Synthesis and optoelectronic property manipulation of conjugated polymer photovoltaic materials based on benzo[d]-dithieno[3,2-b;2′,3′-f]azepine

Article abstract of DOI:10.1016/j.polymer.2018.06.007

To develop new conjugated polymers (CPs) based on heteroepines, a soluble azepine moiety of benzo [d]-dithieno [3,2-b; 2′,3′-f]azepine was copolymerized with different building blocks to obtain a series of CPs. A broad absorption range of 300–850 nm and t

Full text of DOI:10.1016/j.polymer.2018.06.007

Accepted Manuscript
Synthesis and optoelectronic property manipulation of conjugated polymer
photovoltaic materials based on benzo[d]-dithieno[3,2-b;2′,3′-f]azepine
Jian Wang, Pan Yin, Yue Wu, Gangjian Liu, Chaohua Cui, Ping Shen
PII:
S0032-3861(18)30493-2
10.1016/j.polymer.2018.06.007
JPOL 20644
DOI:
Reference:
To appear in: Polymer
Received Date: 25 January 2018
Revised Date: 14 April 2018
Accepted Date: 2 June 2018
Please cite this article as: Wang J, Yin P, Wu Y, Liu G, Cui C, Shen P, Synthesis and optoelectronic
property manipulation of conjugated polymer photovoltaic materials based on benzo[d]-dithieno[3,2-
b;2′,3′-f]azepine, Polymer (2018), doi: 10.1016/j.polymer.2018.06.007.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract

ACCEPTED MANUSCRIPT
Synthesis and optoelectronic property manipulation of conjugated polymer
photovoltaic materials based on benzo[d]-dithieno[3,2-b;2′,3′-f]azepine
1
1
2
1
2
1,
Jian Wang , Pan Yin , Yue Wu , Gangjian Liu , Chaohua Cui , Ping Shen *
1
Key Laboratory for Green Organic Synthesis and Application of Hunan Province, Key Laboratory
of Environmentally Friendly Chemistry and Application of Ministry of Education, College of
Chemistry, Xiangtan University, Xiangtan 411105, China
2
Laboratory of Advanced Optoelectronic Materials, College of Chemistry, Chemical Engineering
and Materials Science, Soochow University, Suzhou 215123, China
*Correspondence to: P. Shen (E-mail: shenping802002@163.com)
Abstract: To develop new conjugated polymers (CPs) based on heteroepines, a soluble azepine
moiety of benzo[d]-dithieno[3,2-b;2′,3′-f]azepine was copolymerized with different building blocks
to obtain a series of CPs. A broad absorption range of 300-850 nm and the desired optical bandgaps
(
1.54-2.42 eV) are easily accessible by varying different copolymerization units. The HOMO levels
of polymers are observed as an almost identical value of -5.21 eV and the LUMO levels range from
2.82 to -3.52 eV. Polymer solar cells (PSCs) with these polymers as donor materials were
-
fabricated to explore application potentials in organic photovoltaics. Preliminary study exhibits the
best power conversion efficiency of 1.28% for PSCs based on these polymers with a relatively high
Voc of 0.84 V. Although photovoltaic performance of these new polymers is not as high as our
expected, the first daring attempt provides some good references to expand application fields of
azepine-based polymers.
Keywords: conjugated polymers; benzodithienoazepine; polymer solar cells
1
. Introduction
Over the past two decades, conjugated polymers (CPs) have received continually increasing
attention and been widely applied in polymer optoelectronic devices, such as polymer solar cells

ACCEPTED MANUSCRIPT
(
PSCs) [1-4], polymer light-emitting diodes (PLEDs) [5-7], polymer field effect transistors (PFETs)
8-10] and photodetectors [11,12], due to their intrinsic semiconducting characteristics. For CPs
[
used in PSCs, strong and broad absorption ability, narrow bandgap, suitable molecular energy level,
high and balanced charge carrier mobility are all of crucial importance for achieving excellent
photovoltaic performance [13]. Additionally, another fundamental issue for designing
high-performance PSCs is the control of the active layer morphology at the nanometer length scale,
which also is of critical importance to the device efficiency [14-16]. The backbone of CPs is the
most important component because it dominates most of the PSC-related optoelectronic properties
mentioned above. The mainly reported conjugated backbones for PSC can be generally classified
into two kinds based on the constitution of the repeating unit: homopolymer and copolymer.
Hundreds of different backbones have been reported so far. Typical conjugated backbone is
constructed by electron-rich and electron-deficient heteroarene units independently or together. The
electron-rich benzo[1,2-b:4,5-b′]dithiophene (BDT) unit emerges as a most sparking tricyclic fused
arene and has been widely used to build high-performance donor-donor (D-D) or donor-acceptor
(
D-A) type copolymers due to its symmetric, rigid and coplanar structure [1]. In 2008, Yang and
Hou et al. reported a series of BDT-based D-D and D-A copolymers for the application of PSCs
17]. The results revealed that the optoelectronic properties, like absorbance, bandgap, molecular
[
energy level, and charge carrier mobility, can be tailored effectively by copolymerizing with
different electron-rich or electron-deficient units. Additionally, our group reported a series of
photovoltaic copolymers based on 6,12-dihydro-diindeno[1,2-b;10,20-e]pyrazine (IPY) group,
which is a multifused pentacyclic systems [18]. The bandgaps and molecular energy levels of the
resulting IPY based CPs have also been controlled successfully by using the similar strategy.
Expect for BDT and IPY, the other electron-rich multifused arenes, such as cyclopentadithiophene,
indacenodithiophene, dithieno[2,3-d:2′,3′-d]benzo[1,2-b:4,5-b′]dithiophene units, are good building
block for preparing efficiently photovoltaic polymers [19]. At present, it is very urgent to develop
new multifused heteroarene unit because they could potentially yield novel photovoltaic polymers,
and as a result that could contribute the further advance of power conversion efficiency (PCE) of
PSCs.
In the past decades, conjugated seven-membered ring systems that have a heteroatom with a lone
pair of electrons, denoted as heteroepines, have received a long-term interest due to their great

ACCEPTED MANUSCRIPT
potential in molecular actuators and other mechanically responsive applications [20,21]. Among
those heteroepines, nitrogen-containing azepine (AP, Chart 1) skeleton has attracted much attention
because it is a peculiar structural motif in many biologically active and medicinally valuable
molecules [22,23]. To improve thermal and electrochemical stability as well as drug activity,
azepine usually is modified by the addition of annulation and steric effects. For example,
benzene-annulated azepines (e.g. dibenz[b,f]azepine) and their derivatives have been variously
proven as antiallergic activity, specifically antihistaminic activity, spansmolytic, serotonin
antagonistic, fungicidal action, et al. [23]. In addition, T. M. Swager and co-workers reported the
synthesis of thiophene-annulated azepine framework and its related conducting polymers [21]. The
results demonstrated that the azepine-based polymers were thermally and electrochemically stable
and retained their redox properties in the solid state, which could be applied in actuating materials
research due to their redox stability and conductivity. Though annulated azepine-containing
materials have shown great potential applications in molecular actuators and medicinally valuable
molecules, there are a few limited examples about this kind of materials for applications in organic
optoelectronic devices. For example, Ding [24] and Suh [25] groups reported on syntheses and
photovoltaic applications of CPs containing an azepine-2,7-dione moiety. In addition, Su et al.
synthesized a dibenz[b,f]azepine-containing sentistizer for dye-sensitized solar cell application [26].
Based on the above considerations, in this paper we present the design, synthesis, and
characterization
of
a
series
of
CPs
based
on
a
soluble
and
stable
benzo[d]-dithieno[3,2-b;2′,3′-f]azepine (BTAP) unit (Chart 1), in which azepine unit is annulated
with one benzene ring and two thiophenes to improve stability and structural modification of
azepine building block. This paper is aimed at not only exploiting new azepine-based copolymers
but also extending their application fields. Herein, three electron-rich donor units of thiophene (Th),
dithiophene (DTh) and BDT and two electron-deficient acceptor building blocks of
2,1,3-benzothiadiazole (BT) and 3,6-dithiophen-2-yl-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione
(
DPP) were selected and copolymerized with annulated azepine motif (BTAP) to construct D-D
PBTAPTh, PBTAPDTh, and PBTAPBDT) or D-A (PBTAPBT and PBTAPDPP) copolymers
(
as shown in Scheme 1. The thermal, optical, electrochemical, and charge carrier properties of these
new copolymers were investigated systematically. Results indicated that the bandgap, molecular
energy levels, and charge carrier mobility of azepine-based CPs can be fine manipulated by

ACCEPTED MANUSCRIPT
selecting different copolymerizing units. Moreover, to tap the potential of optoelectronic devices,
the initial photovoltaic performance of the new polymers were also investigated by preparing the
PSC devices. Though photovoltaic performance of these new polymers is still poor so far, to the
best of our knowledge, this is the first report on design, synthesis and characterization of
benzo[d]-dithieno[3,2-b;2′,3′-f]azepine-based conjugated polymers for application in organic
electronic devices. Our daring attempt provides some good references to expand application fields
of azepine-based polymers and should also be useful for molecular structure design of other kinds
of heteroepine-containing conjugated polymers serving in various optoelectronic devices.
2
3
4
5
1
4
6
11
5
3
S
S
1
2
7
N
10
6
N
H
8
H
9
7
Azepine
AP)
Benzo[d]-dithieno[3,2-b;2′,3′-f]azepine
BTAP)
(
(
Chart 1. Molecular structures of azepine (AP) and benzo[d]-dithieno[3,2-b;2′,3′-f]azepine (BTAP)
with relevant position numbered.
C H
8 17
O
OC
8
H
17
C H17O
OC H
8
8
17
4 9
C H
S
N
N
2 5
C H
S
S
S
S
N
N
O
N
S
N
n
S
O
n
C
2
H
5
4
C H
9
PBTAPDPP
PBTAPBT
2 5
C H
8 17
C H O
OC H
8 17
8
C H17O
OC H
8
C H17O
OC H
8
17
8
17
4 9
C H
O
O
S
S
S
S
S
S
S
N
S
N
N
n
S
S
n
S
n
4 9
C H
2 5
C H
PBTAPTh
PBTAPDTh
PBTAPBDT
Scheme 1. Chemical structures of BTAP-based copolymers.
2
. Experimental section
2.1. Analytical instruments
Nuclear magnetic resonance (NMR) spectra were measured with Bruker AVANCE 400
spectrometer. Ultraviolet-visible (UV-Vis) absorption spectra were measured on PerkinElmer

ACCEPTED MANUSCRIPT
Lamada 25 spectrometer. Molecular mass was determined by flight mass spectrometry
(MALDITOF MS) using a Bruker Aupoflex-III mass spectrometer. Thermal gravimetric analysis
-1
(
TGA) was performed under nitrogen at a heating rate of 20 °C min with TGA Q50 analyzer. The
molecular weight and polydispersity index (PDI) of the polymers were determined using Waters
515 gel permeation chromatoraphy (GPC) analysis with tetrahydrofuran as the eluent and
1
polystyrene as standard. The electrochemical cyclic voltammetry (CV) was recorded on a
electrochemistry work station (CHI830B, Chenhua Shanghai) with a Pt slice electrode (coated with
a polymer film), a Pt ring, and an Ag/AgCl electrode as the working electrode, the auxiliary
-1
electrode and the reference electrode respectively, in a 0.1 mol L tetrabutylammonium
hexafluorophosphate (Bu NPF ) acetonitrile solution. Atomic force microscopy (AFM)
4
6
measurement was carried out on a Digital Instruments (Bruker, Santa Barbara, CA, USA) using a
tapping mode. Transmission electron microscopy (TEM) were performed on a JEM-2100 electron
microscope operating at an acceleration voltage of 100 kV.
2.2. PSC device preparation and characterization
The solar cell devices were prepared with a architecture of ITO/PEDOT:PSS (30 nm)/active
layer/Ca(20 nm)/Al(100 nm), where the active layer was comprised of the synthesized conjugated
polymer as a donor and PC BM or PC BM as an acceptor. The indium tin oxide (ITO) patterned
61
71
glass was cleaned with ultrasonic treatment in detergent, deionized water, acetone, ethanol, and
isopropyl alcohol sequentially, and dried in an ultraviolet–ozone chamber for 15 min. A thin layer
of PEDOT:PSS (~30 nm) was spin-coated on the pre-cleaned ITO glass at 5000 rpm and dried in an
oven at 150 °C for 15 min subsequently. The polymer donor and the acceptor (PC BM or PC BM)
6
1
71
were dissolved in 1,2-dichlorobenzene (DCB) in different weight ratios and the total concentration
of the donor/acceptor blending solution was 20 mg/mL. The solution was stirred at 100 °C for 6 h in
a glovebox prior to use. The device was transferred to a glove box, where the above prepared blend
solution was then spin-coated on PEDOT:PSS surface as the active layer. Subsequently, Ca (~20
nm) and Al (~100 nm) were successively evaporated on the top of the active layer under a vacuum
-4
2
of 5 × 10 Pa. The accurate device area was about 4 mm as defined by shadow mask. The
thicknesses of the active layer were controlled by varying the spin-coating speed and were
measured on an Ambios Technology XP-2 surface profilometer. Photovoltaic performance of
-2
devices was tested under illumination condition with an AM 1.5G (100 mW cm ), and the current

ACCEPTED MANUSCRIPT
density-voltage (J-V) characteristics were measured bya computer controlled Keithley 2602 Source
Meter. The incident light intensity was calibrated using a standard Si solar cell. The hole mobility
was measured by the space charge-limited current (SCLC) method with a hole-only device
configuration (ITO/PEDOT:PSS/polymers:PC BM/MoO /Al) for hole mobility. Hole mobilities
7
1
3
were extracted by fitting measured J-V curves using the empirical Mott–Gurney formula in single
3
2
0.5
carrier SCLC device with the equation of ln( JL /V ) ≈ 0.89(1/E ) (V /L ) + ln(9ε ε µ/8).
0
0 r
2
.3. Materials and synthesis
,2-Benzenediol, 1-bromooctane, p-toluidine, thiophene, N-bromosuccinimide (NBS),
tris(dibenzylideneacetone)dipalladium-chloroform adduct (Pd (dba) ·CHCl ), (t-Bu PH)BF , and
1
2
3
3
3
4
tris(o-tolyl)phosphine (P(o-tol) ), Pd(PPh ) , t-BuONa and n-BuLi were purchased from J&K
3
3 4
Chemical or Energy Chemical and used without further purification. Monomers M2 and M3 were
purchased from Derthon (Shenzhen China) Optoelectronic Materials Science Technology Co., Ltd.
Monomers M4-M6 and 2-(tributylstannyl)thiophene were synthesized according to the published
procedure [27,28]. The synthetic routes of BTAP-based monomer M1 and the related polymers are
outlined in Scheme 2 and Scheme 3, respectively. The synthetic procedure of other compounds are
as follows.
2.3.1. Synthesis of 1,2-dioctyloxybenzene (1)
Under Ar protection, 1,2-benzenediol (5.5 g, 50 mmol), KOH (6.16 g, 110 mmol), and DMF (100
mL) were added into a 250 mL round bottom flask. The mixture was warmed up to 110 °C, then
-bromooctane (21.2, 110 mmol) was added and the reaction mixture was continued to stir at this
1
temperature for overnight. The mixture was cooled to room temperature, quenched with water 50
mL) and extracted with CH Cl three times. The combined organic phase was washed with
2
2
saturated sodium bicarbonate solution and water, then it was dried over anhydrous MgSO . The
4
most of organic solvent was removed under reduced pressure and the residue was purified by silica
gel column chromatography eluted with petroleum ether:CH Cl (4:1, by volume) to obtain a white
2
2
1
crystal as compound 1 (11.5 g, yield 68.9%). H NMR (CDCl , 400 MHz, δ/ppm): 6.88 (s, 4H),
3
3.98 (t, 4H, J = 6.04 Hz), 1.81 (m, 4H), 1.57-1.28 (m, 20H), 0.88 (m, 6H).
2.3.2. Synthesis of 1,2-bibromo-4,5-dioctyloxybenzene (2)
Under Ar protection, to the mixture of compound 1 (9.5 g, 28.5 mmol), CH COOH (30 mL) and
3
CH Cl (100 mL) was slowly added Br (3.3 mL, 62.5 mmol) diluted with 20 mL CH Cl . After
2
2
2
2
2

ACCEPTED MANUSCRIPT
being stirring at room temperature for 4 h, the reaction mixture was quenched with water (50 mL)
and extracted with CH Cl three times. The combined organic phase was washed with sodium
2
2
thiosulfate solution and brine, dried over anhydrous MgSO . The organic layer was evaporated
4
under reduced pressure and the crude product subjected to silica gel column chromatography eluted
with petroleum ether:CH Cl (8:1, by volume) to obtain a white solid as compound 2 (10.3 g, yield
2
2
1
7
1
2
3.6%). H NMR (CDCl , 400 MHz, δ/ppm): 7.05 (s, 2H), 3.93 (t, 4H, J = 6.04 Hz), 1.80 (m, 4H),
3
.56-1.28 (m, 20H), 0.88 (m, 6H).
.3.3. Synthesis of 2,2'-(4,5-dioctyloxy-1,2-phenylene)dithiophene (3)
To a degassed toluene (100 mL) solution of compound 2 (4.92 g, 10 mmol) was added
2-(tributylstannyl)thiophene (20.5 g, 50 mmol) and Pd(PPh
3
)4 (0.231 g, 0.2 mmol). The mixture was
subjected to three freeze-pump-thaw cycles and was then stirred at 120 °C for 24 h. After cooling to
room temperature, toluene was removed by evaporation to get black solid. The solid was dissolved
into a little CH Cl and then added into methanol (150 mL) to form a black solid. The crude product
2
2
subjected to silica gel column chromatography eluted with petroleum ether:CH Cl (10:1, by
2
2
1
volume) to obtain a light green solid as compound 3 (2.17 g, yield 43.6%). H NMR (CDCl , 400
3
MHz, δ/ppm): 7.23 (d, J = 7.60 Hz, 2H), 6.99 (s, 2H), 6.94 (t, J = 3.72 Hz, 2H), 6.83 (d, J = 2.50 Hz,
2H), 4.05 (t, J = 6.40 Hz, 4H), 1.85 (m, 4H), 1.59-1.28 (m, 20H), 0.88 (d, J = 7.70 Hz, 6H); MS
(
MALDI-TOF, m/z) calcd for C H O S : 498.263; found 498.328.
3
0
42
2 2
2.3.4. Synthesis of 5,5'-(4,5-dioctyloxy-1,2-phenylene)bis(2,4-dibromothiophene) (4)
Under Ar protection, compound 3 (1.99 g, 4.0 mmol), CHCl (20 mL) and CH COOH (20 mL)
3
3
was added into a 100 mL round bottom flask. Then NBS (3.2 g, 18 mmol) was added into the
mixture in portions at the darkness. After being stirred at room temperature for 12 h, the reaction
mixture was quenched with water (50 mL) and extracted with CHCl three times. The combined
3
organic phase was washed with sodium bicarbonate solution and brine, dried over anhydrous
MgSO , and evaporated under reduced pressure. The crude product subjected to silica gel column
4
chromatography eluted with petroleum ether:CH Cl (6:1, by volume) to obtain a white solid as
2
2
1
compound 4 (2.96 g, yield 91.4%). H NMR (CDCl , 400 MHz, δ/ppm): 7.28 (s, 2H), 6.94 (s, 2H),
3
4.04 (t, J = 4.32 Hz, 4H), 1.84 (m, 4H), 1.57-1.30 (m, 20H), 0.89 (d, J = 4.36 Hz, 6H); MS
(
MALDI-TOF, m/z) calcd for C H Br O S : 813.901; found 814.016.
3
0
38
4
2 2
2.3.5. Synthesis of 2,2'-(4,5-dioctyloxy-1,2-phenylene)bis(3-dibromothiophene) (5)

ACCEPTED MANUSCRIPT
Under Ar protection, compound 4 (1.22 g, 1.5 mmol) and THF (20 mL) was added into a 100
mL round bottom flask. Then n-BuLi (1.2 mL, 3 mmol) was added into the mixture dropwise at
-78 °C. After being stirring for 0.5 h, methanol (0.5 mL) was added to quench and the reaction
mixture was warmed to room temperature for stirring another 2 h. Water (20 mL) was added into
the reaction mixture and then diethyl ether (50 mL) was added. The organic phase was washed with
brine and water, dried over anhydrous MgSO , and evaporated under reduced pressure. The crude
4
product subjected to silica gel column chromatography eluted with petroleum ether to obtain a
1
white solid as compound 5 (818.9 mg, yield 83.3%). H NMR (CDCl , 400 MHz, δ/ppm): 7.20 (d, J
3
=
5.29 Hz, 2H), 7.02 (s, 2H), 6.92 (d, J = 5.20 Hz, 2H), 4.07 (t, J = 6.52 Hz, 4H), 1.89-1.82 (m, 4H),
1
6
2
.48-1.17 (m, 16H), 0.88 (d, J = 6.60 Hz, 6H); MS (MALDI-TOF, m/z) calcd for C H Br O S :
3
0
40
2
2 2
56.082; found 656.068.
.3.6. Synthesis of 2,3-dioctyloxy-8-(p-tolyl)-8H-benzo[d]dithieno[3,2-b:2'3'-f]azepine (6)
Under Ar protection, compound 5 (4.10 g, 6.25 mmol), p-toluidine (0.668 g, 6.25 mmol),
Pd (dba) •CHCl , (312 mg, 5 mol%), (t-Bu PH)BF (187.5 mg, 12 mol%), t-BuOK (1.75 g, 15.6
2
3
3
3
4
mmol) and toluene (60 mL) was added into a 100 mL round bottom flask. After mixing together,
the mixture was stirred at 100 °C for 24 h. Then the reaction mixture was allowed to cool to room
temperature, and it was filtered through a pad of celite and water (50 mL) was added. The filtrate
was extracted with ethyl acetate three times. The combined organic phase was concentrated under
reduced pressure and purified by column chromatography eluted with petroleum ether:CH Cl (12:1,
2
2
1
by volume) to obtain a light yellow solid as compound 6 (1.75 g, yield 46.5%). H NMR (CDCl ,
3
400 MHz, δ/ppm): 7.40 (d, J = 4.8 Hz, 2H), 7.10 (d, J = 4.80 Hz, 2H), 7.09 (s, 2H), 6.94 (pseudo-d,
J = 8.40 Hz, 2H), 6.74 (pseudo-d, J = 8.40 Hz, 2H), 4.08 (t, J = 6.60 Hz, 4H), 2.24 (s, 3H), 1.87 (t, J
1
3
=
7.20 Hz, 4H), 1.49-1.30 (m, 20H), 0.93 (t, J = 5.60 Hz, 6H); C NMR (100 MHz, CDCl , δ/ppm):
3
148.85, 145.47, 141.73, 135.87, 129.39, 127.91, 127.70, 124.93, 123.93, 113.49, 112.34, 69.37,
31.88, 29.40, 29.34, 29.21, 26.06, 22.74, 20.34, 14.18. MS (MALDI-TOF, m/z) calcd for
C H NO S : 601.305; found 601.333.
37
47
2 2
2
.3.7. Synthesis of 8-(p-tolyl)-2,3-dioctyloxy-6,10-diiodo-8H-benzo[d]dithieno[3,2-b:2'3'-f]azepine
M1)
Under Ar protection, compound 6 (1.20 g, 2.0 mmol) and THF (20 mL) was added into a 100 mL
round bottom flask. Then n-BuLi (1.83 mL, 4.4 mmol) was added into the mixture dropwise at
(

ACCEPTED MANUSCRIPT
-40 °C. After addition of n-BuLi, the mixture was allowed to warm up room temperature and stirred
for 1 h. Then the reaction mixture was cooling to -40 °C again and iodine (1.11 g, 4.4 mmol) was
added. After being allowed to stir at room temperature for 12 h, the mixture was diluted with water
(
50 mL), exacted with diethyl ether and washed with saturated Na S O and brine. The organic
2 2 3
phase was dried over MgSO , evaporated under reduced pressure, and subjected to column
4
chromatography eluted with petroleum ether:CH Cl (10:1, by volume) to obtain a light yellow
2
2
1
solid as compound M1 (900.8 mg, yield 52.8%). H NMR (400 MHz, CDCl , δ/ppm): 7.19 (s, 2H),
3
6.93 (s, 2H), 6.92 (pseudo-d, J = 8.00 Hz, 2H), 6.67 (pseudo-d, J = 8.00 Hz, 2H), 4.01 (t, J = 6.00
1
3
Hz, 4H), 2.21 (s, 3H), 1.82 (t, J = 6.00 Hz, 4H), 1.47-1.29 (m, 20H), 0.89 (t, J = 6.00 Hz, 6H); C
NMR (100 MHz, CDCl , δ/ppm): 149.22, 144.64, 141.69, 141.46, 136.85, 129.53, 128.54, 123.01,
3
112.95, 112.53, 72.93, 69.32, 31.91, 29.41, 29.37, 29.15, 26.05, 22.79, 20.43, 14.26. MS
(
MALDI-TOF, m/z) calcd for C H I NO S : 853.098; found 852.961.
3
7
45 2
2 2
2.3.8. General synthetic procedure of D-A copolymers (for PBTAPDPP and PBTAPBT)
These D-A copolymers were synthesized by Suzuki polycondensation. In a 25 mL two-neck flask,
the dibromo BTAP monomer M1 (0.2 mmol, 1 equiv.), the bis(pinacolato)diboron comonomer (0.2
mmol, 1 equiv.), Pd (dba) (9 mg), P(o-tyl) (15 mg), K PO (848 mg, 4 mmol) and two drop of
2
3
3
3
4
Aliquat 336 were subjected to three cycles of evacuation/argon purging, and then dissolved with 6
mL of degassed toluene after which 2 mL degassed water was added. The reaction mixture was
stirred at 115 °C for 48 h under argon, and then cooled down to room temperature. The mixture was
precipitated in 100 mL of methanol. The dark solids were filtered and Soxhlet extracted with
methanol, acetone, hexane, and chloroform, successively. The chloroform extracts were
concentrated and then subjected to column chromatography eluted with chloroform. The target
polymer was afforded after completely drying in vacuo.
PBTAPDPP: The general Suzuki polymerization procedure was followed using 155.3 mg of
1
monomer M2 and 170 mg of monomer M1 to afford a blue solid. (140.4 mg, yield: 62.6%). H
NMR (400 MHz, CDCl , δ/ppm): 9.01 (br, 2H), 7.36 (br, 3H), 7.06 (br, 4H), 6.79 (br, 3H), 4.06 (br,
3
8
H), 2.21 (s, 3H), 1.84-1.28 (br, 42H), 0.92-0.91 (br, 18H). GPC: M : 67.2 kDa; PDI: 1.76.
n
PBTAPBT: The general Suzuki polymerization procedure was followed using 77 mg of
monomer M2 and 170 mg of monomer M1 to afford a dark brown solid. (87.4 mg, yield: 57.3%).
1
H NMR (400 MHz, CDCl , δ/ppm): 8.25-8.23 (br, 2H), 7.89-7.86 (br, 2H), 7.19-7.17 (br, 2H),
3

ACCEPTED MANUSCRIPT
7.02-6.85 (br, 4H), 4.07-4.05 (br, 4H), 2.21 (s, 3H), 1.84 (br, 4H), 1.48-1.29 (br, 20H), 0.88 (br, 6H).
GPC: M : 7.7 kDa; PDI: 1.25.
n
2.3.9. General synthetic procedure of D-D copolymers (for PBTAPTh, PBTAPDTh, and
PBTAPBDT)
Three D-D copolymers were synthesized by Stille polycondensation. In a 25 mL two-neck flask,
the dibromo BTAP monomer M1 (0.2 mmol, 1 equiv.), the bis(trimethylstannyl) comonomer (0.2
mmol, 1 equiv.), Pd (dba) (9 mg) and P(o-tyl) (15 mg) were subjected to three cycles of
2
3
3
evacuation/argon purging, and then dissolved with 6 mL of degassed toluene. The reaction mixture
was stirred at 115 °C for 48 h under argon, and then cooled down to room temperature. The mixture
was precipitated in 100 mL of methanol. The dark solids were filtered and Soxhlet extracted with
methanol, acetone, hexane, and chloroform, successively. The chloroform extracts were
concentrated and then subjected to column chromatography eluted with chloroform. The target
polymer was afforded after completely drying in vacuo.
PBTAPTh: The general Stille polymerization procedure was followed using 82 mg of monomer
1
M4 and 170 mg of monomer M1 to afford a red solid. (97.3 mg, yield: 68.4%). H NMR (400 MHz,
CDCl , δ/ppm): 7.22-8.15 (br, 4H), 7.03-7.00 (br, 2H), 6.96-6.94 (br, 2H), 6.80-6.78 (br, 2H), 4.05
3
(
br, 4H), 2.21 (s, 3H), 1.84 (br, 4H), 1.68-1.25 (br, 20H), 0.88 (br, 6H). GPC: M : 18.8 kDa; PDI:
n
1.33.
PBTAPDTh: The general Stille polymerization procedure was followed using 98.4 mg of
1
monomer M5 and 170 mg of monomer M1 to afford a red solid. (93.5 mg, yield: 58.8%). H NMR
(
400 MHz, CDCl , δ/ppm): 7.18-7.12 (br, 6H), 7.03-7.00 (br, 2H), 6.96-6.94 (br, 2H), 6.80-6.78 (br,
3
2
H), 4.05 (br, 4H), 2.21 (s, 3H), 1.83 (br, 4H), 1.58-1.29 (br, 20H), 0.89 (br, 6H). GPC: M : 6.4 kDa;
n
PDI: 1.42.
PBTAPBDT: The general Stille polymerization procedure was followed using 154.4 mg of
1
monomer M6 and 170 mg of monomer M1 to afford a red solid. (135.6 mg, yield: 63.2%). H NMR
(
400 MHz, CDCl , δ/ppm): 7.22-8.15 (br, 4H), 7.03-7.00 (br, 2H), 6.96-6.94 (br, 2H), 6.80-6.78 (br,
3
2H), 4.22 (br, 4H), 4.08 (br, 4H), 2.23 (s, 3H), 1.87-1.86 (br, 6H), 1.45-1.30 (br, 36H), 1.08-0.89 (m,
1
8H). GPC: M : 17.4 kDa; PDI: 1.38.
n

ACCEPTED MANUSCRIPT
Bu
Sn Bu
C
8
H
17
O
S
OC
8 17
H
HO
OH
8
C H17O
OC H
C H17O
OC H
8
17
8
8
17
C
8
H
17Br
2 2 2
Br , CH Cl
S
Bu
NBS
KOH,DMF
3
CH COOH
PhMe, Pd(PPh3)4
CHCl , AcOH
3
Br
Br
1
S
2
3
8
C H
17
O
OC
8
H
17
C
8
H
17
O
8 17
OC H
C
8
H
17
O
OC
8
H
17
8
C H17O
8 17
OC H
NH
2
n-BuLi, I
2
S
Br
Br
Br n-BuLi, CH
THF
3
OH Br
Br
S
S
S
Pd
2
(dba)
3
CHCl
3
,(t-Bu
3
PH)BF
4
I
N
THF
I
N
S
S
S
S
t-BuOK
Br
4
5
6
M1
Scheme 2. Synthetic route of BTAP-based monomer M1.
4 9
C H
Sn
Sn
S
C
2
H
5
C
8
H
17
O
OC
8
H
17
PBTAPTh
O
O
M4
N
O
B
S
O
O
PBTAPDPP
PBTAPBT
S
Sn
Sn
Sn
B
S
S
S
S
O
N
Pd (dba)
2 3
I
M5 C
P(o-tol)
I
N
+
3
Pd
2
(dba)
3
, P(o-tol)
3
C
2
H
5
4
H
9
PBTAPDTh
PBTAPBDT
M2
+
toluene
2 5
C H
K
3
PO
4
, Aliquat 336
toluene
4 9
C H
S
N
O
N
S
O
O
O
M1
Sn
B
B
S
O
O
M3
2 5
C H
M6
4 9
C H
Scheme 3. Synthetic routes of BTAP-based D-D and D-A copolymers.
3
. Results and discussion
3.1. Design, synthesis and characterization of the copolymers
The addition of annulation and steric effect has been proved as effective strategies to increase
thermal and redox stability of heteroepines [20,21]. For nitrogen-containing azepine (AP, Chart 1),
steric repulsion is unfavourable the norcaradiene intermediate. The molecule is thermally stable
when bulky groups are introduced at the 2- and 7- positions. If aromatic rings are annulated to AP, a
substantial resonance energy loss accompanies valence-isomerization. Therefore, our design of the
stable AP building scaffold, named as benzo[d]-dithieno[3,2-b;2′,3′-f]azepine (BTAP), incorporates
two thiophenes and one benzene ring (Chart 1). Thiophene ring is chosen as an annulation because
of its intrinsically electrochemical stability and ease of synthetic modification through the
α-position of the two annulated thiophenes (C6 and C10). In addition, two alkoxy groups can also
be easily attached to the annulated benzene moiety to improve the solubility and π-π stacking of
polymers based on BTAP, which is very important for efficient optoelectronic materials. Moreover,

ACCEPTED MANUSCRIPT
N-alkylation of BTAP block can further control the solubility and the degree of planarity of
molecules. Three electron-rich donor units and two electron-deficient acceptor building blocks were
copolymerized with BTAP to construct D-D and D-A type copolymers with the aim to develop new
AP-containing polymers and tailor the optoelectronic properties of the resulting materials.
Monomers M2 and M3 were directly purchased via commercial approach. Monomers M4-M6
were synthesized according to the previously reported procedure [27]. BTAP-based monomer M1
was prepared according to Swager’s procedure with some modifications [21] and the corresponding
synthetic route is depicted in Scheme 2. 1,2-Benzenediol was used as the starting material, that was
easily converted into compound 1 through simple O-alkylation reaction. Compound 3 involving
thiophene and benzene moieties was synthesized through a Stille coupling reaction between
bromide 2 and 2-(tributylstannyl)thiophene with a moderate yield of 43.6%. Next, in order to
introduce bromines into the 3- and 3'-positions, compound 3 was first subjected to bromination with
NBS to form tetrabrominated compound 4, and then debromination to achieve dibrominated
compound 5. This process has been proved to be necessary because bromination occurs at the
5
5
-positions first, and moreover, lithium-bromine exchange also favors the less sterically hindered
-positions. Then, the key intermediate of annulated azepine 6 was successfully synthesized via
coupling dibromide 5 with p-toluidine using Pd (dba) •CHCl /(t-Bu PH)BF catalyst system in an
2
3
3
3
4
acceptable yield of 46.5%. Finally, the annulated azepine monomer M1 was obtained as a light
solid using a lithiation-iodine quenching sequence. The molecular structure of all newly synthesized
1
13
compounds was characterized by H NMR, C NMR, and MS (MALDI-TOF) spectroscopies.
Five BTAP-based conjugated polymers were prepared by the palladium-catalyzed
polycondensations of 6,10-diiodide monomer M1 with functionalized electron-deficient (BT and
DPP) or electron-rich (Th, DTh, and BDT) moieties, as described in Scheme 3. The
polycondensations were carried out under Stille or Suzuki coupling conditions depending on the
nature of comonomers, using Pd (dba) as a catalyst and P(o-tyl) as the corresponding ligand.
2
3
3
Crude copolymers were first purified by extraction using methanol, acetone, hexane and chloroform,
successively. The chloroform extracts were concentrated and then subjected to column
chromatography eluted with chloroform to afford the resulting copolymers. The separated yields
vary depending on the solubility of the polymers. Due to the lack of alkyl chains on the
benzothiadiazole (BT) and dithiophene (DTh), PBTAPBT and PBTAPDTh exhibit relatively low

ACCEPTED MANUSCRIPT
solubility and yields after Soxhlet extraction and column chromatography. The remainder of this
study was carried out on materials out of the chloroform fractions, which have the higher molecular
weights and not been further characterized. Molecular weights and polydispersity indices (PDIs) of
polymers were determined through gel permeation chromatography (GPC) with THF as eluent
against polystyrene standards. PBTAPDPP, PBTAPTh, and PBTAPBDT show relatively high
number-average molecular weight (M ) within 17.4-67.1 kDa, while the recorded M for
n
n
PBTAPBT and PBTAPDTh suffers from poor solubility in THF for GPC analysis, leading to
obvious lower M of 7.7 and 6.4 kDa, respectively. Notably, because the crude polymers subjected
n
not only successive Soxhlet extraction but also column chromatography, all the polymers exhibit
rather small PDIs varied from 1.33 to 1.76. The polymerization results and detailed GPC data are
summarized in Table 1.
Table 1 Polymerization results and thermal properties of these polymers.
a
a
a
b
Polymer
yield (%) M (kDa) M (kDa) PDI Td (°C)
n w
PBTAPDPP
PBTAPBT
PBTAPTh
62.6
57.3
68.4
58.8
63.2
67.1
7.7
118.7
1.76
1.25
1.33
1.42
1.38
286
402
306
392
344
9.6
25.2
9.1
18.8
6.4
PBTAPDTh
PBTAPBDT
17.4
24.0
a
b
Determined by GPC using polystyrene standards in THF. The 5% weight loss temperatures
under an inert atmosphere.
3.2. Thermal stability
Thermogravimetric analysis (TGA) was performed to investigate thermal stability of the
copolymers, as shown in Fig. 1. All polymers demonstrate good thermal stability with the onset
temperatures with 5% weight loss (T ) above 300 °C except for PBTAPDPP with a T only 286 °C
d
d
(
see Table 1). The relatively lower T for PBTAPDPP should be ascribed to the slightly lower
d
stability of DPP unit, which also has been observed in other previously reported DPP-based
copolymers [29]. Obviously, the thermal stability of the copolymers is good enough for the
applications in optoelectronic devices.

ACCEPTED MANUSCRIPT
100
9
0
0
0
0
0
0
PBTAPDPP
PBTAPBT
PBTAPTh
PBTAPDTh
PBTAPBDT
8
7
6
5
4
100
200
300
400
500
600
Temperature
(°C)
-1
Fig. 1. TGA plots of the five copolymers at a heating rate of 20 °C min under nitrogen.
3.3. Absorption spectra and optical bandgaps
The photophysical properties of the new polymers have been evaluated by UV-vis absorption
both in dilute chloroform solution and as thin films (Fig. 2). The optical data including the
opt
absorption peak wavelength (λmax), both in solution and films and optical bandgaps (Eg ) are
summarized in Table 2.
As shown in Fig. 2a, two D-A copolymers PBTAPDPP and PBTAPBT exhibit obviously
broader absorption spectra than those of three D-D copolymers PBTAPTh, PBTAPDTh, and
PBTAPBDT in chloroform solution. This ascribes to the existence of intramolecular charge
transfer (ICT) effect between the donor unit (BTAP) and acceptor unit (BT or DPP) in D-A type
copolymer, resulting a low-energy absorption band at the range of 500-850 nm. In particular,
PBTAPDPP shows a rather wide absorption extended to near-infrared region (> 800 nm) due to a
stronger electron-deficient character of DPP unit. Moreover, one can see that the intensity of ICT
for PBTAPDPP is higher than the π-π* transition of polymeric backbones located in the
high-energy band (300-500 nm). However, for PBTAPBT the result is quite opposite (Fig. 2a).
This difference should be due to the different electron-withdrawing strength of the two acceptors of
BT and DPP. It is well known that the electron-withdrawing ability of DPP is stronger than BT,
leading to an increase of the intensity of ICT absorption band and a remarkably red-shifted λmax for
PBTAPDPP as compared with PBTAPBT. Generally, the three D-D copolymers possess similar
absorption profiles with narrow absorption bands ranged from 300-500 nm. The λmax in solution of
the copolymer PBTAPTh locates at about 435 nm. Compared with PBTAPTh, absorption

ACCEPTED MANUSCRIPT
spectrum of PBTAPDTh displays a red shift by 13 nm (λmax = 448 nm), and the absorption
spectrum of PBTAPBDT further shifts to the red direction by about 20 nm (λmax = 454 nm), which
is probably due to the relatively long π-conjugation length of the BDT-based polymer main chains.
The absorption spectra of these copolymers in films are displayed in Fig. 2b. The shape of the peaks
and the trend of the absorption wavelength are generally similar to those of the polymer solutions.
The λmax of all the polymer films are red-shifted compared with those in solutions, which owes to
the aggregation of the conjugated polymer main chains in the solid films. Furthermore,
PBTAPDPP delivers a broad platform from 600 to 670 nm, meaning that there is a strong
interchain interactions and good π-π stacking in polymer films.
Fig. 2. Normalized UV-vis absorption spectra of the five polymers: (a) in chloroform solution; (b)
as thin films.
The absorption onset (λegde) of the polymer film spectra are estimated to be 805, 625, 520, 534,
and 513 nm for PBTAPDPP, PBTAPBT, PBTAPTh, PBTAPDTh, and PBTAPBDT,
opt
respectively, deducing an optical bandgaps (Eg ) of these polymers in the range of 1.54-2.42 eV
(Table 2). Some conclusions can be deduced from these observations: (i) two D-A copolymers
opt
shows obviously smaller Eg
than those of D-D copolymers; (ii) due to the strong
electron-withdrawing character of DPP unit, PBTAPDPP has a rather low bandgap of 1.54 eV,
agreeing well with many of the reported D-A copolymers based on DPP unit [9]; (iii) the small
difference between the three D-D copolymers indicates the three electron-donating building blocks,
Th, DTh, and BDT, affect the bandgaps slightly; (iv) the large bandgap variations of BTAP-based
polymers (1.54-2.42 eV) demonstrates that the optical bandgaps can be easily tailored to obtain
different desired bandgaps, such as low-bandgap (PBTAPDPP), medium-bangdap (PBTAPBT)

ACCEPTED MANUSCRIPT
and wide-bandgap (PBTAPTh, PBTAPDTh, and PBTAPBDT), by changing the
electron-donating or electron-withdrawing character of a comonomer unit. More importantly, in
view of easily tuned absorption region (300-900 nm), BTAP-based conjugated polymers could be
used as efficient electron donor materials in fullerene or non-fullerene organic solar cells [30,31],
especially their large bandgaps are supposed to be suitable for tandem heterojunction solar cells
[
32].
Table 2 Optical and electrochemical properties of the five copolymers.
λmax
λmax
opt
Eg
CV
λedge
E /HOMO E /LUMO E
ox
red
g
Polymer
a
b
c
d
e
(
nm) (nm) (nm) (eV)
(V/eV)
(V/eV)
(eV)
g
g
e
e
f
1
1
.69
.86
PBTAPDPP 659
629
556
450
461
463
805 1.54 0.75/-5.21 -0.94/-3.52
625 1.98 0.77/-5.23 -0.99/-3.47
520 2.39 0.75/-5.21 none/-2.82
534 2.32 0.74/-5.20 none/-2.88
PBTAPBT
PBTAPTh
537
435
none
none
none
f
PBTAPDTh 448
PBTAPBDT 454
f
513 2.42 0.91/-5.37 none/-2.95
a
b
Measured in dilute chloroform solution. Measured on a quartz plate by polymers cast from
c
d
chloroform solution. The onset wavelength of the thin films. Estimated from the onset
opt
e
wavelength of the absorption spectra: Eg = 1240/λedge. Calculated according to the equation:
f
opt
HOMO/LUMO = -e(Eox/red + 4.46) (eV). Calculated according to the equation: LUMO = Eg
+
g
CV
g
HOMO (eV). Calculated according to the equation: E
= e(E – E ) (eV).
ox red
3.4. Redox properties and molecular energy levels
Electrochemical cyclic voltammetry (CV) experiment was performed on thin films (spin-coated
on indium tin oxide (ITO) substrates) of each polymer to study their redox properties. The CV
curve was measured on a Pt electrode in a 0.1 M Bu NPF –acetonitrile solution and Ag/AgCl
4
6
electrode as a reference that calibrated by the redox potential of a ferrocene/ferrocenium couple
+
(
Fc/Fc measured as 0.34 V vs Ag/AgCl). Thus, the molecular energy levels (HOMO/LUMO) can
be calculated from the following equation: HOMO/LUMO = e(Eox/red + 4.46) (eV), where Eox/red is
the oxidation/reduction onset potential of the polymers. Representative CVs of the five polymers
+
including Fc/Fc are displayed in Fig. 3 and the corresponding electrochemical properties are also
recorded in Table 2.

ACCEPTED MANUSCRIPT
As shown in Fig. 3a, two D-A copolymers give quasireversible oxidation waves and irreversible
reduction waves. The E and Ered values are observed to be 0.75/-0.94 and 0.77/-0.99 V for
ox
PBTAPDPP and PBTAPBT, respectively. Then, the HOMO/LUMO energy levels are calculated
as -5.21/-3.52 eV for PBTAPDPP and -5.23/-3.47 eV for PBTAPBT. Whereas the three D-D
copolymers just display quasireversible oxidation and no apparent reduction processes are observed
from CV curves. Therefore, the resulted HOMO energy levels are deduced as -5.21, -5.20, and
-5.37 eV for PBTAPTh, PBTAPDTh, and PBTAPBDT respectively, according to their relevant
Eox values (see Fig. 3b). Because the reduction potential onsets of the three D-D copolymers could
not clearly be determined from the CVs (Fig. 3b), the LUMO energy levels of these polymers are
opt
opt
estimated from the Eg values and HOMO energy levels (LUMO = Eg + HOMO) as measured by
CV
CV [18,29]. The electrochemical bandgaps (Eg ) are calculated to be 1.69 and 1.86 eV for
PBTAPDPP and PBTAPBT, respectively, which are coincident with those of the corresponding
opt
Eg (Table 2) of the polymer films within the reasonable range of deviation (0.2-0.5 eV) [33-36].
The electrochemical results reveal that all of these copolymers, except for PBTAPBDT,
substantially have an almost identical HOMO energy level of about -5.21 eV, which is not
surprising in view of the same BTAP donor moiety among all the copolymers [18]. Moreover, this
series of BTAP-based polymers deliver the relatively low-lying HOMO energy levels (< -5.20 eV),
implying good stability against oxidization in air [37], that would enhance device stability.
Additionally, among these polymers PBTAPBDT has the lowest HOMO energy level of -5.39 eV,
which would lead to a higher open-circuit voltage (V ) value for the solar cell with respect to the
oc
other four polymers since V of the PSCs is presumed to proportionate to the difference between
oc
the HOMO level of the electron donor material and the LUMO level of the electron acceptor
material [37]. On the other hand, the LUMO energy levels (see Table 2) of these polymers are
found over a wide range: the LUMO level for PBTAPDPP is the deepest at -3.52 eV, while
PBTAPTh has the highest at -2.82 eV. This implies that variation of the electron-related
characteristic of comonomers has a great impact on the LUMOs of these BTAP-based polymers.
Since these polymers will be intended to use as the donor material matching with a fullerene
acceptor (i.e., PC BM or PC BM), they should have LUMO and HOMO energy level offsets at
6
1
71
least of 0.3 eV relative to PC BM or PC BM for the effective exciton dissociation and charge
61
71
transfer [37-39]. To make a more clear comparison, a complete diagram of the HOMO/LUMO

ACCEPTED MANUSCRIPT
energy levels of these polymers and PC BM is presented in Fig. 4. From Fig. 4, one can find that
7
1
the LUMO offsets and HOMO offsets between these polymers and PC BM are large enough to
7
1
meet the above mentioned requirements.
Fig. 3. Cyclic voltammograms of the polymer thin films on ITO glass electrode, recorded at a 100
-1
mV s scan rate in 0.1 M Bu NPF acetonitrile solution: (a) PBTAPDPP, PBTAPBT and
4
6
Ferrocene; (b) PBTAPTh, PBTAPDTh and PBTAPBDT.
Fig. 4. Energy level diagrams of the five polymers and PC BM using data derived from CV.
71
3.5. Photovoltaic properties
In consideration of the aforementioned features of the variable absorption and suitable molecular
energy levels of the BTAP-based polymers, we assume that these polymers should be able to use as
the effective electron donor materials in PSCs. Accordingly, BHJ type PSCs with the conventional
device structure of ITO/PEDOT:PSS/polymer:PCBM (PC BM or PC BM)/Ca/Al were fabricated
61
71
to investigate the photovoltaic properties of the polymers. The photovoltaic data, such as
open-circuit voltage (V ), short-circuit current density (J ), fill factor (FF), and power conversion
oc
sc

ACCEPTED MANUSCRIPT
efficiency (PCE) are summarized in Tables 3 and 4. The typical current density-voltage (J-V)
curves of the photovoltaic performance based on these polymers are described in Fig. 5.
First, we used PC BM as the acceptor material combined with these polymers to prepare the
6
1
photovoltaic device. When the blend ratio of polymer donor to PC BM is determined to be 1:2 (by
61
weight), the highest PCE of 0.51%, 0.34%, 0.25%, 0.38%, and 0.37% is obtained for PBTAPDPP,
PBTAPBT, PBTAPTh, PBTAPDTh, and PBTAPBDT, respectively (see Table 3). Then, the
acceptor is changed as PC BM, the photovoltaic performance is improved apparently,
71
corresponding the best PCE up to 0.87% for PBTAPDPP, 0.93% for PBTAPBT, 0.58% for
PBTAPTh, 0.82% for PBTAPDTh, and 1.00% for PBTAPBDT. The greatly enhanced PCEs
mainly attribute to the substantial increase of J values, as shown in Table 4. This is easily
sc
understood because PC BM has the broader absorption spectrum compared to PC BM, which
7
1
61
benefits the improvements of J [40,41]. According to the data in Tables 3 and 4, one can find that
sc
no matter PC BM or PC BM as the acceptor, PSCs based on PBTAPDPP exhibit the highest J ,
6
1
71
sc
which accords with the relatively broader absorption (Fig. 2). On the other hand,
PBTAPBDT-based PSCs deliver higher V as compared with the other polymers, and moreover
oc
the highest V of 0.84 V can be achieved after optimization of photovoltaic performance discussed
oc
in follow. This is because the V of PSCs is tightly correlated with the energy difference between
oc
the HOMO level of a polymer donor and the LUMO level of an acceptor (PC BM or PC BM)
61
71
[
37]. The polymer PBTAPBDT has the lowest HOMO energy level (see Table 2) and thus the
PSCs containing this polymer displays the highest V . Among these polymers PBTAPBDT
oc
possesses the best PCE, thus, we choose this polymer as a representative to further investigate the
photovoltaic properties of these BTAP-based polymers. Previous studies have demonstrated that
post-solvent treatment with a polar solvent can improve the photovoltaic performance of PSCs
[
42-49]. Here, methanol was chosen as a polar solvent to further optimize the photovoltaic
performance of PBTAPBDT. As shown in Table 4, after the methanol treatment, the better
photovoltaic performance is achieved with the highest PCE up to 1.28%, corresponding the
-2
significantly improved both of V (0.84 V) and J (5.20 mA cm ). Overall, the primary results
oc
sc
indicate that BTAP-based polymer solar cells just can exhibit the moderate performance as a result
of low J and FF, which might be related to the charge mobility of the polymers and the
sc
morphology of the active layer.

ACCEPTED MANUSCRIPT
Table 3 Photovoltaic properties of the PSCs based on these BTAP-based polymers with PC BM as
6
1
the acceptor.
-2
a
polymer
V (V) J (mA cm ) FF (%)
PCE (%)
oc
sc
PBTAPDPP 0.78
2.37
1.70
1.35
1.78
1.65
27.7 0.51 (0.50±0.01)
27.0 0.34 (0.31±0.03)
29.4 0.25 (0.24±0.01)
28.5 0.38 (0.35±0.03)
28.8 0.37 (0.36±0.01)
PBTAPBT
PBTAPTh
0.75
0.62
PBTAPDTh 0.75
PBTAPBDT 0.79
a
Average PCEs were obtained from more than 6 devices shown in parentheses.
Table 4 Photovoltaic properties and hole mobility of the PSCs based on these BTAP-based
polymers with PC BM as the acceptor.
7
1
-2
a
2
-1 -1
polymer
V (V) J (mA cm ) FF (%)
PCE (%)
µ (cm V s )
h
oc
sc
-5
-5
-5
-5
-5
PBTAPDPP
PBTAPBT
PBTAPTh
PBTAPDTh
PBTAPBDT
0.66
0.72
0.69
0.68
0.74
0.84
4.57
4.30
2.96
4.00
4.39
5.20
29.0 0.87 (0.86±0.01)
29.8 0.93 (0.92±0.03)
28.5 0.58 (0.57±0.01)
30.1 0.82 (0.80±0.02)
30.7 1.00 (0.98±0.02)
29.4 1.28 (1.26±0.02)
2.07 × 10
2.33 × 10
3.97 × 10
1.14 × 10
2.96 × 10
-
b
PBTAPBDT
a
b
Average PCEs were obtained from more than 6 devices shown in parentheses. Photovoltaic
property treated with methanol.

ACCEPTED MANUSCRIPT
Fig. 5. Typical J-V curves of PSCs based on the five polymers: (a) with PC BM as the acceptor; (b)
6
1
with PC BM as the acceptor.
7
1
3.6. Hole mobility
To search the possible reason resulted low performance of BTAP-based polymer solar cells, hole
mobility (µ ) of these polymers was measured because it is another important parameter for CPs
h
using as donor materials in PSCs. To evaluate the charge mobility of these BTAP-based polymers,
herein hole mobilities of these polymer:PC BM blend film were measured by using the
7
1
space-charge-limited current (SCLC) method with the hole-only device structure of
ITO/PEDOT:PSS/polymer:PC BM/MoO /Al and estimated through the Mott–Gurney equation
71
3
[
50]. Fig. 6 displays the relationship between current and voltage in the hole-only devices of
-5
-5
polymer:PC BM blend films. The calculated hole mobilities are 2.07 × 10 , 2.33 × 10 , 3.97 ×
7
1
-5
-5
-5
2
-1
-1
1
0 , 1.14 × 10 , and 2.96 × 10 cm V
s
for PBTAPDPP, PBTAPBT, PBTAPTh,
PBTAPDTh, and PBTAPBDT blend film, respectively (see Table 4). Generally, the hole
mobilities of all these BTAP-based polymers are relatively low as compared with other
high-performance conjugated polymers [51-58]. Moreover, the µ value is much lower than electron
h
-3
2
-1 -1
mobilities of PCBMs (∼10 cm V s ). The low and mismatch of charge mobilities should be
partly responsible for the low J and FF, and as a result that limits the performance of photovoltaic
sc
devices [59].

ACCEPTED MANUSCRIPT
Fig. 6. J –V characteristics for the measurement of hole mobility in polymer:PC BM devices by
1
/2
71
the SCLC method.
3.7. Morphology analyses of blend films
As we all know, the photovoltaic performance of the PSCs strongly depends on its morphological
features. To seek the other reasons of the low J and FF of these polymers, atomic force
sc
microscopy (AFM) is employed to study the morphology of the blend film. Among the five
BTAP-based polymers, PBTAPTh and PBTAPBDT give the worst and best PCEs respectively,
hence we have studied the morphology of the two polymer blend films. The AFM height and phase
images of the blend films are described in Fig. 7. The height images of PBTAPTh:PC BM (1:3 by
71
weight, Fig. 7a) and PBTAPBDT:PC BM (1:2 by weight, Fig. 7b) blend films reveal that both of
71
these have rather smooth surfaces with a very small root-mean square roughness (R ) of 0.470 and
q
0.881 nm, respectively. This observation indicates that the active layer has unobvious phase
separation (see Figs. 7c and 7d), which is not good for the effective transport of charge
carriers/collection and unfavorable to the production of high Jsc and FF. Additionally,
PBTAPBDT:PC BM blend film possesses a relatively rougher surface, meaning a relatively
71
efficient exciton dissociation and transport of charge carriers and led to a higher J . Furthermore,
sc
transmission electron microscopy (TEM) was performed to further investigate the composition and
morphology of the polymer:PC BM blends, as shown in Fig. 8. The blend films were prepared
71
with a same condition to the optimized PSC devices. In the TEM images, the light and dark regions
are assigned to polymer-rich and PC BM-rich domains, respectively. From Fig. 8, one can find that
7
1
there is an unobvious polymer aggregation or PC BM aggregation both in PBTAPTh and
71
PBTAPBDT blends, implying a relatively excessive donor/acceptor compatibility for these two

ACCEPTED MANUSCRIPT
blend films. Furthermore, this result shows that both of blends did not form regular interpenetrating
network structure and desired nanoscaled phase separation, leading to a worse charge separation on
the donor/acceptor interfaces. Overall, this non-ideal phase separation should be also responsible for
the low J to some extent.
sc
Fig. 7. AFM height (top) and phase (bottom) images of (a) and (c): PBTAPTh:PC BM (1:3, by
71
weight); (b) and (d): PBTAPBDT:PC BM (1:2, by weight).
71
Fig. 8. TEM images of polymer:PC BM blended films: (a) PBTAPTh:PC BM (1:3, by weight);
7
1
71
(
b) PBTAPBDT:PC BM (1:2, by weight). The scale bar in the TEM images is 200 nm.
71
4. Conclusions
In summary,
a
series
of
new
conjugated
polymers
based
on
benzo[d]-dithieno[3,2-b;2′,3′-f]azepine (BTAP) were designed and synthesized by Stille or Suzuki
cross-coupling reaction. Two electron-withdrawing aromatic copolymerization units (DPP and BT)
and three electron-donating units (Th, DTh, and BDT) were employed to finely manipulate the

ACCEPTED MANUSCRIPT
bandgaps and molecular energy levels of the resulting polymers. These polymers have good thermal
and electrochemical stabilities. The polymers’ absorption spectra can be effectively tuned
throughout a wide range of 300-850 nm and the desired optical bandgaps between 1.54 and 2.42 eV
are easily accessible depending on the electron-rich or electron-deficient character of comonomers.
These polymers have nearly identical HOMO energy levels (~-5.21 eV) and various LUMO levels
in the range of -2.82 ~ -3.52 eV. To expand the application fields of azepine-based conjugated
polymers and tap the applied possibility in various organic electronic devices, BHJ type PSCs with
these polymers as donors and PCBMs as acceptor were fabricated and characterized. Preliminary
studies on PSCs based on these polymers deliver the highest PCE of 1.28% for PBTAPBDT-based
solar cells with a relatively high V of 0.84 V. The moderate performance of BTAP-based PSCs
oc
might attribute to low and unbalanced charge carrier mobility and non-ideal phase separation in
polymer:PCBM blend films. Although photovoltaic performance of the solar cells based on these
new polymers is still rather lower than the state of the art for fullerene-based PSCs (PCE > 10%
[
51,60-64]), to the best of our knowledge, this is the first report on synthesis and characterization of
azepine-based conjugated polymers for application in organic electronic devices. Optimizations on
molecular structures of polymers and morphologies of blend films for the further improvement of
the photovoltaic performance are in progress. Our study open a new window for azepine containing
polymers used as polymeric optoelectronic materials, such as PLEDs, PFETs, and photodetectors,
which our future work is focusing on.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Grant no.
21574111), the Hunan 2011 Collaborative Innovation Center of Chemical Engineering &
Technology with Environmental Benignity and Effective Resource Utilization and the
Undergraduate Investigated-Study and Innovated Experiment Plan of Xiangtan University.
References
[
1] H. Yao, L. Ye, H. Zhang, S. Li, S. Zhang, J. Hou, Molecular design of benzodithiophene-based
organic photovoltaic materials, Chem. Rev. 116 (2016) 7397-7457.

ACCEPTED MANUSCRIPT
[
2] L. Lu, T. Zheng, Q. Wu, A.M. Schneider, D. Zhao, L. Yu, Recent advances in bulk
heterojunction polymer solar cells. Chem. Rev. 115 (2015) 12666-12731.
[
[
3] G. Li, R. Zhu, Y. Yang, Polymer solar cells. Nat. Photonics 6 (2012) 153-161.
4] Z.-G. Zhang, Y. Li, Side-chain engineering of high-efficiency conjugated polymer photovoltaic
materials, Sci. China Chem. 58 (2015) 192-209.
[
[
[
[
[
5] L. Dou, Y. Liu, Z. Hong, G. Li, Y. Yang, Low-bandgap near-IR conjugated polymers/molecules
for organic electronics, Chem. Rev. 115 (2015) 12633-12665.
6] A.J. Heeger, Y. Cao, I.D. Parker, G. Yu, C. Zhang, Improved quantum efficiency for
electroluminescence in semiconducting polymers, Nature 397 (1999) 414-417.
7] H. Sasabe, J. Kido, Development of high performance OLEDs for general lighting, J. Mater.
Chem. C 1 (2013) 1699-1707.
8] C. Wang, H. Dong, W. Hu, Y. Liu, D. Zhu, Semiconducting π-conjugated systems in field-effect
transistors: a material odyssey of organic electronics, Chem. Rev. 112 (2012) 2208-2267.
9] Y. Li, P. Sonar, L. Murphya, W. Hong, High mobility diketopyrrolopyrrole (DPP)-based
organic semiconductor materials for organic thin film transistors and photovoltaics, Energy
Environ. Sci. 6 (2013) 1684-1710.
[
[
[
[
[
10] Z. Yi, S. Wang, Y. Liu, Design of high-mobility diketopyrrolopyrrole-based π-conjugated
copolymers for organic thin-film transistors. Adv. Mater. 27 (2015) 3589-3606.
11] K.J. Baeg, M. Binda, D. Natali, M. Caironi, Y.Y. Noh, Organic light detectors: Photodiodes
and phototransistors, Adv. Mater. 25 (2013) 4267-4295.
12] D.S. Weiss, M. Abkowitz, Advances in organic photoconductor technology, Chem. Rev. 110
(2010) 479-526.
13] S. Xiao, Q. Zhang, W. You, Molecular engineering of conjugated polymers for solar Cells: An
updated report, Adv. Mater. 29 (2017) 1601391.
14] L. Ye, H. Hu, M. Ghasemi, T. Wang, B.A. Collins, J.-H. Kim, K. Jiang, J.H. Carpenter, H. Li,
Z. Li, T. McAfee, J. Zhao, X. Chen, J.L.Y. Lai, T. Ma, J.-L. Bredas, H. Yan, H. Ade,
Quantitative relations between interaction parameter, miscibility and function in organic solar
cells, Nat. Mater. 17 (2018) 253-260.

ACCEPTED MANUSCRIPT
[
15] L. Ye, B.A. Collins, X. Jiao, J. Zhao, H. Yan, H. Ade, Miscibility–function relations in organic
solar cells: Significance of optimal miscibility in relation to percolation, Adv. Energy Mater. 8
(2018) 1703058.
[
[
16] Y. Huang, E.J. Kramer, A.J. Heeger, G.C. Bazan, Bulk heterojunction solar cells: Morphology
and performance relationships, Chem. Rev. 114 (2014) 7006-7043.
17] J. Hou, M.-H. Park, S. Zhang, Y. Yao, L.-M. Chen, J.-H. Li, Y. Yang, Bandgap and molecular
energy level control of conjugated polymer photovoltaic materials based on
benzo[1,2-b:4,5-b′]dithiophene, Macromolecules 41 (2008) 6012-6018.
[
18] P. Shen, X. Liu, P. Tang, B. Zhao, L. Wang, C. Weng, J. Cao, Y. Wu, Y. Chen, S. Tan,
Bandgap and molecular-energy-level control of conjugated-polymer photovoltaic materials
based on 6,12-dihydro-diindeno[1,2-b;10,20-e]pyrazine, Macromol. Chem. Phys. 214 (2013)
1147-1157.
[
[
[
[
19] J.-S. Wu, S.-W. Cheng, Y.-J. Cheng, C.-S. Hsu, Donor-acceptor conjugated polymers based on
multifused ladder-type arenes for organic solar cells, Chem. Soc. Rev. 44 (2015) 1113-1154.
20] C. Song, T.M. Swager, Reactive conducting thiepin polymers, J. Org. Chem. 75 (2010)
999-1005.
21] C. Song, D.B. Walker, T.M. Swager, Conducting thiophene-annulated azepine polymers,
Macromolecules 43 (2010) 5233-5237.
22] J.-M. Yang, C.-Z. Zhu, X.-Y. Tang, M. Shi, Rhodium(II)-catalyzed intramolecular annulation
of 1-sulfonyl-1,2,3-triazoles with pyrrole and indole rings: Facile synthesis of N-bridgehead
azepine skeletons, Angew. Chem. 126 (2014) 5242-5246.
[
23] H.V. Kumar, N. Naik, Synthesis and antioxidant properties of some novel
5H-dibenz[b,f]azepine derivatives in different in vitro model systems, Eur. J. Med. Chem. 45
(2010) 2-10.
[
[
24] D, He, L. Qian, L. Ding, A pentacyclic building block containing an azepine-2,7-dione moiety
for polymer solar cells, Polym. Chem. 7 (2016) 2329-2332.
25] J. Kim, T. Kim, S.Y. Bang, J. Kim, J.Y. Shim, I. Kim, S.Y. Park, H.H. Chun, J.Y. Kim, H. Suh,
Syntheses and solar cell applications of conjugated copolymers consisting of
3,3′-dicarboximide and benzodithiophene units with thiophene and bithiophene linkage, Solar
Energy Materials & Solar Cells 141 (2015) 24-31.

ACCEPTED MANUSCRIPT
[
[
[
[
26] C. Wang, J. Li, S. Cai, Z. Ning, D. Zhao, Q. Zhang, J.-H. Su, Performance improvement of
dye-sensitizing solar cell by semi-rigid triarylamine-based donors, Dyes and Pigments 94
(2012) 40-48.
27] P. Shen, H. Bin, L. Xiao, Y. Li, Enhancing photovoltaic performance of copolymers containing
thiophene unit with D-A conjugated side chain by rational molecular design, Macromolecules
46 (2013) 9575-9586.
28] P. Shen, Y. Tang, S. Jiang, H. Chen, X. Zheng, X. Wang, B. Zhao, S. Tan, Efficient
triphenylamine-based dyes featuring dual-role carbazole, fluorene and spirobifluorene moieties,
Org. Electron. 12 (2011) 125-135.
29] M. Zhang, Y. Sun, X. Guo, C. Cui, Y. He, Y. Li, Synthesis and characterization of
dioctyloxybenzo[1,2-b:4,3-b']-dithiophene-dontaining copolymers for polymer solar Cells,
Macromolecules 44 (2011) 7625-7631.
[
[
[
[
30] C.B. Nielsen, S. Holliday, H.-Y. Chen, S. J. Cryer, I. McCulloch, Non-fullerene electron
acceptors for use in organic solar cells, Acc. Chem. Res. 48 (2015) 2803-2812
31] W. Chen, Q, Zhang, Recent progress in non-fullerene small molecule acceptors in organic solar
cells (OSCs), J. Mater. Chem. C 5 (2017) 1275-1302.
32] O. Adebanjo, B. Vaagensmith, Q. Qiao, Double junction polymer solar cells, J. Mater. Chem.
A 2 (2014) 10331-10349.
33] P. Homyak, Y. Liu, F. Liu, T.P. Russel, E.B. Coughlin, Systematic variation of fluorinated
diketopyrrolopyrrole low bandgap conjugated polymers: Synthesis by direct arylation
polymerization and characterization and performance in organic photovoltaics and organic
field-effect transistors, Macromolecules 48 (2015) 6978-6986.
[
[
34] G. Liu, C. Weng, P. Yin, S. Tan, P. Shen, Impact of the number of fluorine atoms on
crystalline, physicochemical and photovoltaic properties of low bandgap copolymers based on
1,4-dithienylphenylene and diketopyrrolopyrrole, Polymer 125 (2017) 217-226.
35] X. Xu, Y. Wu, J. Fang, Z. Li, Z. Wang, Y. Li, Q. Peng, Side-chain engineering of
benzodithiophene-fluorinated quinoxaline low-band-gap co-polymers for high-performance
polymer solar cells, Chem. Eur. J. 20 (2014) 13259-13271.

ACCEPTED MANUSCRIPT
[
[
[
36] P.T. Wu, T. Bull, F.S. Kim, C.K. Luscombe, S.A. Jenekhe, Organometallic donor-acceptor
conjugated polymer semiconductors: Tunable pptical, electrochemical, charge transport, and
photovoltaic properties, Macromolecules 42 (2009) 671-681.
37] M.C. Scharber, D. Mühbacher, M. Koppe, P. Denk, C. Waldauf, A.J. Heeger, C.J. Brabec,
Design rules for donors in bulk-heterojunction solar cells—Towards 10% energy-conversion
efficiency, Adv. Mater. 18 (2006) 789-794.
38] C.J. Brabec, C. Winder, N.S. Sariciftci, J.C. Hummelen, A. Dhanabalan, P.A. van Hal, R.A.J.
Janssen, A low-bandgap semiconducting polymer for photovoltaic devices and infrared
emitting diodes, Adv. Funct. Mater 12 (2002) 709-712.
[
[
39] L.J.A. Koster, V.D. Mihailetchi, P.W.M. Blom, Ultimate efficiency of polymer/fullerene bulk
heterojunction solar cells, Appl. Phys. Lett. 88 (2006) 093511.
40] R. Ganesamoorthy, G. Sathiyan, P. Sakthivel, Review: Fullerene based acceptors for efficient
bulk heterojunction organic solar cell applications, Solar Energy Materials & Solar Cells 161
(2017) 102-148.
[
[
41] Y. He, Y. Li, Fullerene derivative acceptors for high performance polymer solar cells, Phys.
Chem. Chem. Phys. 13 (2011) 1970-1983.
42] H. Bin, L. Xiao, Y. Liu, P. Shen, Y. Li, Effects of donor unit and π-bridge on photovoltaic
properties of D-A copolymers based on benzo[1,2-b:4,5-c']-dithiophene-4,8-dione acceptor
unit, J. Polym. Sci., Part A: Polym. Chem. 52 (2014) 1929-1940.
[
[
43] Y. Wang, Y. Liu, S. Chen, R. Peng, Z. Ge, Significant enhancement of polymer solar cell
performance via side-chain engineering and simple solvent treatment, Chem. Mater. 25 (2013)
3196-3204.
44] L. Chen, P. Shen, Z.-G. Zhang, Y. Li, Side-chain engineering of benzodithiophene-thiophene
copolymers with conjugated side chains containing the electron-withdrawing ethylrhodanine
group, J. Mater. Chem. A 3 (2015) 12005-12015.
[
[
45] X. Liu, W. Wen, G.C. Bazan, Post-deposition treatment of an arylated-carbazole conjugated
polymer for solar cell fabrication, Adv. Mater. 24 (2012) 4505-4510.
46] K. Kranthiraja, K. Gunasekar, W. Cho, M. Song, Y.G. Park, J.Y. Lee, Y. Shin, I.-N. Kang, A.
Kim, H. Kim, B. Kim, S.-H. Jin, Alkoxyphenylthiophene linked benzodithiophene based