Improving the photovoltaic performance of ladder-type dithienonaphthalene- containing copolymers through structural isomerization

Article abstract of DOI:10.1039/c4ta02585g

A ladder-type angular-shaped dithienonaphthalene (aDTN), an isomer of ladder-type linear-shaped dithienonaphthalene (DTN), was designed and synthesized as an electron-rich unit to construct donor-acceptor copolymers with deep-lying highest occupied molecular orbital (HOMO) energy levels. Benzo[c][1,2,5]thiadiazole (BT) with various substituents were used as electron deficient units for synthesizing the target copolymers (PaDTNBTO, PaDTNBTH, and PaDTNBTF) via the Stille coupling reaction. Incorporating different substituents onto the BT moiety has significant effects on the photophysical and electrochemical properties of the copolymers, as well as on the roughness of the polymer/PC₇₁BM blends. With four solubilizing alkyl chains on the aDTN unit, all its three copolymers have good solubility in common solvents. The synthesized copolymers exhibit deep-lying HOMO energy levels, leading to high open circuit voltages (V_oc ≥ 0.90 V) of the resulting polymer solar cells. The bulk heterojunction solar cell based on the aDTN-containing copolymers (PaDTNBTO) shows an improved efficiency of 6.44% and an increased V_oc of 0.92 V compared to that based on the linear-shaped DTN containing counterpart (efficiency = 4.78%, V_oc = 0.86 V). Whereas, under the same device fabrication conditions, PaDTNBTH- and PaDTNBTF-based devices exhibit efficiencies of 5.22% and 1.73%, respectively. Our results demonstrate that aDTN is a better building block in constructing p-type copolymers for high open circuit voltage devices compared to the linear-shaped DTN. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ta02585g

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Improving the photovoltaic performance of ladder-
type dithienonaphthalene-containing copolymers
through structural isomerization†
Cite this: J. Mater. Chem. A, 2014, 2,
13905
*
Yunlong Ma, Qingdong Zheng, Lixin Wang, Dongdong Cai, Changquan Tang,
Meng Wang, Zhigang Yin and Shan-Ci Chen
A ladder-type angular-shaped dithienonaphthalene (aDTN), an isomer of ladder-type linear-shaped
dithienonaphthalene (DTN), was designed and synthesized as an electron-rich unit to construct donor–
acceptor copolymers with deep-lying highest occupied molecular orbital (HOMO) energy levels. Benzo
[c][1,2,5]thiadiazole (BT) with various substituents were used as electron deficient units for synthesizing
the target copolymers (PaDTNBTO, PaDTNBTH, and PaDTNBTF) via the Stille coupling reaction.
Incorporating different substituents onto the BT moiety has significant effects on the photophysical and
electrochemical properties of the copolymers, as well as on the roughness of the polymer/PC71BM
blends. With four solubilizing alkyl chains on the aDTN unit, all its three copolymers have good solubility
in common solvents. The synthesized copolymers exhibit deep-lying HOMO energy levels, leading to
high open circuit voltages (Voc $ 0.90 V) of the resulting polymer solar cells. The bulk heterojunction
solar cell based on the aDTN-containing copolymers (PaDTNBTO) shows an improved efficiency of
6.44% and an increased Voc of 0.92 V compared to that based on the linear-shaped DTN containing
counterpart (efficiency ¼ 4.78%, Voc ¼ 0.86 V). Whereas, under the same device fabrication conditions,
PaDTNBTH- and PaDTNBTF-based devices exhibit efficiencies of 5.22% and 1.73%, respectively. Our
results demonstrate that aDTN is a better building block in constructing p-type copolymers for high
open circuit voltage devices compared to the linear-shaped DTN.
Received 23rd May 2014
Accepted 18th June 2014
DOI: 10.1039/c4ta02585g
www.rsc.org/MaterialsA
owing to their small band gaps, deep-lying HOMO energy levels,
and high hole mobilities.15–18
1. Introduction
Planarization and rigidization of polyaromatic conjugated
skeletons may facilitate a better p-electron delocalization,
reduce the energetic disorder of the resulting polymers, and
induce strong intermolecular interactions between the polymer
chains for efficient charge transportation.19 Moreover, an
extended p-conjugated aromatic molecule with a multi-ring
fused framework endows a broader, more intense absorption
band, leading to an enhanced solar light harvesting.20,21 Based
on these guidelines, ladder-type structures, which are con-
structed by fastening or fusing adjacent aromatic units such as
benzene or thiophene via a bridging atom, have attracted
signicant attention for photovoltaic applications. Besides their
intrinsic properties, good solubility of these rigid rod-like
polymers can be anticipated through the incorporation of
various alkyl chains into the bridging atoms. Recently, a
number of conjugated polymers based on ladder-type mole-
cules have been reported by some research groups as well as by
us.19,20,22–37 For example, ladder-type indacenodithieno[3,2-b]-
thiophene-based polymers were used to fabricate PSCs with
PCEs up to 7.51%.37 And PSCs with a Voc as high as 1.06 V have
been fabricated, based on ladder-type oligo-p-phenylene con-
taining copolymers.20 All these exciting achievements suggest
Polymer solar cells (PSCs) have emerged as a promising alter-
native for producing green and renewable energy, because of
their low-cost, light-weight, and great potential for fabrication
as large-area exible devices.1–10 Efficient PSCs have been made
using a bulk heterojunction (BHJ) structure, which comprises p-
type polymers blended with n-type fullerene derivatives (such as
PC61BM and PC71BM). Recently, power conversion efficiencies
(PCEs) of PSCs have reached up to 9.35% for single-junction
devices and 10.6% for tandem cells, mainly thanks to the
innovation in tailor-made photoactive materials as well as
device processing technologies.11,12 For the p-type materials, the
most common and successful strategy currently is to design and
synthesize D–A (electron donor–acceptor) copolymers,8,13,14
State Key Laboratory of Structural Chemistry, Fujian Institute of Research on the
Structure of Matter, Chinese Academy of Sciences, Fuzhou, 350002, P. R. China.
E-mail: qingdongzheng@irsm.ac.cn; Fax: +86-591-83721625
† Electronic supplementary information (ESI) available: Figures showing the
absorption spectra of the blended lms, the J–V characteristics for PSCs,
external quantum efficiency curves of the devices, and summary of device
parameters of PSCs. See DOI: 10.1039/c4ta02585g
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13905–13915 | 13905

View Article Online
Journal of Materials Chemistry A
Paper
that ladder-type molecules are an excellent class of building ability.44,45 At the same time, two thiophene groups are anked
blocks for high performance copolymers in photovoltaic on both sides of the BT group to reduce the steric hindrance, if
applications.
any, between the donor and the acceptor units. In order to
The isomeric phenomena generally originate from the further modulate and optimize the performance of the aDTN-
position difference of the heteroatoms/functional groups or the based copolymers, two electron-donating hexyloxy groups and
bond connection styles of the molecular moieties.38 They are two electron-withdrawing uorine atoms are introduced onto
very common in fused aromatic hydrocarbons. Recent studies the BT heterocycle, generating 5,6-bis(hexyloxy)benzo[c][1,2,5]-
indicate that the structural isomerization has a signicant thiadiazole (BTO) and 5,6-diuorobenzo[c][1,2,5]thiadiazole
impact on the properties of organic electronic materials. For (BTF). Using these three monomers, three D–A copolymers
example, as shown in Fig. 1, picene has a better chemical (PaDTNBTH, PaDTNBTO, and PaDTNBTF in Scheme 2) are
stability and device performance than its isomer pentacene in synthesized. The optical, electrochemical, and electrical prop-
air.39 Takimiya et al. discovered that angular-shaped naph- erties, as well as the preliminary photovoltaic performance of
thodithiophenes (NDTs) showed deeper HOMO energy levels these copolymers, are investigated. The results reveal that these
and larger band gaps than their linear counterparts. Interest- copolymers possess deep-lying HOMO energy levels below
ingly, in comparison with their linear counterparts, angular- ꢀ5.34 eV, with medium band gaps in the range of 1.79–1.90 eV.
shaped NDT-based copolymer PNDT3BT could lead to a more PDTNTBT differs from PaDTNBTO in the chemical structure of
ordered molecular packing, which facilitates an efficient carrier the donor unit for D–A copolymers. DTN is used for the former,
transport.40,41 Similar results were also found by Mullen et al. whereas aDTN is used for the latter. However, the PaDTNBTO-
¨
regarding the investigation on the inuence of backbone based device exhibits a PCE of 6.44% with a Voc of 0.92 V, which
curvature on charge-carrier mobilities.42,43 Recently, we reported is higher than that of the PDTNTBT-based device (efficiency of
a series of D–A copolymers based on ladder-type dithieno- 4.78% with a Voc of 0.86 V).
naphthalene (DTN) units. PSCs based on one of these DTN-
containing copolymers (PDTNTBT in Fig. 1) exhibited a PCE of
4.78% with an open circuit voltage (Voc) of 0.86 V, indicating
that DTN is a promising donor unit for constructing high
performance conjugated copolymers.19 Similar to other fused
2. Experimental
2.1. Material
Reagents were purchased from Aldrich Inc., Aladdin-Reagent
aromatics, DTN also contains different isomers, such as linear-
and angular-shaped analogues (aDTN) (Fig. 1). Considering the
abovementioned dependence of material electronic properties
on its chemical structure, we expect that aDTN-based polymers
may exhibit different properties in comparison with their linear-
shaped analogues.
Inc. or Adamas-beta Ltd. and used without further purication
unless otherwise specied. Compounds 1, 9, 10, and 11 were
prepared according to the literature procedures.46–49 Column
chromatography was conducted with silica gel (200–300 mesh).
2.2. Instruments
Herein, we design and synthesize aDTN as a donor unit for
constructing D–A conjugated copolymers. Similar to our
previous studies,19 benzo[c][1,2,5]thiadiazole (BT) is chosen as
an acceptor unit because of its strong electron-accepting
1H and 13C NMR spectra were collected on a Bruker AVANCE-
400 spectrometer operating at 400 and 100 MHz in CDCl3 or
DMSO-d6 with the TMS reference. High-resolution mass spec-
troscopy (HRMS) measurements were performed on an IonSpec
4.7T spectrometer. Gel permeation chromatography (GPC) was
conducted in THF on a Waters 1515 system. The absorption
spectra of the copolymers in chlorobenzene (1 ꢁ 10ꢀ5 M) and in
thin lms on glass substrates were measured by a Lambda 35
UV/vis spectrophotometer. Cyclic voltammetry (CV) was per-
formed on a CHI 700E electrochemical workstation with a three-
electrode system in a solution of 0.1 M Bu4NPF6 in acetonitrile
at a scan rate of 100 mV sꢀ1. The copolymer lms were coated
on a Pt plate electrode by dipping the electrode into the corre-
sponding solutions and then drying. A Pt wire was used as the
counter electrode, and Ag/AgNO3 was used as the reference
electrode. Atomic force microscopy (AFM) was performed by a
Veeco Multimode NS3A-02 Nanoscope III atomic force micro-
scope. Blended lms for AFM were prepared on PEDOT:PSS-
coated ITO substrates prepared in identical fashion to those
prepared for the device fabrication.
2.3. Fabrication of conventional PSCs
Fig. 1 Chemical structures of naphthodithiophene derivatives, NDT1-
4, picene, pentacene, linear-shaped DTN, angular-shaped DTN, and
PDTNTBT.
PSCs were fabricated in the traditional sandwich structure by
using PEDOT:PSS (Baytron PVPAI 4083) as the anode buffer
13906 | J. Mater. Chem. A, 2014, 2, 13905–13915
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
layer. Indium tin oxide (ITO) glass was sequentially cleaned by The thickness of the lm was measured by the Bruker Dektak XT
ultrasonication in detergent, water, acetone, and isopropyl surface prolometer.
alcohol for 30 min each and then dried in an oven at 130 ꢂC
overnight. The ITO/glass substrates were rst treated with
oxygen plasma for 5 min, then PEDOT:PSS was spin-coated on
2.6. Synthesis
the substrates at 3000 rpm for 1 min, and later the substrates
1,5-Dibromo-2,6-bis(dibromomethyl)naphthalene.50
A
were dried at 130 ꢂC in air for 30 min. The active layer was suspension of 1,5-dibromo-2,6-dimethylnaphthalene (1)
prepared by spin-coating the polymer:PC71BM blends at 1000 (2.00 g, 6.4 mmol), N-bromosuccinimide (7.00 g, 39.3
rpm for 1 min in a glove box, and the devices were kept at room mmol), dibenzoyl peroxide (0.95 g, 3.93 mmol) in anhydrous
temperature for 12 h. Before the aluminum deposition, a CCl4 (75 mL) was reuxed for 24 h. Aer the removal of the
conjugated polyelectrolyte PIFB was introduced as an interlayer solvent, the residue was ltered and washed with methanol
to facilitate efficient electron injection. PIFB was prepared and ethyl acetate, respectively. 1,5-Dibromo-2,6-bis(di-
according to our previous report.19 The PIFB layer (ꢃ1 nm) was bromomethyl)naphthalene (3.60 g, 90%) was obtained as a
1
obtained by spin-coating a methanol solution (0.3 mg mLꢀ1 light yellow powder. H NMR (CDCl3, 400 MHz, ppm): 8.42
containing 30 eq. of acetic acid) on the top of the active layer. (d, J ¼ 9.2 Hz, 2H), 8.21 (d, J ¼ 9.2 Hz, 2H), 7.41 (s, 2H).
Ultimately, about 80 nm of aluminum was deposited on the top
1,5-Bibromonaphthalene-2,6-dicarbaldehyde.50 To a solution
of the PIFB layer through shadow masks by thermal evapora- of 1,5-dibromo-2,6-bis(dibromomethyl)naphthalene (10.00 g,
tion. The active area of the devices was xed at 6 mm2. All 15.82 mmol) in 640 mL of ethanol was added 160 mL of an
polymer:PC71BM blends were dissolved in mixed solvents of aqueous AgNO3 solution (10.76 g, 63.28 mmol) and the result-
chlorobenzene and dichlorobenzene (4/1, v/v) at the concen- ing solution was heated to reux. Aer 2 h, a green precipitate
tration of 20 mg mLꢀ1. The blend ratio was xed at 1 : 4 (poly- was ltered by suction from the hot solution. A yellow solid
mer–PC71BM by weight).
appeared upon the removal of the solvent under reduced pres-
sure. The solid was washed with ethanol–water (4 : 1 by volume)
to obtain 1,5-dibromonaphthalene-2,6-dicarbaldehyde (5.23 g,
96%). 1H NMR (CDCl3, 400 MHz, ppm): 8.60 (d, J ¼ 8.8 Hz, 2H),
8.11 (d, J ¼ 8.8 Hz, 2H), 10.69 (s, 2H).
2.4. Measurement of PSC devices
Device characterization was performed under AM 1.5G irradia-
tion (100 mW cmꢀ2) on an Oriel sol3A simulator (Newport) with
a NREL-certied silicon reference cell. The current density–
voltage curves were measured by a Keithley 2440 source
measurement unit.
1,5-Dibromonaphthalene-2,6-dicarboxylic acid (2). To
a
solution of 1,5-dibromonaphthalene-2,6-dicarbaldehyde (3.80
g, 11.1 mmol) in 120 mL of tert-butanol was added 20 mL of 2-
methyl-2-butene. An aqueous solution of NaClO2 (20.10 g, 222
mmol) and NaH2PO4$2H2O (24.20 g, 155 mmol) in 160 mL of
water was added dropwise over 30 min. The resulting mixture
was stirred for 24 h at room temperature. tert-Butanol was
removed by rotary evaporation, and the resulting mixture was
treated with 30 mL of H2O and washed with 30 mL of hexane. A
white solid was precipitated when the aqueous layer was acid-
ied to pH 1. The precipitate was collected by ltration and
washed with cold water to give 1,5-dibromonaphthalene-2,6-
dicarboxylic acid as a white solid (3.60 g, 85%). 1H NMR (DMSO-
d6, 400 MHz, ppm): 13.92 (s, 2H), 8.45 (d, J ¼ 8.4 Hz, 2H), 7.91
(d, J ¼ 8.4 Hz, 2H). 13C NMR (DMSO-d6, 100 MHz, ppm): 169.28,
137.75, 134.11, 129.42, 127.95, 126.36. HRMS (EI) m/z: calcd for
2.5. Mobility measurements
Hole-only devices (ITO/PEDOT:PSS/polymer:PC71BM/Au) and
electron-only devices (ITO/Al/polymer:PC71BM/Ca/Al) were
fabricated in order to estimate the hole and electron mobilities
of these polymers by the SCLC method. The hole-only devices
were prepared following the same procedures used in photo-
voltaic device fabrication, except that the PIFB/Al cathode was
replaced by the Au (40 nm) electrode. For the electron-only
devices, Al (5 nm) was thermally evaporated on the pre-cleaned
ITO; further, the BHJ lm was coated using the same method as
that used for the solar cell fabrication. Finally, Ca (20 nm) and
Al (100 nm) were deposited on the BHJ lm. The current density
(J) was measured by an Agilent 4155C semiconductor parameter
analyzer. The SCLC carrier mobilities were calculated according
to the following equation:
C
12H6Br2O4: 371.8633; found: 371.8630.
Dimethyl 1,5-dibromonaphthalene-2,6-dicarboxylate (3).
Thionyl chloride (24.00 g, 202 mmol) was added dropwise to a
solution of compound 2 (3.80 g, 10.1 mmol) in methanol
ꢂ
(120 mL) at 0 C. Then the resulting mixture was reuxed for
9
V2
L3
12 h, before cooling to 0 ꢂC. The precipitate was collected by
J ¼ 3r30m
(1)
8
ltration and washed with cold methanol to give compound 3
where 30 is the permittivity of free space (8.85 ꢁ 10ꢀ12 F mꢀ1), 3r is (3.87, 94%) as a white solid. H NMR (CDCl3, 400 MHz, ppm):
the dielectric constant of the polymer (assumed to be 3, which is 8.51 (d, J ¼ 8.8 Hz, 2H), 7.81 (d, J ¼ 8.8 Hz, 2H), 4.03 (s, 6H).
a typical value for conjugated polymers), m is the carrier mobility, HRMS (EI) m/z: calcd for C14H10Br2O4: 399.8946; found:
V is the voltage drop across the device (V ¼ Vappl ꢀ Vr ꢀ Vbi, where 399.8944.
1
Vappl is the applied voltage to the device, Vr is the voltage drop due
Dimethyl 1,5-di(thiophen-2-yl)naphthalene-2,6-dicarboxy-
to the contact resistance and series resistance across the elec- late (4). To a suspension of compound 3 (3.85 g, 9.58 mmol) in
trodes, and Vbi is the built-in voltage due to the difference in work 96 mL of N,N-dimethylformamide (DMF) was added tribu-
function of the two electrodes), and L is the polymer thickness. tyl(thiophen-2-yl)stannane (7.86 g, 21.1 mmol). The mixture was
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13905–13915 | 13907

View Article Online
Journal of Materials Chemistry A
Paper
vigorously stirred and bubbled with N2 for 1 h. Then 51 mg of butoxide (2.30 g, 24.0 mmol) in portions. The reaction mixture
Pd(PPh3)2Cl2 was added to this degassed solution. The mixture was heated at 80 ꢂC for 1 h, followed by the dropwise addition of
was then heated at 85 ꢂC under N2 for 24 h. Aer being cooled to 2-ethylhexyl bromide (4.63 g, 24.0 mmol). Aer the addition, the
ꢂ
room temperature, the reaction mixture was poured into water resulting mixture was heated at 85–90 C for 8 h, then poured
and extracted with CH2Cl2 (3 ꢁ 50 mL). The organic layer was into ice-water and extracted with CH2Cl2 (3 ꢁ 50 mL). The
combined, then washed with water and brine, and then dried organic layer was dried over anhydrous MgSO4 and the solvent
over anhydrous MgSO4. Aer ltration, the ltrate was was removed under reduced pressure to get the crude product.
concentrated under reduced pressure. The residue was washed The crude product was puried by column chromatography on
with methanol to afford compound 4 as a white solid (3.50 g, silica, eluting with petroleum ether, to give a yellow viscous oil
90%). 1H NMR (CDCl3, 400 MHz, ppm): 7.97 (d, J ¼ 8.8 Hz, 2H), (1.20 g, 40%). 1H NMR (CDCl3, 400 MHz, ppm): 8.01 (d, J ¼ 8.0
7.84 (d, J ¼ 8.8 Hz, 2H), 7.55 (dd, J1 ¼ 1.2 Hz, J2 ¼ 5.2 Hz, 2H), Hz, 2H), 7.58 (d, J ¼ 8.0 Hz, 2H), 7.37 (d, J ¼ 8.0 Hz, 2H), 7.10 (d,
7.22 (dd, J1 ¼ 3.6 Hz, J2 ¼ 5.2 Hz 2H), 7.11 (dd, J1 ¼ 1.2 Hz, J2 ¼ J ¼ 8.0 Hz, 2H), 2.00–2.07 (m, 8H), 1.26–0.48 (m, 60H). 13C NMR
5.2 Hz, 2H), 3.72 (s, 6H). HRMS (MALDI) m/z: calcd for (CDCl3, 100 MHz, ppm): 157.22, 147.65, 141.15, 135.14, 129.73,
C
22H16O4S2: 408.0490; found: 408.0485.
125.59, 124.12, 121.93, 118.15, 56.25, 44.48, 35.08, 33.45, 28.39,
1,5-Di(thiophen-2-yl)naphthalene-2,6-dicarboxylic acid (5). 26.59, 22.65, 14.03, 10.31. HRMS (MALDI) m/z: calcd for
To a solution of compound 4 (3.45 g, 8.46 mmol) in ethanol
C
52H76S2: 764.5388; found: 764.5383.
(120 mL), an aqueous solution of NaOH (4.74 g NaOH in 30 mL
Compound 8. n-Butyllithium (5.1 mL, 2.5 M in hexanes) was
of H2O) was added. The resulting mixture was heated at reux added dropwise to a solution of compound aDTN (2.45 g, 3.20
ꢂ
overnight. Aer being cooled to room temperature, the solvent mmol) in 20 mL of THF at ꢀ78 C. The mixture was stirred at
was removed under reduced pressure. Then, the residue was ꢀ78 ꢂC for 1 h. A solution of trimethyltin chloride (2.56 g in 10
added to a concentrated hydrochloric acid solution. The mL of THF) was added dropwise. The reaction mixture was
precipitate was collected by ltration and washed with water allowed to warm to room temperature and was stirred over-
then dried in vacuo to afford a white solid (3.10 g, 96%). 1H NMR night. Then the mixture was poured into water and extracted
(DMSO-d6, 400 MHz, ppm): 13.06 (s, 2H), 7.83 (q, J ¼ 8.8 Hz, with ether. The organic layer was washed with water and dried
4H), 7.78 (d, J ¼ 5.2 Hz, 2H), 7.24 (dd, J1 ¼ 3.6 Hz, J2 ¼ 5.2 Hz over MgSO4. Evaporation of the solvent afforded compound 8 as
2H), 7.16 (d, J ¼ 3.2 Hz, 2H). 13C NMR (DMSO-d6, 100 MHz, a light brownish oil (3.10 g, 89%), which was not further puri-
ppm): 169.30, 137.73, 134.15, 133.77, 131.17, 129.51, 127.90, ed but directly used in the next step. 1H NMR (CDCl3, 400
127.71, 127.67, 126.32. HRMS (MALDI) m/z: calcd for MHz, ppm): 8.07 (d, J ¼ 8.0 Hz, 2H), 7.61 (d, J ¼ 8.0 Hz, 2H), 7.07
C
20H12O4S2: 380.0177; found: 380.0172.
(d, J ¼ 8.0 Hz, 2H), 2.43–2.12 (m, 8H), 1.26–0.46 (m, 60H), 0.39
Compound 6. To a suspension of compound 5 (3.10 g, (s, 18H). 13C NMR (CDCl3, 100 MHz, ppm): 157.64, 150.50,
8.16 mmol) in 100 mL of anhydrous CH2Cl2 was added 4.20 g of 146.72, 139.72, 135.22, 129.95, 126.00, 122.08, 121.66, 53.51,
oxalyl chloride (32.64 mmol) and 0.3 mL of DMF. The resulting 43.34, 34.86, 34.29, 28.42, 27.32, 22.81, 13.97, 10.42, ꢀ8.27. MS
mixture was stirred overnight at room temperature. The solvent m/z: calcd for C52H76S2 [M ꢀ 2Sn(CH3)3] 764.5; found: 764.5.
was removed to afford the intermediate dicarboxylic acid chlo-
PaDTNBTO. The monomer 8 (0.32 g, 0.29 mmol), 4,7-bis(5-
ride as a yellow solid. Then the solid was dissolved in 20 mL of bromothiophen-2-yl)-5,6-bis(hexyloxy)benzo[c][1,2,5]thiadiazole
anhydrous CH2Cl2 again and then added to a suspension of (9) (0.193 g, 0.29 mmol) in 20 mL of toluene, and 2 mL of DMF
AlCl3 (6.50 g, 48.96 mmol) in 50 mL of CH2Cl2 at 0 ꢂC. The was degassed by bubbling with nitrogen for 1 h at room
resulting mixture was allowed to warm to room temperature temperature. Aer the addition of 15 mg of Pd(PPh3)4, the
with stirring overnight, and was then poured into ice-cold mixture was heated to reux for 3 days under an argon atmo-
hydrochloride acid solution. The precipitate was collected by sphere. Then, 2-tributylstannylthiophene (54 mg, 0.15 mmol)
ltration and washed with water, then dried under vacuum to was added to the mixture solution and reacted for 3 h. Finally, 2-
afford a deep blue solid (2.74 g, 98%). 1H NMR (CDCl3, 400 bromothiophene (48 mg, 0.29 mmol) was added to the mixture
MHz, ppm): 7.84 (d, J ¼ 8.0 Hz, 2H), 7.55 (d, J ¼ 8.0 Hz, 2H), 7.26 and reacted overnight to complete the end-capping reaction.
(d, J ¼ 8.0 Hz, 2H), 7.02 (d, J ¼ 8.0 Hz, 2H). HRMS (MALDI) m/z: The mixture was cooled to room temperature and added drop-
calcd for C20H8O2S2: 343.9966; found: 343.9960.
wise into 200 mL of methanol. The solid was collected by
Compound 7. A mixture of compound 6 (2.80 g, 8.14 mmol), ltration and extracted in a Soxhlet setup with methanol,
hydrazine monohydrate (8.14 g, 0.16 mol) and KOH (9.20 g, 0.16 acetone, and hexane for 24 h each. The insoluble remainders
mol) in 150 mL of diethylene glycol was heated at 180 ꢂC for 24 were redissolved in chloroform and reprecipitated into meth-
h. Then the mixture was poured into ice-cold hydrochloride anol. The target polymer was collected by ltration and dried in
1
ꢂ
acid. The precipitate was collected by ltration and washed with vacuo at 50 C overnight to give a black solid (0.30 g, 89%) H
water, then dried under vacuum to afford a pale yellow solid NMR (CDCl3, 400 MHz, ppm): 8.62 (br, 2H), 8.08 (br, 2H), 7.66
(2.50 g, 92%). 1H NMR (CDCl3, 400 MHz, ppm): 8.08 (d, J ¼ 8.0 (br, 2H), 7.46 (m, 2H), 7.39 (br, 2H), 4.27 (br, 4H), 2.10 (br, 8H),
Hz, 2H), 7.77 (d, J ¼ 8.0 Hz, 2H), 7.45 (d, J ¼ 8.0 Hz, 2H), 7.24 (d, 1.47 (br, 8H), 1.05–0.59 (m, 64H). Mn ¼ 54.5 kDa, polydispersity
J ¼ 8.0 Hz, 2H), 3.87 (s, 4H). HRMS (MALDI) m/z: calcd for ¼ 1.83. The low molecular weight polymer (PaDTNBTO) was
C
20H12S2: 316.0380; found: 316.0375.
synthesized using the same procedure but a shorter polymeri-
Compound aDTN. To a suspension of compound 7 (1.25 g, zation time (48 h). Mn ¼ 10.3 kDa, polydispersity ¼ 1.66.
4.0 mmol) in 40 mL of anhydrous DMSO was added sodium tert-
13908 | J. Mater. Chem. A, 2014, 2, 13905–13915
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
PaDTNBTH. Polymer PaDTNBTH was prepared according to
Journal of Materials Chemistry A
the same procedure as that for PaDTNBTO, using monomer 8
(0.32 g, 0.29 mmol) and compound 10 (0.134 g, 0.29 mmol).
1
Yield: 0.23 g (79%). H NMR (CDCl3, 400 MHz, ppm): 8.16 (br,
2H), 8.07 (br, 2H), 7.95 (br, 2H), 7.67 (m, 2H), 7.43 (m, 2H), 7.40
(m, 2H), 2.12 (br, 8H), 1.05–0.59 (m, 60H). Mn ¼ 25.5 kDa,
polydispersity ¼ 1.58.
PaDTNBTF. Polymer PaDTNBTF was prepared according to
the same procedure as that for PaDTNBTO, using monomer 8
(0.32 g, 0.29 mmol) and compound 11 (0.145 g, 0.29 mmol).
Scheme 2 The synthetic route for the aDTN-containing copolymers.
1
Yield: 0.19 g (62%). H NMR (CDCl3, 400 MHz, ppm): 8.34 (br,
2H), 8.05 (br, 2H), 7.67 (br, 2H), 7.46 (br, 2H), 7.44 (br, 2H),
2.14 (br, 8H), 1.05–0.60 (m, 60H). Mn ¼ 15.7 kDa, polydispersity
¼ 1.64.
Compound 8 was synthesized in a 89% yield by the lithium–
hydrogen exchange on aDTN and by reaction with trimethyltin
chloride. Stille coupling reactions between monomer 8 and 4,7-
bis(5-bromothiophen-2-yl)-5,6-bis(hexyloxy)benzo[c][1,2,5]thia-
diazole (9), 4,7-bis(5-bromothiophen-2-yl)benzo[c][1,2,5]thia-
3. Results and discussion
3.1. Synthesis and characterization
diazole
(10),
and
4,7-bis(5-bromothiophen-2-yl)-5,6-
diuorobenzo[c][1,2,5]thiadiazole (11) using palladium tetra-
The synthetic routes for the monomer (8) and the copolymers kis(triphenylphosphine) (Pd(PPh3)4) as the catalyst gave the D–A
are depicted in Schemes 1 and 2. All compounds were synthe- copolymers PaDTNBTO, PaDTNBTH, and PaDTNBTF in 89%,
sized similarly to the method previously reported by us except 79% and 62% yields, respectively. These copolymers were
for the key intermediate 1,5-dibromonaphthalene-2,6-dicar- puried by precipitation from methanol and subsequent
boxylic (2). Compound 2 was obtained by using a simplied Soxhlet extraction with methanol, acetone, hexane, and chlo-
procedure, which started from 1,5-dibromo-2,6-dimethylnaph- roform in sequence. Due to the four 2-ethylhexyl chains being
thalene through a three-step process successively involving introduced into both sides of the aDTN unit, all copolymers
benzylic radical dibromination to 1,5-dibromo-2,6-bis(di- show good solubility in chlorinated solvents such as chloro-
bromomethyl)naphthalene, hydrolysis into the corresponding form, chlorobenzene, and dichlorobenzene at room tempera-
1,5-dibromonaphthalene-2,6-dicarbaldehyde, and then oxida- ture. The number-average molecular weights (Mn) of
tion into the corresponding carboxylic acid, to give a 73% PaDTNBTO, PaDTNBTH, and PaDTNBTF, are 54.5 kg, 25.5 kg,
overall yield. Compound 3 was obtained by an esterication and 15.7 kg, respectively, with relatively narrow polydispersity
reaction under the catalytic action of SOCl2 in a 94% yield. The indices (PDIs) in the range of 1.58–1.83 (Table 1), which were
Stille coupling reaction between compound 3 and tributyl(th- estimated by gel permeation chromatography (GPC) using THF
iophen-2-yl)stannane gave compound
4 in a 90% yield. as the eluent and polystyrene as the internal standard.
Compound 4 was hydrolyzed to compound 5, which was
subsequently converted into an acid chloride derivative by
reacting with oxalyl chloride. Further, intramolecular Friedel–
Cras acylation of the acid chloride derivative afforded a ketone
(6) in a 98% yield. Compound 7 was obtained aer a Wolf–
Kishner reduction of compound 6 in a 92% yield. The alkylation
of compound 7 afforded aDTN with four 2-ethylhexyl groups.
3.2. Optical properties
UV-vis absorption spectra of the copolymers in chlorobenzene
solutions (1 ꢁ 10ꢀ5 M) and in thin lms are shown in Fig. 2 and
the optical parameters are summarized in Table 1. All the three
copolymers exhibit one main broad absorption band in both
dilute solutions and thin lms. As shown in Fig. 2a, incorpo-
rating different substituents onto the BT unit have signicant
effects on the optical properties of the copolymers. For example,
the absorption maximum of PaDTNBTH in chlorobenzene is
located at 591 nm. Aer introducing two electron-donating
hexyloxy groups, the absorption maximum of PaDTNBTO blue-
shis about 26 nm (lmax ¼ 565 nm). Interestingly, by adding two
F atoms onto the BT unit, PaDTNBTF also shows a blue shi of
about 5 nm compared to the nonuorinated copolymer
PaDTNBTH, indicating that the uorine substitution on the BT
unit has a certain inuence on the optical properties of the
resulting D–A copolymers. Similar results have also been found
in some other uorinated copolymer systems.12,51–54 In going
from solution to a solid lm (Fig. 2b), negligible changes were
observed in the absorption proles, except that there are 10–32
nm red-shied absorption peaks for the lms. The optical
Scheme 1 Synthesis of the monomer.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13905–13915 | 13909

View Article Online
Journal of Materials Chemistry A
Paper
Table 1 Summary of the molecular weights and optical and electrochemical properties of the copolymers
lmax
lmax
Polymer
Mn (kg molꢀ1
)
PDI
(nm) in chlorobenzene
(nm) in lm
Eogpta (eV)
HOMOb (eV)
LUMOc (eV)
Eegcd (eV)
PaDTNBTO
PaDTNBTH
PaDTNBTF
54.5
25.5
15.7
1.83
1.58
1.64
565
591
586
575
623
617
1.90
1.79
1.81
ꢀ5.40
ꢀ5.34
ꢀ5.38
ꢀ3.57
ꢀ3.58
ꢀ3.59
1.83
1.76
1.79
a
Estimated from the onset of the absorption spectra of thin lms. b EHOMO ¼ ꢀ(4ox + 4.82) eV. c ELUMO ¼ ꢀ(4red + 4.82) eV. d Eegc ¼ electrochemical
band gap (LUMO–HOMO).
Fig. 3 Cyclic voltammogram (CV) of the copolymer films on a plat-
inum electrode at a scan rate of 100 mV sꢀ1
.
were estimated according to the following equations, with the
detailed data summarized in Table 1.
EHOMO ¼ ꢀ(4ox + 4.82)(eV)
ELUMO ¼ ꢀ(4red + 4.82)(eV)
Eegc ¼ (4ox ꢀ 4red)(eV)
(2)
(3)
(4)
Fig. 2 Absorption spectra of the target copolymers in chlorobenzene
(a) and in solid film (b).
As shown in Fig. 3, all the three copolymers exhibit reversible
anodic oxidation and irreversible cathodic reduction curves in
cyclic voltammetry measurements. For PaDTNBTH, the 4ox and
the 4red are 0.52 V and ꢀ1.24 V, respectively, corresponding to a
HOMO level of ꢀ5.34 eV and a LUMO level of ꢀ3.58 eV. Aer
adding two F atoms, PaDTNBTF exhibits a relatively deeper
HOMO level of ꢀ5.38 eV, whereas the LUMO level (ꢀ3.59 eV) is
almost unchanged. Similar trends in the HOMO level change
have also been found in some other uorinated copolymer
bandgaps (Eogpt) deduced from the absorption edges (ledge) of
the polymer lms are 1.90, 1.79, and 1.81 eV for PaDTNBTO,
PaDTNBTH, and PaDTNBTF, respectively.
3.3. Electrochemical properties
The molecular frontier orbitals of the target copolymers were systems.55,56 Interestingly, when the F atom was exchanged to
investigated by the cyclic voltammetry (CV) method in lms on a the hexyloxy group, PaDTNBTO shows a deeper HOMO level of
Pt disk working electrode in 0.1 M Bu4NPF6 acetonitrile solu- ꢀ5.40 eV and an equivalent LUMO level of ꢀ3.57 eV compared
tions at a scan rate of 100 mV sꢀ1. Under this condition, the to that of PaDTNBTF, which could be attributed to the
onset oxidation potential (E1/2ox) of ferrocene was ꢀ0.02 V versus decreased charge transfer from the donor moiety to the acceptor
Ag/Ag+. The CV curves were calibrated using ferrocene as the moiety (BT) with the incorporation of two electron-donating
standard, whose HOMO energy level was set at ꢀ4.80 eV. Fig. 3 hexyloxy groups. Clearly, the HOMO energy levels of all copol-
shows the cyclic voltammetry for these copolymers. From the ymers are low enough to ensure good stability and a high Voc in
onset oxidation (4ox) and the reduction potentials (4red) versus PSCs. As we know, Voc is positively correlated with the offset
Ag/Ag+ in the cyclic voltammogram, the HOMO and LUMO between the HOMO energy levels of donor materials and the
energy levels, as well as the band gaps (Eegc) of the copolymers, LUMO energy levels of acceptor materials.57 Meanwhile, the
13910 | J. Mater. Chem. A, 2014, 2, 13905–13915
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
LUMO level difference between the polymers and PC71BM as Voc, short-circuit current density (Jsc), ll factor (FF), and PCE
(ꢀ4.3 eV) are ꢃ0.7 eV, which means that electron transfer from are summarized in Table 2.
the polymers to PC71BM is energetically favorable. The band
The donor–acceptor (D–A) ratio plays an important role in
gaps calculated from the difference between the onset oxidation the performance of PSCs, because a balance in the hole and
and reduction potentials are 1.83, 1.76, and 1.79 eV for electron transport is necessary to avoid any accumulation of
PaDTNBTO, PaDTNBTH, and PaDTNBTF, in that order. The charge and thus to facilitate the charge transport process during
results are in agreement with its optical band gaps, within the photon-to-current conversion.37,59 Therefore, we fabricated
experimental error.
PSC devices based on PaDTNBTO:PC71BM with different blend
ratios to optimize the device performance. Fig. 5 shows the J–V
curves and the external quantum efficiency (EQE) spectra of the
devices with different blend ratios (D–A ¼ 1 : 2, 1 : 3, 1 : 4, 1 : 5),
with the detailed parameters summarized in Table 2. In Table 2,
the device parameters of the PDTNTBT-based solar cell are lis-
ted as a reference.19 It should be mentioned that the only
difference in PaDTNBTO and PDTNTBT is the donor unit (Fig. 1
and Scheme 2). However, the device based on PaDTNBTO
(blend ratio of 1 : 4) gives the best PCE of 6.44% with a Voc of
0.92 V, a Jsc of 10.91 mA cmꢀ2 and an FF of 64.1%, which is
higher than the best PCE of 4.78% with a Voc of 0.86 V, a Jsc of
10.29 mA cmꢀ2 and an FF of 53.8% for the PDTNTBT-based
device. The EQE spectra (Fig. 5b) show that the device with the
1 : 4 blend ratio has higher EQE values in the range from 360 to
610 nm, compared to those of the devices with other blend
ratios. In order to reveal the impact of D–A ratio on the photo-
voltaic parameters, the Voc, Jsc, FF and PCE data are plotted in
Fig. 6a–d as functions of the blend ratio. It can be seen that the
3.4. Hole mobilities and photovoltaic properties
The hole mobilities of the copolymers were investigated by
using the space-charge limited current (SCLC) method with a
device structure of ITO/PEDOT:PSS (40 nm)/polymer:PC71BM/
Au (40 nm). The linear ts for the plots of J0.5 versus V (V ¼ Vappl
ꢀ Vr ꢀ Vbi, where Vappl is the applied voltage to the device, Vr is
the voltage drop due to contact resistance and series resistance
across the electrodes, and Vbi is the built-in voltage due to the
difference in work function of the two electrodes) are shown in
Fig. 4 and the data are collected in Table S1.† The hole mobil-
ities are estimated to be 2.81 ꢁ 10ꢀ5, 3.69 ꢁ 10ꢀ5, and 7.5 ꢁ
10ꢀ6 cm2 Vꢀ1 sꢀ1 for the blends of PaDTNBTO, PaDTNBTH, and
PaDTNBTF with PC71BM, respectively. Obviously, as donor
materials, PaDTNBTO and PaDTNBTH will have some advan-
tages in the photovoltaic application compared to PaDTNBTF,
due to their higher hole mobilities.
The photovoltaic properties of the resulting copolymers were
investigated with a conventional device structure of ITO/
PEDOT:PSS (40 nm)/polymer:PC71BM/PIFB/Al (80 nm). Here, a
thin layer of PIFB was employed as the cathode interlayer.19
PC71BM was chosen as the electron acceptor due to its broader
and stronger absorption in the visible region, which is
complementary to the absorption valley of the polymer and
benecial for high efficient PSCs.58 The active layers were
prepared by spin coating the polymer:PC71BM blends from
mixed solvents (chorobenzene–o-dichorobenzene ¼ 4 : 1, v/v)
on the top of the PEDOT:PSS layer without any further treat-
ment. The measurements were performed under simulated
AM1.5G, 100 mW cmꢀ2 illumination with an active area of 0.06
cm2. The current density–voltage (J–V) characteristics of these
PSCs are shown in Fig. 5 and 7, and the device parameters such
Fig. 5 (a) Current density (J)–voltage (V) characteristics, and (b) EQE
Fig. 4 J0.5–V characteristics of hole-only devices based on different curves of BHJ solar cells based on PaDTNBTO with various blend
polymers.
ratios.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13905–13915 | 13911

View Article Online
Journal of Materials Chemistry A
Paper
Table 2 Device parameters of PSCs based on three copolymers
D–A
Polymers
ratio (w/w)
Voc (V)
Jsc (mA cmꢀ2
)
FF (%)
PCEavea (%)
PCEmax (%)
Ref.
PaDTNBTO
1 : 2
1 : 3
1 : 4
1 : 5
1 : 4b
1 : 4
1 : 4
1 : 3
0.92
0.91
0.92
0.91
0.89
0.90
0.91
0.86
9.98 ꢄ 0.15
10.62 ꢄ 0.04
10.62 ꢄ 0.29
8.28 ꢄ 0.62
10.55 ꢄ 0.24
11.12 ꢄ 0.05
3.95 ꢄ 0.04
10.29
51.2 ꢄ 0.4
56.7 ꢄ 0.8
63.9 ꢄ 0.2
61.2 ꢄ 0.4
52.0 ꢄ 0.7
51.3 ꢄ 0.5
47.2 ꢄ 0.6
53.8
4.66 ꢄ 0.14
5.47 ꢄ 0.13
6.29 ꢄ 0.15
4.62 ꢄ 0.39
4.81 ꢄ 0.22
5.13 ꢄ 0.09
1.70 ꢄ 0.03
—
4.80
5.60
6.44
5.01
5.03
5.22
1.73
4.78
This work
This work
This work
This work
This work
This work
This work
19
PaDTNBTH
PaDTNBTF
PDTNTBT
a
The average PCE is obtained from eight devices. b PSC devices based on low molecular weight PaDTNBTO (Mn ¼ 10.3 kg molꢀ1; PDI ¼ 1.7).
blend ratio signicantly affects the Jsc and FF, although the Voc contents is probably attributed to the higher degree of PC71BM
is nearly unchanged regardless of the blend ratio. For example, aggregation. For low PC71BM content lm (D–A ¼ 1 : 2,
when the blend ratio increases from 1 : 2 to 1 : 4, the Jsc of the Fig. S4b†), PC71BM is dispersed into the polymer phase, and
devices gradually increases from 10.13 to 10.93 mA cmꢀ2, and thus, the blend lm did not form phase separation with an ideal
the FF of the device increases from 51.6% to 64.1%, while a high domain size. In the case of the 1 : 5 blend ratio (Fig. S4h†), the
Voc of 0.92 V is retained. This effect is probably due to the more phase separation is not optimal because of the over-dominant
balanced hole and electron mobilities in the blend lms with PC71BM in the polymer blend. For the blend lms with D–A ¼
high PC71BM contents, which prevents the accumulation of 1 : 3 (Fig. S4d†) and 1 : 4 (Fig. S4f†) ratios, the lms exhibit
charges. Moreover, the active layer of high PC71BM content phase separation with an optimal domain size of 10–20 nm,
devices may have a bicontinuous phase separation with the which is favorable for efficient exciton dissociation and charge
appropriate domain size, which affords efficient exciton diffu- transport. Moreover, we measured the electron and hole
sion and charge separation. When the blend ratio further mobilities by using the SCLC method to investigate the charge
increases from 1 : 4 to 1 : 5, both the Jsc and FF of the devices transport properties of the blend lms with different blend
decrease, leading to a reduced PCE of 5.01%. Therefore, the ratios. Fig. S3† shows the dark current density–voltage (Jdark–V)
blend ratio of 1 : 4 was chosen as the optimal blend ratio for the characteristics of the hole-only devices and the electron-only
other polymer blends.
devices with different blend ratios, with the detailed results
In order to verify the above hypothesis, the morphology of summarized in Table S2.† It could be found that the hole and
the polymer blends with different blend ratios (D–A ¼ 1 : 2, the electron mobilities are simultaneously improved with an
1 : 3, 1 : 4, and 1 : 5) were investigated using tapping mode increase in the PC71BM content in the PaDTNBTO:PC71BM
AFM. According to the topography images of the blend lms, as blend lms (D–A ratio from 1 : 2 to 1 : 4). When the PC71BM
shown in Fig. S4,† the root-mean-square (RMS) values of the content was further increased (D–A ¼ 1 : 5), the electron
PaDTNBTO:PC71BM blends at weight ratios of 1 : 2, 1 : 3, 1 : 4, mobility continues to increase but the hole mobility starts to
and 1 : 5 are 0.52, 1.37, 1.87, and 2.15 nm, respectively. The decrease. Finally, the electron and hole mobilities of the blend
larger RMS value in the blend lms with higher PC71BM are more balanced at the weight ratio of 1 : 4, and are 5.26 ꢁ
10ꢀ5 and 2.81 ꢁ 10ꢀ5 cm2 Vꢀ1 sꢀ1, respectively. When the FF
values of the PSC devices are correlated with the blend ratios, it
can be concluded that D–A ratios in the blend lm play an
important role in realizing balanced hole and electron mobil-
ities, which is the key to achieving a high FF.
Under the same blend ratio of 1 : 4, as shown in Fig. 7a, the
PSC device based on PaDTNBTF exhibited a quite low PCE of
1.73% with Voc ¼ 0.91 V, Jsc ¼ 3.99 mA cmꢀ2, and FF ¼ 47.8%.
However, dramatically improved photovoltaic performance was
achieved for the device based on PaDTNBTH, which exhibited a
Voc of 0.90 V, a slightly increased FF of 51.8%, and a signicantly
improved Jsc of 11.17 mA cmꢀ2, resulting in a moderate PCE of
5.22%. The best photovoltaic performance was obtained for the
device based on PaDTNBTO, which showed a further improved
FF of 64.1%, Voc of 0.92 V, and a slightly decreased Jsc of
10.91 mA cmꢀ2, leading to a high PCE of 6.44%. It should be
noted that such a promising performance was achieved without
the use of any pre- and/or post-treatments, such as thermal
Fig. 6 The photovoltaic parameters of the PSCs based on PaDTNB-
TO:PC71BM with different blend ratios.
13912 | J. Mater. Chem. A, 2014, 2, 13905–13915
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
3.5. External quantum efficiency
The external quantum efficiency (EQE) curves for the polymer-
based devices under monochromatic light illumination are
shown in Fig. 7b. As shown in the gure, the EQE curves for all
the devices are rather at in the wavelength range from 400 to
600 nm. The device based on PaDTNBTO exhibits a similar
average EQE value to that based on PaDTNBTH (44.4% versus
44.7%) in the range of 300–750 nm, and both of them are much
higher than that based on PaDTNBTF (16.5%). This result is
consistent with the higher Jsc for the PaDTNBTO- or
PaDTNBTH-based devices compared to that for the PaDTNBTF-
based device. In addition, the photoresponse ranges for the
devices based on PaDTNBTH and PaDTNBTF are slightly
broader compared to that for the PaDTNBTO-based device,
which is in agreement with the optical absorption spectra of the
corresponding polymer:PC71BM blends in the same blend ratio
(Fig. S1†).
3.6. Morphology
For polymer solar cells, the morphological control of the BHJ
active layer is of great importance in charge generation, sepa-
ration, and transport, which could inuence nal device
performances. In this work, the surface morphological
Fig. 7 (a) J–V characteristics and (b) EQE curves of BHJ solar cells
based on the three copolymers.
annealing, solvent annealing, or processing additives. In our
case, utilization of the different treatment methodologies only
has negative effects on the device performance for the
PaDTNBTO-based devices. The device parameters for all the
devices fabricated in the different conditions are shown in
Fig. S2 and Table S3.†
It was found that the Jsc and FF values of the PaDTNBTH- and
PaDTNBTO-based devices are much higher than those based on
PaDTNBTF, which could be ascribed to the optimal morphology
with an appropriate domain size of the active layers based on
PaDTNBTH and PaDTNBTO, as well as their higher charge
carrier mobilities that promote charge transport in devices. As
shown in Table 2, an increase in the molecular weight of
PaDTNBTO leads to an improved device performance. When the
Mn increases from 10.3 kg to 54.5 kg, maximum PCE increases
from 5.03% to 6.44%, mainly due to the higher FF (64.1% versus
52.2%). These results have also been observed in previously
reported literature.60 Considering the fact that the molecular
weight plays an important role in the device performance, we
believe that there is still room to further improve the perfor-
mance of the devices based on these copolymers through an
optimization of their molecular weights. As we expected, devices
based on aDTN-containing copolymers show a relatively higher
Voc in the range of 0.90–0.92 V compared to those based on
linear-shaped DTN-containing copolymers (0.75–0.87 V),19
which can be attributed to their deeper-lying HOMO
energy levels.
Fig. 8 Tapping mode AFM topography (left) and phase (right) images
of blend films spin coated from polymer–PC71BM (1 : 4, w/w) (a)
PaDTNBTO:PC71BM, (b) PaDTNBTH:PC71BM, (c) PaDTNBTF:PC71BM
(all image sizes are 0.5 mm ꢁ 0.5 mm).
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13905–13915 | 13913

View Article Online
Journal of Materials Chemistry A
Paper
structures of the polymer–PC71BM (1 : 4, w/w) blend lms were 2012H0044) and the 100 Talents Programme of the Chinese
evaluated by using tapping mode AFM. The AFM topography Academy of Sciences (CAS).
and phase images are shown in Fig. 8. As can be seen in Fig. 8,
phase separation was clearly observed for all the three polymer
blends. Due to the structural differences in the three copoly-
mers, the phase separation images for the blended lms look
Notes and references
different. For example, the lm of the PaDTNBTF:PC71BM blend
does not show uniform phase separation with many domain
sizes over 40 nm, which is much larger than the typical exciton
diffusion length of ꢃ10 nm.1,61 This can be accounted for by the
low FF of 47.8% found in the PaDTNBTF-based device. The
phase image of the lm from PaDTNBTH blended with PC71BM
displays a very smooth surface without any clear phase sepa-
ration, which is also unfavorable for effective charge separation
and transportation. In the case of the PaDTNBTO:PC71BM
blend, a more clear phase separation could be observed with a
domain size of ꢃ20 nm. This is benecial for the efficient
exciton diffusion to the donor-acceptor interface, and thus
contributes to the high FF of 64.1% for the PaDTNBTO-based
device. The obtained root-mean-square roughness (RMS) values
are 1.87, 0.72, and 0.25 nm for the blend lms of PaDTNB-
TO:PC71BM, PaDTNBTF:PC71BM and PaDTNBTH:PC71BM,
respectively. The difference in the RMS value can be attributed
1 G. Yu, J. Gao, J. Hummelen, F. Wudl and A. J. Heeger, Science,
1995, 270, 1789–1790.
2 J. Huang, Z. Yin and Q. Zheng, Energy Environ. Sci., 2011, 4,
3861–3877.
3 Y.-J. Cheng, S.-H. Yang and C.-S. Hsu, Chem. Rev., 2009, 109,
5868–5923.
4 Z. Yin and Q. Zheng, Adv. Energy Mater., 2012, 2, 179–218.
5 J. Chen and Y. Cao, Acc. Chem. Res., 2009, 42, 1709–1718.
¨
6 O. InganAs, F. Zhang and M. R. Andersson, Acc. Chem. Res.,
2009, 42, 1731–1739.
¨
7 Z. B. Henson, K. Mullen and G. C. Bazan, Nat. Chem., 2012, 4,
699–704.
8 G. Li, R. Zhu and Y. Yang, Nat. Photonics, 2012, 6, 153–161.
9 Y. Li, Acc. Chem. Res., 2012, 45, 723–733; I. McCulloch,
R. S. Ashraf, L. Biniek, H. Bronstein, C. Combe,
J. E. Donaghey, D. I. James, C. B. Nielsen, B. C. Schroeder
and W. Zhang, Acc. Chem. Res., 2012, 45, 714–722.
to the different substituents in the BT unit. Because the longer 10 L. Ye, S. Zhang, L. Huo, M. Zhang and J. Hou, Acc. Chem. Res.,
and bulkier side chains will create more disorder in the lm 2014, 47, 1595–1603.
self-organization, it is not surprising that the lm based on the 11 S.-H. Liao, H.-J. Jhuo, Y.-S. Cheng and S.-A. Chen, Adv.
copolymers with two hexyloxy groups is the coarsest, and that
Mater., 2013, 25, 4766–4771.
based on the copolymers without any substitutes is the 12 J. You, L. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty,
smoothest.
K. Emery, C.-C. Chen, J. Gao, G. Li and Y. Yang, Nat.
Commun., 2013, 4, 1446.
13 Z. He, C. Zhong, X. Huang, W.-Y. Wong, H. Wu, L. Chen,
S. Su and Y. Cao, Adv. Mater., 2011, 23, 4636–4643.
4. Conclusions
In summary, a novel ladder-type angular-shaped electron-rich 14 L. Wang, D. Cai, Q. Zheng, C. Tang, S.-C. Chen and Z. Yin,
unit (aDTN) has been designed and synthesized. Three D–A ACS Macro Lett., 2013, 2, 605–608.
copolymers based on aDTN exhibit medium band-gaps in the 15 R. Qin, W. Li, C. Li, C. Du, C. Veit, H.-F. Schleiermacher,
¨
range of 1.79–1.90 eV, along with deep-lying HOMO levels
below ꢀ5.34 eV. As a result of the deep-lying HOMO level, BHJ
M. Andersson, Z. Bo, Z. Liu, O. Inganaas, U. Wuerfel and
F. Zhang, J. Am. Chem. Soc., 2009, 131, 14612–14613.
photovoltaic devices derived from these polymers have a 16 S. H. Park, A. Roy, S. Beaupre, S. Cho, N. Coates, J. S. Moon,
´
relatively high Voc over 0.9 V. In particular, the photovoltaic
device based on PaDTNBTO exhibits the best efficiency of
D. Moses, M. Leclerc, K. Lee and A. J. Heeger, Nat. Photonics,
2009, 3, 297–302.
6.44%, which is higher than that of the device based on linear- 17 C. E. Small, S. Chen, J. Subbiah, C. M. Amb, S.-W. Tsang,
shaped DTN-containing copolymers (PDTNTBT).19 These
preliminary results demonstrate that aDTN is a more prom-
T.-H. Lai, J. R. Reynolds and F. So, Nat. Photonics, 2012, 6,
115–120.
ising electron-rich building block in constructing p-type 18 S. Albrecht, S. Janietz, W. Schindler, J. Frisch, J. Kurpiers,
copolymers for solar cells with high open circuit voltages
J. Kniepert, S. Inal, P. Pingel, K. Fostiropoulos, N. Koch
compared to the linear-shaped DTN. Compared with other
and D. Neher, J. Am. Chem. Soc., 2012, 134, 14932–14944.
large band-gap materials such as P3HT, PaDTNBTO shows an 19 Y. Ma, Q. Zheng, Z. Yin, D. Cai, S.-C. Chen and C. Tang,
improved performance in Voc, which makes it a promising Macromolecules, 2013, 46, 4813–4821.
candidate for the short wavelength absorbing materials used 20 Q. Zheng, B. J. Jung, J. Sun and H. E. Katz, J. Am. Chem. Soc.,
for tandem PSCs.
2010, 132, 5394–5404.
21 Y. Ma, Y. Sheng, C. Tang, X. Hong, S.-C. Chen, D. Zhu and
Q. Zheng, Sens. Actuators, B, 2013, 176, 132–140.
22 D. Cai, Q. Zheng, S.-C. Chen, Q. Zhang, C.-Z. Lu, Y. Sheng,
D. Zhu, Z. Yin and C. Tang, J. Mater. Chem., 2012, 22,
16032–16040.
Acknowledgements
This work was supported by the National Science Foundation of
China (no. 51173186, 61325026), the National Basic Research
973 Program (no. 2011CB935904), the Key Project of Depart- 23 Q. Zheng, S. Chen, B. Zhang, L. Wang, C. Tang and
ment of Science and Technology of Fujian Province (no.
H. E. Katz, Org. Lett., 2010, 13, 324–327.
13914 | J. Mater. Chem. A, 2014, 2, 13905–13915
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
24 B. C. Schroeder, Z. Huang, R. S. Ashraf, J. Smith, P. D'Angelo, 43 R. Rieger, D. Beckmann, W. Pisula, W. Steffen, M. Kastler
¨
and K. Mullen, Adv. Mater., 2010, 22, 83–86.
S. E. Watkins, T. D. Anthopoulos, J. R. Durrant and
I. McCulloch, Adv. Funct. Mater., 2012, 22, 1663–1670.
¨
44 D. Muhlbacher, M. Scharber, M. Morana, Z. Zhu, D. Waller,
25 Z. Fei, R. S. Ashraf, Z. Huang, J. Smith, R. J. Kline,
R. Gaudiana and C. Brabec, Adv. Mater., 2006, 18, 2884–2889.
P. D'Angelo, T. D. Anthopoulos, J. R. Durrant, I. McCulloch 45 J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses,
and M. Heeney, Chem. Commun., 2012, 48, 2955–2957. A. J. Heeger and G. C. Bazan, Nat. Mater., 2007, 6, 497–500.
26 B. C. Schroeder, R. S. Ashraf, S. Thomas, A. J. P. White, 46 S. Zeng, L. Yin, C. Ji, X. Jiang, K. Li, Y. Li and Y. Wang, Chem.
L. Biniek, C. B. Nielsen, W. Zhang, Z. Huang, Commun., 2012, 48, 10627–10629.
P. S. Tuladhar, S. E. Watkins, T. D. Anthopoulos, 47 X. Zhao, C. Piliego, B. Kim, D. A. Poulsen, B. Ma, D. A. Unruh
´
J. R. Durrant and I. McCulloch, Chem. Commun., 2012, 48,
7699–7701.
27 C.-P. Chen, S.-H. Chan, T.-C. Chao, C. Ting and B.-T. Ko, J.
Am. Chem. Soc., 2008, 130, 12828–12833.
and J. M. J. Frechet, Chem. Mater., 2010, 22, 2325–2332.
48 N. Cho, K. Song, J. K. Lee and J. Ko, Chem.–Eur. J., 2012, 18,
11433–11439.
49 C. B. Nielsen, J. Arnbjerg, M. Johnsen, M. Jørgensen and
P. R. Ogilby, J. Org. Chem., 2009, 74, 9094–9104.
28 Y.-J. Cheng, J.-S. Wu, P.-I. Shih, C.-Y. Chang, P.-C. Jwo,
W.-S. Kao and C.-S. Hsu, Chem. Mater., 2011, 23, 2361–2369. 50 E. Preis and U. Scherf, Macromol. Rapid Commun., 2006, 27,
29 Y.-J. Cheng, C.-H. Chen, Y.-S. Lin, C.-Y. Chang and C.-S. Hsu,
Chem. Mater., 2011, 23, 5068–5075.
30 Y.-L. Chen, C.-Y. Chang, Y.-J. Cheng and C.-S. Hsu, Chem.
Mater., 2012, 24, 3964–3971.
1105–1109.
51 H. Bronstein, J. M. Frost, A. Hadipour, Y. Kim, C. B. Nielsen,
R. S. Ashraf, B. P. Rand, S. Watkins and I. McCulloch, Chem.
Mater., 2013, 25, 277–285.
31 Y.-X. Xu, C.-C. Chueh, H.-L. Yip, F.-Z. Ding, Y.-X. Li, C.-Z. Li, 52 D. Dang, W. Chen, R. Yang, W. Zhu, W. Mammo and
X. Li, W.-C. Chen and A. K. Y. Jen, Adv. Mater., 2012, 24,
6356–6361.
E. Wang, Chem. Commun., 2013, 49, 9335–9337.
53 R. Kroon, R. Gehlhaar, T. T. Steckler, P. Henriksson,
¨
32 J. J. Intemann, K. Yao, H.-L. Yip, Y.-X. Xu, Y.-X. Li,
P.-W. Liang, F.-Z. Ding, X. Li and A. K. Y. Jen, Chem.
Mater., 2013, 25, 3188–3195.
C. Muller, J. Bergqvist, A. Hadipour, P. Heremans and
M. R. Andersson, Sol. Energy Mater. Sol. Cells, 2012, 105,
280–286.
33 X. Xu, P. Cai, Y. Lu, N. S. Choon, J. Chen, B. S. Ong and X. Hu, 54 Y. Zhang, S.-C. Chien, K.-S. Chen, H.-L. Yip, Y. Sun,
Macromol. Rapid Commun., 2013, 34, 681–688.
34 H. Usta, C. Risko, Z. Wang, H. Huang, M. K. Deliomeroglu,
J. A. Davies, F.-C. Chen and A. K. Y. Jen, Chem. Commun.,
2011, 47, 11026–11028.
A. Zhukhovitskiy, A. Facchetti and T. J. Marks, J. Am. Chem. 55 S. C. Price, A. C. Stuart, L. Yang, H. Zhou and W. You, J. Am.
Soc., 2009, 131, 5586–5608. Chem. Soc., 2011, 133, 4625–4631.
35 P. Sakthivel, H. S. Song, N. Chakravarthi, J. W. Lee, Y.-S. Gal, 56 H. Zhou, L. Yang, A. C. Stuart, S. C. Price, S. Liu and W. You,
S. Hwang and S.-H. Jin, Polymer, 2013, 54, 4883–4893. Angew. Chem., Int. Ed., 2011, 123, 3051–3054.
36 J. Cao, Q. Liao, X. Du, J. Chen, Z. Xiao, Q. Zuo and L. Ding, 57 M. C. Scharber, D. Wuhlbacher, M. Koppe, P. Denk,
Energy Environ. Sci., 2013, 6, 3224–3228.
37 X. Guo, M. Zhang, J. Tan, S. Zhang, L. Huo, W. Hu, Y. Li and
J. Hou, Adv. Mater., 2012, 24, 6536–6541.
38 Z. Zhao, F. Zhang, X. Zhang, X. Yang, H. Li, X. Gao, C.-a. Di
and D. Zhu, Macromolecules, 2013, 46, 7705–7714.
C. Waldauf, A. J. Heeger and C. L. Brabec, Adv. Mater.,
2006, 18, 789–794.
58 M. M. Wienk, J. M. Kroon, W. J. H. Verhees, J. Knol,
J. C. Hummelen, P. A. van Hal and R. A. J. Janssen, Angew.
Chem., Int. Ed., 2003, 115, 3493–3497.
39 H. Okamoto, N. Kawasaki, Y. Kaji, Y. Kubozono, A. Fujiwara 59 P. W. M. Blom, V. D. Mihailetchi, L. J. A. Koster and
and M. Yamaji, J. Am. Chem. Soc., 2008, 130, 10470–10471. D. E. Markov, Adv. Mater., 2007, 19, 1551–1566.
40 I. Osaka, T. Abe, S. Shinamura and K. Takimiya, J. Am. Chem. 60 T.-Y. Chu, J. Lu, S. Beaupre, Y. Zhang, J.-R. Pouliot, J. Zhou,
´
Soc., 2011, 133, 6852–6860.
41 S. Shinamura, I. Osaka, E. Miyazaki, A. Nakao,
A. Najari, M. Leclerc and Y. Tao, Adv. Funct. Mater., 2012, 22,
2345–2351.
M. Yamagishi, J. Takeya and K. Takimiya, J. Am. Chem. 61 J. J. M. Halls, C. A. Walsh, N. C. Greenham, E. A. Marseglia,
Soc., 2011, 133, 5024–5035.
42 R. Rieger, D. Beckmann, A. Mavrinskiy, M. Kastler and
R. H. Friend, S. C. Moratti and A. B. Holmes, Nature, 1995,
376, 498–500.
¨
K. Mullen, Chem. Mater., 2010, 22, 5314–5318.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13905–13915 | 13915