue, J. Zhu, Z.B. Zhang, N.C. Zhou, Y.F. Tu, X.L. Zhu, Macromolecules 43
(
2010) 2704–2712.