Tuning optical and electronic properties of star-shaped conjugated molecules with enlarged π-delocalization for organic solar cell application

Article abstract of DOI:10.1039/c3ta11001j

Three structurally related conjugated molecules (BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2) in star shape have been designed and synthesized as donor materials for small molecule based bulk heterojunction (BHJ) solar cells. The structural features of these molecules include a planarized benzo[1,2-b:3,4-b′:5,6- b″]trithiophene (BTT) with a C_3h symmetry as the central core and three conjugated arms incorporating electron deficient benzo[2,1,3] thiadiazole (BTD) units, with arms being linked to the core via different number of thiophene connecting units (e.g., 0, 1, 2 corresponding to BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2, respectively). Comparative analyses of optical and electronic properties indicate that the molecules bearing more thiophene units between the BTT core and the BTD arms possess higher-lying HOMO levels while their LUMO levels remain almost unchanged. The improvement of BHJ device performance, with [6,6]-phenyl-C₆₁-butyric acid methyl ester (PC ₆₁BM) as the acceptor, is observed with increasing number of thiophene units between the BTT core and BTD arms, from BTT-BTD-0 to BTT-BTD-1 and BTT-BTD-2. The BTT-BTD-2:PC₆₁BM based BHJ devices show the highest power conversion efficiency (PCE) of 0.74%, with an open-circuit voltage (V_oc) of 0.69 V, a short-circuit current density (J_sc) of 2.93 mA cm^-2, and a fill factor (FF) of 0.37 under 1 sun (100 mW cm^-2) AM 1.5G simulated solar illumination. The PV performance of BTT-BTD-2 is further improved when [6,6]-phenyl-C₇₁-butyric acid methyl ester (PC₇₁BM) is used as the electron acceptor, yielding the best device performance with J_sc of 4.13 mA cm^-2, V _oc of 0.72 V, FF at 0.46 and PCE of 1.36%. The effect of the different number of thiophenes linking the BTT core and the conjugated BTD arms has been clearly demonstrated on regulating optical and electrochemical properties of the three molecules and their BHJ device performances.

Full text of DOI:10.1039/c3ta11001j

Journal of
Materials Chemistry A
View Article Online
View Journal
PAPER
Tuning optical and electronic properties of star-shaped
conjugated molecules with enlarged p-delocalization
for organic solar cell application†
Cite this: DOI: 10.1039/c3ta11001j
a
a
a
a
a
b
Youyu Jiang, Di Yu, Luhua Lu, Chun Zhan, Di Wu, Wei You, Zhizhong Xiec
*
a
*
and Shengqiang Xiao
Three structurally related conjugated molecules (BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2) in star shape
have been designed and synthesized as donor materials for small molecule based bulk heterojunction
(BHJ) solar cells. The structural features of these molecules include a planarized benzo[1,2-b:3,4-b0:5,6-b00]-
trithiophene (BTT) with a C3h symmetry as the central core and three conjugated arms incorporating
electron deficient benzo[2,1,3]thiadiazole (BTD) units, with arms being linked to the core via different
number of thiophene connecting units (e.g., 0, 1, 2 corresponding to BTT-BTD-0, BTT-BTD-1 and BTT-
BTD-2, respectively). Comparative analyses of optical and electronic properties indicate that the molecules
bearing more thiophene units between the BTT core and the BTD arms possess higher-lying HOMO levels
while their LUMO levels remain almost unchanged. The improvement of BHJ device performance, with
[6,6]-phenyl-C61-butyric acid methyl ester (PC61BM) as the acceptor, is observed with increasing number of
thiophene units between the BTT core and BTD arms, from BTT-BTD-0 to BTT-BTD-1 and BTT-BTD-2. The
BTT-BTD-2:PC61BM based BHJ devices show the highest power conversion efficiency (PCE) of 0.74%, with
an open-circuit voltage (Voc) of 0.69 V, a short-circuit current density (Jsc) of 2.93 mA cmꢀ2, and a fill factor
(FF) of 0.37 under 1 sun (100 mW cmꢀ2) AM 1.5G simulated solar illumination. The PV performance of
BTT-BTD-2 is further improved when [6,6]-phenyl-C71-butyric acid methyl ester (PC71BM) is used as the
electron acceptor, yielding the best device performance with Jsc of 4.13 mA cmꢀ2, Voc of 0.72 V, FF at 0.46
and PCE of 1.36%. The effect of the different number of thiophenes linking the BTT core and the
conjugated BTD arms has been clearly demonstrated on regulating optical and electrochemical properties
of the three molecules and their BHJ device performances.
Received 10th March 2013
Accepted 10th May 2013
DOI: 10.1039/c3ta11001j
www.rsc.org/MaterialsA
conversion efficiency of such polymer–fullerene based BHJ cells
has been progressively improved to 7–9% for single junction
Introduction
cells,2–13 while over 10% has been reported for the multi-junc-
tion design (tandem cells).14 Impressive as these efficiency
values are, these semiconducting polymers pose outstanding
issues such as structural polydispersity, end-group contamina-
tion and trace impurities originating from their syntheses.15–17
All these catalyzed the emergence of solution processed small
molecule-based BHJ cells, since small molecules can have well-
dened structures, extremely high purity (e.g., defect-free), and
superior batch-to-batch reproducibility. For these reasons,
solution processed small molecule-based BHJ solar cells have
recently gained signicant momentum, with efficiency
numbers rivalling those achieved by polymer-based ones.18–24
However, active research on small molecules for solution
processed BHJ cells poses different challenges, as well as
opportunities. For example, it appears that very subtle changes
on the molecular structure can have signicant inuence on the
performance of related solar cells, while the photovoltaic
properties of polymers are quite insensitive or much more
tolerant to these minor structural changes. One such example is
Research efforts on polymer-based solar cells have been mainly
focused on polymers as the electron-donating material and
fullerene derivatives as the electron-accepting material, both of
which are typically blended in a bulk heterojunction (BHJ)
conguration to achieve the photovoltaic effect.1 The power
aState Key Laboratory of Advanced Technology for Materials Synthesis and Processing,
WUT-USG Joint Laboratory of Advanced Optoelectronic Materials and Devices, Centre
for Chemical and Material Engineering, Wuhan University of Technology, Wuhan
430070, P. R. China. E-mail: shengqiang@whut.edu.cn; Fax: +86-27-87879468; Tel:
+86-27-87870537
bDepartment of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill,
North Carolina 27599-3290, USA. E-mail: wyou@unc.edu
cDepartment of Applied Chemistry, School of Science, Wuhan University of Technology,
Wuhan, 430070, P. R. China
† Electronic supplementary information (ESI) available: Reagents and
instrumentation information as well as results of TGA, DSC, CV and GIXRD
measurements, AFM topography and phase images of blended thin lms and
J–V plots for hole mobility measurement of three molecules. See DOI:
10.1039/c3ta11001j
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
given by Bazan et al. recently,24 where they showed that the conjugated arms were constructed with a “donor–acceptor” motif
relative orientation of two anking [1,2,5]thiadiazolo[3,4-c]- of one electron decient benzo[2,1,3]thiadiazole (BTD) and one
pyridines had signicant impact on the efficiencies of devices electron rich a-alkylated thiophene. All three conjugated arms
based on these molecules (from 3% to 7%). In contrast, when were then connected to the central core via un-substituted thio-
these [1,2,5]thiadiazolo[3,4-c]pyridines were employed in a phenes (0, 1, or 2). We envisioned that such star-shaped chemical
related polymer – presumably in a random fashion,25 You et al. construction could provide tunable p-conjugation regulated by
demonstrated an efficiency of 6.3% in an earlier study, without the number of thiophene units used as the linkage between the
observing such an “orientation” effect (likely non-existing). BTT core and BTD arms. Furthermore, the “donor–acceptor” or
Furthermore, the early studied molecular species largely “push–pull” molecular framework on the arms can reduce the
emerged from the design of polymers for BHJ solar cells. band-gap through the partial intramolecular charge transfer
However, recently a variety of molecular structures – quite between the electron rich unit and the electron decient one on
specic to small molecules
– have appeared, including the arms. Finally, these aliphatic side chains at a-positions of
dendritic, star-shaped, in addition to more traditional, linear these terminal thiophenes would offer the solubility of resultant
conjugated molecules, with reported device efficiencies ranging molecules, facilitating material characterization and BHJ device
from 1–7%.23,26–40 In particular, star-shaped conjugated mate- fabrication. Through a combination of synthesis, calculation and
rials are a very intriguing class of organic materials. The high characterization, we found that the number of un-substituted
structural symmetry and planarity of star-shaped conjugated thiophenes, i.e., the connecting units, signicantly impacts the
molecules may promote the p–p stacking of conjugated arms observed efficiency of BHJ devices. The highest efficiency of 1.4%
due to a broad p-delocalization of armed molecules. On the was obtained from BTT-BTD-2 that contains the greatest number
other hand, the increased steric hindrance associated with of “connecting” thiophenes in this studied series of molecules.
these armed molecules may also prevent the formation of an
ordered, long-range and coplanar p–p stacking, which could be
benecial to charge transport across the thin lm based
Experimental
devices. Such interesting effects for star-shaped conjugated
materials, stemming from both their molecular structures and
morphology at the solid state, contribute to their interesting
optical and electronic properties. All these properties, together
with the morphological features in the solid state, will deter-
mine the device performance of these materials when applied in
organic electronics and photonics.41
General information of reagents, instrumentations and condi-
tions on optical spectroscopy and electrochemistry testing and
quantum chemical calculations can be found in the ESI.† Benzo-
[1,2-b:3,4-b0:5,6-b00]trithiophene was synthesized according to the
procedure reported in the literature.42
Synthesis
In the spirit of further exploring the structure–property rela-
tionship of these star-shaped molecules for organic solar cells, we
[5-(2-Hexyldecyl)-thiophen-2-yl]-tributyl-stannane. To a mix-
investigated a family of three star-shaped conjugated molecules ture of thiophene (7 mL, 110 mmol) and 50 mL of anhydrous THF
of different degree of conjugation, BTT-BTD-0, BTT-BTD-1 and under nitrogen was added dropwise 2.5 M n-BuLi in hexane
BTT-BTD-2 as shown in Fig. 1, and their application as electron (40 mL, 100 mmol) under stirring at 0 ꢁC. The mixture was
donors in combination with fullerene acceptors in typical BHJ allowed to warm to room temperature aer 1 hour and 1-bromo-
devices. We chose planarized benzo[1,2-b:3,4-b0:5,6-b00]trithio- 2-hexyldecane (27.5 g, 90 mmol) was added before being heated
phene (BTT) with C3h symmetry as the central core. Three to reux overnight. Aer cooling down to room temperature, the
mixture was poured into 200 mL of cold water and extracted with
petroleum ether (3 ꢂ 50 mL). The combined organics was dried
over anhydrous magnesium sulfate and concentrated under
vacuum. The crude product was puried by silica gel column
chromatography eluted with petroleum ether to afford 17.7 g of
pure 2-(2-hexyldecyl)thiophene (compound 1, 64%). 1H-NMR
(400 MHz, CDCl3, ppm): 7.11 (dd, J ¼ 5.1; 1.2 Hz, 1H), 6.91
(dd, J ¼ 5.1; 3.4 Hz, 1H), 6.78–6.70 (dd, J ¼ 3.3; 1.2 Hz, 1H),
2.75 (d, J ¼ 6.6 Hz, 2H), 1.60 (m, 1H), 1.27–1.25 (m, 24H), 0.90–
0.86 (m, 6H). 13C-NMR (400 MHz, CDCl3, ppm): 144.26, 126.48,
124.93, 122.87, 39.99, 34.17, 33.11, 31.91, 29.96, 29.66, 29.65,
29.37, 26.55, 22.72, 14.20. EI-MS for C20H36S: calcd, 308.25;
found, 308 (M+). To a mixture of 2-(2-hexyldecyl)thiophene
(6.16 g, 20.0 mmol) and 40 mL of anhydrous THF was added
dropwise 2.5 M n-BuLi in hexane (8.2 mL, 20.5 mmol) under
stirring at room temperature under N2. Bu3SnCl (7.5 g, 22.0
mmol) was then added 30 minutes later. Aer stirring for
another 1 hour, the mixture was poured into 100 mL of water and
Fig. 1 Molecular structures of BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2.
extracted with petroleum ether (3 ꢂ 20 mL). The combined
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
organics was dried over anhydrous magnesium sulfate and 124.70, 124.29, 123.74, 114.02, 39.96, 34.58, 33.15, 31.89, 29.98,
concentrated under reduced pressure to give the crude product 29.67, 29.63, 29.35, 26.54, 22.68, 14.14. EI-TOF MS for
1
as yellow oil. H-NMR spectrum of the crude oil indicated that
C
30H39BrN2S3: calcd, 602.15; found, 602/604 (1 : 1, M+).
the conversion efficiency of the reaction was almost quantitative
Compound 5. Compound 5 was prepared from compound 4
and the crude product was then used directly for the next step (2.72 g, 4.5 mmol) and 2-(tributylstannyl)thiophene (2.02 g, 5.4
without further purication. mmol) according to the same procedure for compound 3. The
4-Bromo-7-((5-(2-hexyldecyl))-thiophen-2-yl) benzo[2,1,3]thia- crude product from the reaction was puried by silica gel column
diazole (2). To a ame-dried ask containing 4,7-dibromo-benzo chromatography eluted by petroleum ether–dichloromethane
[2,1,3]thidiazole (6.4 g, 22.0 mmol) and Pd(PPh3)4 (115 mg, 0.1 (6/1, v/v) to afford 1.84 g of the target compound (yield, 68%). 1H-
mmol, 0.5% equiv.) under N2 was transferred a solution of NMR (400 MHz, CDCl3, ppm): 7.99 (d, J ¼ 3.8 Hz, 2H), 7.83 (d, J ¼
compound 1 (20.0 mmol) in 50 mL of anhydrous THF through a 7.7 Hz, 1H), 7.80 (d, J ¼ 7.7 Hz, 1H), 7.19 (d, J ¼ 3.9 Hz, 1H), 7.04
cannula. The mixture was then heated to reux overnight. Aer (d, J ¼ 3.8 Hz, 1H), 7.02 (d, J ¼ 3.8 Hz, 1H), 6.87 (d, J ¼ 3.6 Hz, 1H),
removal of the solvent under reduced pressure, the mixture was 2.81 (d, J ¼ 6.7 Hz, 2H), 1.71 (m, 1H), 1.33–1.20 (m, 24H), 0.90–
redissolved in 40 mL of petroleum ether and ltered to remove 0.85 (m, 6H). 13C-NMR (400 MHz, CDCl3, ppm): 152.50, 146.68,
unconsumed 4,7-dibromo-benzo[2,1,3]thidiazole. The ltrate was 138.47, 138.19, 137.31, 136.94, 127.94, 127.60, 126.38, 126.27,
concentrated by rotoevaporation and was puried by silica gel 125.31, 124.89, 124.71, 124.49, 123.95, 40.05, 34.72, 33.21, 31.70,
column chromatography eluted by gradient petroleum ether– 29.97, 29.63, 29.33, 26.60, 22.68, 14.12. MALDI-TOF MS for
dichloromethane (from 6/1 to 1/1, v/v) to afford 3.8 g of the target
C
34H42N2S4: calcd, 606.22; found, 606.97 (M+).
Compound 6. Compound 6 was prepared from compound 5
compound (yield, 37%). 1H-NMR (400 MHz, CDCl3, ppm): 7.95 (d,
J ¼ 3.7 Hz, 1H), 7.82 (d, J ¼ 7.7 Hz, 1H), 7.64 (d, J ¼ 7.7 Hz, 1H), 6.85 (1.76 g, 2.91 mmol) and NBS (0.52 g, 2.91 mmol) by the same
(d, J ¼ 3.7 Hz, 1H), 2.80 (d, J ¼ 6.7 Hz, 2H), 1.70 (m, 1H), 1.32–1.26 procedure for compound 4. The resulted crude mixture was
(m, 24H), 0.89–0.85 (m, 6H). 13C-NMR (400 MHz, CDCl3, ppm): loaded onto a silica gel column eluted by petroleum ether–
153.59, 151.52, 146.95, 135.95, 132.10, 128.07, 127.19, 126.40, dichloromethane (8/1, v/v) to present 1.62 g of pure product
1
124.77, 111.30, 39.95, 34.58, 33.11, 31.87, 31.56, 29.93, 29.60, 29.32, (yield, 81%). H-NMR (400 MHz, CDCl3, ppm): 7.95 (d, J ¼ 3.7
26.53, 22.66, 14.10. EI-MS for C26H37BrN2S2: calcd, 520.16; found, Hz, 1H), 7.74 (d, J ¼ 4.0 Hz, 1H), 7.71 (s, 1H), 7.70 (s, 1H), 7.12
520/522 (1 : 1, M+).
(d, J ¼ 4.0 Hz, 1H), 6.84 (d, J ¼ 3.7 Hz, 1H), 2.82 (d, J ¼ 6.7 Hz,
Compound 3. To a ame-dried 100 mL two-necked ask was 2H), 1.71 (m, 1H), 1.45–1.15 (m, 24H), 0.92–0.85 (m, 6H). 13C-
sequentially added compound 2 (5.53 g, 10.6 mmol), 2-(tribu- NMR (400 MHz, CDCl3, ppm): 152.41, 146.75, 138.79, 138.59,
tylstannyl)thiophene (3.68 g, 11.6 mmol), Ph(PPh3)4 (61 mg, 137.26, 136.66, 130.73, 127.76, 127.69, 126.40, 125.34, 125.33,
0.053 mmol, 0.5% equiv.) and 40 mL of anhydrous toluene 124.74, 124.58, 123.87, 111.27, 40.05, 34.71, 33.22, 31.90, 29.97,
under N2. The resulted mixture was then stirred and heated to 29.63, 29.34, 26.59, 22.68, 14.12. MALDI-TOF MS for
reux for 7 hours. The solvent was removed under reduced
pressure and the crude product was puried by silica gel
C
34H41BrN2S4: calcd, 685.87; found, 686.1 (M+).
1,3,5-Tri(trimethylstannyl) benzo[1,2-b:3,4-b0:5,6-b00]trithio-
column chromatography eluted by petroleum ether–dichloro- phene (7). To a solution of benzo[1,2-b:3,4-b0:5,6-b00]trithiophene
methane (3/1, v/v) to afford 5.0 g of the target compound (yield, (0.246 g, 1.0 mmol) in 50 mL of anhydrous THF was added
90%). 1H-NMR (400 MHz, CDCl3, ppm): 8.09 (d, J ¼ 3.0 Hz, 1H), dropwise 2.5 M n-BuLi in hexane (2.4 mL, 6 mmol) at 0 ꢁC under
7.97 (d, J ¼ 3.6 Hz, 1H), 7.84 (d, J ¼ 7.6 Hz, 1H), 7.78 (d, J ¼ 7.6 N2. The resulting mixture was then warmed up to room temper-
Hz, 1H), 7.44 (d, J ¼ 4.6 Hz, 1H), 7.20 (dd, J ¼ 3.8; 5.0 Hz, 1H), ature and stirred for 1 hour. 1 M trimethyltin chloride in anhy-
6.85 (d, J ¼ 3.6 Hz, 1H), 2.82 (d, J ¼ 6.7 Hz, 2H), 1.71 (m, 1H), drous THF (7.0 mL, 7 mmol) was then added to the mixture. Aer
1.33–1.27 (m, 24H), 0.90–0.85 (m, 6H). 13C-NMR (400 MHz, stirring for another 1 hour, the mixture was diluted with 50 mL of
CDCl3, ppm): 152.50, 152.37, 146.45, 139.41, 136.87, 127.84, ethyl ether and washed with saturated ammonium chloride and
127.50, 127.13, 126.39, 126.27, 126.17, 125.68, 125.09, 124.75, water consequently. The organic layer was dried over anhydrous
40.01, 34.65, 33.18, 31.89, 29.96, 29.62, 29.33, 26.57, 22.67, MgSO4 and concentrated by rotoevaporation. The crude solid was
14.12. EI-MS for C30H40N2S3: calcd, 524.24; found, 524 (M+).
washed with a small amount of methanol and then dried under
Compound 4. To a mixture of compound 3 (2.62 g, 5.0 mmol) vacuum to give 0.6 g of pure product (yield, 82%). 1H-NMR
and 80 mL of chloroform–acetic acid (1/1, v/v) was added NBS (400 MHz, CDCl3, ppm): 7.70 (s, 3H), 0.48 (s, 27H). 13C-NMR
(0.89 mg, 5.0 mmol) in several parts under stirring at room (400 MHz, CDCl3, ppm): 137.94, 135.78, 132.70, 130.40, ꢀ8.15.
temperature. When the reaction completed, the mixture was MALDI-TOF MS for C21H30S3Sn3: calcd, 734.79; found, 734.1 (M+).
washed with water, solution of 10% sodium hydroxide and
BTT-BTD-0. To a ame-dried two-necked round bottom ask
saturated brine consecutively. The organic layer was dried over containing 1,3,5-tri(trimethylstannyl) benzo[1,2-b:3,4-b0:5,6-b00]-
anhydrous MgSO4 and concentrated under vacuum to afford trithiophene (0.367 g, 0.5 mmol) and Ph(PPh3)4 (18.0 mg, 0.015
2.72 g of pure product ( yield, 90%). 1H-NMR (400 MHz, CDCl3, mmol) under nitrogen was transferred a solution of compound
ppm): 7.95 (d, J ¼ 3.7 Hz, 1H), 7.74 (d, J ¼ 4.0 Hz, 1H), 7.71 (s, 2 (0.94 g, 1.8 mmol) in 40 mL of anhydrous toluene via a
1H), 7.70 (s, 1H), 7.12 (d, J ¼ 4.0 Hz, 1H), 6.84 (d, J ¼ 3.7 Hz, 1H), cannula. The resulting mixture was then stirred and heated to
2.81 (d, J ¼ 6.7 Hz, 2H), 1.71 (m, 1H), 1.40–1.18 (m, 24H), 0.92– reux overnight. Aer removal of the solvent under reduced
0.81 (m, 6H). 13C-NMR (400 MHz, CDCl3, ppm): 151.95, 151.86, pressure, the crude product was washed with 50 mL of meth-
146.54, 140.60, 136.62, 130.35, 127.65, 126.49, 126.28, 126.20, anol and dried. The solid was then further puried by silica gel
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
column chromatography eluted by petroleum ether–dichloro- poly(styrenesulfonate) (PEDOT:PSS)/donor-PCBM/Ca (20 nm)/Al
methane (2/1, v/v) to give 0.35 g of the target molecule (yield, (80 nm). Glass substrates with patterned ITO were sonicated for
44%). 1H-NMR (400 MHz, CDCl3, ppm): 7.62 (s, 3H), 7.38 (d, J ¼ 10 minutes consecutively in DI water, acetone and isopropyl
2.4 Hz, 3H), 6.94 (d, J ¼ 7.0 Hz, 3H), 6.84 (d, J ¼ 7.0 Hz, 3H), 6.48 alcohol, followed by drying under a stream of N2. The substrates
(s, 3H), 2.69 (d, J ¼ 5.7 Hz, 6H), 1.69 (m, 3H), 1.36 (m, 66H), were then treated over UV–ozone for 10 minutes. A layer (45 nm)
1.02–0.87 (m, 18H). 13C-NMR (400 MHz, CDCl3, ppm): 151.02, of PEDOT:PSS lm was then spin-coated onto the substrates at
145.03, 136.93, 136.40, 130.98, 130.40, 127.62, 125.95, 124.92, 3000 RPM for 60 s and baked at 150 ꢁC for 30 minutes. The
124.23, 123.40, 122.90, 121.44, 77.32, 77.00, 76.68, 39.90, 34.58, substrates were then transferred into a glove box under N2.
33.16, 32.04, 30.19, 30.18, 29.84, 29.81, 29.50, 26.61, 26.57, Subsequently, the active layer was spin-cast from a different
22.81, 14.26. MALDI-TOF MS for C90H114N6S9: calcd, 1566.66; blend ratio (weight-to-weight, w/w) of donor (10 mg mLꢀ1) and
found, 1568.0 (M + H+). Elemental analysis for C90H114N6S9: PC61BM in 1,2-dichlorobenzene (o-DCB) onto the ITO/
calcd C, 68.92; H, 7.33; N, 5.36; S, 18.40. Found: C, 68.40; H, PEDOT:PSS substrate and dried naturally in the glove box.
7.20; N, 5.35; S, 19.05%.
Finally, Ca (20 nm) and Al (80 nm) were sequentially deposited on
Compound BTT-BTD-1. The compound was synthesized by the top of the active layer as cathode at a pressure of ꢃ2 ꢂ 10ꢀ6
the same method for BTT-BTD-0 through the reaction of mbar through a shadow mask that denes 8 devices with each
compound 4 with 1,3,5-tri(trimethylstannyl) benzo[1,2-b:3,4- active area of 18 mm2. The thicknesses of active layers were
b0:5,6-b00]trithiophene (0.35 mmol). Aer the reaction nished recorded by a DEKTAK XT prolometer.
as indicated by TLC, the mixture was added dropwise into 250
Current–voltage measurements were carried out in a
mL of acetone and precipitated. The precipitate was extracted nitrogen-lled glove box under AM 1.5G irradiation with the
using acetone and hexane sequentially on a Soxhlet extractor. intensity of 100 mW cmꢀ2 (from a 450 W solar simulator,
The extraction from hexane was concentrated and further Newport 94023A-U) calibrated by a NREL certied standard
puried by silica gel column chromatography by petroleum silicon cell. Current versus potential ( J–V) curves were recorded
ether–dichloromethane (1/1, v/v) to give 0.35 g of pure product with a Keithley 2420 digital source meter. For external quantum
(yield, 54%). 1H-NMR (400 MHz, CDCl3, ppm): 7.10 (s, 3H), 6.95 efficiency (EQE) tests, devices were transferred by a self-made
(s, 3H), 6.64–5.87 (m, 15H), 2.58 (s, 6H), 1.56 (m, 3H), 1.50–1.11 testing box under the protection of nitrogen atmosphere into
(m, 66H), 1.07–0.78 (m, 18H). 13C-NMR (400 MHz, CDCl3, ppm): the sample chamber of a 7-SCSpec Spectral Performance Solar
150.75, 150.54, 144.66, 137.76, 136.50, 135.17, 127.11, 125.81, Cell Test System consisting of a 500 watt SCS028-7ILX500 xenon
123.94, 123.87, 123.48, 123.07, 115.77, 39.78, 34.50, 33.17, light source, a 7ISW301 vertical grating spectrometer, a 7IFW6
31.95, 30.08, 29.71, 29.40, 26.54, 22.72, 14.14. MALDI-TOF MS lter wheel, a SR540 chopper and a SR810 lock-in amplier. The
for C102H120N6S12: calcd, 1812.62; found, 1813.0 (M+). Elemental calibration of the incident monochromatic light was carried out
analysis for C102H120N6S12: calcd C, 67.50; H, 6.66; N, 4.63; S, with a Hamamatsu S1337-1010 BQ Silicon photo detector.
22.20. Found: C, 66.50; H, 6.44; N, 4.37; S, 22.69%.
Morphology of lms was characterized through traditional
BTT-BTD-2. The compound was synthesized by the same Tapping Mode on an Agilent 5500 AFM (Agilent Technologies,
method for BTT-BTD-0 through the reaction of compound 6 Santa Clara, CA) under ambient conditions. Si3N4-coated silicon
with 1,3,5-tri(trimethylstannyl) benzo[1,2-b:3,4-b0:5,6-b00]trithio- AFM tips (NSC19/Si3N4, Mikromasch, Estonia) with a spring
phene (0.35 mmol). When the reaction nished, the mixture constant of 0.6 N mꢀ1 and a resonance frequency of 80 kHz were
was added dropwise into 100 mL of acetone and precipitated. used in imaging. Powder X-ray diffraction (XRD) and BHJ thin
The precipitate was loaded onto a Soxhlet extractor and lm grazing incident X-ray diffraction (GIXRD) data were
extracted using acetone, hexane and methylene dichloride recorded in reection mode with a D8 ADVANCE X-ray diffrac-
ꢀ
consequently. The extraction from methylene dichloride was tometer using Cu Ka radiation (l ¼ 1.540 598 A) at room
concentrated under reduced pressure and the resulted solid was temperature. BHJ thin lms of BTT-BTD-x (x ¼ 0, 1, 2 respec-
then dried under vacuum to give 503 mg of pure product (yield, tively)/PC61BM were spun-coated on a silicon substrate covered
70%). With conventional liquid phase NMR spectroscopy by a layer of PEDOT:PSS before the measurements.
aromatic protons of BTT-BTD-2 could not be efficiently detec-
Hole mobilities of donor-PC61BM (w/w, 1 : 1) blended lms
ted, presumably because of strong shielding effects of aromatic were measured in conguration of ITO/PEDOT:PSS (45 nm)/
ring current caused by enlarged p-conjugation in BTT-BTD-2. As donor-PC61BM (w/w at optimized ratio)/Pd (40 nm) by taking
a result, isotope-resolved MS with mild ionization methods (e.g., dark current–voltage in the range of 0–15 V and tting the
MALDI-TOF) and elemental analysis were the only applicable results to a space charge limited form.43
techniques for its structure characterization. MALDI-TOF MS
for C114H126N6S15: calcd, 2058.59; found, 2060.2 (M + H+).
Elemental analysis for C114H126N6S15: calcd C, 66.43; H, 6.16; N,
Results and discussion
1
Synthesis
4.08; S, 23.33. Found: C, 66.20; H, 6.20; N, 4.06; S, 23.54%.
The detailed synthetic route to the series of BTT-BTD-0, BTT-
BTD-1, BTT-BTD-2 molecules is depicted in Scheme 1. As the key
building block of the conjugated arms, asymmetric 4-bromo-7-[5-
BHJ device fabrication and characterization
BHJ solar cells were fabricated with a general device congura- (2-hexyldecyl)-thiophen-2-yl]-benzo[2,1,3]thiadiazole (compound
tion of indium tin oxide (ITO)/poly(3,4-ethylenedioxythiophene) 2) was prepared from a Stille coupling reaction between
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
Scheme 1 Synthetic route to BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2.
4,7-dibromo-benzo[2,1,3]thiadiazole and [5-(2-hexyldecyl)-thio-
phen-2-yl]-tributyl-stannane as an orange liquid. Iterative Stille
coupling reactions of compound 2 with stannylated thiophene
and subsequent bromination with NBS were then executed to
afford the other two building blocks of conjugated arms as
compounds 4 and 6, respectively. Compound 2, 4 or 6 was then
reacted with 2,5,8-tris(trimethylstannyl)benzo[1,2-b:3,4-b0:5,6-
b00]trithiophene (compound 7) via the Stille coupling reaction
condition to afford the target material of BTT-BTD-0, BTT-BTD-1,
or BTT-BTD-2, respectively. These star-shaped materials showed
good solubility in common organic solvents, such as chloroform,
toluene and chlorinated benzene.
Fig. 2 Normalized UV-Vis absorption spectra of BTT-BTD-0 (gray solid), BTT-
BTD-1 (black solid) and BTT-BTD-2 (gray dash) in (a) dichloromethane solution
and (b) thin films.
respectively. For thin lms cast from the chlorobenzene
solution, BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2 exhibited
absorption maxima at 528, 545 and 547 nm, red-shied by 31,
35 and 24 nm, respectively, when compared with those of
corresponding solutions. This absorption red-shi suggests
the presence of weak intermolecular interactions from BTT-
BTD-0 and BTT-BTD-1 to BTT-BTD-2 in lms. It is worth
noting that the absorption maxima of these molecules in
solution exhibit a bathochromic shi (Dl) of 13 nm for every
added thiophene connecting unit, conrming the effect of
spectral tuning by these inserted thiophene units between the
BTT core and the BTD arms. This progressing red shi of lmax
can be attributed to a more extended p-conjugation as the
number of thiophene connecting units increases. For
absorption maxima in lms, a bathochromic shi of 17 nm
was also observed from BTT-BTD-0 to BTT-BTD-1. However,
adding one more thiophene unit – converting BTT-BTD-1 to
2
Thermal stability
Thermal properties of the three molecules were investigated by
thermogravimetric analysis (TGA) and differential scanning
calorimetry (DSC). The decomposition temperatures (Td, refer-
ꢁ
ring to 5% weight loss) were all above 400 C (Fig. S1a in the
ESI†), indicating good thermal stability of all three molecules. A
good thermal stability would minimize any possible morpho-
logical deformation and degradation of the active layer in BHJ
devices. DSC data also indicate that increasing the number of
thiophene spacer leads the melting point dropping from 194 ꢁC
for BTT-BTD-0 to 158 ꢁC for BTT-BTD-2 (ESI, Fig. S1b†), while no
obvious phase transition was detected from the DSC curve of
BTT-BTD-1. The higher melting point of BTT-BTD-0 indicates
stronger intermolecular interactions in the solid state.39
3
Spectroscopic properties
Steady-state UV-Vis absorption spectroscopy was carried out to BTT-BTD-2 – only leads to negligible red-shi in the absorp-
gain insight into the electronic structure of these star-shaped tion maximum. It appears that the band gap of BTT-BTD
molecules. The normalized UV-Vis absorption spectra both in based materials would reach its lower limit with just two of
solution and thin lm are plotted in Fig. 2. The absorption thiophene connecting units between the BTT core and the
maxima of BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2 solution in BTD arms, implying additional thiophenes would not be
dichloromethane are located at 497, 510 and 523 nm, necessary if only for the purpose of narrowing the band gap.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
The optical gaps (Eg,opt) of BTT-BTD-0, BTT-BTD-1 and BTT- insights can be obtained from computational simulation of
BTD-2, calculated from the absorption onset of thin lms, are these molecules (vide infra).
ꢃ2.02, 1.90 and 1.85 eV, respectively (Table 1). In comparison
with the lm absorption of a related star-shaped molecule with
5
Quantum chemical calculations
only bithiophene in the three arms linked to the BTT core,44
these three molecules show signicant red-shi of absorption
maxima by 108 nm, 125 nm and 127 nm from BTT-BTD-0 to
BTT-BTD-2, respectively. This clearly demonstrates the benet
of incorporating the electron decient BTD unit into arms: BTD
can lead to strong charge transfer characteristics, resulting in
broad optical absorption spectra extending to 700 nm.
The frontier molecular orbitals of the three molecules were
calculated by density functional theory (DFT) at the B3LYP/6-
31G (d) level to gain better insights into their electronic struc-
tures (Fig. 3). The optimized geometry of arms tends to take a
planarized structure, which promotes the p-electrons delocal-
ization. As indicated in Fig. 3, while the electron density of the
HOMO of BTT-BTD-0 is delocalized on the whole molecule,
such a delocalization across the entire molecule is much less for
BTT-BTD-1 and BTT-BTD-2. On the other hand, the electron
density of the LUMO of BTT-BTD-0 is mainly localized on the
BTD moieties, whereas it is slightly more delocalized on the
entire molecules for BTT-BTD-1 and BTT-BTD-2. These results
help to explain the negligible change of the LUMO levels for all
three molecules and a slight rise (0.15 eV) of the HOMO levels of
BTT-BTD-1 and BTT-BTD-2 in comparison with that of BTT-
BTD-0 (Table 1).
4
Electrochemical characteristics
The electrochemical properties of the three molecules in thin
lms were investigated by cyclic voltammetry (CV) to determine
their frontier orbital energy levels and the potentials were cali-
brated to Fc/Fc+ (Fig. S2 in the ESI†). All three molecules showed
irreversible anodic oxidation processes and no cathodic reduc-
tion processes were recorded. Therefore the rst onset oxida-
tion potential was used to calculate the energy level of the
highest occupied molecular orbital (HOMO) (i.e., HOMO ¼
ꢀ(Eox + 4.8) eV (ref. 43)), while the energy level of the lowest
molecular orbital (LUMO) was estimated by adding the optical
band gap (Eg,opt) to the HOMO energy level. For example, the
HOMO level of BTT-BTD-0 was found to be ꢀ5.41 eV, and the
LUMO level was estimated to be ꢀ3.39 eV aer considering
the band gap of BTT-BTD-0 is 2.02 eV. All data are summarized
in Table 1.
Table 1 clearly shows that both of the HOMO energy levels of
BTT-BTD-1 and BTT-BTD-2 are almost identical, and higher
than the HOMO level of BTT-BTD-0. This not only indicates that
the extension of the conjugation length in arms through adding
thiophene units makes BTT-BTD-1 and BTT-BTD-2 easier to
oxidize, but also implies that just one thiophene unit is suffi-
cient to raise the HOMO level for this series of materials. In
contrast, the LUMO levels of all three molecules remain almost
unaffected by increasing the conjugation length of arms
through additional thiophene units from BTT-BTD-0 to BTT-
BTD-2, i.e., the LUMO level is “pinned” around ꢀ3.4 eV. This
observation, i.e., a “tunable” HOMO level and a LUMO level with
negligible variation for structurally related organic materials, is
consistent with other similarly constructed “donor–acceptor”
polymers.43,45 This is because the LUMO level is typically local-
ized on the “acceptor”, which would determine the LUMO level
of the entire molecule, whereas the HOMO level is more delo-
calized along the entire molecular system and signicantly
impacted by the electronic nature of the “donor”.1,46 Further
6
Solar cell characteristics
To probe the photovoltaic properties of this series of star-sha-
ped molecules, we rst prepared the solution for the BHJ device
fabrication by blending BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2
as the electron donor with PC61BM as the electron acceptor in
1,2-dicholorobenzene (o-DCB). Then we spun-coated a thin lm
from such a blended solution in a typical device conguration
of ITO/PEDOT:PSS (45 nm)/BTT-BTD-x (x ¼ 0, 1, 2):PC61BM/Ca
Fig. 3 The frontier molecular orbitals of BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2
based on optimized geometries (calculated with DFT at the B3LYP/6-31G level).
Table 1 Optical and electrochemical property of BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2
lmax,solution
lmax,lm
lonset,lm
Compound
(nm)
(nm)
(nm)
Eg,opt (eV)
Eox,onset (V)
HOMO (eV)
LUMO (eV)
BTT-BTD-0
BTT-BTD-1
BTT-BTD-2
497
510
523
528
545
547
614
657
670
2.02
1.90
1.85
0.61
0.46
0.46
–5.41
–5.26
–5.26
–3.39
–3.36
–3.43
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
(20 nm)/Al (80 nm). Different weight ratios of BTT-BTD-
x:PC61BM were attempted, but the concentrations of BTT-BTD-x
in all solutions were xed at 10 mg mLꢀ1 in o-DCB. We found
that PV performance of BTT-BTD-0 based devices did not vary
much in regard to the weight ratios of BTT-BTD-0:PC61BM, with
the observed Voc around 0.55 ꢄ 0.05 V. Such a ratio-insensitivity
was also observed for the PV performance of BTT-BTD-1 based
devices, with Voc around 0.60 ꢄ 0.05 V. However, in the case of
BTT-BTD-2, the ones fabricated from the 1 : 1 weight ratio in the
blend showed the best PV properties with Voc approaching 0.7 V
(all these data are tabulated in the ESI, Table S1†). Therefore, we
decided to use the weight ratio of 1 : 1 of BTT-BTD-x:PC61BM to
compare the PV performance for all three molecules. Repre-
sentative current density–voltage ( J–V) characteristics of
the three molecules blended with PC61BM at a weight ratio of
1 : 1 are shown in Fig. 6a, and the related data are tabulated
in Table 2.
Fig. 4 Determined energy diagram with HOMO/LUMO levels of BTT-BTD-0,
BTT-BTD-1, BTT-BTD-2 and PCBM in relation to the work functions of the elec-
trode materials ITO/PEDOT:PSS and Ca in a BHJ OSC device.
increase the dark saturation current and result in a lower
Voc.48,50
Based on the current understanding of the origin of the Voc
in organic solar cells, we next attempt to offer an explanation of
the seemingly unusual trend of Voc for the studied series of
molecules. The high structural symmetry and planarity of these
star-shaped conjugated molecules may promote the p–p
stacking of conjugated arms due to a broad p-delocalization of
armed molecules. On the other hand, the increased steric
hindrance associated with these armed molecules may also
prevent the formation of an ordered, long-range and coplanar
p–p stacking. All these would have a signicant impact on the
intermolecular interaction between the donor molecule and the
fullerene at the donor–acceptor (D–A) interface, thereby
affecting the Voc. However, the intermolecular interaction
between the donor molecule and PCBM at the D–A interface is
very difficult to probe directly in a BHJ blend. Fortunately, the
easily accessible intermolecular interaction among the donor
molecules can be investigated instead, on the assumption that
greater intermolecular interactions between donor molecules
would give rise to stronger interactions between the donor
molecules and PCBM at the D–A interface. To this end, X-ray
diffraction (XRD) characterizations of donor molecules and the
BHJ blend can provide insights into qualitative understanding
and comparison of these intermolecular interactions. Similar
approaches have been employed in previous reports.48
Typically, the Voc of BHJ devices roughly linearly depends on
the interface energy difference between the HOMO level of the
electron donor and the LUMO of the electron acceptor.15,47
Therefore, if PC61BM is used as the common electron acceptor
for all the devices (e.g., in our study), the Voc would increase as
the HOMO levels decrease. With the energy level alignment
presented in Fig. 4, one would project the highest Voc should
appear for BTT-BTD-0 based devices. Interestingly, despite the
lower-lying HOMO of BTT-BTD-0 at ꢀ5.41 eV and similarly
higher HOMO level at ꢀ5.26 eV for BTT-BTD-1 and BTT-BTD-2,
devices based on BTT-BTD-2 deliver a higher Voc of around 0.70
ꢄ 0.05 V while BTT-BTD-0 and BTT-BTD-1 provide lower values
around 0.55 ꢄ 0.05 V and 0.60 ꢄ 0.05 V, respectively. This is
because the energetics is just one factor (though dominating)
that decides the Voc. Others, including charge recombination at
the donor–acceptor interface, resistance related to the thickness
of the active layer, degree of phase-separation between the
components and intermolecular interactions of active materials
in the blend, have also been demonstrated to signicantly
modify the energetically predicted Voc value.48–50 For example,
donor materials that show evidence for strong aggregation by
their thin-lm absorption spectra and polycrystallinity by
their thin lm X-ray diffraction result in a lower Voc than the
simple estimated value (i.e., Voc ꢃ ELUMO(acceptor) ꢀ EHOMO(donor)).49
It was believed that such a strong aggregation/polycrystallinity
indicates strong intermolecular interactions between active
components, leading to the observed high saturation dark
current density and thereby reducing the Voc according to the
Shockley equation. Similarly, an increased electronic coupling
between the donor polymers and the fullerene acceptors would
We thus conducted powder XRD experiments on these three
molecules without being blended with PC61BM, and grazing
incident X-ray diffraction (GIXRD) measurements of both BHJ
lms of these three molecules to gain insight into the solid state
structures of the three molecules. The data from powder
XRD indicate that there are indeed different intermolecular
Table 2 Average photovoltaic parameters of BTT-BTE-0, BTT-BTD-1 and BTT-BTD-2 blended with PC61BM at the optimized weight ratios (1 : 1) and BTT-BTD-2 with
PC71BM at the weight ratio of 1 : 2
Active layer
Ratio
Thickness (nm)
Voc (V)
Jsc (mA cmꢀ2
)
FF
PCE (%)
mh (cm2 Vꢀ1 sꢀ1
)
BTT-BTD-0:PC61BM
BTT-BTD-1:PC61BM
BTT-BTD-2:PC61BM
BTT-BTD-2:PC71BM
1 : 1
1 : 1
1 : 1
1 : 2
90
65
75
55
0.54
0.55
0.69
0.72
0.95
1.59
2.93
4.13
0.31
0.31
0.37
0.46
0.16
0.27
0.74
1.36
2.2 ꢂ 10ꢀ6
2.2 ꢂ 10ꢀ6
3.0 ꢂ 10ꢀ6
—
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
interactions for these three molecules (Fig. 5a). BTT-BTD-0, the observation from powder XRD (vide supra). Again, the
BTT-BTD-1 and BTT-BTD-2 show a diffraction peak at 2q about presumably stronger donor–acceptor interaction in the BTT-
ꢁ
25.7ꢁ, 25.4ꢁ and 24.7 , corresponding to a distance of 3.45 A, BTD-0 based BHJ lms would contribute to a larger saturation
ꢀ
ꢀ
ꢀ
3.50 A and 3.60 A from the Bragg equation, respectively. The dark current density and thereby a lower Voc value, based on the
reections are a typical p-stacking spacing of conjugated same reasoning as presented by Forrest and Thompson et al. in
backbones. BTT-BTD-0 shows the strongest intermolecular their study.49
interaction through p–p stacking and the shortest p-stacking
Jsc, on the other hand, is closely related to the morphology of
distance, as indicated in Fig. 5a, whereas BTT-BTD-2 shows the the donor:acceptor blend.51 A nanoscale phase separation in the
weakest p–p stacking interactions (e.g., lowest intensity of the active layer of BHJ devices has been proven to be essential to
diffraction peak and longest p-stacking distance). It appears enable a large interface area for exciton dissociation and to
that inserting more connecting thiophene units between the provide a continuous percolating path for charge transport,
BTT core and the BTD arms leads to a decrease of structural both are crucial for achieving a high Jsc. In this work, tapping-
rigidity and less planarity of the entire molecule, which would mode atomic force microscopy (AFM) studies were carried out
explain the reduced p–p intermolecular interaction from BTT- to investigate the inuence of lm morphology of different
BTD-0 to BTT-BTD-2 in the solid state (please note that the p–p active layers on their PV performances. These AFM topography
intermolecular distance is commonly assigned to the (010) images are shown in Fig. S4 of the ESI.† A very rough surface
distance as shown in Fig. S3†). The reduced p–p stacking with broad and deep grooves (and even some pinholes with the
interactions among donors from BTT-BTD-0 to BTT-BTD-1 depth of 10–20 nm) was observed for the blended lm of BTT-
imply a weaker interaction between donor molecules and BDT-0:PC61BM according to the AFM topography image.
PC61BM (acceptor) in the BHJ blend, which can (at least partly) Showing a slightly better surface morphology, BTT-BDT-
account for the observed lower Voc in BTT-BTD-0 based devices. 1:PC61BM blended lm still has signicant surface roughness.
GIXRD data offer further insights into the structural order Furthermore, phase images of both BTT-BDT-0:PC61BM and
(e.g., crystallinity) of these three molecules in their BHJ blends BTT-BDT-1:PC61BM blended lms do not indicate any desirable
with PC61BM. Specically, GIXRD spectra of these BHJ lms phase separation. On the other hand, the average roughness of
indicate that BTT-BTD-0:PC61BM and BTT-BTD-1:PC61BM the blend lms of BTT-BTD-2 with PC61BM is ꢃ5 nm. We also
exhibit a reection peak at 2q around 3.33ꢁ and 3.07ꢁ, respec- observed clusters for the blend lm based on BTT-BTD-
ꢀ
tively (Fig. 5b), indicative of a spacing value of 26.40 A and 28.72 2:PC61BM. Given the similar hole mobilities for these molecules
ꢀ
ꢀ
A. Such a difference of ꢃ2.3 A between BTT-BTD-0 and BTT- in their BHJ blends (Table 2), the much improved morphology
BTD-1 BHJ lms is close to the size of a unit thiophene, which is for the BTT-BTD-2:PC61BM based blend, in contrast to those of
ꢀ
around 2.7 A calculated with the MM2 force eld method. It is the other two molecules based BHJ blends, indicates a better
thus very likely that these spacing values correspond to the solid state structure for exciton dissociation, resulting in
distance between conjugated backbones, which is determined noticeably higher Jsc for BTT-BTD-2:PC61BM based BHJ cells
by the length of conjugated arms and the terminal long alkyl than those of the other two molecules based counterparts.
chains in this set of molecules (e.g., (100) or (001) distance as
With the highest Voc and Jsc (and FF), the BTT-BTD-2 based
proposed in Fig. S3†). It is worth noting that the intensity of the BHJ device offers the highest observed efficiency of 0.74% in all
reection peak of the BTT-BTD-0:PC61BM lm is much stronger studied three molecules when blended with PC61BM. We then
than that of BTT-BTD-1:PC61BM (Fig. 5b), indicating more chose to blend BTT-BTD-2 with the electron acceptor PC71BM to
polycrystalline domains in the BTT-BTD-0:PC61BM BHJ lm further optimize its PV performance, since PC71BM has a
than the BTT-BTD-1 based BHJ lm. Furthermore, no reection broader absorption and a larger extinction coefficient than
peaks were observed for the BHJ lm of BTT-BTD-2:PC61BM, those of PC61BM. All results from different attempts (e.g.,
indicating little evidence of aggregation of BTT-BTD-2 in its BHJ varying the ratio of donor:acceptor) are listed in the ESI
lm. The largest amount of polycrystalline domains in the BTT- (Table S2),† and the typical J–V curves of champion
BTD-0 based BHJ blend (among all blends) likely leads to devices are presented in Fig. 6b. AFM study on the lm of
strongest donor–PC61BM interactions, which is consistent with
Fig. 6 Current–voltage (J–V) characteristics of BHJ solar cells. (a) Devices based
Fig. 5 Powder XRD patterns of BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2 (a).
GIXRD patterns of the BHJ films of BTT-BTD-0, BTT-BTD-1 and BTT-BTD-2 with
PC61BM (b).
on molecular donor: PC61BM blended films at an optimized weight ratio (1 : 1)
and (b) devices based on various weight ratios of BTT-BTD-2:PC71BM under
illumination of 1 sun at AM1.5G (100 mW cmꢀ2).
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
expected, incorporating the electron decient BTD units into
arms leads to intramolecular charge transfer characteristics,
resulting in broad optical absorption spectra extending to
700 nm. Nevertheless, it appears that band gaps and HOMO
levels of this system will not be signicantly affected by incor-
porating more than two of these thiophene connecting units
between the BTT core and the BTD arms. However, the more
thiophene connecting units there are, the better the perfor-
mance of related solar cells. With just PC61BM, BTT-BTD-2 based
BHJ devices already show the highest Voc, Jsc and FF among all
devices based on these molecules. Furthermore, with PC71BM as
the electron acceptor, the PV performance of BTT-BTD-2 based
devices can be further improved to 1.36%, with Jsc of 4.13 mA
cmꢀ2, Voc at 0.72 V and FF of 0.46. This discovered structure–
property relationship provides important insights into the future
design of conjugated molecules for photovoltaic applications.
Fig. 7 (a) EQE of preliminarily optimized devices of BTT-BTD-0, BTT-BTD-1 and
BTT-BTD-2 and (b) the active layer absorption spectra of devices in (a).
BTT-BTD-2:PC71BM indicates further intimate mixing and a
smoother surface with an average roughness around 1 nm (ESI,
Fig. S4†). It appears that BTT-BTD-2 is more homogeneously
distributed in the blend with PC71BM in comparison to the BTT-
BTD-2:PC61BM blend. These morphological differences might
account for the better photovoltaic performance of BTT-BTD-
2:PC71BM than that of the BTT-BTD-2:PC61BM blend. For
example, the PV performance was much enhanced for devices
based on BTT-BTD-2:PC71BM at an optimized donor:acceptor
weight ratio of 1 : 2, which showed relatively larger values of Voc
(0.72 V), Jsc (4.13 mA cmꢀ2) and FF (46%), leading to an
increased average PCE of 1.36%.
Fig. 7a presents the external quantum efficiency (EQE) as a
function of wavelength for all the optimized BHJ devices, while
the absorption features of the active layers are shown in Fig. 7b.
The EQE curves of all devices match the optical absorption of
active layers well. The calculated Jsc values obtained by inte-
grating the EQE spectra with the standard AM 1.5G solar spec-
trum match those (within ꢃ5% error) obtained from the J–V
measurement, indicating these measurements are quite accu-
rate. When increasing the number of connecting thiophene
units between the BTT core and these BTD arms, progressively
increased EQE values were observed. Devices of the BTT-BTD-
2:PC61BM blend show more efficient photo conversion effi-
ciency in the range between 300 and 550 nm with EQE value
around 16%, which was improved to 25% when using PC71BM
as the electron acceptor. It is worth noting that there exists a
broad plateau around the maximum in EQE curves between 400
and 550 nm for devices based on BTT-BTD-2 due to stronger and
wider absorption of BTT-BTD-2 and PC71BM in that wavelength
region. Stronger and broader absorption together with better
lm morphology of blended lms explain the better PV
performance of BTT-BTD-2 based BHJ devices over the other two
molecules based ones.
Acknowledgements
The authors are grateful to the National Natural Science Foun-
dation of China for funding through Grant 51073124 and
21031006. We also want to acknowledge nancial support from
self-determined and innovative research funds of WUT (2011-
II-009) and SKLWUT (2009-PY-1), the Research Fund for the
Doctoral Program of Higher Education of China (201001
43120002), the Scientic Research Foundation for the Returned
Overseas Chinese Scholars, Ministry of Education of China
(2011508) and the Natural Science Foundation of Hubei Prov-
ince (2011CDA102 and 2010CDA078).
Notes and references
1 H. Zhou, L. Yang and W. You, Macromolecules, 2012, 45, 607.
2 H.-Y. Chen, J. Hou, S. Zhang, Y. Liang, G. Yang, Y. Yang,
L. Yu, Y. Wu and G. Li, Nat. Photonics, 2009, 3, 649.
3 S. H. Park, A. Roy, S. Beaupre, S. Cho, N. Coates, J. S. Moon,
D. Moses, M. Leclerc, K. Lee and A. J. Heeger, Nat. Photonics,
2009, 3, 297.
4 Y. Liang, Z. Xu, J. Xia, S.-T. Tsai, Y. Wu, G. Li, C. Ray and
L. Yu, Adv. Mater., 2010, 22, E135.
5 C. M. Amb, S. Chen, K. R. Graham, J. Subbiah, C. E. Small,
F. So and J. R. Reynolds, J. Am. Chem. Soc., 2011, 133, 10062.
´
6 T.-Y. Chu, J. Lu, S. Beaupre, Y. Zhang, J.-R. Pouliot, S. Wakim,
J. Zhou, M. Leclerc, Z. Li, J. Ding and Y. Tao, J. Am. Chem.
Soc., 2011, 133, 4250.
7 Z. He, C. Zhong, X. Huang, W.-Y. Wong, H. Wu, L. Chen,
S. Su and Y. Cao, Adv. Mater., 2011, 23, 4636.
8 L. Huo, S. Zhang, X. Guo, F. Xu, Y. Li and J. Hou, Angew.
Chem., Int. Ed., 2011, 50, 9697.
Conclusion
In summary, we have investigated the synthesis and photovoltaic
application of three structurally closely related molecules, BTT-
9 S. C. Price, A. C. Stuart, L. Yang, H. Zhou and W. You, J. Am.
Chem. Soc., 2011, 133, 4625.
BTD-0, BTT-BTD-1 and BTT-BTD-2, with 0, 1, 2 indicating the 10 M.-S. Su, C.-Y. Kuo, M.-C. Yuan, U. S. Jeng, C.-J. Su and
number of thiophene connecting units in between the BTT core
K.-H. Wei, Adv. Mater., 2011, 23, 3315.
and the BTD arms. Increasing the number of thiophene con- 11 Y. Sun, C. J. Takacs, S. R. Cowan, J. H. Seo, X. Gong, A. Roy
necting units elevates the HOMO energy level of the BTT-BTD and A. J. Heeger, Adv. Mater., 2011, 23, 2226.
based molecule, while the LUMO level remains almost 12 H. Zhou, L. Yang, S. C. Price, K. J. Knight and W. You, Angew.
unchanged because of the common “acceptor”, BTD. As
Chem., Int. Ed., 2010, 49, 7992.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
13 C. E. Small, S. Chen, J. Subbiah, C. M. Amb, S.-W. Tsang, 33 Y. Liu, X. Wan, F. Wang, J. Zhou, G. Long, J. Tian and
T.-H. Lai, J. R. Reynolds and F. So, Nat. Photonics, 2012, 6,
115.
14 J. You, L. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty,
Y. Chen, Adv. Mater., 2011, 23, 5387.
34 S. Loser, C. J. Bruns, H. Miyauchi, R. o. P. Ortiz, A. Facchetti,
S. I. Stupp and T. J. Marks, J. Am. Chem. Soc., 2011, 133, 8142.
K. Emery, C.-C. Chen, J. Gao, G. Li and Y. Yang, Nat. 35 G. C. Welch, L. A. Perez, C. V. Hoven, Y. Zhang, X.-D. Dang,
Commun., 2013, 4, 1446.
15 B. C. Thompson and J. M. J. Frechet, Angew. Chem., Int. Ed.,
A. Sharenko, M. F. Toney, E. J. Kramer, T.-Q. Nguyen and
G. C. Bazan, J. Mater. Chem., 2011, 21, 12700.
36 D. Bagnis, L. Beverina, H. Huang, F. Silvestri, Y. Yao, H. Yan,
G. A. Pagani, T. J. Marks and A. Facchetti, J. Am. Chem. Soc.,
2010, 132, 4074.
´
2008, 47, 58.
16 W. L. Leong, G. C. Welch, L. G. Kaake, C. J. Takacs, Y. Sun,
G. C. Bazan and A. J. Heeger, Chem. Sci., 2012, 3, 2103.
17 Z. B. Henson, K. Mullen and G. C. Bazan, Nat. Chem., 2012, 4, 37 T. Rousseau, A. Cravino, T. Bura, G. Ulrich, R. Ziessel and
699.
J. Roncali, Chem. Commun., 2009, 1673.
18 J. Roncali, Acc. Chem. Res., 2009, 42, 1719.
19 Y. Li, Q. Guo, Z. Li, J. Pei and W. Tian, Energy Environ. Sci.,
2010, 3, 1427.
38 Y.-H. Chen, L.-Y. Lin, C.-W. Lu, F. Lin, Z.-Y. Huang, H.-W. Lin,
P.-H. Wang, Y.-H. Liu, K.-T. Wong, J. Wen, D. J. Miller and
S. B. Darling, J. Am. Chem. Soc., 2012, 134, 13616.
20 B. Walker, C. Kim and T.-Q. Nguyen, Chem. Mater., 2011, 23, 39 R. Fitzner, E. Mena-Osteritz, A. Mishra, G. Schulz,
¨
E. Reinold, M. Weil, C. Korner, H. Ziehlke, C. Elschner,
470.
¨
21 Y. Lin, Y. Li and X. Zhan, Chem. Soc. Rev., 2012, 41, 4245.
K. Leo, M. Riede, M. Pfeiffer, C. Uhrich and P. Bauerle,
¨
22 A. Mishra and P. Bauerle, Angew. Chem., Int. Ed., 2012, 51,
J. Am. Chem. Soc., 2012, 134, 11064.
2020.
40 J. Zhou, X. Wan, Y. Liu, Y. Zuo, Z. Li, G. He, G. Long, W. Ni,
C. Li, X. Su and Y. Chen, J. Am. Chem. Soc., 2012, 134, 16345.
41 A. L. Kanibolotsky, I. F. Perepichka and P. J. Skabara, Chem.
Soc. Rev., 2010, 39, 2695.
23 Y. M. Sun, G. C. Welch, W. L. Leong, C. J. Takacs, G. C. Bazan
and A. J. Heeger, Nat. Mater., 2012, 11, 44.
24 C. J. Takacs, Y. Sun, G. C. Welch, L. A. Perez, X. Liu, W. Wen,
G. C. Bazan and A. J. Heeger, J. Am. Chem. Soc., 2012, 134, 42 Y. Nicolas, P. Blanchard, E. Levillain, M. Allain, N. Mercier
16597. and J. Roncali, Org. Lett., 2004, 6, 273.
25 H. Zhou, L. Yang, A. C. Stuart, S. C. Price, S. Liu and W. You, 43 S. Xiao, H. Zhou and W. You, Macromolecules, 2008, 41, 5688.
Angew. Chem., Int. Ed., 2011, 50, 2995.
44 R. de Bettignies, Y. Nicolas, P. Blanchard, E. Levillain,
J. M. Nunzi and J. Roncali, Adv. Mater., 2003, 15, 1939.
45 H. Zhou, L. Yang, S. Stoneking and W. You, ACS Appl. Mater.
Interfaces, 2010, 2, 1377.
26 W. Zhang, G. M. Ng, H. L. Tam, M. S. Wong and F. Zhu,
J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 1865.
27 H. X. Shang, H. J. Fan, Y. Liu, W. P. Hu, Y. F. Li and
X. W. Zhan, Adv. Mater., 2011, 23, 1554.
´
46 J.-L. Bredas, J. E. Norton, J. Cornil and V. Coropceanu, Acc.
28 F. Lincker, N. Delbosc, S. Bailly, R. De Bettignies, M. Billon,
Chem. Res., 2009, 42, 1691.
¨
A. Pron and R. Demadrille, Adv. Funct. Mater., 2008, 18, 3444. 47 M. C. Scharber, D. Muhlbacher, M. Koppe, P. Denk,
¨
29 U. Mayerhoffer, K. Deing, K. Gruß, H. Braunschweig,
C. Waldauf, A. J. Heeger and C. J. Brabec, Adv. Mater.,
¨
K. Meerholz and F. Wurthner, Angew. Chem., Int. Ed., 2009,
2006, 18, 789.
48, 8776.
48 L. Yang, H. Zhou and W. You, J. Phys. Chem. C, 2010, 114,
30 B. Walker, A. B. Tamayo, X.-D. Dang, P. Zalar, J. H. Seo,
16793.
A. Garcia, M. Tantiwiwat and T.-Q. Nguyen, Adv. Funct. 49 M. D. Perez, C. Borek, S. R. Forrest and M. E. Thompson,
Mater., 2009, 19, 3063. J. Am. Chem. Soc., 2009, 131, 9281.
31 Y. Liu, X. Wan, B. Yin, J. Zhou, G. Long, S. Yin and Y. Chen, 50 K. Vandewal, K. Tvingstedt, A. Gadisa, O. Inganas and
J. Mater. Chem., 2010, 20, 2464. J. V. Manca, Nat. Mater., 2009, 8, 904.
32 H. M. Ko, H. Choi, S. Paek, K. Kim, K. Song, J. K. Lee and 51 G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery
J. Ko, J. Mater. Chem., 2011, 21, 7248.
and Y. Yang, Nat. Mater., 2005, 4, 864.
J. Mater. Chem. A
This journal is ª The Royal Society of Chemistry 2013