Effect of substituent groups on quinoxaline-based random copolymers on the optoelectronic and photovoltaic properties

Article abstract of DOI:10.1016/j.polymer.2016.08.076

In this study, three low band gap donor-acceptor (D-A) type quinoxaline-based random copolymers were synthesized. The effect of varying substituent groups on quinoxaline groups on optoelectronic properties and the performance of bulk-heterojunction polyme

Full text of DOI:10.1016/j.polymer.2016.08.076

Polymer 101 (2016) 208e216
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Effect of substituent groups on quinoxaline-based random copolymers
on the optoelectronic and photovoltaic properties
a
a
b
c
Sevki Can Cevher , Gonul Hizalan , Cansel Temiz , Yasemin Arslan Udum ,
a,
b,
d,
e
a, b, e, f, *
Levent Toppare
, Ali Cirpan
a
Department of Chemistry, Middle East Technical University, 06800 Ankara, Turkey
b
c
d
e
f
Department of Polymer Science and Technology, Middle East Technical University, 06800 Ankara, Turkey
Institute of Science and Technology, Department of Advanced Technologies, Gazi University, Ankara 06570, Turkey
Department of Biotechnology, Middle East Technical University, 06800 Ankara, Turkey
The Center for Solar Energy Research and Application (GUNAM), Middle East Technical University, 06800 Ankara, Turkey
Department of Micro and Nanotechnology, Middle East Technical University, 06800 Ankara, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 June 2016
Received in revised form
In this study, three low band gap donor-acceptor (D-A) type quinoxaline-based random copolymers were
synthesized. The effect of varying substituent groups on quinoxaline groups on optoelectronic properties
and the performance of bulk-heterojunction polymer solar cells was investigated. Electrochemical and
optical studies indicate that these copolymers are promising materials for both electrochromic and
polymer solar cells applications. Two random polymers were found as excellent candidates for NIR
electrochromic devices owing to their 60% and 94% optical contrast values respectively in NIR region with
less than 1 s switching times. Polymer solar cells were constructed using the copolymers as the donor
components together with PC71BM as the acceptor in the active layer. Among all, the highest PCE was
reported as 2.13%.
16 August 2016
Accepted 23 August 2016
Available online 24 August 2016
Keywords:
Quinoxaline
Benzotriazole
Polymer solar cell
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
transport separated free charges into corresponding domains
resulting in enhanced efficiency in bulk heterojunction PSCs
Polymer solar cells (PSCs) based on semiconducting polymers
have drawn great attention both in academia and industry due to
their low cost, flexibility, light weight, ease of fabrication and fine
tuning of polymer structure [1e5]. However, low power conversion
efficiency (PCE) of PSCs has been the greatest obstacle [6]. One
possible way to increase efficiency is to introduce bulk-
heterojunction as photoactive layer where donor conjugated
polymers and acceptor fullerene derivatives are blended [1e7]. In
PSCs, absorbed incident photons generate tightly bounded
electron-hole pairs, namely excitons, which then dissociate into
free charges only at the donor-acceptor (D-A) interface [8e10].
Therefore, bulk heterojunction takes the leading part in effective
charge separation owing to bicontinuous interpenetrating network
with large donor-acceptor interfacial area. In this manner,
bicontinuous-interpenetrating area creates two channels to
[3,9,11,12]. PCEs have significantly improved up to 11% with the
advances in device fabrication and versatility of organic chemistry
[13,14].
Another way to increase the efficiency of PSCs is to develop
novel low band gap polymers with D-A approach [15]. Combination
of D-A units has become an attractive methodology to tune optical
and electrochemical properties of desired polymers and to extend
absorption spectra to longer wavelengths [5,7,12,16]. Among a wide
range of acceptor units, quinoxaline unit is shown as conspicuous
moiety to attain low band gap conjugated polymers owing to two
electron-withdrawing imine nitrogens in quinoxaline unit. They
can be easily structurally diversified via introducing substituent
groups on the two and three positions [4,6,16,17]. E. Wand and
coworkers reported that 5,8-dithien-2-yl-2,3-diphenylquinoxaline
based polymers revealed up to 6% PCE [18]. Later on Jen group re-
ported a quinoxaline derivative containing fused-phenyl rings and
6
.24% PCE was achieved [19]. It was proven that fused-phenyl rings
reduce the energetic disorder of the polymer backbone after facil-
itating polymer coplanarity and interchain * interactions which
*
Corresponding author. Department of Chemistry, Middle East Technical Uni-
versity, 06800 Ankara, Turkey.
p-p
E-mail address: acirpan@metu.edu.tr (A. Cirpan).
http://dx.doi.org/10.1016/j.polymer.2016.08.076
0
032-3861/© 2016 Elsevier Ltd. All rights reserved.

S.C. Cevher et al. / Polymer 101 (2016) 208e216
209
ꢀ
ꢀ
enable high FF and PCE in PCSs [6,20].
atmosphere with a 10 C/min heating rate up to 300 C. For cyclic
voltammetry studies polymers were dissolved in CHCl (5 mg/mL)
Benzotriazole (BTz) containing polymers revealed superior
properties in electrochromic and photovoltaic applications. Most of
the BTz derivatives showed multichromic character with highly
transmissive states which are very essential for electrochromic
devices [21]. Due to the diamine unit, BTz was used as the
moderately electron-deficient unit in D-A type conjugated poly-
mers [22]. Low electron accepting property of BTz leads to high
LUMO energy level and high band gap consequently [21]. BTz is not
accepted as a strong acceptor as benzothidiazole for polymer solar
cell structures [23]. However, it can be easily modified with alkyl
groups at one of the nitrogen atoms on the triazole unit to achieve
high solubility [24]. Hence, in this study different substituents on
quinoxaline unit were selected to combine with BTz to improve
electrochemical and photovoltaic cell properties of conjugated
polymers in terms of optical contrast, switching times, process-
ability, low band gap and high power conversion efficiency. Two
absorption peaks in the visible region located at around 400 nm
and 600 nm result in green color. In literature, copolymers of qui-
noxaline and thiophene derivatives showed such property [25e28].
Moreover, Yongfang Li and coworkers reported quinoxaline and
thiophene based alternating copolymer (PT-QX(0F)) which has dual
absorption peaks at red and blue regions [29]. The aim of this study
was to obtain polymers which have broad absorption peak covering
red, blue and green regions. To compensate the lack of absorption
maxima at green region, benzotriazole and thiophene based poly-
mers can be incorporated into polymer backbone [30]. Optical
properties of resulting polymers can be tuned via rational design
strategies to harvest more sunlight. In this study, regarding these
information, random copolymers of quinoxaline, thiophene and
benzotriazole moieties were synthesized.
3
and spray coated onto an ITO coated glass substrate. Cyclic vol-
tammetry studies were carried out in a solution of 0.1 M of tetra-
butylammonium hexafluorophosphate in anhydrous acetonitrile
(ACN) solution at a scan rate of 100 mV/s. A Gamry 600 potentiostat
was used with a three-electrode cell consisting of an ITO-coated
glass substrate as the working electrode, Pt wire as the counter
electrode, and Ag wire as the reference electrode.
2.3. Fabrication and characterization of polymer solar cell
The BHJ polymer solar cell device was fabricated with the
structure of ITO/PEDOT:PSS/Polymer:PC71BM/LiF/Al. ITO coated
glass substrates were cleaned with toluene, detergent, water and
isopropyl alcohol in ultrasonic bath for 15 min and dried with N
gun. After the plasma cleaning (Harrick Plasma Cleaner),
PEDOT:PSS was filtered through 0.45 m PVDF syringe filter.
2
m
PEDOT:PSS was coated onto the ITO surface at 4000 rpm for 1 min
by spin coating. To evaporate the water substrates were dried at
ꢀ
1
35 C for 15 min polymer:PC71BM mixtures with different ratios
were prepared and filtered with 0.2
mixtures were coated on PEDOT:PSS layer in the glove-box system
filled with nitrogen (O concentration <0.1 ppm and H O concen-
mm PTFE syringe filter. Filtered
2
2
tration <0.1 ppm). LiF (0.6 nm) and Al layer (100 nm) were evap-
orated through a shadow mask in sequence under a vacuum of
ꢁ6
2
10
mbar. The active area was determined as 6 mm . Current
density-voltage (J-V) characteristics were measured using a Keith-
ley 2400 source meter under AM 1.5G irradiation.
2
2
.4. Synthesis
2
. Experimental part
.4.1. Synthesis of 9-(bromomethyl)nonadecane (1)
2
PPh
-Octyl-1-dodecanol (10 g, 33.5 mmol) and triphenylphosphine
2
.1. Materials
(
3
) (9.22 g, 35.15 mmol) were dissolved in 75 mL dichloro-
ꢀ
methane (DCM) and reaction mixture was cooled down to 0
Scheme 1). Bromine (1.8 mL, 35 mmol) was added to reaction
mixture and mixture was stirred for 30 min at 0 C. Then, reaction
was warmed to room temperature and stirred for additional
0 min. NaHSO
compensate for excess Br
solution turns into faint yellow. Organic phase was washed with
DCM, water and brine. After the extraction, organic phase was dried
4
over MgSO and evaporated to obtain a white solid. Purification of
crude product was performed with column chromatography on
C
All chemicals used in the synthesis of monomers and polymers
(
were purchased from Sigma-Aldrich Chemical Co. Ltd. PC71BM was
purchased from Solenne. 9-(Bromomethyl)nonadecane (1) [21],
4
ꢀ
,7-dibromo-1H-benzo[d][1,2,3]triazole (4) [31], 4,7-dibromo-2-
3
3
solution was added to the reaction mixture to
. Via addition of NaHSO the dark yellow
(
2-octyldodecyl)-2H- benzo[d][1,2,3]triazole (5) [21], 2,5-
2
3
bis(trimethylstannyl)thiophene
di(thiophen-2-yl)quinoxaline
(6)
(7)
[32],
[33]
5,8-dibromo-2,3-
5,8-dibromo-2,3-
diphenylquinoxaline (8) [33], 10,13-dibromodibenzo[a,c]phena-
zine (9) [33] with some modifications were synthesized according
to literature reports. Tetrahydrofuran (THF) and toluene were dried
over sodium and benzophenone and freshly used in the reactions.
Other solvents were used as received without any further drying
operation. Reactions were performed under argon atmosphere
unless otherwise mentioned.
silica gel by using hexane and 9-(bromomethyl)nonadecane was
obtained as a colorless oil (12 g, 94%) H NMR (400 MHz, CDCl
1
3
),
d
(ppm): 3.37 (d, J ¼ 4.7 Hz, 2H), 1.52 (m, 1.46e1.57, 1H), 1.20 (m,
13
1
d
.15e1.32, 32 H), 0.81 (t, J ¼ 6.58 Hz, 6H). C NMR (100 MHz, CDCl
3
),
(ppm): 39.6, 39.5, 32.59, 31.9, 29.8, 29.7, 29.6, 29.5, 29.4, 29.3,
6.6, 22.7, 14.1.
2
2.2. Measurements
Chemical structures of the monomers and polymers were
2.4.2. Synthesis of 4,7-dibromobenzo[c][1,2,5]thiadiazole (2)
1
13
proven by H and C NMR spectra in CDCl
3
. They were recorded on
Benzo-1,2,5-thiadiazole (5.0 g, 36.7 mmol) was dissolved in
a Bruker Spectrospin Avance DPX-400 Spectrometer with respect to
trimethylsilane (TMS) as the internal reference. Average molecular
weights of the polymers were determined by gel permeation
chromatography (GPC) with respect to polystyrene standards in
40 mL HBr (47%) in room temperature. Later, a solution of Br
(17.6 g, 110 mmol) in 30 mL HBr was added drop wise to reaction
2
mixture. After the addition of Br
2
, the reaction mixture was heated
ꢀ
to 135 C and refluxed for overnight. Then the mixture was cooled
chloroform (CHCl
3
). Thermal analysis of the polymers were studied
down to room temperature and NaHSO
3
solution was added to get
by thermogravimetry analysis (TGA) and differential scanning
calorimetry (DSC). Perkin Elmer Pyris 1 TGA was used under ni-
rid of excess Br . The mixture was filtered and orange solid was
2
washed with cold diethyl ether and water several times to obtain
4,7-dibromobenzo[c][1,2,5]thiadiazole (9.53 g, yield 88%) as a yel-
ꢀ
ꢀ
trogen (N
In order to clarify melting point (T
ature (T ), Perkin Elmer DSC Diamond was used under N
2
) atmosphere with a 10 C/min heating rate up to 950 C.
1
13
m
) and glass transition temper-
low solid. H NMR (400 MHz, CDCl
3
) d (ppm): 7.66 (s, 2H). C NMR
g
2
(100 MHz, CDCl (ppm): 152.9, 132.3, 113.9.
3
) d

210
S.C. Cevher et al. / Polymer 101 (2016) 208e216
Scheme 1. Schematic representation of synthesis of monomers and random copolymers.
1
2
.4.3. Synthesis of 3,6-dibromobenzene-1,2-diamine (3)
[1,2,3]triazole as a pink powder (80% yield). H NMR (400 MHz,
CDCl
13
4
,7-Dibromobenzothiadiazole (5.0 g, 17 mmol) was dissolved in
3 3
) d (ppm): 7.78 (d, 2H). C NMR (100 MHz, CDCl ) d (ppm):
5
0 mL ethanol (EtOH) in 1 L-round flask. The reaction mixture was
153.8, 137.8, 112.2.
ꢀ
cooled down to 0 C and NaBH
4
(1.7 g, 0.51 mol) was added to re-
action mixture in portions. The reaction was stirred in the ice bath
until gas evolution was stopped and then stirred for 20 h in room
temperature. After the evaporation of ethanol, extraction was per-
formed with water and diethyl ether. Extraction was carried out
once again with brine to obtain faint yellow organic phase. Later,
2
[
.4.5. Synthesis of 4,7-dibromo-2-(2-octyldodecyl)-2H-benzo[d]
1,2,3]triazole (5)
,7-Dibromo-2H-benzo[d][1,2,3]triazole (3.0 g, 7.22 mmol) was
4
dissolved in 10 mL dimethylformamide (DMF) under argon atmo-
sphere and reaction medium was cooled down to 0 C. Then, NaH
ꢀ
the organic phase was dried over MgSO
reduced pressure and a faint yellow solid, 3,6-dibromobenzene-
,2-diamine(3.5 g, yield 80%) was obtained. The product is time-
4
, evaporated under
ꢀ
(312 mg, 8.75 mmol) was added to reaction mixture at 0 C. After
ꢀ
completion of addition, the reaction was heated to 60 C. 9-(bro-
momethyl)nonadecane (4.68 g, 12.96 mmol) was added to reaction
medium. The reaction was stirred at 65e70 C overnight. The re-
1
1
sensitive, in couple of days, it may disintegrate. H NMR
ꢀ
1
3
(
(
400 MHz, CDCl
100 MHz, CDCl
3
)
)
d
d
(ppm): 6.78 (s, 2H), 3.82 (s, 4H). C NMR
action mixture was extracted with chloroform (CHCl
The organic phase was extracted with brine and chloroform once
again. The organic residue was dried over MgSO and evaporated.
Column chromatography on silica gel with hexane and CHCl (1:3)
yielded 4,7-dibromo-2-(2-octyldodecyl)-2H-benzo[d][1,2,3]tri-
azole as a faint yellow oil (2.3 g, yield 40%). H NMR (400 MHz,
CDCl ),
(ppm): 7.32 (s, 2H), 4.59 (d, J ¼ 7.3 Hz, 2H), 2.25 (m,
2.19e2.30, 1H), 1.13 (m, 1.08e1.30, 32H), 0.78 (m, 0.75e0.80, 6H).
3
) and water.
3
(ppm): 133.5, 123.2, 109.5.
4
2.4.4. Synthesis of 4,7-dibromo-2H-benzo[d][1,2,3]triazole (4)
3
To a solution of 3,6-dibromobenzene-1,2-diamine (0.8 g,
1
3
3
mmol) in 12 mL acetic acid (AcOH), a solution of NaNO
2
(0.3 g,
.3 mmol) in 6 mL H O was added drop wise at room temperature.
2
3
d
After 30 min stirring, the powder was filtered and washed with
distilled water several times to afford 4,7-dibromo-2H-benzo[d]
13
3
C NMR (100 MHz, CDCl ), d (ppm): 141.4, 127.2, 107.8, 58.9, 36.8,

S.C. Cevher et al. / Polymer 101 (2016) 208e216
211
2
11.9.
9.7, 29.6, 29.4, 29.0, 27.6, 27.4, 27.3, 27.2, 27.1, 27.0, 23.8, 20.5, 20.4,
0.02 g) and co-catalyst tris(o-tolyl)phosphine (11.0 mol %, 0.015 g)
were added into solution mixture and refluxed at 70 C for 36 h
ꢀ
under argon atmosphere. 2-Bromothiophene (0.216 g, 1.327 mmol)
and catalytic amount of extra catalyst were added and the solution
mixture was stirred for 6 h. Later on, trimethyl(thiophen-2-yl)
stannane (0.99 g, 2.654 mmol) and catalytic amount of extra
catalyst were added and stirred for 6 h under same conditions. The
reaction was cooled to room temperature and precipitated in cold
methanol. Filtered mixture was purified by Soxhlet apparatus. Pu-
rification was made firstly with acetone and then with hexane to
remove oligomers, small molecules and excess catalyst and co-
catalyst. The polymer CoP1 was collected by chloroform and then
the solvent was evaporated under vacuum. The solid residue was
precipitated in cold methanol. Pure polymer was collected by
vacuum filtration. (GPC: Mn: 14.9 kDa, Mw: 18.9 kDa, PDI: 1.27).
2
.4.6. Synthesis of 2,5-bis(trimethylstannyl)thiophene (6)
2
,5-Dibromothiophene (1.5 g, 6.2 mmol) was dissolved into
ꢀ
4
0 mL dry THF at ꢁ78 C under argon atmosphere. 2.5 M n-butyl-
lithium (n-BuLi) solution (5.95 mL, 14.88 mmol) was added drop
wise into the solution mixture and stirred for 1 h under the same
conditions. Later, 1 M trimethyltin chloride (SnMe
3
Cl) (15.5 mL,
ꢀ
1
5.5 mmol) was added and stirred for 5 min at ꢁ78 C. Solution
mixture was stirred overnight at room temperature under argon
atmosphere. After the evaporation of the solvent, the resulting
mixture was extracted with diethyl ether and water several times.
4
The organic phase was dried with MgSO and evaporated under
vacuum. The yellowish solid was washed with cold methanol a few
times to obtain white crystal solid. (1 g; yield 39.2%) H NMR
1
13
(
CDCl
3
, 400 MHz):
d
(ppm) 7.46 (s, 2H), 0.34e0.52 (m, 18H).
(ppm): 136.01.
C
NMR (100 MHz, CDCl
3
) d
2.4.11. Synthesis of CoP2
The same procedure used for the synthesis of CoP1 was fol-
2
.4.7. Synthesis of 5,8-dibromo-2,3-di(thiophen-2-yl)quinoxaline
lowed to synthesize CoP2. In this manner, 4,7-dibromo-2-(2-
octyldodecyl)-2H-benzo[d] [1e3]triazole (5) (0.388 g,
.696 mmol) and 5,8-dibromo-2,3-diphenylquinoxaline (8)
0.306 g, 0.696 mmol) and 2,5-bis(trimethylstannyl)thiophene (6)
0.571 g, 1.392 mmol) in 8 mL THF in room temperature under
(7)
3
,6-Dibromobenzene-1,2-diamine (3) (200 mg, 0.752 mmol)
0
(
(
and 1,2-di(thiophen-2-yl)ethane-1,2-dione (418 mg, 1.88 mmol)
were dissolved in the 25 mL ethanol at room temperature. The
solution mixture was heated for few minutes and later on a cata-
lytic amount of PTSA was added. The desired solid was precipitated
with PTSA addition. The mixture was refluxed at 100 C overnight.
The precipitate was filtered and washed with ethanol to obtain
argon atmosphere. The solution mixture was bubbled with argon
for a while in order to remove air. After bubbling, catalyst tris(di-
benzylideneacetone)dipalladium(0) (5.0 mol %, 0.032 g) and co-
catalyst tris(o-tolyl)phosphine (11.0 mol %, 0.023 g) were added
ꢀ
1
yellow solid. (88% yield) H (400 MHz, CDCl
3
)
d
(ppm): 7.77 (s, 2H),
ꢀ
into solution mixture and refluxed at 70 C for 36 h under argon
13
3
7.49 (dd, 2H), 7.41 (dd, 2H), 6.98 (s, 2H). C NMR (100 MHz, CDCl )
atmosphere. 2-Bromothiophene (0.340 g, 2.088 mmol) and a cat-
alytic amount of extra catalyst were added and the solution mixture
was stirred for 6 h. Later on, trimethyl(thiophen-2-yl)stannane
d
(ppm): 152.01, 147.01, 143.1, 131.8, 129.2, 128.5, 126.5, 123.3.
2
.4.8. Synthesis of 5,8-dibromo-2,3-diphenylquinoxaline (8)
(
1.56 g, 2.654 mmol) and a catalytic amount of extra catalyst were
3
,6-Dibromobenzene-1,2-diamine (3) (200 mg, 0.752 mmol)
added and stirred for 6 h under same conditions. The reaction was
cooled to room temperature and precipitated in cold methanol.
Filtered product was purified by Soxhlet apparatus. The polymer
CoP2 was collected by chloroform and then the solvent was
evaporated under vacuum. The solid residue was precipitated in
cold methanol. Pure polymer was collected by vacuum filtration.
and benzil (395 mg, 1.88 mmol) were dissolved in the 25 mL
ethanol at room temperature. The solution mixture was heated for
few minutes and later on catalytic amount of PTSA was added. The
desired solid was precipitated with the PTSA addition. The mixture
ꢀ
was refluxed at 100 C overnight. The precipitate was filtered and
1
washed with ethanol to obtain yellow solid (94% yield).
400 MHz, CDCl
.05 (s, 2H). 13C NMR (100 MHz, CDCl
H
(
GPC: Mn: 18.4 kDa, Mw: 22.6 kDa, PDI: 1.23).
(
7
3
)
d
(ppm): 7.84 (s, 2H), 7.56 (dd, 2H), 7.48 (dd, 2H),
(ppm): 150.9, 143.1, 137.3,
3
) d
131.8, 129.7, 123.3.
2.4.12. Synthesis of CoP3
2.4.9. Synthesis of 10,13-dibromodibenzo[a,c]phenazine (9)
The same procedure was followed to synthesize CoP3. In this
3
,6-Dibromobenzene-1,2-diamine (3) (200 mg, 0.752 mmol)
manner,
4,7-dibromo-2-(2-octyldodecyl)-2H-benzo[d][1,2,3]tri-
and phenanthrene-9,10-dione (391 mg, 1.88 mmol) were dissolved
in the 25 mL ethanol at room temperature. The solution mixture
was heated for few minutes and later on a catalytic amount of PTSA
was added. The desired solid was precipitated with the PTSA
azole (5) (0.254 g, 0.456 mmol) and 10,13-dibromodibenzo[a,c]
phenazine (9) (0.200 g, 0.456 mmol) and 2,5-bis(trimethylstannyl)
thiophene (6) (0.374 g, 0.912 mmol) were mixed in 8 mL THF at
room temperature under argon atmosphere. The solution mixture
was bubbled with argon in order to remove air. After bubbling,
catalyst tris(dibenzylideneacetone)dipalladium (0) (5.0 mol %,
0.021 g) and co-catalyst tris(o-tolyl)phosphine (11.0 mol %, 0.015 g)
ꢀ
addition. The mixture was refluxed at 100 C overnight. The pre-
cipitate was filtered and washed with ethanol to obtain yellow solid
1
(
8
88% yield). H (400 MHz, CDCl
3
)
d
(ppm): 9.41 (d, J ¼ 1.5 Hz, 2H),
13
ꢀ
.50 (dd, J ¼ 1.5, 2H), 7.96 (s, 2H), 7.75 (t, 4H). C NMR (100 MHz,
were added into solution mixture and refluxed at 70 C for 36 h
CDCl (ppm): 143.4, 131.8, 130.07, 126.7, 123.3.
3
)
d
under argon atmosphere. 2-Bromothiophene (0.223 g, 1.368 mmol)
and catalytic amount of extra catalyst were added and the solution
mixture was stirred for 6 h. Later on, trimethyl(thiophen-2-yl)
stannane (1.02 g, 2.736 mmol) and catalytic amount of extra cata-
lyst were added and stirred for 6 h under same conditions. The
reaction was cooled to room temperature and precipitated in cold
methanol. Filtered product was purified by Soxhlet apparatus. The
polymer CoP3 was collected by chloroform and then the solvent
was evaporated under vacuum. The solid residue was precipitated
in cold methanol. Pure polymer was collected by vacuum filtration.
(GPC: Mn: 4.2 kDa, Mw: 5.4 kDa, PDI: 1.3).
2.4.10. Synthesis of CoP1
CoP1 was synthesized via Stille coupling reaction by mixing 4,7-
dibromo-2-(2-octyldodecyl)-2H-benzo[d]
[1e3]triazole
(5)
(
0.247 g, 0.442 mmol) and 5,8-dibromo-2,3-di(thiophen-2-yl)qui-
noxaline (7) (0.200 g, 0.442 mmol) and 2,5-bis(trimethylstannyl)
thiophene (6) (0.362 g, 0.883 mmol) in 8 mL dry THF in room
temperature under argon atmosphere. The solution mixture was
bubbled with argon for 5 min in order to remove air. After bubbling,
catalyst tris(dibenzylideneacetone)dipalladium(0) (5.0 mol %,

212
S.C. Cevher et al. / Polymer 101 (2016) 208e216
3
. Results and discussion
Cyclic voltammograms are shown in Fig. 2 and electrochemical
properties are summarized in Table 1. All copolymers have only p-
3.1. Optical studies
doping character. Oxidation potential of CoP1 was observed at
0.96 V with a dedoping peak at 0.50 V. Oxidation of CoP2 was
Optical properties of random copolymers in o-dichlorobenzene
and in thin film are presented in Fig.1. All three copolymers showed
broad absorption in visible region. Copolymers have dual absorp-
observed at a higher potential than CoP1 due to benzene ring and
seen at 1.25 V/0.60 V. As expected oxidation potential of CoP3 was
0.31 V lower than CoP2 owing to its fused structure (0.94 V/0.90 V).
HOMO energy levels of copolymers were found as ꢁ5.19, ꢁ5.10
and ꢁ5.29 eV for CoP1, CoP2 and CoP3 respectively. All copolymers
have suitable HOMO energy levels for polymer solar cell applica-
tions. Due to absence of n-type dopable character, LUMO energy
levels of copolymers were calculated from the difference of optical
tion peaks that could be assigned to
lecular charge transfer (ICT) [17].
p-p* transition and intermo-
Maximum absorption wavelengths of CoP1, CoP2 and CoP3 in
solution were observed at 585 nm, 590 nm and 545 nm respec-
tively. Thin film absorptions for CoP1, CoP2 and CoP3 were
revealed at 550 nm, 580 nm and 540 nm correspondingly. The
maximum absorption spectra of the copolymers CoP1, CoP2 and
CoP3 in the film were red shifted by 35-10-5 nm respectively
compared with those recorded in solution. This was attributed to
decrease in conformational freedom, aggregation or solvent-
polymer interactions [34]. Among all, CoP3 showed the least red-
shift absorption (5 nm) which may result from less tendency to
aggregation and less solvent-polymer chain interaction. CoP3 is
expected to show more red shift compared to CoP1 and CoP2 due
to its fused structure that increases the planarity and conjugation
length [16]. This situation could be attributed to its low molecular
weight.
As it was intended broad absorption spectrum for copolymers
were achieved. Broad absorption in the visible region due to
combination of red and green color was observed however ab-
sorption maxima were so close to each other to identify easily. As a
result, one obstacle for low PCE of polymer solar cells has been
solved with the rational polymerization of benzotriazole, thio-
phene and quinoxaline moieties.
3.2. Electrochemical studies
6
Fig. 2. Single-scan cyclic voltammograms of copolymers films in 0.1 M TBAPF /ACN
electrolyte solution.
Cyclic voltammetry (CV) studies were performed to analyze
oxidation and reduction potentials of copolymers and to calculate
HOMO and LUMO energy levels from their oxidation and reduction
onset potentials. HOMO and LUMO energy levels were determined
according to following equations:
Table 1
Summary of electrochemical of optical properties of CoP1, CoP2 and CoP3.
Electrochemical properties
Optical properties
Thin film Solution
op
g
(eV)
Polymer
HOMO
(eV)
LUMO (eV)
E
À
Á
onset
ox
HOMO ¼ ꢁ 4:75 þ E
l
max (nm) max (nm)
l
CoP1
CoP2
CoP3
ꢁ5.19
ꢁ5.10
ꢁ5.29
ꢁ3.74
ꢁ3.50
ꢁ3.81
550
580
540
585
590
545
1.45
1.60
1.48
op
E
¼ HOMO ꢁ LUMO
g
Fig. 1. Absorption spectra of CoP1, CoP2 and CoP3 in (a) o-dichlorobenzene solution and (b) film states.

S.C. Cevher et al. / Polymer 101 (2016) 208e216
213
Fig. 3. UV-Vis-NIR spectra of (a) CoP1 potential between ꢁ0.2 V and 1.5 V (b) CoP2 potential between ꢁ0.4 V and 1.6 V (c) CoP3 between 0.0 V and 1.5 V in 0.1 M TBAPF
6
/ACN
electrolyte solution couple and the colors of corresponding polymers and their L, a and b values. (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)

214
S.C. Cevher et al. / Polymer 101 (2016) 208e216
band gap and HOMO energy levels.
calculated as 1.45 eV from the onset of lmax. CoP1 was observed as
According to HOMO levels of copolymers, CoP2 shows the
highest lying HOMO energy level. High lying HOMO energy level
provides lower Voc value which negatively affects the power con-
version efficiency. However, it ensures to obtain low band gap
polymers. Additionally, HOMO energy level is not the only criteria
for obtaining low band gap polymers. LUMO energy level is also
important for achieving a low band gap. In order to have better
charge separation, LUMO energy level of donor polymer should be
blackish blue (L: 53.492, a: 2.700, b: 4.39) in its neutral state. Upon
oxidation, the color shifted to light purple-blue (L: 51.267,
a: ꢁ3.963, b: 2.478) and light blue-green (L: 63.754, a: ꢁ4.699, b:
2.938). CoP2 revealed a lmax at 580 nm with an onset of 775 nm.
Optical band gap of CoP2 was calculated as 1.60 eV from the onset
of max. Optical band gap of CoP2 was found higher than CoP1 due
l
to more aromatic benzene compound contribution instead of less
aromatic thiophene ring. Electron donor ability of thiophene is
larger than benzene ring which increases HOMO energy level with
increasing electron donor ability. Moreover, having thiophene ring
on quinoxaline moiety indicates the prolonged conjugation length
of copolymer.
Colorimetry studies were carried out and given in Fig. 3. In CIE
(Commission Internationale de l'Eclairage) coordinates, L repre-
sents the brightness of a color, a represents the color between green
and red, whereas b represents the color between blue and yellow.
0
.3e0.5 eV higher than LUMO energy level of acceptor PC71BM
(ꢁ4.2 eV) [17]. According to LUMO energy levels of copolymers, the
most effective charge transfer is expected for CoP1 and CoP3
with ꢁ3.74 eV and ꢁ3.81 eV LUMO energy levels, respectively.
Comparing the band gaps of copolymers (Table 1), the lowest
band gap was calculated for CoP1 with 1.45 eV. Cyclization of the
fused phenyl groups on quinoxaline unit exhibits increased
planarity and conjugation length along the polymer backbone than
the separated-phenyl rings [16,17]. Therefore, CoP2 possesses
higher band gap than CoP3. Thiophene rings gives CoP1 more
prolonged conjugation length than benzene rings locating in CoP2.
Moreover, aromatic resonance stabilization of benzene is higher
than thiophene unit which makes thiophene unit to adapt quinoid
form more easily which lowers the band gap [4]. In consideration of
the foregoing; the band gap of CoP1 with 1.45 eV was found lower
than the band gap of CoP2 with 1.60 eV.
Table 2
Summary of optical contrast and switching time values of CoP1, CoP2 and CoP3.
Polymer
Optical contrast (%T)
Switching time (s)
CoP1
12% (550 nm)
14% (880 nm)
1.2
1
2
9% (1500 nm)
35% (580 nm)
9% (830 nm)
0.7
0.4
1.0
0.4
1.0
1.5
0.8
CoP2
CoP3
2
3.3. Spectoelectrochemical studies
60% (1400 nm)
28% (540 nm)
3
9
4% (860 nm)
4% (1540 nm)
CoP1 revealed a maximum absorption band (lmax) at 550 nm
with an onset of 854 nm (Fig. 3). Optical band gap of CoP1 was
6
Fig. 4. Percent transmittance change of (a) CoP1 (b) CoP2 (c) CoP3 in 0.1 M TBAPF /ACN electrolyte solution at maximum wavelengths of copolymers.

S.C. Cevher et al. / Polymer 101 (2016) 208e216
215
The neutral state of CoP2 was observed as dark purple color (L:
8.376, a: 6.001, b: ꢁ12.786). Through increasing potential, CoP2
became dark grey (L: 55.887, a: ꢁ3.718, b: 1.122) and later on sea
green (L: 60.159, a: ꢁ10.083, b: 6.475). Lastly, max was calculated at
40 nm with the absorption onset 840 nm for CoP3. CoP3 was
purple (L: 46.042, a: 13.222, b: ꢁ13.392) in its neutral state and
through increasing voltage, it became lilac grey (L: 48.884, a: 5.384,
b: ꢁ3.973) and sea green (L: 53.059, a: ꢁ8.978, b: 7.198). The lack of
transparent color can be attributed to the tails of polaron peaks in
the visible region of the spectrum.
were performed to determine the switching times of copolymers at
their maximum absorption wavelengths and to monitor the
percent transmittance changes as a function of time. During the
experiment, percent transmittance changes were recorded by UV-
Vis-NIR spectrophotometer within 5 s time intervals (Fig. 4). As
summarized in Table 2, optical contrast values were found as 12% at
550 nm, 14% at 880 nm and 29% at 1500 nm for CoP1 while
switching times were calculated as 1.2, 1 and 0.7 s respectively.
CoP2 showed 35% at 580 nm, 29% at 830 nm and 60% at 1400 nm
optical contrast values within 0.4, 1.0 and 0.4 s switching times
which are quite fast switching times. CoP3 optical contrast values
were found as 28% (545 nm), 34% (860 nm) and 94% (1540 nm)
along with 1.0, 1.5 and 0.8 s switching times. Kinetic studies indi-
cated that except CoP1, CoP2 and CoP3 might be excellent candi-
dates for near IR electrochromic device applications such as
environmental heat gain and loss control systems in buildings,
telecommunication windows or variable optical attenuators with
their 60% and 94% optical contrast values respectively in NIR region
3
l
5
3.4. Kinetic studies
It's known that human eye is sensitive up to 95% of the full
contrast. In this manner, the time needed for copolymers to switch
from their neutral states to fully oxidized states, i.e. switching time,
was defined at 95% of contrast value. Chronoamperometry studies
[35]. Among all, CoP3 achieved an outstanding 94% optical contrast
with only 0.8 s switching time.
3.5. Polymer solar cell applications
Current density-voltage (J-V) characteristics (Fig. 5) were
investigated using the device configuration of ITO/PEDOT:PSS
40 nm)/polymer:PC71BM/LiF/Al. For all copolymers, polymer
(
PC71BM ratios were optimized. Following active layer coating, LiF as
the charge collection layer and cathode [36,37], were evaporated.
Due to its asymmetrical structure PC71BM has higher absorption
coefficient than PC61BM [38]. In order to compensate the valley in
the absorption spectra of the copolymers, PC71BM was preferred.
The best performance polymer solar cells showed power conver-
sion efficiency (PCE) of 2.13%, 1.72% and 1.54% for CoP1, CoP2 and
CoP3, respectively as summarized in Table 3. The highest PCE value
was obtained with CoP1 incorporated device with a Voc of 0.53 V, Jsc
8.05 of and FF of 50%. For all copolymers, Voc values was found in the
range of 0.53e0.57 V in correlation with the HOMO energy levels of
the copolymers.
HOMO levels of the polymers are ꢁ5.19, ꢁ5.10, ꢁ5.29 eV, and
the Voc values are 0.50, 0.57 V, 0.57 V. Although Voc is related to the
energy difference between the HOMO energy level of the donor and
the LUMO energy level of the acceptor, it may differ with different
donor acceptor ratios. It can be related to the energy level of the
intermolecular charge transfer (CT) state mediated by the interac-
tion of the HOMO of the donor and the LUMO of the acceptor. In
Fig. 5. Current density-voltage characteristics of CoP1, CoP2 and CoP3 based devices.
Table 3
Summary of photovoltaic parameters.
Polymer
Polymer:
V
oc (V)
J
sc (mA/cm2)
FF (%)
PCE (%)
PC71BM ratio
CoP1
CoP2
CoP3
1:2
1:3
1:3
0.53
0.57
0.57
8.05
5.70
4.36
50
53
62
2.13
1.72
1.54
Fig. 6. AFM images of the best performance devices based on a) CoP1:PC71BM (1:2) b) CoP2:PC71BM (1:3) c) CoP3:PC71BM (1:3) (Scale bar is 400 nm).

216
S.C. Cevher et al. / Polymer 101 (2016) 208e216
excellent candidates for NIR electrochromic device applications
owing to their 60% and 94% optical contrast values respectively in
NIR region with less than 1 s switching times. Among all co-
polymers, the highest PCE was reported as 2.13% for CoP1 with high
2
J
sc value (8.05 mA/cm ), Voc of 0.53 V, FF of 50%.
References
[
1] Y. Liang, Z. Xu, J. Xia, S. Tsai, Y. Wu, G. Li, C. Ray, L. Yu, Adv. Mater. 22 (2010)
35e138.
2] Z. Zhang, B. Peng, B. Liu, C. Pan, Y. Li, Y. He, K. Zhou, Y. Zou, Polym. Chem. 1
2010) 1441e1447.
1
[
(
[
[
3] Y. Cheng, S. Yang, C. Hsu, Chem. Rev. 109 (2009) 5868e5923.
4] Z. Gao, B. Qu, H. Wu, C. Gao, H. Yang, L. Zhang, L. Xiao, Z. Chen, Q. Gong, J. Appl.
Polym. Sci. 26 (2014) 5670e5677.
[5] H. Unay, N.A. Unlu, G. Hizalan, S.O. Hacioglu, D.E. Yildiz, L. Toppare, A. Cirpan,
J. Polym. Sci. Pol. Chem. 53 (2015) 528e535.
[
[
6] H. Song, D. Kim, E. Lee, D. Moon, J. Mater. Chem. 1 (2013) 6010e6020.
7] L. Zhang, C. He, J. Chen, P. Yuan, L. Huang, C. Zhang, W. Cai, Z. Liu, Y. Cao,
Macromolecules 43 (2010) 9771e9778.
Fig. 7. IPCE curves for the CoP1, CoP2 and CoP3 based devices.
[8] J. Brabec, N.S. Sariciftci, J.C. Hummelen, Adv. Funct. Mater. 11 (2001) 15e26.
9] T.M. Clarke, J.R. Durrant, Chem. Rev. 110 (2010) 6736e6767.
10] J. Kniepert, M. Schubert, J.C. Blakesley, D. Neher, J. Phys. Chem. C 2 (2011)
00e705.
[
[
[
7
addition to this morphology of the active layer may affect the Voc
values [39].
11] K. Colladet, S. Fourier, T.J. Cleij, L. Lutsen, J. Gelan, D. Vanderzande, Macro-
molecules 40 (2007) 65e72.
Although polymers used in this work exhibit not very high ef-
ficiencies, promising results were obtained. In literature, a qui-
noxaline and thiophene bearing polymer revealed a Voc of 0.63 V,
Jsc of 6.76 mA cm and a PCE of 1.47% [29]. In this study least
efficient polymer, CoP3, gave 1.54% and the most efficient polymer,
CoP1, gave 2.13% efficiency.
[12] D. Kitazawa, N. Watanabe, S. Yamamoto, J. Tsukamoto, Appl. Phys. Lett. 95
2009) 053701.
(
[13] J. You, L. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C. Chen,
J. Gao, G. Li, Y. Yang, Nat. Commun. 4 (2013) 1446.
[14] M.A. Green, K. Emery, Y. Hishikawa, W. Warta, E.D. Dunlop, Prog. Photovolt.
ꢁ
2
Res. Appl. 23 (2015) 805e812.
[
[
15] Y. Li, Acc. Chem. Res. 45 (2012) 723e733.
16] H.J. Song, J.Y. Lee, E.J. Lee, D.K. Moon, Eur. Polym. J. 49 (2013) 3261e3270.
Morphological properties of the active layers were investigated
with Atomic Force Microscopy (AFM). According to AFM results
[17] Y. Lee, Y.M. Nam, W.H. Jo, J. Mater. Chem. 21 (2011) 8583.
[
[
[
[
18] E. Wang, L. Hou, Z. Wang, S. Hellstrom, F. Zhang, O. Inganas, M.R. Andersson,
Adv. Mater. 22 (2010) 5240.
19] Y. Zhang, J. Zou, K.-S. Chen, D.F. Zeigler, Y. Sun, Chem. Mater. 23 (2011)
2289e2291.
20] T. Yamamoto, K. Sugiyama, T. Kanbara, H. Hayashi, H. Etori, Macromol. Chem.
Phys. 199 (1998) 1807.
21] S.C. Cevher, N.A. Unlu, A.C. Ozelcaglayan, D.H. Apaydin, Y.A. Udum, L. Toppare,
A. Cirpan, J. Polym. Sci. A Polym. Chem. 51 (2013) 1933e1941.
(
Fig. 6aec) active layer root mean square roughnesses (rms) were
determined as 1.48 nm, 2.10 nm and 0.34 nm for the CoP1:PC71BM
1:2), CoP2:PC71BM (1:3) CoP3:PC71BM (1:3) blends, respectively.
(
As depicted in Fig. 6b, aggregates (about100e200 nm in diameter)
is an indication of non-optimal morphology and adversely affecting
charge separation and charge transport. This may be responsible for
obtaining low Jsc in photovoltaic studies. On the contrary, finer
phase separation was observed in Fig. 6a, which is crucial for charge
transport. Low PCE and Jsc values obtained in CoP3 based solar cells
could be related to the low molecular weight of polymer. Since
intermolecular charge-hopping is enhanced by the length of the
polymer chain, charge carrier mobilities are enhanced with
increasing molecular weight [40].
[22] J.L. Banal, J. Subbiah, H. Graham, J.K. Lee, K.P. Ghiggino, W.W.H. Wong, Polym.
Chem. 4 (2013) 1077.
[23] W.W.H. Wong, J. Subbiah, S.R. Puniredd, W. Pisula, D.J. Jones, A.B. Holmes,
Polym. Chem. 5 (2014) 1258e1263.
[24] S.C. Price, A.C. Stuart, L. Yang, H. Zhou, W. You, J. Am. Chem. Soc. 133 (2011)
625e4631.
4
[25] S. Gunes, D. Baran, G. Gunbas, A. Durmus, A. Fuchsbauer, N.S. Sariciftci,
L. Toppare, Polym. Chem. 1 (2010) 1245e1251.
[26] G. Gunbas, L. Toppare, Polym. Chem. 297 (2010) 79e86.
[
[
[
[
27] J.E. Carl eꢀ , B. Andreasen, T. Tromholt, M.V. Madsen, K. Norrman, M. Jørgensen,
F.C. Krebs, J. Mater. Chem. 22 (2012) 24417e24423.
28] J.E. Carle, M. Jørgensen, M. Manceau, M. Helgesen, O. Hagemann,
R. Søndergaard, F.C. Krebs, Sol. Energy Mat. Sol. Cells. 95 (2011) 3222e3226.
29] Z. Qiao, M. Wang, M. Zhao, Z.G. Zhang, Y. Li, X. Li, H. Wang, Polym. Chem. 6
By using incident photon to current efficiency (IPCE) measure-
ment, the ratio of number of charges collected by electrodes to the
number of incident photons was obtained for polymer solar cells
which were prepared by 1:2 CoP1:PC71BM, 1:3 CoP2:PC71BM and
(2015) 8203e8213.
30] A. Balan, D. Baran, L. Toppare, J. Mater. Chem. 20 (2010) 9861e9866.
1
:3 CoP3:PC71BM weight ratio (Fig. 7). The measurements were
[31] A. Tanimoto, T. Yamamoto, Adv. Synth. Catal. 346 (2004) 1818e1823.
[32] J. Choi, K.-H. Kim, H. Yu, C. Lee, H. Kang, I. Song, Y. Kim, J.H. Oh, B.J. Kim, Chem.
Mater. 27 (2015) 5230e5237.
done in the range of 300e1000 nm using monochromic light. The
maximum IPCE for CoP1 was found as 38.3%, 23.2% for CoP2 and
[33] T. Yamamoto, K. Sugiyama, T. Kushida, T. Inoue, T. Kanbara, J. Am. Chem. Soc.
118 (1996) 3930e3937.
10.5% for CoP3.
[
[
34] G. Hizalan, A. Balan, D. Baran, L. Toppare, J. Mater. Chem. 21 (2011) 1804.
35] K. Aydemir, S. Tarkuc, A. Durmus, G.E. Gunbas, L. Toppare, Polymer 49 (2008)
4
. Conclusions
2029e2032.
[36] C.J. Brabec, S.E. Shaheen, C. Winder, N.S. Sariciftci, P. Denk, Appl. Phys. Lett. 80
We have synthesized quinoxaline-based donor polymers with
(2002) 1288e1290.
[
[
37] H.J. Park, Y. Lee, J.W. Jo, W.H. Jo, Polym. Chem. 3 (2012) 2928e2932.
38] M.M. Wienk, J.M. Kroon, W.J.H. Verhees, J. Knol, J.C. Hummelen, P.A. van Hal,
R.A.J. Janssen, Angew. Chem. Int. Ed. 42 (2003) 3371e3375.
varying substituent groups on the quinoxaline ring, and focused on
the effect of those substituents on the performance of bulk-
heterojunction polymer solar cells. Incisive modification of
quioxaline moiety CoP2 to CoP3, in order to enhance the quinoid
character of polymer, revealed better results for the NIR applica-
tions. Kinetic studies indicated that CoP2 and CoP3 might be
[39] N. Blouin, A. Michaud, D. Gendron, S. Wakim, E. Blair, R. Neagu-Plesu,
M. Bellet e^ te, G. Durocher, Y. Tao, M. Leclerc, J. Am. Chem. Soc. 130 (2008)
732e742.
[
40] C. Liu, K. Wang, X. Hu, Y. Yang, C.-H. Hsu, W. Zhang, S. Xiao, X. Gong, Y. Cao,
ACS Appl. Mater. Interfaces 5 (2013) 12163e12167.